ESE's works engine tuner

[90GTVert]

I've used a main jet extender in desperation on long tracks before I realised that my carbs internal drilling feeding the float needle was actually the restriction.

I think the extender is a good thing. Not saying there's not other problem, but the jet seems way too high without it. It should be a little closer in the standard bowl, but the setup puts the carb at an angle and I like keeping more fuel in there.

Here's the standard setup with a larger clear bowl :


My homemade extender :

But let's start at the beginning, or rather the end. What size was your stinger? As you raise in power it needs to increase. The perfect size to 20hp may kill a 25hp bike.

I will get back to you on that. It's the pipe made for the setup. It's a Malossi RC-One (MHR 94cc) with the Malossi MHR 94cc exhaust. Same pipe that the two other people I have rode with using the same setup and 34mm PWKs use. Pretty much everything on their bikes is the same setup, but it all works easily for them with no drama.

Oh and sorry for the ribbing at the start, but to be fair we did have a marvelous time.

LOL Glad you guys had a good time. No worries. I'm very appreciative for all the help. I can't tell you how frustrating it has been to be the guy that actually has a spreadsheet of 600 1/8 mile runs with EGT, RPM, temp and DA (every run with this engine for the last 2 years) and to try my best to be thoughtful and thorough and fail and fail while watching everyone else pop a 34mm on theirs, swap a few jets, use the popular needles (JJL/JJK) and live happily ever after. I thought for sure if I bought the exact carb that my friend uses on his, with his setup, I'd have little to do but fine tune and wheelie off into the sunset. WRONG. For most people that I talk to, this goes beyond any normal level of tuning and they are lost like me or I have some claimed pros giving me advice, but on YouTube comments and such you really have no idea who is suggesting things and most of the time another "pro" will contradict what they say.

One more variable to throw in here - main jet sizing.
I have found it is absolutely necessary to use a set of pin gauges to size the main jets , the Dellorto ones are all over the place and are nowhere near what is marked on them.
Keihin QC may be OK, but I trust no one and would check anyway.
You can use an injector test rig as well to make flow comparisons.

I didn't have anything for that size, but I did work a brass rod down as a sort of go/no-go gauge after the 210 jet was an outlier in the temp and RPM data collected. It was indeed somewhere between the 205 and 215, which makes it even more strange to me that it acted richer than the 215. It was about 200RPM lower on average during acceleration runs with 70-80F cooler EGT than the jets on either side of it. I don't have access to any flow equipment, but I did intend to take a close look and see if there's anything obvious like taper on an end or something different. I forgot about it after the engine failed again and I was just trying to source a piston.

[F5 Dave]

An extender is only any good if your float level is too low. It just buys you some more time, but the real problem is that the float level should not drop.

Running on a big angle, well, maybe. . But I'd want to get the carb a bit more level and get it working right then consider more angle if you free the demons.

So PWK is keihin. Oko and several others make clones. Their cloning machine is a bit variable. Some people get away with it , but I've spent a bit of time at the track helping the less fortunate ones. Usually with float and float jet issues I'd never really seen on my Japanese keihins.

[Pursang]

Looks like a blob of water, in the well at the bottom of the jet pickup! That would move about under accell. & decell., and restrict the fuel flow through the jet!

[41juergen]

Thanks for the links and advice. Believe it or not, I'm not a total noob to two-strokes. I'm a scooter guy that's been messing with them for roughly 20 years. Predominantly bolt-on and pretty mild porting and I do cut trenches in cases, which was how I was introduced to Wobbly's presence online. I've read Bell and Jennings long ago and got very little into Blair before realizing it was over my head, and I know of the legends like some of the present company. I have had this and Pit Lane bookmarked for quite some time now, but I run my own scooter forum and YouTube in addition to some locals that are looking for help and by the time I get done reading and replying to other people's problems I make excuses and don't want to spend more time reading daunting page counts worth of undoubtedly valuable info. I used to google search this thread and could find some answers easier that way, but that no longer seems to work.

Hopefully this gets posted soon, along with my previous reply. I've moderated and ran scooter forums for nearly 20 years now so it's really weird being the new guy awaiting approval.

Have you ever been on a dyno with Lambda sensor measurement? Could give you a quick feedback whats going on with fueling over the rpm range...

[Betaversio]

Hello everyone!

Dear guru's, I would like to ask question.

We have been discussing with friends about pipe max diameter. It's quite clear that narrow pipe just produces weak pulses and when you increase diameter, effect of pipe increases. At some point, more diameter is not better any more, and question is, what is mechanism what defines that max diameter in particular engine with current scavenging?

There have been few ideas in thought. First, and most obivious, what have been around since these old books discussed above, is that too deep of a diffuser will take too much energy away from main exhaust pulse, and big positive return wave is too weak to be effective enough.

Second idea is that during scavenging, when exhaust pulse is about to start scavenging, too big of a pressure difference between exhaust port and part in cylinder near it without (at this point) proper flow from transfer port makes short circuit happening. Important thing in this case is that pressure difference comes from exhaust side and transfer flow is not yet fast enough to keep up scavenging loop working. If it was given more time, then there would not be any difference between making "vacuum" with exhaust or making "boost" with crankcase.

Third idea is about the same than last one, but more simpler way. It is just "too high flow velocity is too turbulent to make proper scavenging".

Second and third idea tells you, that in many cases bad scavenging needs weak pipe, becouse bad scavenging is so easy to mix up to be even worse. Of course, there is also cases when scavenging is so bad that anything you can flush through is just better. I have tried this one with one cylinder with horrible transfers.

Any thoughts? Thank you

P.S. this have been wonderful topic to read

[koenich]

I can't tell you how frustrating it has been to be the guy that actually has a spreadsheet of 600 1/8 mile runs with EGT, RPM, temp and DA (every run with this engine for the last 2 years) and to try my best to be thoughtful and thorough and fail and fail while watching everyone else pop a 34mm on theirs, swap a few jets, use the popular needles (JJL/JJK) and live happily ever after. I thought for sure if I bought the exact carb that my friend uses on his, with his setup, I'd have little to do but fine tune and wheelie off into the sunset. WRONG. For most people that I talk to, this goes beyond any normal level of tuning and they are lost like me or I have some claimed pros giving me advice...

Just saw the Polini logo on the carb - are you sure it's not an underlying carb issue? I was at a similar point than you, although put in less effort.

Had my go with 3 different Polini PWK30 carbs and no matter which jets, needle or slide I couldn't get one of them to run even close to a good tune. After wasting a full day changing parts I questioned my own abilities or if the engine is the problem, so to rule out the engine being the problem switched back to the trusty OKO PWK30 and tada - bike ran like a champ again. Same story with a few friends Polini PWKs...

[Condyn]

I find the whole idea of “short circuiting with a big sucky pipe” to be misleading to some extent. Yes, there is a limit, but even on mediocre “developed” elevator shaft transfer geometry engines, I have had very very good results with increasing cone angles to what many would consider “ too steep “. Would I add a 4+ degree header that is 25 percent to this? No, but there is way more to the equation than just belly diameter.

[Frits Overmars]

Have you ever been on a dyno with Lambda sensor measurement? Could give you a quick feedback whats going on with fueling over the rpm range...

..........

[wobbly]

I believe the first documented case of a belly diameter that was "too big " was documented by Erv Kanemoto , and he described its effect as " overscavenging".
He was working on an early TZ750 to be run at Daytona , and designed the largest pipes that could actually be fitted on the bike without issues of grounding them.
No matter what he did with these 110 diameter belly pipes , they always made less power.

This engine was very oversquare , had elevator shaft transfer ducts , and the A port front wall was not angled back well past bore center.
Thus the more efficient diffuser was, very early in the cycle , pulling mixture around the corner from the A port , as it had limited directional control or rearward bias.
As it turned out , this was confirmed later in the same series of cylinders with the TZ350 , and it wasnt until the 6 port G series , that 110 diameter pipes would be capable of
increased power output, and were fitted to the bike from the factory.

Regarding the use of available wave energy , this I believe is less of an issue , as a fatter pipe can be fitted with steeper rear cones , that does not act in the normal manner of severely restricting overev power.
A 24* included angle rear on a 110 pipe acts steep , on a 120 diameter belly it acts very shallow in its powerband effect, with more front side , less peak , and more overev.
Lastly , the idea of increasing scavenging via diffuser action efficiency has a very real and dramatic effect on power production.
But so the called " pumping " action of the case in being able to increase the Delivery Ratio is a very minor element in the overall engine efficiency , as increasing the case volume obviously
reduces this effect , but this is more than offset by that cases extra volume being available for an increased diffuser effect to act upon , thus increasing Scavenging Efficiency as well as the Delivery Ratio..

Moving along in development history we now have very narrow range of efficient case volume ratio's that work with the much better scavenging regimes as seen in later model designs.
For a reed low to mid 1.3 ratio , and for a rotary valve low to mid 1.2 ratio, and yes , the crappyer the duct geometry is , the smaller case , then works with a less efficient diffuser and that pairing
shows more power potential overall.
As always no free lunch , due to the action of transfer duct geometry being the final arbiter of what combination of case volume /pipe volume is the most synergistic.

[Condyn]

Wob, Great information, especially interesting is your explanation regarding the use of available wave energy, which has opened my eyes up even more to the fact that cone angles only paint part of the picture.

I also read the same Kevin Cameron article regarding the fat pipes, and Erv Kanemotos findings. This was early on in my pipe building journey, and because I had a crappy 70s Yamaha 4 port turd, I thought 110mm could be the limit. Slowly I started breaking this trend by pumping up the pipes. The A port has in fact been massaged for more rearward bias, and I do see the importance.
We are currently at 130mm in the belly section with more power across the entire curve.

With modern developed transfer geometry, or at least semi modern developed, Do you see any racing application where a smaller case than 1.2-1.3 could benefit on the track?

[Betaversio]

This belly diameter question came into think when pipes designed using EngMod did not work expectedly. Simulation calibration was done accordingly, and good pipe found in simulator was always "the fatter the better". Some of these pipes were built (in many engines, some bad and some good) and results were similar. Fat pipes did not rev as these supposed to be - and did not make power.

Then we started to think about this, and came into conclusion that becouse simu and scavenging model is 1D, it cannot tell when pipe is about to be too fat. You can of course change scavenging parameters, but it does not help when you don't know how much and in wich point to change them. So, we did not realize that and made too fat pipes.

At some point I was using kind of a "free search script" on EngMod to look after good pipe and it gave as result 140mm fat 660mm long (short..) pipe for 80cc engine that should rev around 12.5-13k peak power due to lack of time-area. In EngMod it gave great power. This did help when figuring out scavenging model problem as that pipe was so silly that only M.F. would believe in it

[wobbly]

The 1.3 for reeds seems to be a hard fact. I tested a TM KZ10B model by machining 5mm off the reed block mounting face on the case.
So this gave alot smaller volume up around 1.32 - result , it lost about 1.5 Hp almost everywhere , and I tried softer and heavier reed backups and rejetted to achieve the same egt.
Net result - NFG , and no the reeds were not masking the boost port entry.
So next I added a 10mm spacer , this gave from memory about 1.285 CCR - again a power loss of closer to 2 Hp , but by using softer petals I got most of the front side back.
But by going a tiny bit softer by grinding small 1/2 moon cutouts just in front of the reed clamp plate , it absolutely shat it pants and all the peak and overev disappeared.

In the rotary valve engine I have tested ( in EngMod) a 110cc bucket race engine that TeeZee was working on , and he had machined off the flywheel OD and fitted a long rod.
This gave a 1.2 ratio , and my first reaction was , thats way too big as the Aprilia had around 1.24.
Well dream on - it lost a heap of power everywhere in the sim when I made it that smaller figure.
In the real engine , it would also have benefitted from less parasitic flywheel drag as well , a win /win.

Re the pipe wave energy usage. When I did the TM R1 pipe design the sim really liked a super steep last diffuser cone on a 134 Dia entry into the mid section.
This pumped up the front side dramatically , especially when I went under the 66% diffuser length guideline - longer made less power everywhere.
But going above 30* included on the rear cone did not seem to work as expected , so I reduced the mid exit to 132 and put in a huge series of 3 cones , with progressively increasing angles, as per the Aprilia layout.
This ended up with 28/31/33* angles and made more power everywhere , but I was skeptical and built and tested 3 different rear cone variations , including the 30* single angle.
Well , bow down to Uncle Neels , the sim was absolutely on the money.

[Condyn]

In the Sim, with your level of understanding and experience, I would be interested in what you observe at 600-800 below peak when shrinking the CCR to around 1.4 vs 1.3 on a reed motor.

[wobbly]

Man , the easiest question this week. 20 secs later here is the answer.
Wouldn't go there if I was you.
I could get some front side back with thin petals , but then they would go spastic up top , been there done that - no free lunch today.

[Condyn]

Thank you for taking the time.