ESE's works engine tuner [Archive] - Page 152

husaberg

9th February 2023, 19:37

Hi Frits!

If I remember right, your 360 was a Bultaco Bandido powered monocoque racer? Once you posted a picture of the combustion chamber, and I think to remember it had two plugs, so it would be a model 61. I know it is air-cooled and for "agricultural racing" but I'm building an engine like that for flat track use, is there any tip or suggestion in your experience, of where to start modifying this engine?

I believe a bigger carburetor would help, since the original is a 32mm, and the TSS that used the same engine used a 35.
Also, reshaping the cylinder head would be an easy mod I think. All the cylinder heads I have are full of detonation marks.

About the cylinder head, you wrote before that you prefer lower compression engines, as they produce hotter exhaust gases with more energy to use on the pipe. What compression ratio would you recommend? This type of engine would require a different pipe to make use of that extra energy, I suppose with a bigger belly? Would it be a good idea, or you think because the transfer design on this engine is not the best, the bigger diameter pipe is going to create more short-circuiting?

Thanks in advance!
Bert Flood prepped an el bandito for road racing in the late 60's or ealy 70's it was pretty successful
On two wheels detailed it.
i have a copy somewhere.

TZ350

9th February 2023, 20:57

352307 352308

F81M 18:1 Meth + Aggressive advance.

352310 352311 352312

EngMod 2T's simulation of what it thinks Meth plus 18:1 at 32 deg advance would look like at the back wheel.
I graph in Kw when looking for RWHP as that is this lazy mans way of accounting for drive train losses.
I had previously entered 32 deg adv to get EngMod's opinion.

Yea TeeZee , that's WAY too much timing with 18:1 - I would guarantee its detonating and its the shock thru the cylinder thats loosening off the nuts.
Running rich on Meth masks many evils , but deto creates havoc no matter what.

352309

TUmax graph in deg C. It is hitting 960. Would this be detonation territory? Maybe Ok for Av Gas but I have no idea what the limit is for Methanol. I understand Meth is prone to pre-ignition.

TZ350

9th February 2023, 21:13

.
All the other details of the 1971 Kawasaki F81M 250.

Lef16

9th February 2023, 23:11

If meth works at ~450* EGT as stated before,wouldnt this mean that you need an exhaust TL of about 950mm for 9500rpm and 200* exhaust port instead of 1000+?

Frits Overmars

9th February 2023, 23:31

https://www.youtube.com/watch?v=nkBabPAMIsc

skako

10th February 2023, 07:18

.
All the other details of the 1971 Kawasaki F81M 250.

Hallo,
While I'm studying what I forgot or don't know, I tried to design an exhaust pipe for simulation during my pause.
I was not able to find all the necessary technical data for the Kawasaki F81M 250, so I partially evaluated them.
I'd love to see a graph of how it looks in EngMod2T with this exhaust pipe and symmetrical intake. Duration 180°, opening 90° btdc, closing 90° atdc, full opening 45° btdc, intake length 120mm, carb. dia 38mm.
Everything else unchanged.
Thank you,

352319

TZ350

10th February 2023, 11:45

If meth works at ~450* EGT as stated before,wouldnt this mean that you need an exhaust TL of about 950mm for 9500rpm and 200* exhaust port instead of 1000+?

Yes, I would have thought so too. Not sure what is happening here, just looking to EngMod for guidance. I suspect I may have messed up the data input when designing the pipe.

Francisco Erjavec

10th February 2023, 13:45

https://www.youtube.com/watch?v=nkBabPAMIsc

Hi Frits!
Whats your opinion about the belly butterfly valve shown in the video?

TZ350

10th February 2023, 16:31

.
352322

Standard 250 head. 11:1 com ratio swept. Will get a chance to run this up maybe mid next week. A tropical cyclone is coming, and I need to get ready for that, we got pretty wet in the last storm.

F5 Dave

10th February 2023, 17:59

Um why are you showing a picture of a pressure gauge? What has that got to do with compression ratio?
Please tell me you used a burette. Two threads up from the bottom works well.

TZ350

10th February 2023, 20:48

Um why are you showing a picture of a pressure gauge? What has that got to do with compression ratio?

There had been a bit of talk about dynamic compression pressure so I was curious to see what the cold cranking pressure for 11:1 looked like.
If I had of had more time I would have popped the other higher compression head back on for comparison.
Also EngMod-2T gives a cranking pressure for your chosen com ratio. A handy check. If you'er not seeing something close to the expected pressure its time to ask yourself some questions.

Please tell me you used a burette. Two threads up from the bottom works well.

No Burette, although I have two glass laboratory one's. We find it much more convenient and safer to use plastic syringe's.
They are cheap, readily available, quick and easy to use with good accuracy. They come in a variety of suitable sizes. Team ESE achieve results with them that we are confident in.

F5 Dave

10th February 2023, 20:56

Ok so at least you measured it static. Pressure gauge, one way valve or not, tells you limited information, often disinformation.

Burette is cheap enough from online suppliers. Easy to use. I favour 5w fork oil.

Measure thrice. Cut once.

DoldGuy

12th February 2023, 05:41

352312

TZ, I’ll go ahead and confirm my idiotic thinking by asking the question, in the EngMod2T Head file shows 300 mm dome radius???

TZ350

12th February 2023, 07:31

352312 TZ, I’ll go ahead and confirm my idiotic thinking by asking the question, in the EngMod2T Head file shows 300 mm dome radius???

352352

300mm dome radius refers to the curve of the piston crown. In the lower right box you can also see the calculated compression pressure for the chosen com ratio.

Orengo98

12th February 2023, 09:08

Hi Orengo, my first reaction was: if you want to learn a lot, go ahead and do what I did. If you want to ride, don't.
My engine came with one central spark plug. I fitted a head with two plug holes after the original head had cracked, because that was all I could find. But I could only fit the forward plug; my monocoque frame got in the way of the rear plug.
It taught me the difference a plug position can make. Maximum power was not affected but with the forward plug the engine idled much nicer than with the central plug.

Hi Frits,

Thanks for your answer! really interesting!

Part of the idea is to try to make parts for the engine, experiment and learn so it is fine.
Thanks for sharing your experience with the plug placement, the ignition was another place where I had doubts. This engine is extremely over square, and reading a paper about the bore/stroke ratio, they found that as the ratio increased, the combustion efficiency dropped. The dual plug design maybe has some part on solving this problem, as Bultaco stated a slight power increase with the dual plugs. Do you believe this design is advantageous?

In my experience, low compression ratios make it easier to extract power from an engine. But for an air cooled big single with a shrunk-in cast-iron sleeve I would not go too low; relatively cool exhaust gases are more important than power. About 12:1 should be a good value.
The transfer design on my engine was rather different from the original MX-engine. I had enlarged the A-transfers and added a big third port where the piston-controlled inlet port used to be.
The pipe belly was considered huge at the time, but rather modest with 51 years of hindsight: 112 mm if I remember correctly.
I think nowadays that would be a good belly diameter for your more-or-less standard engine.
352306

Thanks for the tip, I'll try a 12:1 head with a concentric toroidal chamber design to start with, seems like an easy mod to try.

Your engine must have been quite powerful, the gears on that engine are tractor like, really wide!
The problem of the primary gears seems to be usual, once the factory power is surpassed, the axial force of the helical gears is greater than the press fit of the clutch basket bearing. The TSS had and advantage there as well as the 5 speeds. I have the port map of the factory racer, really similar, wider transfers, taller exhaust port, and two extra transfers carved behind the sleeve. Maybe I can post it here if it is of some use for someone building a similar engine?

The pipe is kinda strange. It was wide for the era as you say, and also has a huge stinger, looks like a canon hahaha. It also has two big "dents" to clear the frame and rear wheel, that I'm sure they produce a noticeable pressure trace.

Thank you a lot for the details and insight in your learnings and engine!

Bert Flood prepped an el bandito for road racing in the late 60's or ealy 70's it was pretty successful
On two wheels detailed it.
i have a copy somewhere.

Hi Husaberg!

If I remember right, it was a bike for speed records? Is the one in the pictures I attach?

352353
352354

Flettner

12th February 2023, 20:45

What to take heed of when running two separate pipes off a single cylinder?
Fit a joining tube from the two mid sections?
Make the reverse cones adjustable so as to be sure the tuned length is the same in both pipes. Adjust them on a dyno to get them in sync?

husaberg

12th February 2023, 20:46

Hi Husaberg!

If I remember right, it was a bike for speed records? Is the one in the pictures I attach?

352353
352354
Started with a road racer thats the one i most remember.
It was very mildly tuned.
i started to look on Friday.

ken seeber

12th February 2023, 21:41

Frits Overmars

12th February 2023, 22:41

Hi Frits.... Thanks for sharing your experience with the plug placement, the ignition was another place where I had doubts. This engine is extremely over square, and reading a paper about the bore/stroke ratio, they found that as the ratio increased, the combustion efficiency dropped. The dual plug design maybe has some part on solving this problem, as Bultaco stated a slight power increase with the dual plugs. Do you believe this design is advantageous? The dual plug setup may have been advantageous in that Bultaco combustion chamber (now there's a real bathtub chamber for all you guys who keep calling a bowler head chamber a bathtub). But as I wrote, I had only room for one plug, so I cannot answer from experience. But never mind, stick to a single-plug toroidal chamber like you planned.

What to take heed of when running two separate pipes off a single cylinder? Fit a joining tube from the two mid sections? Make the reverse cones adjustable so as to be sure the tuned length is the same in both pipes. Adjust them on a dyno to get them in sync?KISS and KISS again Neil. No joining tube. And you can use adjustable reflector cones if you wish, but I don't think you'll need anything that fancy to synchronize the pipes.
Sticking to the design dimensions should be accurate enough.

wobbly

13th February 2023, 09:52

I dont understand , one minute TeeZee you are running 18:1 that would be considered a good spot for a 125 water cooled race engine - and completely mental in a 250 aircooled , then you go off the planet the other way
with 11:1 that would be slow on petrol.
Whats the point of testing that at all - except wasting a pile of time generating ignition and fueling numbers that are of no use whatever in a properly setup engine.
Either way is not even close for a 250 aircooled engine on Meth.

Meth pipe design has been done and dusted. A normal petrol pipe wall temp that works is 325 at the bottom of the power band and 425 at peak power - where egt would be in the 600's.
The Methanol numbers are 250/350.

And a header length way past a normal max of 33% at 37% is also not even close to optimal, and is always going to give weird arse anomalies in the powerband shape.

A CCR of 1.2 without the transfer duct volumes included also simply cant be right , shit the Aprilia was at 1.24 WITH the transfer ducts - and that was considered large.

TZ350

13th February 2023, 10:40

I dont understand , one minute TeeZee you are running 18:1, then you go off the planet the other way with 11:1 that would be slow on petrol. Either way is not even close for a 250 aircooled engine on Meth.

Meth pipe design has been done and dusted. A normal petrol pipe wall temp that works is 325 at the bottom of the power band and 425 at peak power - where egt would be in the 600's.
The Methanol numbers are 250/350.

A CCR of 1.2 without the transfer duct volumes included also simply cant be right , shit the Aprilia was at 1.24 WITH the transfer ducts - and that was considered large.

My 125 would have had a larger case volume than the Aprilia but you might be right about the 250's CCR, I will have to check by CC'ing it. The 250's case volume was increased with a slightly longer rod and the crank inner faces slimmed down to the thrust faces.

Thanks for the pipe temperature info, very helpful.

It was an easy swap to go back to a standard head with its 11:1. 18:1 was too much. What would you suggest as a more appropriate ratio for a 250 on meth.

wobbly

13th February 2023, 12:25

My take would be 15:1 as a first try for a 250 Aircooled.
This will give a good ratio of much needed cylinder /head cooling Vs increasing the power due to compression.

Be aware that the default EngMod process is you input the case CC via a hole in the piston at TDC and when you input this number into the calculator it removes the transfer ducts
and displays this ratio. This allows the code to recalculate the duct volume when a port is changed.
You can off course input your own ratio and include the transfer ducts as per normal practice.

Re doing compression CC testing.
To eliminate any variables trying to peer down a threaded hole and guess the meniscus position . I use a gauge that screws into the plug threads and bottoms such that the end is flush with the chamber.
A normal plug has 18.3mm of actual thread engagement taking the crushed washer into account , and virtually all 10 heat range race plugs displace 2.25cc in the threads - so this is the volume of
the gauges bore.
Its then real easy to fill to within 10mm of the gauge top , turn the rotor back and forth to hit TDC exactly , then fill it to the top and remove 2.25cc from the reading.
I use a digital burette , but a glass one is fine as long as you leave it to sit for 15 mins while the fluid clinging to the glass slowly runs down.
The fluid has always been ATF.

TZ350

13th February 2023, 16:09

.
352365

Hopefully I have got this right. To mimic Methanol in a 15:1 air cooled 250 cc engine.

I setup the temperature file by selecting the water cooled cylinder and crankcase default. And the pipe temperature varied with rpm 250 at 6000 to 350 at 9500 rpm.

352364

Simulated two runs. One with 32 deg ignition straight line and the other starting at 34 and reducing to 7 deg. An ignition curve suggested by EngMod.

Hopefully later next week I will get a chance to do this for real on the dyno.

wobbly

13th February 2023, 18:36

So why would you select a watercooled option when the engine is aircooled - makes no sense.
32* straight line is insane and completely impossible , why even think about using it as no one else has done this , let alone got away with it , ever.
And the Exhaust duct temp is not going to be any thing like 250* as on petrol , I would try 180* as a start point and reduce the crank and transfer wall temps by at least 20*.
I have no idea why , but in this case EngMod is talking rubbish about the ignition , im not.

And one of the things Methanol does is to allow very rich mixtures to be used without power loss.
This translates into negating the usual effect on petrol , that when the engine is over the top of the pipe , the air flow thru the carb keeps going up , but power is going down.
Thus less fuel is needed in reality , so the carb is forcing the mixture into going rich - exactly what solenoid power jets fix so elegantly.
On petrol , going rich over the top looses power - on Meth it doesn't.

One way to help the situation on petrol is to retard the ignition over the top , thus dumping more heat into the pipe.
But Methanol basically makes the engine insensitive to fueling , and the pipe is already way shorter due to the inherent drop in wall temperature.

Thus retard over the top is not needed , as richness doesn't have the same negative effect as on petrol.
The only odd case , that proves the rule , is that when using 20:1 compression on Meth in a 125 , this needs a hugely rich mixture to live , and a solenoid power jet then heats the pipe enough to make rev on power better.
I would look at retarding as usual to 15* at peak power , then flat lining it.

TZ350

13th February 2023, 20:05

So why would you select a watercooled option when the engine is aircooled - makes no sense.
32* straight line is insane and completely impossible , why even think about using it as no one else has done this , let alone got away with it , ever.
And the Exhaust duct temp is not going to be any thing like 250* as on petrol , I would try 180* as a start point and reduce the crank and transfer wall temps by at least 20*.
I have no idea why , but in this case EngMod is talking rubbish about the ignition , im not.

Made good sense to me as there is not a Methanol default option. As I had no idea what the meth simulation temps should be. Water cooled was selected to mimic the engine cooling effect of Meth.

I will use the lower temperatures you suggest with the air cooled option.

32* might be insane but was starting to look possible. Top end power was increasing as I straitened up the ignition curve into a 30 deg straight line. 32 or even 34 looked like it might have been possible and EngMod agreed.

352373352374 DynoJet dyno and Ignitec ignition.

But I was suspicious of the whole thing, it was quite harsh until about 9,000 rpm. A bit like a diesel model aero engine that is being over compressed.

Was it the 18:1 com ratio, the advanced ignition although it was making more power each 2 deg step of extra adv, or was the ever richer fueling at the top end masking something??? Maybe worth trying a std 11:1 head to see what changes.

Frits Overmars

14th February 2023, 00:24

32* straight line is insane and completely impossible , why even think about using it as no one else has done this , let alone got away with it , ever.... I would look at retarding as usual to 15* at peak power, then flat lining it.I have tried all kinds of crazy things but I never had the guts to try 32° straight line ignition advance in the powerband of any engine with a decent BMEP.
Flat lining at 15° would be a safe way to start your ignition quest. Subsequently you can run flat lines with even less advance, en finally carefully try more advance at the rpm points that suffered from retarding.

Maybe worth trying a std 11:1 head to see what changes.Now that is something you can safely try. You may be generating ignition and fueling numbers that are of no use whatever in a properly setup engine, as Wob says, but I would not call it a complete waste of time. It will be educational either way and you won't have to get rid of broken engine parts afterwards.

I've long been a proponent of low compression ratios. In my experience, a low-compressed engine reacts much more clearly to changes. But to take advantage, you'll need an exhaust pipe that deals extremely efficiently with the energy contained in the exhaust gases of such an engine. And the engine will need a transfer layout that can handle that amount of pipe suction.

A CCR of 1.2 without the transfer duct volumes included also simply cant be right , shit the Aprilia was at 1.24 WITH the transfer ducts - and that was considered large.It was considered large compared to what had been done before.
The Aprilia RSW had a TDC crankcase volume of 650 cc and a Crankcase Compression Ratio of 1,238 .
The Aprilia RSA had a TDC crankcase volume of 675 cc and a CCR of 1,227, which gave a power improvement over its predecessor.
But is 1,227 the optimum? I don't know and Jan Thiel doesn't know either. You won't discover a border until you cross it, and Jan retired before it came to that.

DoldGuy

14th February 2023, 03:48

I’m so enjoying this learning experience that is being shared, most trials and errors are tucked away. TZ I’ve learned more about EngMod2T with your posts and others replies in regards!
Thanks again for the education…

Condyn

14th February 2023, 06:08

The gauges Wob is referring to are sold by L.A.D and often called L.A.D gauges. They work excellent as described above. The small bore of the gauge minimizes surface tension discrepancies. A tenth of a CC is not as critical on a 70mm bore as it is on a 54mm bore. Trying to fiddle around counting threads is foolish, especially on small bore cylinders. I feel many builders simply estimate on compression ratio, as it is tricky without the L.A.D gauge.

Just yesterday I attempted to check CCR for engmod modeling on my parallel twin piston controlled inlet engine. I plugged the inlet port with modeling clay as Wob mentioned a while ago. Next I smeared a bit of grease around the piston ring to insure nothing would seep by. Then started filling through a hole cut in a used piston. It was a learning experience. As soon as I was about 400cc into it, the adjacent cylinder case pulse line fitting started pissing. Apparently labyrinth seals do not function until the crank is spinning. (Or my labyrinth is not working) I think I will attempt to inject a wad of grease between the center bearing and labyrinth seal in order to get the measurement.

Vannik

14th February 2023, 06:41

Apparently labyrinth seals do not function until the crank is spinning.

Measuring crankcase volume without a labyrinth seal is both a learning and a trying experience. Can quickly turn into a very messy situation.

A labyrinth seal has a controlled but very small amount of leakage when in motion. Stationary it does not seal.

wobbly

14th February 2023, 08:54

Reducing the duct temp has a large effect on pipe bulk temp , so lowering it will need a shorter pipe for the same peak rpm.
As I said use the age old guide of 30 to 33% for the header and 64 to 68% for the diffuser.

Having no power valve , the front side power can be helped by a quite long and steep last diffuser section.
To gain more overev if needed , a shorter header , and a steeper/shorter 1st diffuser is the go - but this will be less of an issue on Meth due to the reduced adverse reaction to the fuel curve going rich over the top.

I completely agree with Frits's ideology , but still think 11:1 is too low even as a first try , as that would be conservative on shit pump gas.
Lastly my experiences lead to winding up the MSV with Meth , into the mid 40M/s, with a sharp corner on the bowl as this mimics having more advance by increasing flame speed due to squish turbulence
But that technique makes more power , than ignition advance does - maybe due to less compression pumping losses after the spark has fired BTDC , another reason I think anything above 30* is mental.

diesel pig

14th February 2023, 12:58

I completely agree with Frits's ideology , but still think 11:1 is too low even as a first try , as that would be conservative on shit pump gas.

Interesting, What compression do you think we nutter's wanting to build dirt bike engines to run on 91 petrol as one should be able get it no matter where we nutter's go?

wobbly

14th February 2023, 13:27

An aircooled 250 MX on 91 would go 12.5 all day , as you are not hard on the gas for long at all - and need plenty of mid snap to get off berms etc , not screaming top end.
Not that you can generate big bmep from one of those old shitters anyway, without basically starting over with welded ports and hours on the sim.

diesel pig

14th February 2023, 14:11

diesel pig

14th February 2023, 14:24

Also, There is no way I would start re engineering things I just wanted to know what value to look for when re building a motor. Cheers

husaberg

14th February 2023, 17:48

Interesting, What compression do you think we nutter's wanting to build dirt bike engines to run on 91 petrol as one should be able get it no matter where we nutter's go?

352376
................i assume 700 octane refers to 100 octane....

pretty sure Robinson had a table as likely did jennings

TZ350

14th February 2023, 19:24

.

352376

Ok thanks Husa, so from Bells book 15.7:1 pretty close to Wob's suggestion of 15:1 for Methanol. I will aim for 15:1.

Please tell me you used a burette. Two threads up from the bottom works well.

Here we go, a 10cc burette and 10cc syringe.

The burette can easily be read to 0,1cc and the syringe 0,2cc. Probably 0,2cc is good enough for a 250 or even a 125 but I agree being able to easily read 0,1cc or with care 0,05cc is much better for a 50.

352397

At 15:1 for the 250 I will be looking for 17.75cc for the combustion chamber including the 0,8 squish.

With CNC machining someone could probably get that sort of accuracy. But with my hand carving in "mister rattly", our old faithful lathe, I will be very lucky if I can get close to 0.2cc of the target either way.

wobbly

15th February 2023, 08:26

Getting the cc correct has nothing to do with the equipment , just your patience.
Spinning the chamber in a battery drill would do , as using 80 grit emery cloth under your finger ( form tool No1 ) you can easily remove 0.05cc at a time by polishing alone.

And as an aside to all this talk of compression , and the looming threat of detonation - fuel quality is , in racing , of utmost importance.
As I tell every customer I have in karting or track racebikes , if you leave 98 pump fuel in a 1/2 empty drum , or in a vented tank overnight - its fucked the next day.
Its now 91 if you are lucky , as every time the lid is removed all the light front ends evaporate to atmosphere.
The Reid Vapor Pressure has dropped thru the floor , and this results in huge un atomized raw fuel droplets entering the engine.
These do not burn during combustion , and the tune appears rich on the piston and the egt, so you naturally lean it down and bang - siezure or deto.

Thus all race fuel should be bought or immediately decanted into 5L containers , with no airspace , and only one mixed and used at a time.
Absolutely vital if you are reading a weather gauge ( your phone App and the local airport ) and have jetting records based on RAD or Density Altitude.

So yes a 250MX will be happy with 12:1 but not on 91 fuel thats been in the tank for 3 weeks.

porttiming124

15th February 2023, 09:18

Hello everyone . I have a question about the shape of the slight hook of my b port.
If I measure before the little hook it gives me 20 mm . If I measure at the liner at the bore it gives me 23 mm.
IF i look on the aprilia rsw(pic CC) the tunnel seems more regular until the end.
Am I correct in thinking that the port should shrink gradually to the edge of the bore?
b352398352399352400352401ecause in my case it opens before arriving at the edge of the bore.
I would love to hear your opinion on what you would do.

wobbly

15th February 2023, 09:41

This is the age old problem , you want to use Aprilia like technology , but your Transfer Ducts are nothing like one at all.
Your hook has virtually no radius on it and in the Aprilia both front and rear wall are perpendicular to the bore center line.

porttiming124

15th February 2023, 09:55

Ok perfect then I will forget to try to compare with the aprilia.
To come back to my old B duct what do you see to improve?
Should I modify the front and rear wall more perpendicular to the center of the bore or am I still wrong?
thanks.

wobbly

15th February 2023, 10:40

The best B port configuration for max power and overev is both walls perpendicular and a full radius in the hook corner , but for example a TM Kart , where front side
is important due to no PV , the walls are approx 15* but both parallel , plus the corner radius.
And the hooks should point to near 1/3 from the boost to bore center.

EDIT - and the A port front wall does not look angled back enough , where does its radial intersect.

porttiming124

15th February 2023, 14:43

It's an acceleration snowmobile engine with a cvt clutch so the front side doesn't really matter.

The front walls of the A ports intersect in the center of the bore.

352403

porttiming124

15th February 2023, 15:37

I just came across a recommendation from tsr software that dates back to 1999.
What do you think of it 24 years later?352404

Frits Overmars

15th February 2023, 23:14

It's an acceleration snowmobile engine with a cvt clutch so the front side doesn't really matter.
The front walls of the A ports intersect well in the center of the bore.It seems you've already made up your mind, so posting the image below won't be any use. But I'll do it anyway.
The radial angles of the A-port front walls are more important than any other radials, in any two-stroke. And look where they intersect the center line.
352405

porttiming124

16th February 2023, 01:38

Hello Frits.
No I haven't made a decision.
I removed the word well in my sentence. (conflict between me and google translation).
Thanks for posting the image anyway.
Indeed from what I see on your image the intersection of the front wall of duct A must be done about more towards the back.
Thank you.

Condyn

16th February 2023, 02:16

PT, I do think a Very small amount of focus is to be put towards front side power with a cvt. After all, clutch engagment is lower than maximum power rpm. From data logs, I can see this area is used for 4-6 tenths of a second on my 440cc. Not much… but enough to be important to critical 60 foot times. And much more important than over-rev the way I see it.

porttiming124

16th February 2023, 03:25

Hello Condyn.
Actually, I totally agree with what you are saying.
This engine will drag on (soft) snow.
For this discipline (215+hp) we are constantly looking for traction.
I notice that the stronger the engine is at 6000rpm the bigger the traction problem.
Example a 600cc of 120 hp does not experience this problem.

This is why I feel like the front end of the power curve doesn't matter as much.

lohring

16th February 2023, 03:44

Below are a series of posts by Wayne I copied years ago. They may be helpful.

Lohring Miller

352406

porttiming124

16th February 2023, 04:18

Thanks Lohring. Much appreciated.

wobbly

16th February 2023, 08:06

My take would be that in this case ( very unusual ) where the initial acceleration phase is traction limited , then the " front side " is important once the thing is hooked up and the clutches are locking.
This area may only last for a few tenths of a second , but is all important for the 60ft times.
A port front wall intersection points have been gradually moving back toward the boost port over many years , this being a function of the initial realization that fat pipes can easily
overcome the A port exit stream coherency , and second , we have had visual wave simulators for long enough now to have got a handle on what area of each diffuser section affects the power
band shape.
Having the intersection point well back near the boost , means steeper angles can be employed , and this is just as easy to bias toward peak power for a CVT as it is to generate front side.

porttiming124

16th February 2023, 09:34

Okay thanks Woobly.
Here are the dimensions of a popular pipe for this type of engine that I measured last week and reproduced in engmod.
352407

I know you don't have a crystal ball.
With the angles of this pipe for example can I point the front wall of duc A backwards as you recommend or else I will have to make a pipe with it more aggressive angles?

wobbly

16th February 2023, 10:55

Yes I have a crystal ball , I believe the A port front wall pointed further back would work well if the pipe was modified.
33% and 67% are too long for optimum performance in a CVT.
Narrow the header angle to steepen the 1st diffuser and shorten the header - the amount you shorten the header closer to 30%, add this to the mid section.
Thus the whole diffuser is moved leftward to something around 30 - 64% with a steeper 1st section.
This adds a bunch of peak and upper front side - as would a 3 angle rear cone.

wobbly

16th February 2023, 10:57

Anyone got a TD3 cylinder I could use to check what is possible going mental on it , adding reeds etc.
Please PM me if there is one out there.

TZ350

16th February 2023, 12:39

TZ350

16th February 2023, 13:32

.
Amal alcohol needles needed some modification to be used successfully in the Yamaha TD3 34mm Mikuni carbs.

352412

This is a chart I made of the flow characteristics of a 6F9 Mikuni needle vis an Amal alcohol needle.

The curve on the graph is an un modified Amal needle at less than 20% throttle. The top line shows a modified Amal needle.

The shoulder on the Amal needle where the taper meets the parallel section at the top needs to be blended out other wise it obstructs fueling at low throttle.

This blending needs to be about where the curve touches the Amal needle in the picture.

The Amal Concentric needles are to short for the Mikuni 34's. So the clip needs to be silver soldered to the very top of the needle itself.

Otherwise the needle can pull out of the needle jet and hang the slide on full throttle.

I did this was work to help some racers get past seizing issues they were having at a time when the Amal alcohol needle was popular.

Unfortunately I don't have any further detailed notes on the rest of the setup for TD3's running Methanol.

Flettner

16th February 2023, 14:38

https://www.youtube.com/watch?v=nkBabPAMIsc

Interesting, just what I proposed, butterfly in the exhaust reverse cone.
Its what I have in store for the AG100 twin piper.
Thats why I wanted a single reverse cone but it will be easier to make two, one for each pipe.
Im not sure I saw a dyno run with it installed?

F5 Dave

16th February 2023, 18:41

352409 Ok, time to put these bad boys to use. 352411 After a bit of lathe work a 15:1 head by test (17.8cc) 0.75mm for 40m/sec max squish velocity at 9K rpm.

352410 12:1 head vis 352408 15:1 head. Bearing in mind that the motor is cold and there is a lot of oil and grease around the ring land of the piston.

I'm going to grumble again. That is a picture of a pressure gauge with a one way valve.

It is not a good indicator of compression ratio.

For example, and for giggles, I tested my 200cc dirtbike after planning the head. 193psi. It measured at something silly like 14.5:1. Oh dear. Took some metal out and was 13.4 after a couple of iterations.

Great. Same guage, still hadn't started the bike. 195psi.

Ohh I guess the compression ratio has gone up despite metal removed and measured. <_< (spoiler alert, it hadn't).

Put the gauge in a box.

TZ350

16th February 2023, 20:06

I'm going to grumble again. That is a picture of a pressure gauge with a one way valve. It is not a good indicator of compression ratio. Put the gauge in a box.

Maybe I should have pointed out more clearly the gauge was just an interesting shits and giggles comparison between 12:1 and 15:1 heads that had been CC'd using the measuring devices seen in the first photo.

352413 EngMod-2T gives a calculated pressure of 167 psi for a 15:1 head.

Pretty much what I am seeing on my gauge so it is in the ball park. Reassuring, but you are right, I should not rely on the gauge, a CC ing test is best. Although the pressure gauge reading may come in handy later for trouble shooting.

F5 Dave

16th February 2023, 20:24

That's what it is. I pulled it out last month to test a mates RS125 Aprilia roadbike which stopped running. 30psi. Clear what happened there. Course a finger over the plug could have told me that. Bit weird with electric only start mind.

Frits Overmars

16th February 2023, 23:35

A-port front wall intersection points have been gradually moving back toward the boost port over many years, this being a function of the initial realization that fat pipes can easily overcome the A-port exit stream coherency... Having the intersection point well back near the boost , means steeper angles can be employed , and this is just as easy to bias toward peak power for a CVT as it is to generate front side.I could not have said it any better, not even in Dutch.

I've noticed that some people find it easier to work with lengths rather than angles. That is why I'm considering a graphic representation in which, instead of the various axial angles of the transfer ports, their positions and target points are expressed as percentages of the cylinder bore.
This may also be helpful in drawing the desired flow paths on a piston dome.
Like the man said, I'll be back.

wobbly

17th February 2023, 08:36

Just an interesting factoid for those fiddling with butterfly things in the rear cone.
This setup was used in direct drive karts ( chainsaw engines ) many years ago but they had a simple fixed flat plate @ 50% area about 1/2 way down the rear cone - this was called a plate pipe ( funny that )
These were quickly replaced by the "new" technology of having a series of perforations spaced radially down the rear cone, within an enclosed muffler tube that was simply an extension of the mid section.
This methodology has been standard fare on kart pipes for 30 years.

Even today this is used in several direct drive International classes.
When it came to modelling the things in EngMod , the assumption ( wrong ) was that there were two functions involved.
Firstly was that as the bulk flow thru the pipe increased the effect of the holes gradually reduced , and the volume of the parallel " muffler " tube would resonate at a particular rpm , effectively then
making the holes disappear.
Thus the return wave was smeared out in amplitude at low rpm , reducing the effect of the pipe being too short down low, and causing the return wave to arrive at the port too soon.

But then a customer , working for the Tony Kart factory , bought the TFX system and it became clear this is not how it works. ( Frits is now using this same piece of gear ).
At low rpm the rightward wave is transmitted thru the perforations , and is reflected back , at a much longer TL by the flat plate at the end of the " muffler " tube - well past the rear cones exit hole.
Then at high rpm the tubes resonance effect overcame this , and the holes do virtually disappear.
Neels changed the code to reflect this new line of thought , and boom , we had way better correlation with the TFX results.

The system works unbelievably well , in situations where the crank is connected directly to the load - be that a wheel or a propeller.
And is KISS to the extreme.

Flettner

17th February 2023, 13:24

There must be some compromises?

ken seeber

17th February 2023, 13:43

After a 12 months hiatus, no, not jail or anything like that, I now have the opportunity to get into some 2 stroke stuff again. This is before we’re all forced to be 100% electric from 2035.

So, it’s back onto DCI (Direct Cylinder Induction). However the previous one, which ingested into the A ports, this version will be into the B ports. Why? Well going thru the A ports meant that at TDC or so, the DCI intake also communicated with the crankcase, offering both crankcase piston port induction plus the DCI function. So it didn’t really tell me whether it actually worked or not as a pure DCI. Even if it did, the poor old pressure waves in the passage would have got confused and possibly self-cancelling…dunno.

Anyways, as it turned out, the B ports were always covered by the piston skirt (see pic of skirt with some trial coating), other than when at BDC. So, it’s obvious, to give it a go with the A ports feeding the cylinder via the standard piston port inlet system.

As previous, will weld on aluminium blocks (still need to be drilled and tapped) after fixing to the with the little fixture plates shown and then, with Devcon F, fill all the gaps etc such that I can create a direct passage to the B ports, blocking off the original B passage to the crankcase.

Intend to leave the port timing as it is, nominally exh 178, transfer 120 & inlet164 durations. The B ports of the liner can be seen in the pics.

Would be interested in any thorts on the poughts in terms of their entry angle if this could be improved. Easy to do when the liner is out before the welding takes place.

352414352415352416352417

ken seeber

17th February 2023, 13:45

Couple more pics....seem to be limited to 4 max.

352418352419

husaberg

17th February 2023, 19:32

Couple more pics....seem to be limited to 4 max.

5,but it spits out repeats....

TZ350

17th February 2023, 20:21

.

Using EngMod2T to save wasted hours on the dyno.

If I have modeled the F81M engine correctly in EngMod and that is a big if then I can develop an ignition curve using EngMod.

352422

Did multiple simulated runs with fixed line ignition advances.

And by using the pointer I could pick out the best timing for any RPM.

352420

The best ignition points make a curve that starts of very advanced and drops off as it starts making power and then there is an uptick in the past peak power area.

The shape of the initial big advance that then drops off to sensible values looks very much like the curve people have described to me for methanol.

352421

Power trace using the developed ignition curve.

Using KW's at the crankshaft for the curve is my lazy way of accounting for loses in the drive train and indicating what RWHP should look like.

I will try this ignition curve as soon as I can get back to working on the 250 again.

TZ350

18th February 2023, 07:21

husaberg

18th February 2023, 10:26

The details of his inertia dyno is here.

Yep
316720

Thanks!

All I did was tear out the windigs.
316742
316743
316744
follows on from here.

TZ350

19th February 2023, 18:45

.

352429 352428

Ignition curve and performance simulation developed using EngMod2T.

352431 352430

Ignetec Ignition curve and dyno results so far.

More work to do to get rid of the bumps at 7k rpm and the lower rpm range. More advance and/or needle.

EngMod shows an advance bump at 7k rpm. I will try that.

When the timing is pretty much spot on I will try playing with re sizing the main jet.

352432

I took a 100ml sample of my Methanol mix, 90% meth, 10% acetone and 30:1 oil.

I added 2.5cc of water to the 100ml in the jar and let it sit to see if there was any fuel separation. Looks good, no separation.

Water in Methanol greatly adds to its cooling and anti knock effect. Might try a batch of this brew in the bike next and crank up the advance.

When I am in the ball park with the ignition and jetting I will strip the motor and see what can be done about the balance factor and getting rid of that horrendous vibration.

Vannik

19th February 2023, 23:03

352433

For those that are interested in a biography of Jan Thiel in English, send an email to:

So far we have received 83 emails from people that are interested in the English version and 22 for the Spanish edition.
We need at least 200 requests to start printing, so we have to wait a little longer... Mail to English@LegacyJanThiel.com for English prenotations

Frits Overmars

20th February 2023, 01:45

I've noticed that some people find it easier to work with lengths rather than angles. That is why I'm considering a graphic representation in which, instead of the various axial angles of the transfer ports, their positions and target points are expressed as percentages of the cylinder bore.
This may also be helpful in drawing the desired flow paths on a piston dome.And here it is: the ability to draw transfer port positions and flow directions without having to use a protractor.

There is also a drawback: for ports close to the front and rear cylinder walls, for example the C-port, specifying position points by means of bore percentages is not as accurate as using angles. That is why I did not quote an optimum position percentage for the C-port.
352458

wobbly

20th February 2023, 08:32

TeeZee , I see you are running the Turbulent Model in the Combustion page.
You only need to do this once , when all the Combustion related parameters are finalized , ie Ignition , Compression , MSV . You then redo it if any of these are changed.
Then save this as a new name and transfer all the Delay /Vibe numbers into your sim.
Run the Turb sim , open that iteration and its says Write New Combustion File , do that and save it with a new name then open that as an Existing File ( transfer it out of the project folder into Two Stroke ).
The Turbulent sim is god awful slow , as its recalculating all the values each time ,and Neels has recently made the sim speed way faster.

porttiming124

20th February 2023, 09:47

And here it is: the ability to draw transfer port positions and flow directions without having to use a protractor.

There is also a drawback: for ports close to the front and rear cylinder walls, for example the C-port, specifying position points by means of bore percentages is not as accurate as using angles. That is why I did not quote an optimum position percentage for the C-port.
352436

Thank you Frits

wobbly

20th February 2023, 10:42

Frits , thought it might be helpful to add 33% as a nominal width of the C port as many people have no idea where to start with this.

diesel pig

20th February 2023, 12:22

Frits Overmars

20th February 2023, 12:51

Frits , thought it might be helpful to add 33% as a nominal width of the C port as many people have no idea where to start with this.Thanks Wob, you do have a point (pun intended). I must think of a way to incorporate a port width percentage within the position points approach.
Then again, the C-port has to fit between the B-ports rear position points, so there isn't much anyone can do wrong there.
Hopefully I'll get a brain wave. It's early days yet, and all comments are very welcome.

wobbly

20th February 2023, 14:28

Frits Overmars

20th February 2023, 22:52

I just thought that as you are proposing optimum port wall % crossing points , and the C port doesn't cross this line , having a reference % width is in essence the same thing when its walls are parallel to the centerline.Yes it is. I just need to give some thought to presenting it clear and simple. The ability of some people to misread and/or misunderstand something is amazing :facepalm:.

JanBros

21st February 2023, 00:58

Frits Overmars

21st February 2023, 03:03

so automaticaly a question pops up in my head : optimum for what ? only absolute high revving power seakers or all 2-strokes ?All two-strokes Jan. They will all benefit from depositing the fresh mixture where it is needed, while minimizing short-circuiting losses.

philou

21st February 2023, 03:38

the only thing that continues to interest people is the hight compression of the crankcase...

porttiming124

21st February 2023, 07:37

I just took an old piston to figure out the arrangement of the ports as you suggested. I realize that there is a very big difference.352443

wobbly

21st February 2023, 16:39

Philou , there is no black magic going on with CCR.
Bigger is better , but the limit for reeds is 1.3 as any bigger and the petals become too thin to suit the case Helmholtz and will flutter.
Rotary valve is around 1.24. TeeZee has been bigger , but we have no way of knowing if that was " better ".
Very high performance piston ports ( eg - full house snowmobiles ) all have made more power when made bigger up to 1.3 , but made no more power when going bigger , and the carburation would probably be impossible.
There is a raft of way more interesting things I am keen to learn about - over to you.

Frits Overmars

22nd February 2023, 00:28

...there is no black magic going on with CCR.
Bigger is better , but the limit for reeds is 1.3 as any bigger and the petals become too thin to suit the case Helmholtz and will flutter.
Rotary valve is around 1.24. TeeZee has been bigger , but we have no way of knowing if that was " better ".
Very high performance piston ports ( eg - full house snowmobiles ) all have made more power when made bigger up to 1.3 , but made no more power when going bigger .Wob, in the first line of your above quote you mention CCR: Crankcase Compression Ratio. But that collides with everything that follows...

I briefly thought about suggesting CV for Crankcase Volume. But that would steer people in the direction of the already-existing Curriculum Vitae.
Writing clear and simple texts can be a hassle...

Condyn

22nd February 2023, 03:08

Mixing up Case Volume and Curriculum Vitae surely would not confuse me Frits, hahaha.

skako

22nd February 2023, 04:03

Frits, I would like to ask.
Can you set the vertical (axial) slopes of the transfer channel depending on the stroke, bore and angle of the transfer duration.
As an example, the transfer duration of 120° and the other extreme of 144°.
These are some possible times for a moped with two transfer channels of the Schnuerle type with a piston-controlled intake.
Are your axial tilts valid for all transfer durations?

Peter1962

22nd February 2023, 06:19

I have also a question for Frits and Neels : have the optimum axial and radial angles that you kindly published been integrated in Engmod2T software ?

And a more general question : why the heck have all the manufacturers chosen for 66,4 x 72 for their MX engines when research in the GP world clearly showed that oversquare is less good than square, and not a single undersquare engine ever won a GP ?

I know that KTM had a square 250 MX engine, back in the days that they were still a rather small company, but did they change that just to copy the japanese brands who were at that point in time the market and technological leaders in the MX scene ? Or was there a more rational reason ?

In the Enduro world, there seems to be a broad consensus that square engines like the 300's from KTM, or Beta, Sherco and TM are the optimum way to go. BUT they all use 66,4 x 72 for their 250's... Would a factory that would design a new 2T 250 and that has NO plans to make a 300cc version be better off with a square design ?

Vannik

22nd February 2023, 06:46

I have also a question for Frits and Neels : have the optimum axial and radial angles that you kindly published been integrated in Engmod2T software ?

Yes, it has the option to design a 5-port scavenging systems based on Frits' recommendations, giving the results in either Blair or FOS geometry.

wobbly

22nd February 2023, 07:41

The undersquare 250MX engines were developed specifically around bottom and mid range torque production , as a higher reving square 250 is not needed at all.
Yamaha had a square 250MX for one year ( 68 X 68.8 )
- its great for 250 Superkarting but useless on an MX track.
And im lost Frits.
CCR is the ratio of the Case Volume/ Case Volume less the Swept Volume ie for a 250cc engine with a 900cc case we get 900/900-250 = 1,38 CCR.
Yes its counter intuitive that a smaller numerical ratio is a bigger case volume , but thats how its always been done and I thought everyone knew that.
An assumption being an error waiting to be revealed ?

husaberg

22nd February 2023, 18:15

The undersquare 250MX engines were developed specifically around bottom and mid range torque production , as a higher reving square 250 is not needed at all.
Yamaha had a square 250MX for one year ( 68 X 68.8 )
?

With the YZ's as far as i know 83-97 or so were all 68x68mm?
prior to this it was 70mm x 64
after 98 it was whatever its is now 66.?x72.
Honda as far as i know went to the long stroke in the Works RC's in about 79.
Villers and greeves had the LS dimensions sorted back to the 50's :devil2:
Std Villiers never worried about pesky rollers being oversped, as they has plain bearings (bronze bushes) on big and small ends just like a modern MX foul stroke.

How long did they last on a STD bottom end Grump? was it the load that killed them or the the milage?

Grumph

22nd February 2023, 18:34

Re Villiers. It was only the very early engines which had plain big ends. But plain bronze bush small ends lasted nearly till the end. And surprisingly they did last quite well. What usually broke was the gearbox shell once you got any HP out of them. Or the rod. I remember seeing a DOT with the rod wrapped around a frame member - but still with a sound big end.

Still watching with interest Rob - and reading Wob's contributions too. Still learning.

Frits Overmars

22nd February 2023, 23:05

im lost Frits. CCR is the ratio of the Case Volume/ Case Volume less the Swept Volume ie for a 250cc engine with a 900cc case we get 900/900-250 = 1,38 CCR.
Yes its counter intuitive that a smaller numerical ratio is a bigger case volume , but thats how its always been done....That's how I do it too. A bigger crankcase volume for a given cylinder capacity yields a smaller Crankcase Compression Ratio; nothing wrong with that.
But I feared that your post might confuse some folks. I do not want to ramble on but let's look at it one more time, for clarity.
Philou , there is no black magic going on with CCR.
Bigger is better , but the limit for reeds is 1.3 as any bigger and the petals become too thin to suit the case Helmholtz and will flutter.
Rotary valve is around 1.24. TeeZee has been bigger , but we have no way of knowing if that was " better ".
Very high performance piston ports ( eg - full house snowmobiles ) all have made more power when made bigger up to 1.3....You mention CCR and then you write "Bigger is better...etc" without ever using the word Volume, so a layman would read it as "Bigger CCR is better".
And I'm sure that is not what you wanted to communicate.

why the heck have all the manufacturers chosen for 66,4 x 72 for their MX engines when research in the GP world clearly showed that oversquare is less good than square, and not a single undersquare engine ever won a GP ?

In the Enduro world, there seems to be a broad consensus that square engines like the 300's from KTM, or Beta, Sherco and TM are the optimum way to go. BUT they all use 66,4 x 72 for their 250's... Would a factory that would design a new 2T 250 and that has NO plans to make a 300cc version be better off with a square design ?Long-stroke two-strokes have better blowdown and transfer angle.areas than square or short-stroke engines. With the same exhaust and transfer port timings they will rev easier.
Or, to look at it from the other side: they can make their power without the need for need high timings that would spoil rideability.

Undersquare means: the bore is smaller than the stroke. So an engine with 54 mm bore and 54,5 mm stroke is (granted: just) undersquare, or long-stroke as I prefer to call it.
And long-stroke Honda, Yamaha, Suzuki and Aprilia engines did win a GP or two....

You write that square engines like the 300's from KTM, or Beta, Sherco and TM are the optimum way to go.
Sure, if you already have a twofifty on the shelve and all you have to do is overbore the cylinder. From a technical point of view it's not the best solution, but try telling that to the bean counters. A factory that would design a new two-stroke 300 and that has NO plans to make a 250cc version, would definitely come up with a long-stroke 300.

Frits, Can you set the vertical (axial) slopes of the transfer channel depending on the stroke, bore and angle of the transfer duration.
As an example, the transfer duration of 120° and the other extreme of 144°. These are some possible times for a moped with two transfer channels of the Schnuerle type with a piston-controlled intake. Are your axial tilts valid for all transfer durations?I did, Skako. You can find the axial angle calculations below, and as you'll see, transfer durations do not play a role. But a transfer timing of 144°? :eek5:
For an engine with only two transfer ports I think it would be best to use the radial angles of the A-ports below and the average of the axial angles of the A-ports and the B-ports.
352448

philou

22nd February 2023, 23:27

skako, have you tried the wave simulator that I sent you?

a comparison with the one you developed?

skako

23rd February 2023, 01:25

skako, have you tried the wave simulator that I sent you?

a comparison with the one you developed?
I couldn't get it to run. Sorry I didn't get back to you. It accepts the input data, but I don't get the output. I've stopped, I'm learning and looking at how to rearrange and expand my programming code. If I manage something, I will contact you.

philou

23rd February 2023, 01:41

it only works in MS-DOS. works under win xp.

programmed in the years 1996

F5 Dave

23rd February 2023, 06:19

Yet Beta just last year changed their 300 from 72x72 to 73x69.9. Not really much of a change, but why go that way rather than the other?
And they hardly needed to, I have a 21 engine and it is fantastic for purpose.

wobbly

23rd February 2023, 09:07

Thanks Frits , the disparity in my CCR description never even occurred to me - criticism fully accepted.
The " bigger is better " thing came from the fact I have never ever , had to make a case smaller - its always been longer rods , cylinder spacer plates , pockets around the mains , reed spacers
etc trying to correct a small volume.

skako

23rd February 2023, 11:16

......
I did, Skako. You can find the axial angle calculations below, and as you'll see, transfer durations do not play a role. But a transfer timing of 144°? :eek5:
For an engine with only two transfer ports I think it would be best to use the radial angles of the A-ports below and the average of the axial angles of the A-ports and the B-ports.
352448
Thank you Frits,
I will try to post a video of the simulations. The simulator simply took me to angles of 144°. I combined different exhaust pipe designs and transfer times. The goal was to achieve greater power and elasticity (wider power range combined with gear ratios).

https://streamable.com/279h89

Frits Overmars

24th February 2023, 01:52

The " bigger is better " thing came from the fact I have never ever , had to make a case smaller -
its always been longer rods , cylinder spacer plates , pockets around the mains , reed spacers etc trying to correct a small volume.Same here :D.
And it's my pleasure to show your input below:
352457

skako

24th February 2023, 03:23

Same here :D.
And it's my pleasure to show your input below:
352453

Frits,
Do you have perhaps a scavenging efficiency graph as a function of scavenging (volumetric) ratio for your type of transfer channels.
Which known and measured scavenging type best describes your transfer channels.

Frits Overmars

24th February 2023, 04:29

Frits, Do you have perhaps a scavenging efficiency graph as a function of scavenging (volumetric) ratio for your type of transfer channels.
Which known and measured scavenging type best describes your transfer channels.What kind of graph would that be Skako? Efficiency along the Y-axis, OK. But what about the X-axis? Primary compression ratio? Revs? Throttle position? Mixture strength? Engine temperature? Environmental temperature? Environmental air pressure? We can handle 2-D graphs and 3-D maps. But 7-D...?

The 'known and measured scavenging types' you mention remind me of the Yahama papers, the various SAE-papers and Blairs "Design and Simulation of Two-stroke Engines".
But a lot of research has been done since then. My name appears on some of the publications, but it really was Jan Thiel who carried it all as far as we are now.
So if you feel the need to give it a name, think of him.

Peter1962

24th February 2023, 04:56

...
Undersquare means: the bore is smaller than the stroke. So an engine with 54 mm bore and 54,5 mm stroke is (granted: just) undersquare, or long-stroke as I prefer to call it.
And long-stroke Honda, Yamaha, Suzuki and Aprilia engines did win a GP or two....

Frits, I know see that I wrote my question extremely clumsy...
Of course in GP racing, the bulk of GP's were won on 54 x 54,5 bore to stroke engines (what i should have called 'square')
In gp mx, in the 250 cc class all engines are long stroke. (66,4 x 72 )

You write that square engines like the 300's from KTM, or Beta, Sherco and TM are the optimum way to go.

That was what I thought, since all the modern 300cc two strokes are 72 x72 and being very succesful...

Sure, if you already have a twofifty on the shelve and all you have to do is overbore the cylinder. From a technical point of view it's not the best solution, but try telling that to the bean counters. A factory that would design a new two-stroke 300 and that has NO plans to make a 250cc version, would definitely come up with a long-stroke 300.
...

Thank you for a your answer. I assumed (wrong) that square ratio would be the optimum in a 250 or 300 MX engine.

Peter1962

24th February 2023, 04:59

Yes, it has the option to design a 5-port scavenging systems based on Frits' recommendations, giving the results in either Blair or FOS geometry.

Thank you, Neels.

skako

24th February 2023, 06:36

What kind of graph would that be Skako? Efficiency along the Y-axis, OK. But what about the X-axis? Primary compression ratio? Revs? Throttle position? Mixture strength? Engine temperature? Environmental temperature? Environmental air pressure? We can handle 2-D graphs and 3-D maps. But 7-D...?

The 'known and measured scavenging types' you mention remind me of the Yahama papers, the various SAE-papers and Blairs "Design and Simulation of Two-stroke Engines".
But a lot of research has been done since then. My name appears on some of the publications, but it really was Jan Thiel who carried it all as far as we are now.
So if you feel the need to give it a name, think of him.

I meant something like this graph
352456

Vannik

24th February 2023, 16:19

I meant something like this graph
352456

That is the single most difficult thing to find! There used to be a special test bench at QUB to measure it but is has since been scrapped as QUB has moved away from engine development. They printed over a thousand cylinders to test but I have not been able to find the results anywhere. Honda built their own but according to Prof Blair they could never get it to function correctly. Today the characteristic is an output of unsteady moving piston CFD simulations which is another ball game altogether.

F5 Dave

24th February 2023, 18:05

I was going to say it looks uncomfortably like something from a Eugenics site. But I decided against it. Until Friday night happened. :drinknsin

PS search Adam Ruderfords(?) Bad Blood podcast BBC fir a balanced history from a Biologist.

TZ350

25th February 2023, 08:55

pete376403

25th February 2023, 11:15

TwoStroke Stuffing .......

Soeedway bike frame? Many years back a guy I competed against was seriously looking into putting a KX500 motor into a Jawa DT frame but rules wouldnt allow two stroke in speedway

TZ350

25th February 2023, 15:19

.
352459

Many thanks to Speedpro for his help with turbine housing for this.

A turbo charged Suzuki RG with a aftermarket 65cc big bore cylinder. The next project after I get the Kawasaki F81M 250 running properly on Methanol.

edgefinder

26th February 2023, 04:49

The beauty of the turbo rg50 is you can use all the power a lot and it can't get out of hand try to hurt you. Good for tuning driveability

TZ350

26th February 2023, 14:41

.

352460

So much for a quiet afternoon trying different needles to improve mid range throttle response.

You know you are getting close to success when Murphy (Murphy's Law) finds other ways unrelated to your project to stuff you up.

Last week it was an explosion in the extractor fan, sounded like a massive back fire from a big old truck. This time the battery burst with a loud bang when I rolled the engine over, internal hydrogen explosion I guess.

diesel pig

26th February 2023, 14:58

The beauty of the turbo rg50 is you can use all the power a lot and it can't get out of hand try to hurt you. Good for tuning driveability

Could not of said it better myself.

Haufen

26th February 2023, 21:40

https://www.youtube.com/watch?v=liBRL6Hov-Y&t=10s

transparent two-stroke cylinder

OopsClunkThud

27th February 2023, 04:15

That is the single most difficult thing to find! There used to be a special test bench at QUB to measure it but is has since been scrapped as QUB has moved away from engine development. They printed over a thousand cylinders to test but I have not been able to find the results anywhere. Honda built their own but according to Prof Blair they could never get it to function correctly. Today the characteristic is an output of unsteady moving piston CFD simulations which is another ball game altogether.

Assume you are you referring to the "QUB single-cycle gas scavenging apparatus".

A CFD simulation of this setup has been done and could be way more useful than the physical setup in speed of results as well as insight of what happens inside the ports throughout the cycle. Sadly it combines all the really hard and non-standard CFD corner cases of unsteady flow, moving geometry, and mixing gases. Each of these has some limited support in the CFD packages but are often mutually exclusive. The customization and setup of the software to do it seems almost harder than the physical test rig.

But to have that CFD process in hand with a somewhat efficient workflow for setting up the test cylinder mesh would be a dream.

Vannik

27th February 2023, 18:55

A nice video on the uses and explanation of WaveViewer2T:

https://www.youtube.com/watch?v=jaAE-Bq2lvo

skako

28th February 2023, 04:03

A nice video on the uses and explanation of WaveViewer2T:

https://www.youtube.com/watch?v=jaAE-Bq2lvo

Many years ago, at the very beginning of my work, in a company that made trucks in the north of our region, I found a book in the library by Prof. Jante and Dr. Hofmann "Uber Verbrennungsmotoren und Kraftfahrwesen" Band 2, edition from 1959.
In it, Dr. Hofmann in his Dissertation calculates and clearly shows the waves in the intake pipe of a 2T engine with a piston-controlled intake. Dr. Hofmann's presentation about waves in the intake pipe is accompanied by many graphs that show the size of the outgoing wave from the intake port and the size of the return reflected wave (which is delayed by some time delta_t depending on the length of the intake pipe and the sonic velocity of the atmosphere).
The graphs clearly show the superposition of those two waves and their effect on the intake port. Then I learned that the superposition wave is the only one that we can measure at a certain position with very fast electronic sensors, and also that it is composed of two waves that we cannot measure but can approximately calculate.
Based on that procedure, I made a suction Basic-program on my first home 8-bit computer (Schneider CPC 464) where I could see the first suction graphs. I was very happy about that, because I saved some costs about measurements.

rgvbaz

28th February 2023, 05:00

A nice video on the uses and explanation of WaveViewer2T:

https://www.youtube.com/watch?v=jaAE-Bq2lvo

Thanks Neels :)

I hope it can be of some use.

Edit - please read the pinned comment in the comments about my error in stating that the left and right waves are theoretical. They aren't, they can't be measured as they both merge and form the superposition pressure - they are real.

Cheers

Dave

rgvbaz

28th February 2023, 05:09

Many years ago, at the very beginning of my work, in a company that made trucks in the north of our region, I found a book in the library by Prof. Jante and Dr. Hofmann "Uber Verbrennungsmotoren und Kraftfahrwesen" Band 2, edition from 1959.
In it, Dr. Hofmann in his Dissertation calculates and clearly shows the waves in the intake pipe of a 2T engine with a piston-controlled intake. Dr. Hofmann's presentation about waves in the intake pipe is accompanied by many graphs that show the size of the outgoing wave from the intake port and the size of the return reflected wave (which is delayed by some time delta_t depending on the length of the intake pipe and the sonic velocity of the atmosphere).
The graphs clearly show the superposition of those two waves and their effect on the intake port. Then I learned that the superposition wave is the only one that we can measure at a certain position with very fast electronic sensors, and also that it is composed of two waves that we cannot measure but can approximately calculate.
Based on that procedure, I made a suction Basic-program on my first home 8-bit computer (Schneider CPC 464) where I could see the first suction graphs. I was very happy about that, because I saved some costs about measurements.

That sound like a really cool program - did you use it to develop intakes and exhausts?

This is an extract from a SAE paper, it's 4 stroke but seems to give reason to why Dr Blair was keen to be able to predict these waves despite not being able to actually measure them too.

352461

Cheers,

Dave

wobbly

28th February 2023, 08:23

Also many years ago I had a long discussion by snailmail with Dr Fleck at QUB about the left/right waves and their superposition.
My assertion was that if you had a pair of sensors at a set distance apart , a fast processor could detect each wave and then display its direction / speed /amplitude from the time delta of the signal across this sensor pair.
He did arrange to have a student work this up as a part of a dissertation , but in the end it proved "too difficult " in reality , and the student went on to complete his Phd on a new code
methodology that ultimately was used in a commercial engine sim.

skako

28th February 2023, 11:04

That sound like a really cool program - did you use it to develop intakes and exhausts?

This is an extract from a SAE paper, it's 4 stroke but seems to give reason to why Dr Blair was keen to be able to predict these waves despite not being able to actually measure them too.

352461

Cheers,

Dave

Dr. Hofmann gave a calculation procedure for a suction pipe with subsonic flow.
I wrote the Qbasic code according to that procedure, for intake into the engine crankcase. It looks can be seen in the appendix.
The procedure was applied to a straight pipe of constant section.
Exhaust pipes are much more complex and they were processed by Prof. Blair.

Also many years ago I had a long discussion by snailmail with Dr Fleck at QUB about the left/right waves and their superposition.
My assertion was that if you had a pair of sensors at a set distance apart , a fast processor could detect each wave and then display its direction / speed /amplitude from the time delta of the signal across this sensor pair.
He did arrange to have a student work this up as a part of a dissertation , but in the end it proved "too difficult " in reality , and the student went on to complete his Phd on a new code
methodology that ultimately was used in a commercial engine sim.

There would be no use in measuring the left outgoing wave because it is generated again as a right return wave.
With a certain delay, the right return wave is superimposed in front of the observed port and somewhat cancels the original (left wave).
This means that again you can only measure the superimposed magnitude of those two waves at that point at that time.
At any point on the exhaust pipe, the pressure sensor will measure some pressure value that is currently the sum of the value of the left wave with some of the pressure values of the right wave. These are the values that found themselves in that position at that moment.
When you look at the left outgoing wave, think of it as the composition of a train with many wagons. Each of the wagons has a certain size of load (pressure) that it transports.
The measured size is the sum of these two left and right wave sizes at that moment in that position. Let's imagine that the 76th wagon of the outgoing wave was in that position (as expansion pressure) and that it met the returning 13th wagon of the right wave (which carries a strong compression pressure during the blow down time).
Therefore, all measured quantities are always superimposed wave quantities.

I'm sorry I can't express myself better because I have to use a translator.

352462352463

Vannik

28th February 2023, 17:58

Also many years ago I had a long discussion by snailmail with Dr Fleck at QUB about the left/right waves and their superposition.
My assertion was that if you had a pair of sensors at a set distance apart , a fast processor could detect each wave and then display its direction / speed /amplitude from the time delta of the signal across this sensor pair.
He did arrange to have a student work this up as a part of a dissertation , but in the end it proved "too difficult " in reality , and the student went on to complete his Phd on a new code methodology that ultimately was used in a commercial engine sim.

This was eventually done at a French university. They used 2 high frequency response transducers and fairly advanced mathematics to separate the left and right moving waves from the signal, but it only confirmed the theory and did not show anything new.

rgvbaz

28th February 2023, 23:34

Here's an open access paper - The application of PIV to the study of unsteady gas dynamic flow within pipes and at pipe discontinuities by VD Thornhill, R Fleck, H Li and J Woods

https://iopscience.iop.org/article/10.1088/1742-6596/85/1/012002

The apparatus used produces discrete waves that were observed without being superposed.

Regards

Dave

Gradella23

1st March 2023, 23:10

ken seeber

1st March 2023, 23:47

hello everyone,
has someone ever tried a 3d printed reed valve? I found some fuel-resistant materials, but i don't know how to replace the viton coating (that needs a mold etc.).
Are there some sort of spray coatings or paints than can resemble that effect?

thanks so much for all the knowledge shared here!

Just get some raspberry juice, thongs and walk around for a bit....

https://www.youtube.com/watch?v=sKC2027ynMg

husaberg

2nd March 2023, 09:48

hello everyone,
has someone ever tried a 3d printed reed valve? I found some fuel-resistant materials, but i don't know how to replace the viton coating (that needs a mold etc.).
Are there some sort of spray coatings or paints than can resemble that effect?

thanks so much for all the knowledge shared here!

google FKM liquid. you will come up with products. Like below.
https://www.thermodynglobal.com/product/viton-coating-thp-2000/
Vitons a trade name. Like Vaseline

Gradella23

2nd March 2023, 22:22

Just get some raspberry juice, thongs and walk around for a bit....

https://www.youtube.com/watch?v=sKC2027ynMg

google FKM liquid. you will come up with products. Like below.
https://www.thermodynglobal.com/product/viton-coating-thp-2000/
Vitons a trade name. Like Vaseline

Thanks guys, i'll take a look.
This is what i'm working on: a 50cc Cagiva Prima racer.
This prototype reed valve is made to fit an Ibea 30mm carb on it, which has a 33,5mm exit diameter. It's a little bit large but i tried to arrange an acceptable internal duct.
This engine already has a nice inlet shape, and was born to also fit a 75cc so it has a 47mm large crankcase (bigger than a Minarelli or Derbi).
The cylinder base is with external studs so it's been possible to design a cylinder with proper transfer ducts.

ken seeber

3rd March 2023, 00:45

Thanks guys, i'll take a look.
This is what i'm working on: a 50cc Cagiva Prima racer.
This prototype reed valve is made to fit an Ibea 30mm carb on it, which has a 33,5mm exit diameter. It's a little bit large but i tried to arrange an acceptable internal duct.
This engine already has a nice inlet shape, and was born to also fit a 75cc so it has a 47mm large crankcase (bigger than a Minarelli or Derbi).
The cylinder base is with external studs so it's been possible to design a cylinder with proper transfer ducts.

Being 3D printed, it might be possible that you don't need any coating at all, depending on the choice of filament. I was thinking of TPU, but not sure of its resistance to petrol.
However it sounds like you might know all of this. Another thought is that with a coarse Z height, say 0.28 (a Cura setting) that the voids in each layer, might offer some form of oil retention and subsequent dampening for the impact of closure of the reeds.
You can be our experiment. Looking forward to your durability test results tomorrow.::rolleyes:

koenich

3rd March 2023, 01:57

I have a few PTFE CNC milled reed cages w/o FKM coating. I was worried at first but the reeds surprisingly last at least 15h racing in a 20HP 50cc before showing minimal signs of wear.

wobbly

3rd March 2023, 07:41

I have printed around 30 reed inserts , the best have much more 1/2 moon rounded side fill volume than the one you show.
Here is an example with a thin divider added for a test.

monkeyfumi

3rd March 2023, 08:34

I think he is looking at printing not just the stuffer, but the reed cage itself.

I'm not sure printed TPU would be stiff enough for a cage, PETG or Nylon would be a better material to try.

skako

3rd March 2023, 11:39

Here is a calculation of the outgoing and return (reflected) waves acting on the opening port.
A return wave delay to the intake port can be seen.
I will try to put together a graphic for better understanding.
352526

Here, after several hard days, I transferred the program. I put together a graphic where you can clearly see individual waves that are created during the process of "breathing" the engine.
352527

Gradella23

3rd March 2023, 20:57

Being 3D printed, it might be possible that you don't need any coating at all, depending on the choice of filament. I was thinking of TPU, but not sure of its resistance to petrol.
However it sounds like you might know all of this. Another thought is that with a coarse Z height, say 0.28 (a Cura setting) that the voids in each layer, might offer some form of oil retention and subsequent dampening for the impact of closure of the reeds.
You can be our experiment. Looking forward to your durability test results tomorrow.::rolleyes:

I'm investigating with my 3d printing expert colleagues to understand if there is something both fuel resisting and "soft" enough to avoid any additional coating.
the company where i'm working has an entire building of printers (they used to grant public money here for companies to invest in new tecnologies) so i'm trying to take advantage.. they also have a DMLS but unfortunately we can only print steel and titanium and no aluminium.

I have a few PTFE CNC milled reed cages w/o FKM coating. I was worried at first but the reeds surprisingly last at least 15h racing in a 20HP 50cc before showing minimal signs of wear.

this is nice to hear, i hope we'll reach the 20hp goal :drool:

I have printed around 30 reed inserts , the best have much more 1/2 moon rounded side fill volume than the one you show.
Here is an example with a thin divider added for a test.

thanks Wob, i have plenty of your works in my gallery. My cylinder is obviusly using a wobbly exhaust duct!
Regarding this reed cage, the problem is that is made to fit the Ibea 30mm carb (which i already have but i think it's too big for this duct) and that has a 33,5mm exit D.
To maintain a constant cross area i needed to enlarge the base section of the "insert" and i don't have any more room to create the 1/2 moon shape i was looking for (see pics). I guess the situation will improve a lot with the other carbs i plan to test (mikuni tmx30 and tm28ss).
The only way i have to reach that result with this setup is to squeeze the duct laterally but i don't know if it's worth...

porttiming124

4th March 2023, 03:04

Hello everyone.
A friend of mine set out to make a turbo methanol drag sled.
the original engine is a 2 cylinder 1000 cc
with a standard compression ratio of 12.5@1
the end goal would be to have about 15 lbs of boost.
Do any of you have experience with 2 stroke turbo methanol engines to line it up for starters?
compression, timing, air fuel, egt...
Thank you

wobbly

4th March 2023, 07:53

Just one more small detail re the stuffer to add. Even though Im dealing with a very small carb on a big reed ( 30mm/125 reed cage ) you would think that the carb bore was the limiting factor.
Not so , the best stuffers have a 2mm smaller venturi around 10mm into the block , where normally the area would be increasing or remaining equal.
This idea may work where your carb exit is " too big " ie neck the area down into a venturi within the stuffer.

F5 Dave

4th March 2023, 08:44

So it would go 30 at the carb slide, out to say 32 at the carb exit and back to 28 in the Reed? That is counter intuitive.

wobbly

4th March 2023, 10:27

Of course its counter intui it intuit , that big word.
Its a 2T , who would have thought making the Ex duct exit 75% would be WAY faster.
Or that too much Blowdown would wreak the scavenging due to LESS pressure at TPO.

husaberg

4th March 2023, 10:36

Two strokes were way more fun when we were just raising the ex port 2mm and cutting out 25mm of the belly...:whistle:

F5 Dave

4th March 2023, 20:12

Look, I'm just going to keep my polishing mops primed up for max shine and be done with it yeah?

husaberg

4th March 2023, 20:44

The shorter old model straight rubber is best.
Here is a very old dyno test I did on a TM125MX motor we used to race in 125 National class karts, this was used with a VeeForce 2 and a bored SPJ powerjet carb..
Its a crap printout - used to be orange and red, but the short manifold looses no mid power, but is near 2 Hp up around peak.
The short rubber is obsolete and all but impossible to find.

https://italkit.com/epages/Italkit.sf/en_GB/?ObjectPath=/Shops/Italkit/Products/TA.26.51
Not sure re the length it looks half way between

size is odd as well at 32mm and 36mm

katinas

5th March 2023, 01:42

So it would go 30 at the carb slide, out to say 32 at the carb exit and back to 28 in the Reed? That is counter intuitive.

Honda CR 125 2003 could hold the record for the biggest difference. For 2003 Honda made the biggest ever CR 125 reed cage ( 66 mm wide vs. 59 mm of 2002 ) with the smallest insert. Bigger air box, new carburetor 38 mm ( 36 mm 2002) huge reed cage and all this against 28 mm reed cage insert at narrowest point. Reports from testers:

"After a few hours riding the 2003 CR125R, we are happy to report that Honda has delivered a substantially improved motor, and that the chassis is just as sweet as last year. While last year’s bike required massive clutch abuse exiting corners, the ’03 model has a much broader, easier-to-use powerband. While “better breathing” might lead you to believe Honda produced a top end screamer with no low-to-mid power, this is not the case.

Indeed, the 2003 CR125R seems to pull harder from bottom to top when compared to the 2002 model (also available for our sampling at Carlsbad Raceway). For the first time in several years, Honda has a very competitive stock 125 motor to go along with its superb handling and suspension."

2004 same big 66 mm reed cage, but the insert dia. 38mm, same as carb .
From 2005 to the last 2007 CR 125, they return to narrower 59 mm reed cage with usual three petals and 38 mm insert.

Husa, other very short intake manifolds are used on the Yamaha YZ 125 from 2005 to 2022? Shorter could be to insert the carb directly to the plate that holds the reed cage.

Splatter

5th March 2023, 02:22

352539352540352541352542352543

We have been running these for 7-8 years with no issues. The cages are Nylon/fiberglass, the spacer is nylon/carbon fiber. This cage is for the KX500/CR500. There are some concessions made to the build due to the material. It doesn't do well with supports so the entry of the splitters has too much of a sharp edge before it radiuises because that has been the best way to get it to build off the plate.

Gradella23

5th March 2023, 10:19

husaberg

5th March 2023, 10:21

Honda CR 125 2003 could hold the record for the biggest difference. For 2003 Honda made the biggest ever CR 125 reed cage ( 66 mm wide vs. 59 mm of 2002 ) with the smallest insert. Bigger air box, new carburetor 38 mm ( 36 mm 2002) huge reed cage and all this against 28 mm reed cage insert at narrowest point. Reports from testers:

"After a few hours riding the 2003 CR125R, we are happy to report that Honda has delivered a substantially improved motor, and that the chassis is just as sweet as last year. While last year’s bike required massive clutch abuse exiting corners, the ’03 model has a much broader, easier-to-use powerband. While “better breathing” might lead you to believe Honda produced a top end screamer with no low-to-mid power, this is not the case.

Indeed, the 2003 CR125R seems to pull harder from bottom to top when compared to the 2002 model (also available for our sampling at Carlsbad Raceway). For the first time in several years, Honda has a very competitive stock 125 motor to go along with its superb handling and suspension."

2004 same big 66 mm reed cage, but the insert dia. 38mm, same as carb .
From 2005 to the last 2007 CR 125, they return to narrower 59 mm reed cage with usual three petals and 38 mm insert.

Husa, other very short intake manifolds are used on the Yamaha YZ 125 from 2005 to 2022? Shorter could be to insert the carb directly to the plate that holds the reed cage.

Cheers some odd spacings put might be the go for a smaller cadge as o hasroom for redrilling
352553352554352555

http://www.transcanimports.com/downloads/Essentials%20Catalogue/Combo%20-%20Mikuni%20Mounting%20Flanges.pdf

https://www.allensperformance.co.uk/carb-kit/carburettor-mounting-rubber-manifold-dimension-information-page/

http://jbmindustries.com/MZ_Socket_Boot.html

wobbly

5th March 2023, 12:33

Did a ton of dyno tests on the RS125/CR125 cage and stuffers as they fitted the TM 125MX when we had 125 National in karts ( bike engines only )
The best manifold by far was the NX4 - 000 but that has been NLA for years.
The 780 is the later model RS125 but its way too long
The angled one looses 2 Hp.

EDIT - re the reed stuffer venturi - I have no idea what would work better in your case , best to have a go with varying sizes.

Muhr

6th March 2023, 06:36

Grumph

6th March 2023, 09:19

Muhr - not knowing what the OE setup looks like, what are the changes ? Spacer or venturi at the entry point ?

Is the angled entry a factor ?

thanks

Muhr

6th March 2023, 10:20

Muhr - not knowing what the OE setup looks like, what are the changes ? Spacer or venturi at the entry point ?

Is the angled entry a factor ?

thanks

A spacer was a necessary "evil" because a kz reed cage is too big to fit on an mx engine.

husaberg

6th March 2023, 11:20

Tried getting a little more power from a tm mx 125 a while ago.
Where I have experienced that the most limiting thing is intake/reed.
I tried these Vforce 3,4 and 4r without any results worth mentioning.
Did as below in the end, which made a big difference for us. (from a 144cc kz kit)

Did you try VF2?
My understanding is that was the one that got results at least with Hondas.
I think that came from mr Waynewright

wobbly

6th March 2023, 11:35

Yep , the VF2 ( original bolted together assembly ) was far superior to the later VF versions on the TM 125MX.
I think the KTM had it from the factory.

Muhr

6th March 2023, 11:42

Gradella23

6th March 2023, 21:48

Unfortunately, nothing I've tested. I also can't say that it was reed it self or an increase in crankcase volume, or maybe both. But there was a big difference

So this is the OG? https://www.ebay.ca/itm/332619174160?hash=item4d71a5d110:g:j3gAAOSwmWVaz-nT

hello muhr, i've seen your 50cc project while reading some pages back.. have you finally tried it? that looked nasty.

352539352540352541352542352543

We have been running these for 7-8 years with no issues. The cages are Nylon/fiberglass, the spacer is nylon/carbon fiber. This cage is for the KX500/CR500. There are some concessions made to the build due to the material. It doesn't do well with supports so the entry of the splitters has too much of a sharp edge before it radiuises because that has been the best way to get it to build off the plate.

amazing quality. have you also experienced an improvement restricting the internal cross section instead of maintaining the carb exit cross section?
Have you tested different lenghts for the horizontal splitter? i see that this goes all the way from the base to the top of the cusp.
I even saw someone on the dutch forum (deraceheldenvanweleer.nl) hypothesizing a manifold with a continuation of the splitter, but i can't find any feedback on that.

wobbly

7th March 2023, 08:00

TZ350

7th March 2023, 11:11

.
F81M balance factor post 1.

352574 Reciprocating weight is 372g

352575 50% balance factor would be 186g.

352576 And a picture of my F81M's rotary valve compared to a standard unit for anyone interested in such things.

TZ350

7th March 2023, 11:11

.
F81M balance factor post 2.

352577 Ok, I see why the F81M vibrated like a basted. I had machined away so much of the counter weight there is negative balance factor.

352578 I hung 50% of the reciprocating weight on the rod and added extra counter weight until it all balanced out.

352579 So if I have got this right. I need to add 227g to the counter weight to achieve a balance factor of 50%.

Grumph

7th March 2023, 12:22

.
F81M balance factor post 2.

352579 So if I have got this right. I need to add 227g to the counter weight to achieve a balance factor of 50%.

Or - take 227g off the pin side of the wheels.

Why 50% ? Personally I'd be going for somewhere around 60% as a starting point. Do you know the OE balance factor ?

F5 Dave

7th March 2023, 12:28

Can't help you there but your gudgeon pin looks dreadful. Hope that's an old scrap one.

TZ350

7th March 2023, 12:53

Can't help you there but your gudgeon pin looks dreadful. Hope that's an old scrap one.

You are right. It is a re used one. looks much worse in the photo than real life. Camera picks up all the little discolorations.

wobbly

7th March 2023, 15:04

Dead correct , for a race engine 58% is proven technology.

husaberg

7th March 2023, 17:50

You are right. It is a re used one. looks much worse in the photo than real life. Camera picks up all the little discolorations.

Out of interest, What is the balance factor now in its current mondo vibrating form?

F5 Dave

7th March 2023, 17:52

You are right. It is a re used one. looks much worse in the photo than real life. Camera picks up all the little discolorations.

I think you are flattering your camera. Looks like it was taken with a Nokia flip phone. :whistle:

TZ350

7th March 2023, 19:21

Out of interest, What is the balance factor now in its current mondo vibrating form?

I did not bother to work it out. All I know for sure is that it vibrates really really bad at 9-10,000 rpm. When I looked into it I found the reciprocating weight side was notably heavier than the counter weight side.

P.S. Thanks for the link, read up on the 10,000 rpm 250 single and that they dynamically balanced the crank. I have seen car cranks done that way but how to make something home made to test a single crank looks challenging.

Lots of interesting reading here on bikes and adventure:- https://www.oldbikemag.com.au/

TZ350

7th March 2023, 19:32

I think you are flattering your camera. Looks like it was taken with a Nokia flip phone. :whistle:

Yep you are probably right. I should dig around in the old tea chest again and see if I can find a better piston pin. I suppose you would be horrified to know it's the same place I found the little end and cir-clips....... :laugh:

A great deal of this engine came out of a Cowshed and old wooden Tea chest.

352588

Frits Overmars

7th March 2023, 23:03

.
F81M balance factor post 1.
352582Reciprocating weight is 372g

35258350% balance factor would be 186g.Where is your con rod in that second picture? The weight of its small end should be part of the 186 g.
186 g plus the small end weight will give a notably higher balance factor (which in practice may not be a bad thing).

Gradella23

8th March 2023, 01:06

Where is your con rod in that second picture? The weight of its small end should be part of the 186 g.
186 g plus the small end weight will give a notably higher balance factor (which in practice may not be a bad thing).

Hello Frits, it's been an honour to read your posts on the various forums during the years.
Regarding the balance factor, i'd like to ask you which is, in your opinion, the best route to follow when manufacturing a crankshaft from scratch: once I know the alternating mass, and designed the crank webs according to the engine case... how should I continue? simmetric holes on each big end side? a simulation?

my engine also has its oem balance shaft (pic below): if I, for example, use the 58% balance factor Wobbly suggested on the crankshaft, is there a way to properly adjust the balance shaft factor? (other than trying i guess, but just to have a starting point). Tbh I've also read Jan Thiel saying that you must just get rid of cagiva type balance shaft, but there's always time to throw it away after testing :lol:

thanks in advance.

Frits Overmars

8th March 2023, 03:42

Hello Frits, it's been an honour to read your posts on the various forums during the years.
Regarding the balance factor, i'd like to ask you which is, in your opinion, the best route to follow when manufacturing a crankshaft from scratch: once I know the alternating mass, and designed the crank webs according to the engine case... how should I continue? simmetric holes on each big end side? a simulation?

my engine also has its oem balance shaft (pic below): if I, for example, use the 58% balance factor Wobbly suggested on the crankshaft, is there a way to properly adjust the balance shaft factor? (other than trying i guess, but just to have a starting point). Tbh I've also read Jan Thiel saying that you must just get rid of cagiva type balance shaft, but there's always time to throw it away after testing :lol:
thanks in advance.Grazie Gradella.
A full description of how to manufacture a crankshaft would take more time than I have available. But I would strive for symmetry as much as possible.
You can also take a look at an Aprilia RSA125 crankshaft; not at the RSA250 crankshaft, because that is asymmetrically balanced.

For an engine with a balance shaft I would give the crankshaft a 50% balance factor en let the balance shaft take care of the other 50%.

The Cagiva balance shaft in your photo evokes memories. In Calusco d'Adda, where Jan Thiel and I lived for a while, we knew a Sport Production rider with a Cagiva Mito. According to the regulations you were not allowed to do anything to the engine but when I saw that balance shaft, I saw possibilities. It rotated at the bottom of the crankcase, submerged in gearbox oil, like a cream whipper. That must have cost power, so a cylindrical sleeve was fitted around it, smoothing its circumference. This brought 2 hp and no one ever thought of checking not only the cylinder but also the balance shaft.

Vannik

8th March 2023, 05:20

.

352576 And a picture of my F81M's rotary valve compared to a standard unit for anyone interested in such things.

I am surprised no-one commented on this yet!

pete376403

8th March 2023, 07:53

I am surprised no-one commented on this yet!

Wish I had known that 50 years ago. I had a 72 F9 (Bighorn) 350 that did that twice. Once with the standard rotary valve, and once with the disc I had trimmed a bit off. Wear on the port was the least of the damage, the stainless bits made a real mess of the piston.

TZ350

8th March 2023, 08:11

352586 I am surprised no-one commented on this yet!

I had rounded the edge's of the disk so that the port was completely opened/closed at the same instance in the hope that would improve the induction pulse. Wondered why the port edge was wearing. Thanks for the heads up.

TZ350

8th March 2023, 08:25

Where is your con rod in that second picture? The weight of its small end should be part of the 186 g.
186 g plus the small end weight will give a notably higher balance factor (which in practice may not be a bad thing).

Thanks for the heads up. I thought weighing the reciprocating mass and then hanging a percentage of its mass off the end of the conrod was the way to go. I will have to re visit how I am doing this.

wobbly

8th March 2023, 08:49

The best disc shape to help wear is to angle cut the closing side as per this Aprilia pic.
I tested every shape possible doing Hines 250 Rotax Superkart , and nothing made any more power than that - with the added bonus of less wear on the case.

Grumph

8th March 2023, 09:49

Thanks for the heads up. I thought weighing the reciprocating mass and then hanging a percentage of its mass off the end of the conrod was the way to go. I will have to re visit how I am doing this.

Method is fine. Arithmetic needs tweaking.

TZ350

8th March 2023, 13:57

Where is your con rod in that second picture? The weight of its small end should be part of the 186 g.
186 g plus the small end weight will give a notably higher balance factor (which in practice may not be a bad thing).

Method is fine. Arithmetic needs tweaking.

Ok. Looks like I incorrectly calculated the mass to be balanced.

I had included the little end of the con rod in the reciprocating mass calculation Step-1, but by not including it in Step-2 I had inadvertently included it's effect twice when calculating the mass required to be balanced in Step-2.

Looks like I should have made my calculation:-

Step-1 Reciprocating Mass X B/F% = "B/F Mass required"

Step-2 Weigh con rod little end plus hook and washers until you have the"B/F Mass required" to hang on the end of the rod.

Step-3 add/subtract mass from/to the cranks counter weight side to achieve static balance.

352589 I think they make the same mistake in this description of how to balance a single cylinder crankshaft.

2 Stroke Stuffings take on finding the existing balance factor. https://youtu.be/0hoE_MuFpPI

Grumph

8th March 2023, 18:12

husaberg

8th March 2023, 18:29

Moments like this I always go back to good old Phil Irving. Pretty well every book he ever wrote - except his autobiography - had the method and formulas for crank balancing included. Pre internet references for a process which predates the internet.

Probably showing my age there.....

.

There has been a bit of talk about Crank Shaft Balance Factors on the Two Stroke Tuners thread, resulting in more heat than light. As there is a bit of a rowdy crowd over there I am going to make my posts here.

I intend to keep it simple, so for what its worth, this is our interpritation of Phill Irvings work.

Phill Irving talks about CrankShaft Balancing and Balance Factors on pages 107 to 110 of the book which can be found on page 61 of the PDF.

http://www.scribd.com/doc/15392252/Tuning-for-Speed-P-E-Irving-1965-Tuning-Racing-Motorcycle-Engines

Anyone can open a free account with Scribed where they can download "Tuning for Speed" by Phill Irving.

"xwhatsit" is the guy who uploaded Tuning for Speed to Scribd in the first place, thanks xwhatsit.

I will use a 2-Stroke single cylinder TF100 crank to demonstrate my explanations and I need to break it down into sections as I can't cover it all in one or two posts. Thomas gave me a hand, taking the photos and gave me a few tips along the way.

My plan is to:-

(1) Post photos of all the cranks I have lying around and quickly describe the features I can see on them.

(2) Describe how to find the reciprocating weight.

(3) What the "Balance Factor" really is.

(4) I will answer this question, Which is heavier, (A) the reciprocating weight of the rod piston assembly or (B) the counter weight of the crankshaft.

(5) Why there is a "Balance Factor" and what changing it does.

(6) What to do with your "Balance Factor"

(7) A simple but very effective jig for staticaly balancing a crank, that you can make yourself in 10min's. (Thanks Thomas)

(8) What a rocking couple is, and how the term is used in crank shaft balancing.

(9) Just for Skunk, Some of the problems associated with balancing a V2 90 degree twin with one horizontal cylinder.

This is going to be a very simple "KISS" look at single cylinder crankshaft balancing, based on Phill Irvings book. If anyone thinks I've missed something or wants a more in-depth treatment they are welcome to explore the internet for what they can find. Or dust of data from their own experiments and post it themselves. Then we can all enjoy reading their interpretation of what they have learnt.

.
plus 10.....

For those too lazy
352590352591

F5 Dave

8th March 2023, 19:19

I'm definitely too lazy to exercise blandishment in the right quarters.

Splatter

8th March 2023, 23:20

hello muhr, i've seen your 50cc project while reading some pages back.. have you finally tried it? that looked nasty.

amazing quality. have you also experienced an improvement restricting the internal cross section instead of maintaining the carb exit cross section?
Have you tested different lenghts for the horizontal splitter? i see that this goes all the way from the base to the top of the cusp.
I even saw someone on the dutch forum (deraceheldenvanweleer.nl) hypothesizing a manifold with a continuation of the splitter, but i can't find any feedback on that.

352592352593

The splitter is full length for several reasons. The material used does not work well using supports so it's easier and better quality to print the splitter off the print bed. The area of the splitter is considered when figuring out the total flow area. The cross section is smaller than the carb size. The other reason for the full length cage splitter is that it connects to the splitter in the intake. The intake splitter is curved with the angle of the intake to try and maintain equal flow to the top and bottom of the cage.

Splatter

8th March 2023, 23:34

Your 1/2 moon fillers look really nice , but I now make them with no step up to the petal seal surface on the sides.
This does not apply to the end of the splitter at the reed tips , a right angle step here , on the flow bench lifts the petals higher and faster - resulting in more power on the dyno as well.
Completely counter intuitive to having a triangle divider under the petal tips - but hey its a 2T so reality doesn't match anyone's pre conceived ideas.

Also the splitter is shortened , it starts where the venturi ends , further in from the block entry face.
As we also bias the flow upward toward the piston by having asymmetric reed backups , im about to try moving the front edge of the airfoil splitter downward , too force more flow into the top pair of reed block ports.

Edit - the guys who make the UFO for Mikuni round slides also make a short splitter that divides the carb exit in 1/2 at the engine side of the slide.
This gives way better response at low throttle openings - so the idea could be extended so the reed splitter is continuous right up to the throttle slide - maybe ?

That is an older cage pic. The current cage has no step on the sides. The splitter extends into the intake where it curves with the angle of the intake, but only to the carb boot. Maybe I should try extending it to the slide. This cage has no lip at reed exit and I'm trying to use the case pressure changes to lift and hold open the reeds. Data aq and lap times show it works, but I haven't dyno'd it. I have limited dyno access for the bigger engines. I do have access to a smaller engine dyno and am making one of these to test on a TM R2 just for proof of concept. I'm using soft reeds (30.5) with no stiffeners and they are working well with decent longevity, but we only rev these to 9500.

Frits Overmars

9th March 2023, 00:16

Ok. Looks like I incorrectly calculated the mass to be balanced..... I think they make the same mistake in this description of how to balance a single cylinder crankshaft.Yes they do. First they make it more complicated than it needs to be by working with 100% minus the balance factor instead of simply working with the balance factor itself.
And then they forget that 100% of the small end weight (I should really say mass instead of weight) is already there, hanging from the big end pin.
352597

2 Stroke Stuffings take on finding the existing balance factor. https://youtu.be/0hoE_MuFpPII like this video. From 3:30 Aleks Degnes (mr. 2Stroke Stuffing) says: "Place the weight you added to the con rod on your scale, remember to include the wire hook,
add the con rod weight to hung weight....."
So Aleks is in the clear. Just as well, because he's coming over next week to sniff some two-stroke fumes here. Now I won't need to lecture him :msn-wink:

wobbly

9th March 2023, 11:02

monkeyfumi

9th March 2023, 13:27

The company I used originally to print stuffers used a process of putting the part in a box with I think acetone fumes , this fused the surface into a smooth shiny finish.

The process is Vapour polishing. Be careful if you attempt it though, heating acetone can be dangerous

https://www.youtube.com/watch?v=PegrypbCwag

philou

9th March 2023, 19:15

for crankshaft balancing. an old idea is circulating. on the primary side only insert 2 weights at 90 and 270 degrees from the crank pin to change its vibration frequency in addition to bringing additional inertia

41juergen

9th March 2023, 19:59

just a question maybe mainly to Wob and Frits: when talking about the crankcase compression ratio you use Neels definition (meaning excluded the transfer duct volumes), correct?
cheers
Juergen

philou

9th March 2023, 20:16

in my opinion the software adds them.

they are part of the volume since they communicate directly with

Frits Overmars

9th March 2023, 21:38

when talking about the crankcase compression ratio you use Neels definition (meaning excluded the transfer duct volumes), correct?My definition for crankcase compression ratio is

(volume below the piston dome at TDC) / (volume below the piston dome at TDC - cylinder capacity)

and for me, volume below the piston dome at TDC includes the transfer duct volumes.

Of course this is just a geometric approach. One can argue that crankcase compression can only begin when the inlet is closed, and it ends when the transfers start opening.
So the crankcase compression ratio is just a number. The crankcase volume, in combination with the transfer ducts and above all the exhaust system, determines the engine's pressure fluctuations and resonance frequency.

Splatter

10th March 2023, 00:35

Splatter , yea buddy that cage looks dead cool , now try the step under the reed tips , it worked well for me in the TM R1 and now the R2 as well.
The company I used originally to print stuffers used a process of putting the part in a box with I think acetone fumes , this fused the surface into a smooth shiny finish.

Thanks, I will try a step. I haven't found a chemical solution to smoothing this material. The fiberglass in it might be a problem anyway. It can be sanded, but I'm not that dedicated.

41juergen

10th March 2023, 07:17

My definition for crankcase compression ratio is

(volume below the piston dome at TDC) / (volume below the piston dome at TDC - cylinder capacity)

and for me, volume below the piston dome at TDC includes the transfer duct volumes.

Of course this is just a geometric approach. One can argue that crankcase compression can only begin when the inlet is closed, and it ends when the transfers start opening.
So the crankcase compression ratio is just a number. The crankcase volume, in combination with the transfer ducts and above all the exhaust system, determines the engine's pressure fluctuations and resonance frequency.

Of course that makes sense (as always for your statements :niceone:)
@Wob: when you say, a well designed reed engine should have a CCR of approx 1.3, is that with the transfer volume included?

philou

10th March 2023, 07:43

Have you ever measured a crankcase volume?

41juergen

10th March 2023, 08:15

Have you ever measured a crankcase volume?

Yes, my 3XV / 4DP engine has on the left cylinder 565 ccm and on the right cylinder 510 ccm. On my 125ccm engine with a 4DP cylinder and a spacer plate (to match the 110 mm rod) I got 630 ccm (needs a re-work to get the CCV down)..

wobbly

10th March 2023, 08:27

philou

10th March 2023, 08:43

565 cc = 1,28
510 cc = 1,32
630 cc = 1,25

41juergen

10th March 2023, 08:45

Neels CCR " definition " isn't different than the accepted method.
In the code it automatically removes the duct volumes when you use the calculator.
Thus if the duct or port area changes it will update the CCR.
Alternately you can input the actual CCR as measured manually, and use the Transfers Included radio button.

Re the CCR with reeds - yes 1.3 is the measured ratio with transfers included.

Thank's a lot, so the CCR number shown in Neels code is slightly different to the "accepted method" (what of course is not a problem, plus the benefit of the automatic update when changing the goemetry of the transfers)..

wobbly

10th March 2023, 14:47

Just to be clear , although the code displays a " non standard " output it is using the normal CCR ratio including the transfers when running the sim.

Vannik

10th March 2023, 17:47

Although not 2T but 4T, this video showing the inlet pulses is very insightful. It shows how through an rpm sweep the number of pulses in the inlet system decreases as the rpm value increases and the inlet system goes in and out of tune. What I find interesting is how the last pulse slowly disappears until the next pulse becomes the dominant pulse. This effect is what causes the fluctuations in the torque curve of a 4T engine.

https://www.youtube.com/watch?v=Axro8WRa_dQ

Gradella23

10th March 2023, 20:06

Splatter , yea buddy that cage looks dead cool , now try the step under the reed tips , it worked well for me in the TM R1 and now the R2 as well.
The company I used originally to print stuffers used a process of putting the part in a box with I think acetone fumes , this fused the surface into a smooth shiny finish.

if i understood you're using something like this? more or less angle in relation to the reed surface?

Frits Overmars

10th March 2023, 22:12

My definition for crankcase compression ratio is (volume below the piston dome at TDC) / (volume below the piston dome at TDC - cylinder capacity)
Of course this is just a geometric approach. One can argue that crankcase compression can only begin when the inlet is closed, and it ends when the transfers start opening.
So the crankcase compression ratio is just a number. The crankcase volume, in combination with the transfer ducts and above all the exhaust system, determines the engine's pressure fluctuations and resonance frequency.Come to think of it, I do not use the crankcase compression ratio nor the combustion chamber compression ratio anywhere in my sim.
Instead the code calculates the crankcase and combustion chamber volumes for each crankshaft degree.

For me, these compression ratios are only useful for comparing engines of different displacements. But even then I do not use the Crankcase Compression Ratio CCR.
I use the Crankshaft Volume Ratio CVR between TDC crankcase volume and cylinder displacement.
I find this just as useful in comparing engines with different displacements, but the math is much simpler.
Here is an example, using the Aprilia RSA values of 124,8 cc cylinder displacement and 675 cc TDC crankcase volume.
The Crankcase Compression Ratio CCR is 675 / (675 -124,8) = 1,2268
The Crankcase Volume Ratio CVR is 675 / 124,8 = 5,409
Now suppose we want the same crankcase ratio in a 100 cc engine. So cylinder displacement CC=100, and TDC crankcase volume CV must be calculated.

When using CCR, calculating the required TDC crankcase volume CV goes like this:
CCR = CV / (CV - CC)
CV / (CV - CC) = CCR
CV = CCR x (CV - CC)
CV = CCR x CV - CCR x CC
(CCR -1) x CV = CCR x CC
CV = CCR x CC / (CCR - 1)
now we enter the current vallues:
CV = 1,2268 x 100 / (1,2268 -1)
CV = 122,68 / 0,2268 = 540,9 cc

For those who are still with me, let's repeat the calculation using the Crankcase Volume Ratio CVR. I promise it won't take long.
CVR = CV / CC
CV = CVR x CC
entering the current values:
CV = 5,409 x 100 = 540,9 cc
Piece of cake, isn't it?

ranasada

10th March 2023, 22:53

come to think of it, i do not use the crankcase compression ratio nor the combustion chamber compression ratio anywhere in my sim.
Instead the code calculates the crankcase and combustion chamber volumes for each crankshaft degree.

For me, the ratios are only useful for comparing engines of different displacements. But even there i do not use the crankcase compression ratio ccr.
I use the crankshaft volume ratio cvr between tdc crankcase volume and cylinder displacement.
I find this just as useful in comparing engines with different displacements, but the math is much simpler.
Here is an example, using the aprilia rsa values of 124,8 cc cylinder displacement and 675 cc tdc crankcase volume.
The crankcase compression ratio ccr is 675 / (675 -124,8) = 1,2268
the crankcase volume ratio cvr is 675 / 124,8 = 5,409

now suppose we want the same ratio in a 100 cc engine. So cylinder displacement cc=100 and crankcase volume cv must be calculated.
When using ccr, calculating the required tdc crankcase volume cv goes like this:
Ccr = cv / (cv - cc)
cv / (cv - cc) = ccr
cv = ccr x (cv - cc)
cv = ccr x cv - ccr x cc
(ccr -1) x cv = ccr x cc
cv = ccr x cc / (ccr - 1)
now we enter the current vallues:
Cv = 1,2268 x 100 / (1,2268 -1)
cv = 122,68 / 0,2268 = 540,9 cc

for those who are still with me, let's repeat the calculation using the crankcase volume ratio cvr. I promise it won't take long.
Cvr = cv / cc
cv = cvr x cc
entering the current values:
Cv = 5,409 x 100 = 540,9 cc
piece of cake, isn't it?

x : 100 = 675 : 124,8
more simple

Frits Overmars

10th March 2023, 23:31

x : 100 = 675 : 124,8
more simpleIf you say so Ranasada :D.
But something else puzzles me.
What magic did you use to convert most uppercase letters in my above quoted post to lowercase? Only some capital letters at the beginning of sentences have survived.
A conversion like this doesn't exactly make the formulas any easier to read. So why ?
Initially I suspected the quoting algorithm of the forum to be the culprit. But after looking at a number of other quotes, the forum seems to be innocent...

ranasada

11th March 2023, 03:44

If you say so Ranasada :D.
But something else puzzles me.
What magic did you use to convert most uppercase letters in my above quoted post to lowercase? Only some capital letters at the beginning of sentences have survived.
A conversion like this doesn't exactly make the formulas any easier to read. So why ?
Initially I suspected the quoting algorithm of the forum to be the culprit. But after looking at a number of other quotes, the forum seems to be innocent...

not my fault...
maybe for having modified my post?

Grumph

11th March 2023, 06:00

Of course that makes sense (as always for your statements :niceone:)
@Wob: when you say, a well designed reed engine should have a CCR of approx 1.3, is that with the transfer volume included?

Further to this : What is left in or removed from the inlet tract when reed CCR is measured ? Everything out and an external plate across the cavity ? Leave the reeds and stuffer in and put a plug in the carb mount rubber ?

What's accepted practice ?

41juergen

11th March 2023, 07:30

Further to this : What is left in or removed from the inlet tract when reed CCR is measured ? Everything out and an external plate across the cavity ? Leave the reeds and stuffer in and put a plug in the carb mount rubber ?

What's accepted practice ?

I use tape to make sure the reeds are closed and no oil can escape through the reed plates...and have the reed cage installed. And also make sure no oil can escape through the aux ex ports or power valves.

wobbly

11th March 2023, 08:56

Yep , tape up around the reeds and use plasticine around the piston edges.
Re the reed block , make the spitter as thin as you can under the petal tips with a right angle step as you had on the 1/2 moon sides previously.

Grumph

13th March 2023, 15:43

Had a visit today from a pensioner who lives around the corner. Quite liked the latest project. Plenty of ideas and memories discussed.

wobbly

13th March 2023, 16:28

That pensioner has forgotten more than most of us have learned yet.

mprestwi

15th March 2023, 15:19

Wobbly sir, with this recent discussion around stepped reed cages you had mentioned in the past that you had other examples where non-aerodynamic applications went against expected results and made more hp. Any chance you could elaborate on this?

wobbly

16th March 2023, 10:02

Only ones that come to mind are the 2mm smaller venturi inside the reed block , that is smaller then the already too small carb , and , not having a slow taper exiting from a stinger insert.
This should have a 45* step , and reduces the effect of the waves bouncing up and down the stinger , from the atmospheric end ( the muffler entry in the case of a KZ ).
These can interfere with the main positive reflection off the reverse cone , as the + wave bouncing off the open ended pipe reverses sign and travels back to the stinger entry.

Edit - the other I just thought of was the geometry of what Neels refers to as the " contact discontinuity " at the interface of the spigot transition from round ( at the header start ) to oval ( with ears in a 3 port ).
Where this very tapered transition meets the cylinder duct , the best geometry is to have the taper on the floor of the spigot colinear with the duct floor.
This in effect means the spigot centerline is moved up , such that most of the transition taper is in the roof.

Again counter intuitive , in that you would think " helping " the return wave front along the roof would give the best return pulse amplitude , when its needed , as the piston is nearing closure.
Not so , I built 3 spigots in CAD and had them CNC cut , one with the male spigot centered on the duct exit , one with it lowered 3mm , and one with it lifted 3mm.
Guess which one was 3Hp up at 14500 over the worst.

wobbly

16th March 2023, 11:06

Sir - that assumes I am a gentleman , this proves otherwise.

jfn2

16th March 2023, 12:18

Wobbly
A few pages back you were talking about reeds and UFO's and other air directional devices. Back in the late nineties, early 2000's a couple of us guy's were on the MacDizzy forum talking about how to get better air flow through the carburetor and we were working with placing a divider (UFO) in the venturi and that instead of using a straight center bar that using an up-side down air foil worked much better. We also found that the placement of the horiz bar didn't necessarily have to be in the middle of the bore. In fact if it was placed about 10mm above the top of the needle jet spout it worked better. Of course it would (I think) depend on the size of the carb. I was using it with a Mikuni 38mm AS carb. We placed this UFO right up to the carb throttle valve. We also tried using 2 bars, one in the middle of the bore and one down a little bit. We also used a honeycombed alumn waffer in the front of the carb air horn. In fact we milled a slot for it to set into, at the very front of the air horn. Both of these together worked very well. But we didn't have a dyno or flowbench back then to test with. I was building bikes to run in the GNCC on the east coast of the US. Made for very good throttle response. And we could run a size bigger carb. Why couldn't a airfoil be used in the reed cages? Or a bent airfoil? Also, I used to work for an asphalt co. that made asphalt for road use and in their plants they use a very big fan and a 48" duct work to clean and dry the stone. There is a big cost associated with running a big electric motor to turn the fan. To make it more efficient the accepted practice is to put dividers in the air duct in the, usually, 90 deg turn into the fan. These dividers are bent at 2- 45 degs and if only using one divider placed in the middle of the flow, or if using 2, are placed at the 1/3 and 2/3 part of the flow. It works. Why couldn't this be done in the intake manifold to the reed cage?
Also wanted to ask you about the rear cone of the expansion chamber pipe. What %'s do you suggest to use for a 2 and 3 part cone? Thanks Jeff

wobbly

16th March 2023, 13:22

The thingy called a UFO ( Ultimate Flow Optimizer ) from Thunder Products goes into the bottom of a Mikuni round slide.
What you are talking about is an even bigger marketing masturbation called a Quad Flow Torque Wing.
Both work , the first to increase the idle and transition signal at low throttle openings , and to reduce turbulence from the slide cavity at WOT.
The second , effectively divides the carb in 1/2 at part throttle , again , increasing response in a carb with shit part throttle response.
This is all fine if you want ( or have to ) use Mikuni round slides - not needed at all in a flat slide of recent manufacture.

The next idea of a wing cross section in the middle of a reed block entry has been used forever by VeeForce , only because that is the only way to build one.
I have tried all manner of splitters/wings etc in the KZ application where a small carb feeds a big reed block.

After 3D printing more blocks than my lifetimes hot meals total , I have given up.
A simple horizontal divider made from carbon reed material or thin sheet metal works the best with the stepped end under the reed tips.
As usual being a 2T designed to disappoint at every opportunity , I got real clever and made a wing section with the forward end tipped down to help bias the flow upwards as it exits into the case chamber.
We already do this with much differing petal and backup stiffness on the top and bottom reed sets.
Did it work - of course not - picture of failure number 9576 enclosed.

Lastly the rear cone geometry.
Forget a 2 cone - its not even in the same area code as a 3 cone.
What the 3 cone concept allows is you can ameliorate the annoying trend , that as you increase the rear cone included angle , you increase the upper front side and peak power , at the direct expense of overev
bandwidth.

Each engine application is usually highly specific re overev requirement - look at a drag race CVT vs a Roadrace bike.
One needs NO overev , the other as much as can be generated without sacrificing too much front side and peak.
There are no guidelines to get you in the ballpark with this.

Suffice to say the length and angle of the last cone sets the peak and overev potential , and the middle cone sets the front side bandwidth.
The only way is to run a shit ton of EM2T sims , with maybe a little guidance from the R1 pipe design I have published - but that is super end use specific design , of no use whatever in other applications.

mprestwi

16th March 2023, 16:49

Thanks for the reply Wobbly. No disrespect intended.

wobbly

16th March 2023, 17:14

None taken , just been dying to show the sign I put up in the kitchen when the mother in law arrives.

jfn2

17th March 2023, 01:03

Thanks Wobbly for the reply.
As they say.... No rest for the weary.
I do appreciate the time taken. And the info.
Jeff

Pagi

17th March 2023, 05:41

Wobbly, sorry to ask again a question about reeds ( my english is poor....and google translate too!!!!), but what do you mine by " the divider with a stepped end under the reeds tips" ?....do you have a drawing or picture....:yes:

Condyn

17th March 2023, 06:05

Edit - the other I just thought of was the geometry of what Neels refers to as the " contact discontinuity " at the interface of the spigot transition from round ( at the header start ) to oval ( with ears in a 3 port ).
Where this very tapered transition meets the cylinder duct , the best geometry is to have the taper on the floor of the spigot colinear with the duct floor.
This in effect means the spigot centerline is moved up , such that most of the transition taper is in the roof.

Again counter intuitive , in that you would think " helping " the return wave front along the roof would give the best return pulse amplitude , when its needed , as the piston is nearing closure.
Not so , I built 3 spigots in CAD and had them CNC cut , one with the male spigot centered on the duct exit , one with it lowered 3mm , and one with it lifted 3mm.
Guess which one was 3Hp up at 14500 over the worst.

The complexity of that transition makes my brain hurt, the counter-intuitive aspect as well. Does anyone have any theories on why the roof taper works better?

Pagi

17th March 2023, 06:24

Thanks Wobbly for the reply.
As they say.... No rest for the weary.
I do appreciate the time taken. And the info.
Jeff

Wobbly, sorry to ask again a question about reeds ( my english is poor....and google translate too!!!!), but what do you mine by " the divider with a stepped end under the reeds tips" ?....do you have a drawing or picture....:yes:

sorry ( my poor english again... i just see my error in my question.....i need to return at school:rolleyes ), but what do you mean with a step at the end of the divider under the reeds tips.

wobbly

17th March 2023, 08:53

Super secret squirrel shit this.
Seems to work well with the asymmetric reed tensions biasing flow upward under the piston.

wobbly

22nd March 2023, 14:16

TZ350

22nd March 2023, 15:47

352660 Dane Rowe

https://rocinantemecanico.blogspot.com/2012/04/surprise-guest-my-interview-with-dane.html

352661 TZ750. One of Rudi and Danes chairs.

Pagi

23rd March 2023, 00:05

The only logic I can muster re the transition taper , is that when the floor is colinear the returning wave front easily ramps up the reverse ski jump and arrives early , with maximum
amplitude whilst the piston is still low over the port - just after BDC.
Thus more of the finite plugging energy is seen at the port face , with the least interference from the rising piston.

Does your exhaust duct is still relevant ? ( except perhaps the percentage of surface of the exit of the conduit as you specify it in a previous message). Have you or can you give us more details on the dimensions of the reverse ski jump ?(length, angle)352667

Vannik

23rd March 2023, 01:59

Does your exhaust duct is still relevant ? ( except perhaps the percentage of surface of the exit of the conduit as you specify it in a previous message). Have you or can you give us more details on the dimensions of the reverse ski jump ?(length, angle)

Latest picture:

352676

Wobbly can comment further.

Pagi

23rd March 2023, 02:27

Latest picture:

352676

Wobbly can comment further.

thanks Vannik....:yes:

Kickaha

23rd March 2023, 06:15

352661 TZ750. One of Rudi and Danes chairs.

TZ750 ? are you sure about that because I didn't know Yamaha made a triple :msn-wink:

TZ350

23rd March 2023, 07:24

TZ750 ? are you sure about that because I didn't know Yamaha made a triple :msn-wink:

You are right. After a better look I now think it was a TZ350 with an extra cylinder added.

https://www.flickr.com/photos/teamheronsuzuki/albums/72157608302512322/with/2966770061/

The days when someone with talent, energy and a modest workshop could build something to compete on the world stage with.

wobbly

23rd March 2023, 07:31

Only thing to add is the ideal port exit roof angle has been found to be 25* , and the ski jump floor should be perpendicular to the bore then turn down , such
that the main duct smallest CSA is coincident with the increase in area created by the Auxiliary ducts entering from each side.

katinas

23rd March 2023, 08:07

You are right. After a better look I now think it was a TZ350 with an extra cylinder added.

https://www.flickr.com/photos/teamheronsuzuki/albums/72157608302512322/with/2966770061/

The days when someone with talent, energy and a modest workshop could build something to compete on the world stage with.

Very nice photo album collection. And nice short message from Dane's daughter Karin Hasen: Dane Rowe (Kurth) is my mum. She‘s not on facebook (she thinks facebook is evil) she‘s still alive and kicking and says „thank you for the nice comments“ and „I am very busy with ancient coins these days“

Add Yamaha TZ 525 triple. Maybe have been ridden by Gene Remero.

Pagi

23rd March 2023, 08:11

Wobbly, thank you for the clarification:niceone:

Kickaha

23rd March 2023, 17:01

You are right. After a better look I now think it was a TZ350 with an extra cylinder added.

Apparently one of the first people to do it, I was already well aware of Rudy and Dane, he built some interesting stuff

Condyn

24th March 2023, 06:37

**Edited** So just to clarify, the whole entire exhaust duct (up until the tapered transition) is supposed to be 75% of the total window area? Before seeing this new drawing, I always thought the duct was supposed to taper down from port window face to the cylinder exit, and then taper back up into 1:1 at 2x bore from the window. Now it seems that the entire point of the initial flat floor, leading into the ski jump (with roof at 25 degrees) and later the convex floor at aux entry into the duct is to maintain the 75% all the way down. Pardon my ignorance of gas dynamics, but this makes more sense, as the main purpose of this special duct is thought to be because it keeps the fresh slug more dense via the smaller surface contact.

wobbly

24th March 2023, 07:53

Completely wrong end of the stick .
Nowhere ever , is it mentioned that the duct at any point is @ 75% CSA except at the exit face of the cylinder.
The floor is a convex ski jump with the main ducts CSA area reduced down from 100% at the port face, to be a minimum at the Aux duct intersection point , then the combined Main / Aux area is gradually reduced down
toward the exit such that it ends at 75%.

And who said that the intrinsic idea was to reduce the duct surface contact area of the retained return slug of overscavenged A/F mixture - not me.
Thats an assumption , and they are usually errors waiting to be revealed.
I have stated dozens of times the whole concept was based on achieving 0.8 Mach at the duct exit point , then have a highly tapered transition out to the 100% CSA header entry point.
This highly tapered divergent section of course reflects a changed sign wave ( negative ) very early back to the port , increasing scavenging whilst the piston controlled port is still open.
Just another piece of the puzzle.

But the lengths shown in all cases I have seen , are longer than the norm - giving more surface area , not less.
And of course , I have done the dyno experiment ( as proof of concept ) by lengthening the cylinder duct using a spacer with slots to allow water to cool right out to the spigot face , and the 1.5 X bore
( based on square engines ) makes more power , combined with a modified pipe to keep the Lt the same.
And EngMod agrees - funny that.

Condyn

24th March 2023, 08:34

Thank you for setting me straight. I misinterpreted something I had read.

F5 Dave

24th March 2023, 12:14

I don't want to derail this conversation as certainly back in the day I ruined my good 50 cylinder by making a nice taper exhaust port before I had any clue of this.

But to hark back to the 250MX long stroke advantage just for a segway. I've got back into Trials. And bothered to read the spec of my new bike. Which is industry standard.

250cc.
Bore 72.5. Stroke 60mm! So the other way around, and these are not high revving (or high power) engines.

The 280s and 300s keep the same stroke. Ok these are low BMEP engines with hardly a conventional pipe, but it seems odd that they would be so short stroked. We Dynod a Beta 250 20 years ago and it puffed out about 13hp. But was dynamite to ride at least for a beginner.

Pursang

24th March 2023, 15:37

I've got back into Trials. And bothered to read the spec of my new bike. Which is industry standard.

250cc. Bore 72.5. Stroke 60mm! So the other way around, and these are not high revving (or high power) engines.

The 280s and 300s keep the same stroke. Ok these are low BMEP engines with hardly a conventional pipe, but it seems odd that they would be so short stroked.

It was 'Industry Standard in the Sixties and Seventies too! Bultaco Sherpa T. :niceone:

The stroke was inherited from the original 125cc engines.

It was the same in Metralla road bikes, Pursang MX and TSS roadracers. The 60mm stroke was used in Bultaco engines up to 325cc.

Grumph

24th March 2023, 18:32

It was 'Industry Standard in the Sixties and Seventies too! Bultaco Sherpa T. :niceone:

The stroke was inherited from the original 125cc engines.

It was the same in Metralla road bikes, Pursang MX and TSS roadracers. The 60mm stroke was used in Bultaco engines up to 325cc.

Once you'v e got a boring machine set up to do the job on that center, the design department tends to use what is already there. If that happens to be your only boring machine as I'm sure was the case with some of the smaller outfits, then everything gets designed around that. You may also have a stock of crank forgings of the size needed for that stroke. Got to pay for that tooling.

I read in one of Phil Irving's books that the stroke of the Velocette Roarer came about because the factory had just taken delivery of a Heald Bore-matic machine.
One of the settings was fairly close to what they had wanted so that's what got used.

husaberg

24th March 2023, 18:53

Once you'v e got a boring machine set up to do the job on that center, the design department tends to use what is already there. If that happens to be your only boring machine as I'm sure was the case with some of the smaller outfits, then everything gets designed around that. You may also have a stock of crank forgings of the size needed for that stroke. Got to pay for that tooling.

I read in one of Phil Irving's books that the stroke of the Velocette Roarer came about because the factory had just taken delivery of a Heald Bore-matic machine.
One of the settings was fairly close to what they had wanted so that's what got used.

One of the mags i was reading that came from you the other day had the Irving twin (model O of similar concept to the Roarer that used a heap of parts from other road velos, i never knew he designed the adjustable top shock mount either but makes sense in hinsight it was him.

Pursang

24th March 2023, 21:21

https://images1.bonhams.com/image?src=Images/live/2022-03/21/25187523-1-7.jpg

Love to give one of these some two stroke hindsight 'updates'.

Frits Overmars

24th March 2023, 22:10

Love to give one of these some two stroke hindsight 'updates'.Don't.
I was approached in a similar situation by a friend who wished to 'update' his factory MZ racers. They were not very reliable 55 years ago with 30 hp per cylinder. They really don't need the extra 10 hp per cylinder that we could add today without much effort.
At the time, the factory constructed a new eight-speed gearbox three years in a row. I suppose they did not do that because the first version was so reliable. And I'm afraid the latest version wasn't either.
Then there was the engine cooling, or rather the lack of it: thermosyphon cooling, which I usually refer to as thermosyphilis cooling. A coolant temperature of 90°C was not unusual so it shouldn't surprise you that the pistons had an extremely short life.
Leave those old machines in their former glory. They are suitable for demos, not for competition any more.
352690

Pursang

25th March 2023, 00:31

Don't.352690

Don't..try this at home! Thank you Frits, for the sane, sensible, experienced, voice of reason.

I do understand and accept the need to not over-challenge the vintage mechanicals... but....with a shed full of 50 -80 y/o projects and now, finally, some time to play with them, the path of the factory faithful restorer (cough, spit) is not for me!

I appreciate every piece of of technical advice, tips, warnings that have been shared here. Thank-you everyone.
May the wisdom contained in this thread and the others, be available forever.

Cheers, Daryl.

katinas

25th March 2023, 08:54

Another thermo-symphony, Dave Simmonds Kawasaki 125 and Guy, what a happy guy. https://www.paddock-gp.com/people-a-quoi-marche-guy-coulon/

Grumph

26th March 2023, 13:31

husaberg

26th March 2023, 14:12

Here's an unusual question for the hive mind.
I find myself in posession of three cylinder heads for the current project.

A genuine toroid shape. With squish band. Plug is dropped into the chamber slightly as it should be.
A head with squish and a central cylindrical shape with basically square edges and a flat roof.
A head with the current conventional squish and part spherical chamber

Initially I'll be aiming for around 12:1 compression. No significant difference in fin area between them. All are central plug.

Which is the favourite ?
i knbow you were thinking reed but
just saying

http://progress-is-fine.blogspot.com/2019/10/another-world-beater-1967-fhs-merlin.html
the later one which peter crew raced was as far as i remember starmaker based.
paging neil...
352702352703

Grumph

26th March 2023, 14:38

i knbow you were thinking reed but
just saying

the later one which peter crew raced was as far as i remember starmaker based.
paging neil...

No I've seen the articles on them and none were Starmaker based. You're the one with a binful of Villiers crankcases - and a Greeves barrel, lol.
There actually was a disc valve converted Villiers here in ChCh late 60's. Built in the University workshops. Came out 2 or 3 times and disappeared.
I suspect the lack of the now common Japanese rod kits and low expansion pistons meant it was prone to meltdowns.

F5 Dave

26th March 2023, 15:40

Well the answer is quite simple.

On the middle cylinder run the Toroidal, on the left cylinder the conventional, on the right cylinder. . Well you get the picture.

Glad to be of help.

Grumph

26th March 2023, 15:47

Well the answer is quite simple.

On the middle cylinder run the Toroidal, on the left cylinder the conventional, on the right cylinder. . Well you get the picture.

Glad to be of help.

Thanks Dave. i don't quite have the bits to build a triple. I will admit to feeling slightly inadequate there, I'm aware there are people on here who would not let that stop them.

F5 Dave

26th March 2023, 15:52

More power to you that you are still spinning spanners. Presumably in the laundry.

husaberg

26th March 2023, 16:49

No I've seen the articles on them and none were Starmaker based. You're the one with a binful of Villiers crankcases - and a Greeves barrel, lol.
There actually was a disc valve converted Villiers here in ChCh late 60's. Built in the University workshops. Came out 2 or 3 times and disappeared.
I suspect the lack of the now common Japanese rod kits and low expansion pistons meant it was prone to meltdowns.

I will have to look for it in the pile unless someone has it handy?

My recollection was the original was very much greeves based but the later looked more starmaker like and even had one o f those 6 speeds added.
Crew ended up with a few incl a 350 but eiter sold or stopped racing them about 10 years ago.
i found a FB page that looks to be hois but it only has TZ and Manx plus nothering before 2012.

352706

wobbly

26th March 2023, 19:50

A toroid will win hands down , but that only " works " when pushing the envelope , and at 12:1 that hardly defines pushing anything.

Grumph

27th March 2023, 05:55

A toroid will win hands down , but that only " works " when pushing the envelope , and at 12:1 that hardly defines pushing anything.

Thanks Wob. Refer Frit's post 24/3 above. It's a 60's design so I'm pushing the probabilities. I'll admit once i found there was an original toroid head there the temptation to use it became nearly irresistable.

jfn2

27th March 2023, 08:13

What is the design importance of the bottom of the exhaust port, ( from BDC to transfer opening)? And if there is not much how do you calculate for the exhaust duct design and pipe openings? As in Wobbly's exhaust port design or port area calc's? And if it is important, what shape would you use, or size? I am speaking of the port opening or face area.

wobbly

27th March 2023, 08:40

The floor of the main duct was tested by Jan @ 3mm ABDC and a gain realized.
I also have done this as its the only proven example to follow , but is tempered by the fact it only works if the blowdown STA is fully developed.
The exit area I used was calculated @ 75% assuming the port was at BDC , as using the smaller area gave a Mach number closer to 0.9 and this lost power in the sim.
The higher floor also makes the main duct pinch point smaller where the Aux ducts enter from the side , and this may be a factor in the designs success , as well as the possible reduction in
short circuiting from the A port front wall.

jfn2

28th March 2023, 14:38

Wobbly
Thank you for the reply. What I'm trying to ask is about the shape of the bottom of the exhaust. On engines I've seen with a single exhaust port the bottom has a elliptical bottom edge and the edges raise straight or with a slight elliptical shape to the top edge where the widest point is the middle of the vertical lines. In the engines with a winged exhaust port the bottom is not as wide and the sides are slanted towards the widest point, or to a point level with the top of the transfer ports, then flare out into the wing shape. Sort of 'V' shaped. If I were to make a replacement sleeve and were to cut the port openings into it what shape would be the best? The 'V' shape, I think, would help with short circuiting. That opens another question, where does the most short circuiting occur from, the top of the 'A' port or the back side of the 'A' port? With Frits' suggestion about the directional angle of the 'A' port I would guess the back side. What do you think?