ESE's works engine tuner

[F5 Dave]

I had some bored out 28s on my RZ (back in the day) & every time I put them on I developed an electrical fault. Ok it was only twice, but enough to make me lose interest in them again.

[TZ350]

Page 530 ....

interesting article:
http://motortecmagazine.net/article.asp?AID=1&AP=1
Far to many strokes for my liking; but horses for courses.

Poly Quad 4 Valve Head .....

Nice little riddle: this video was sent to me by a flabbergasted tuner who wonders how a running engine can blow back more air than it inhales...
https://www.youtube.com/watch?v=aob8dgrnCjg

The inhaled air enters the duct from all sides (hence the positive effect of a bellmouth), but the blow-back shoots back out in a single column that hits the piece of paper in the video.



(Of course no running engine can blow back more air than it inhaled ).

... at low revs the crankcase is filled long before a piston-controled or disk-controled inlet port closes. The kinetic energy of the blow-back may even create an underpressure in the crankcase. And if the revs are really low, that underpressure can cause a restart of the induction process so that the engine inhales once, blows back once and inhales a second time, all within one inlet open period.

Yes, but that's all in the game of finding a jetting that makes the engine happy.

You must take into consideration the fuel being used, and how the factory engineers approached the problem.
Unleaded ( FIM that needs gloves and a respirator ) will go to 15:1 but must be run VERY rich to achieve best power with that level of com.
Leaded was run up at 19:1 in GP engines and at the end of the day the result was that the advance curves changed very little.
The big advance was in using the powerjet solenoids.
The unleaded scenario uses BIG powerjets ( 60 + ) to turn off the rich mixture past peak power.
The leaded setup was very lean at peak power, so only needed small ( 35 ) powerjets to keep the pipe temp up, over the top of the pipe.
Later of course that bastard Thiel came along and did the clever thing of PWM controlling the powerjet, thus even closer continuously matching the A/F ratio to the pipe
temp needed for max power.

Yep, easy to construct a 3D map with tps in the Ignitech by doing runs at 1/4, 1/2 and 3/4 throttle.
Or use the technology properly and use the deto sensor output as well to tell the ecu to retard at any time/rpm when the deto is happening , on the fly.

Using the det sensor with lights makes it dead easy to shape the curve to suit the engine, you dont need a brake at all.
You can dyno load cycle the engine as fast as you like, and for the short period it may deto, in a small band of rpm - the lights go off like a Xmas tree.
Takes a bit of fiddling with sensitivity to get in the ballpark and read actual deto - not normal noise, as the det frequency is
determined by the bore diameter.
Just make sure the M8 bolt thru the Bosch sensor is tightly fixed to the head or case, with no washers to crush etc.
Data logging the sensor output against rpm is the trick way to go, but simply watching the revcounter and the lights to see where deto is happening during a pull, is easy.
Works just like the deto warning when running EngMod - see the warning and fix the excess advance before a rod comes thru your screen, your choice.

In this case we are dealing with an aircooled scenario - remember two things here, Avgas loves com to make power, and the pipe loves retard to get heat for revon power.
Whenever you have high com or lots of advance a larger % of the heat of combustion is directed out from the chamber into the finning, via mainly the ring contact with
the bore, reduce either or both and more heat ends up in the pipe, making it appear shorter due to the increased wave speed..
Thus when you are thermally limited by the finnings capability to reject heat, you must reduce the heat input by limiting the com.
This then means more heat is available in the pipe, so more power can then be made by upping the advance - just another balancing act.
Its a Catch 22 when there is high dynamic com - ie, when the engine is on the pipe,there is danger lurking, but below that point - the cylinder filling is so piss poor the only way to get good
throttle response off slow corners is to pile in plenty of advance - its safe as houses as there is very little effective com.
This engine has no powervalve so the cylinder filling efficiency comes on slowly, so you roll of the advance slowly, then pull it out quicker as the system starts to work well.
With a powervalve its the opposite - as soon as the valve starts to lift the efficiency increases dramatically, so the advance needs to be pulled out much quicker and sooner.
Any well set up race engine will love 28* below the pipe when running Avgas.

In one of the World Champ skis I did we had 35* of mid advance, as the only time it was in that band was at full throttle from a standing start.
It would jump real hard using that advance to generate cylinder pressure - but troll around on part throttle at that rpm and it would kill the engine by loosening the barrel nuts with rampant deto.

Andrew down here has some head phones that run through a little amp. You just attach a little alligator lip to the head and listen at different rpm under load. You can hear it det loud and clear. You can run A wide band lambda with 2 stroke's on Ave gas apparently. The main thing is to let it heat up to operating temp before starting the bike. Also if the bike is silly rich it will kill it fast.

I've got a link somewhere (I'll dig it out) that shows one being built from Jaycar bits; but yes its a mic attached to the Alligator clip (on the ones I seen; not Andrew's one).

Here's a good link:
http://www.youtube.com/watch?v=L57fnOlWdaI

I can't see how you will be able to set up a safe ignition with no brake on the dyno and not listening for knock. Or as mentioned a wide band lambda.

Could be reving over a potential engine killer. Could be wrong but seams thats what has killed a few engines lately.

I will be there myself soon my ignition arrived yesterday.

The combination of port and pipe is firmly set at 12,000 so this is what I would use as a baseline.
Its initial drop off is to get it to idle well from the big static, plenty of mid advance for throttle response and then enough for a good peak ( assuming correct com cc ) but falling further to help revon.

As we are talking about Small Hondas, heres how to get rid of the total loss battery ignition and have a push and go.

Attachment 264470

This is the typical Total Loss Battery Ignition System and is very similar to an Energy Transfer or Magneto Ignition Sytem and could be converted fairly easily.

Attachment 264416

Here are details on how to convert the Hondas alternator to a Energy Transfer Magneto system.

And how Gordon Jennings converted a CB450 to Energy Transfer Magneto.
http://www.race-uscra.com/linksfiles/engineeringanddesignfiles/HondaIgnition/Jennings Honda Magneto Cycle 1969.pdf

Attachment 264418

Hondas version of a Low Tension Magneto System.

Attachment 264419

And Energy Transfer Mag.

Aladdins Cave

In reality the Store is the local Hose & Hydraulic Centre. Phone Fax (03) 768 9068 specializing in Hydraulink fluid connectors, Robin Engines and everything to do with hydraulic hose and coupling service.

Attachment 264350

The shop has plenty of Hydraulic Stock, everything you could need, but even at the counter there is other interesting things to see.

Attachment 264353

Like the Willys jeep

Attachment 264354

and Bradford van

Attachment 264351

and old radios

Attachment 264355

stationary engine and something else buried under there

There used to be lots of interesting places around the country like this, and this has to be the best of them that's left.

Made all the more interesting because its not just some sort of musem but a real working enviroment with the biggest range of hydraulic hose and fittings you could imagine.

Attachment 264352

I am very grateful to Robin Gray for taking the time to show me around. It was a great morning looking at some very interesting stuff and talking about motorcycles and racing.

At 01.30 AM it took me some thinking to kick out the inches and the thou's. But I finally worked it out.
That rule of thumb says: ring gap = 0,4% of bore. Jan Thiel found this to be critical; we use 0,5% of bore as a minimum.
Using too wide a gap may loose you some power. Using too tight a gap may turn the ring into a very effective oil scraper which can lead to a giant seizure.

Very old rule of thumb was 4 thou per inch of bore, so at 72 its near as damn it 3"= 12 thou = 0.3mm = perfect ring gap.

Hey here's an odd one. I pulled out the piston from a mates GG dirty bike with 72mm bore. He bought it low hours & has ridden it quite a bit. Piston shows some wear but pretty good.

But checked the ring end gap: 0.3mm! Hells teeth. That's tight for a 300. It didn't look like anyone had been in there before so we have a bike with decent hours on it & rings are super tight close. std spec is 0.5 with 0.8 limit. Wonder how close they were from new?

I would have thought this close on a big bore would have closed the tips up & bulged them. Above ports are perfect with no scoring & cylinder in general looks sweet.

Putting a Wossner in this weekend. Did one in mine a few months back & I went to the bother of filing the end gap to 0.5. Wonder if I should bother?

Are these clearances over protective?

The wings also lessen another problem with bent manifolds. Without the wing the flow will centrifuge to the outside which will overload the outside reeds so they need to be thicker than the inside reeds that seem to have eternal life. These hard outside reeds create a lot of flow resistance. The soft inside reeds would offer less resistance to the flow, if only there was any flow to speak of on the inside of the bend...
Adding the wing (preferably curved) allows the use of softer outside reeds and makes better use of the inside reeds' flow area.
But nothing beats a dead straight inlet tract without a wing.

I have done plenty of reed blocks fitted with alloy wing shaped splitters and then filled the cavity with Devcon.
The Devcon also can be shaped with a wet finger, so it doesnt stick.
I make the wing from 5mm plate with a full front radius tapering to a point that sits under the petal tips.
The real trick with the wings is to position the foil offset vertically a few mm, this forces the flow in a bent manifold to equalise the lift on all the reed petals instead of pushing most to the top.
In a back to back on a flow bench test this gained around 12% of flow at the same pressure drop, and was worth around 2 Hp in a 44 Hp old KX125 - this is why dead straight intakes/downdraft work so well.

Try to keep the length down, Dave. That is especially important if you are using a small-diameter carb.Good approach.I used to use the rubbery stuff that is used to fit car windows. You can model it with a wet finger and it stays somewhat elastic.

Just been thinking about trying to make a new manifold for a reedblock, so that's simple enough, a mikuni type spigot bolted to a block of ally that bolts to the reedblock.

But I'm trying to think of a clever way to make a few different stuffers to try out. I'm using a TZR250 1KT 6 petal block & trying to get a block of acetal to fit tight in there & then bore & port it sounds like a wind up. . .

Any other ideas for fabrication?

The following could be an option (about the same $$ as a FCR carb and spares; Damm the exchange rate) and offers a programmable ignition as well:
http://www.mbe-motorsports.com/catal...h1kh1lp0dlbdl3

In the MBE forum they talk about a two stroke kit (but it has gone from the website) using two injectors (theory was: one very small-bottom end, normal size-rest of the range, both together at the topend). largest throttle body looks to be 28mm.
and if you look hard enough there is a turbo kit as well...

But it would still take some work to get it functioning with a two stroke.
I'm guessing the best starting point would be to log a map sensor on a standard carb to get an idea of air flow rates vs. RPM.
then start building a fuel map around the ideal A/F ratios.

[Grumph]

In the light of this, does it make sense to think this could also happen with a membrane inlet?

This is the second go on a different RZ with a given carb and its again the same deal, have to use a realy lean pilot jet (because of double dipping ?) but then when opening the throttle the slide needs to be really rich (since no more double dipping, then the pilot shows it's true colours?)

And, if all that is true... what would one do about it or is it just a case of jet accordingly?

edit: seems the case only with this type of (modded) carb btw

Are you running the carbs on a downdraft angle ? VM's don't like anything above about a 15 degree downdraft as they will weep fuel through the pilot circuits...

[F5 Dave]

Back to det sensors

. . . The 3 pole ones have 3 wires so dont know how these would be connected.
Most are 2 wire and are common to most late model cars, Audi, VW ,BMW, Subi etc, the plug is standard injector type.
Easiest is the one with 8mm hole thru the middle, they are rugged as hell, so rarely bust, whip down to wrecker and get one for jack shit with the plug connected.

Well I got a Subi one. But now I pull the munted plug off it only seems to have one wire, so I'm assuming that it just grounds on the base & the piezo generates a voltage & the ignitech senses that.

Hmm, throw it on the scope & give it some quick taps & it can produce ~ 7V p-p.

[kel]

Back to det sensors



Well I got a Subi one. But now I pull the munted plug off it only seems to have one wire.

I made the same mistake
Let us know how you get on interfacing to the Igni.

[F5 Dave]

So there is a 2 wire one? how could it be different? Presumably when you configure the igni input to retard it configures that port to an A-D, so senses the voltage spikes and interprets that as detonation. I guess there is some algorithm to ignore engine vibration. it would have been nice if one could have an output from this to drive a LED indicator so when it is retarding you know about it.

I suspect if you connected a LED to the det sensor there would be enough noise you wouldn't believe (ie no algorithm) it & there may not be enough omph to drive it without extra cctry, though it did sink into a 1k resistor & still produce voltage. However I have no idea how much it will produce under real running detonation conditions rather than me hitting it with a small screwdriver blade.

Can you see on the PC when it is retarding?

[TZ350]

Hi Dave, I think it all works on the frequency of the vibration, an engine at 12,000rpm is 200 hz and detonation rings at a much higher frequency, like 1-10kHz. The detector attached to the knock sensor must be responding to the frequency of the vibration, and not the p-p amplitude.

http://www.bosch.com.au/content/language1/html/4563.htm

[F5 Dave]

12k/60, yeah, so just passes out low freq stuff

My subi peizo is a Unisia Jecs, not a Bosch after all, can't find any info, but surely a simple device.

I bought a Subi TPS at the same time for giggles as there is no room for like an RGV splitter box in the RS. I'll have to pull it with a cable, so just have to screw it to a plate & make a pivot arm. Possibly not as simple as I hoped now I have it in my hand, but possible & can hid it under the seat out of the way.

[wobbly]

Seems to be some confusion here.
The det sensor has to be connected to the knock gauge I posted the link to on here a while ago.
This has the filter on board as well as led indication.
And if you buy the option of a current sink to ground wire it will drive the retard input on an Ignitech, when deto is seen within the sensitivity range..
Yes any Ignitech input config is shown on the PC screen when it is grounded.

[kel]

Knock sensor needs to pass through a tuneable band pass filter, seem to remember Jaycar sell a tuneable unit. Will see what I can find.

[dinamik2t]

Yeah, I 'm with Kel; it's about frequency, not voltage amplitude. So, logically, there should be a Fourier transf circuitry and the band-pass filter - all either in physical or pc program form.
I believe it can be done if you know some electronics -and internet browsing-, but the result might not be worthy of the time spent.
The one Wob informed us about has a reasonable price, so why not?


edit :
I don't quite remember at the moment, but there must be some knock signal examples at the Labview website at NI. And some (a little more sophisticated) knock program examples in their forums.

I had the Jaycar page saved in my bookmarks, but it's seems to be no more: http://www.jaycar.co.nz/productView....k&form=KEYWORD
It must have been this one : http://www.rhinopower.org/knock/knock.html

[F5 Dave]

ahh, I see, the inputs aren't as smart as I thought. I have a plan B, but we'll see where that goes.

Heck I should really be in the garage sorting out plates for the engine mounts or I won't even be able to start it once the engine is done. Pretty cold though.

[diesel pig]

Seems to be some confusion here.
The det sensor has to be connected to the knock gauge I posted the link to on here a while ago.
This has the filter on board as well as led indication.
And if you buy the option of a current sink to ground wire it will drive the retard input on an Ignitech, when deto is seen within the sensitivity range..
Yes any Ignitech input config is shown on the PC screen when it is grounded.

On the subject of Ignitechs, I wonder if I am missing a trick with the coil. I am using a old VT250 coil with my DC-CDI-P2 but is there a better coil to use?

[dinamik2t]

Just an addition here:
There this "inexpensive" Selettra ignition coil from the IAME X30 karts, here: http://www.ebay.de/itm/ws/eBayISAPI....ht_3433wt_1139
You can ask the seller to send around (he sent mine in Greece at least).

I don't know whether it's a top choice coil, but based on the fact that it comes from a high revving competition engine and that Selettra is a known company, it could be. Perhaps the elders can confirm -or not!

[koba]

On the subject of Ignitechs, I wonder if I am missing a trick with the coil. I am using a old VT250 coil with my DC-CDI-P2 but is there a better coil to use?

Funny, I've got a few of those too, I guess they are pretty common considering the engine is likely to die well before anything else.


EDIT: I once bought a complete "running with a knock" VT250 for a packet of biscuits. I got it, took of the mirrors and a few other bit I wanted off it then contacted the local wrecker.
I said: come and get it for free.
The reply was: "shoulda kept the bikkies".