ESE's works engine tuner

[TZ350]

If you are running a regulated return system what if you plumb the top side of the regulator (boost reference) to the pipe??? This is how a lot of turbo 4T adjust the fueling as boost rises and falls...maybe it would work the same with the pipe on the 2T. As pipe pressure rises, so would fuel pressure and hence more fuel. If the pipe collapses then pressure would also, so the fuel pressure would drop as a result.

Yes I once thought so too.

I have tried measuring the pressure in the expansion chamber and header at different points.
It turns out that the average pressure at the measuring point falls when the motor is on song and making real power.
Makes sense when you think that more power means more suck and that the suck side of the outgoing wave is stronger than the reflected return wave. From my measurements I found the average to be lower than the outside ambient air pressure.

Unlike a pressurized vessel like say a spray-paint can, an expansion chamber is not filled with a uniformly consistent pressure but has dynamic patches of high and low pressure throughout at the same time.

And as you get on the power and the strength of the suction pulse increases and the average pressure at any point decreases.

In an active expansion chamber there are two different pressure scenarios going on at the same time, "Average" pipe pressure and "Dynamic" pressure at any singular point.

Unless you have a way of timing the sampling at any point chosen point in time it is just random whether you are seeing a high or low patch.

Currently I am trying to read a cylinder pressure pulse just before exhaust port opening as a way of determining if the motor has fired or not. The timing of the pulse and pressure will hopefully be more consistent.

[Yow Ling]

Can anyone help me with an ignition timing map for a TZ250 5F7

Stock 5F7 is 1.1mm BTDC

[2T Institute]

I think one problem with using KT100 pistons is the process where a piston which is only a very small amount larger is fitted after a hone, and then the next one and the next one and so on. The bore changes shape slightly each time it's used and the hone simply changes the surface texture. I think it would be worthwhile to have the bore ground in a cylindrical grinder every 2nd piston to maintain it's cylindrical shape and remove any taper that may develop. I quite like the look of the Burris pistons but don't know that they are actually any better. They can certainly extend the life of a cylinder due to the increased clearance required versus say a Yamaha piston. there might be some slight difference in the shape of the crown requiring a different head.

Erm no not by a long shot. Concentricity was maintained by a quality Sunnen hone, an air cooled cylinder should have a taper. He's a bit crook now but erstwhile forum member Ken Seeber manufactured pistons in a 0.01mm step. Kept old KT100 S and J engines for years and years. Piston to bore clearance can be adjusted to circuit and weather conditions.

[paul gane]

Stock 5F7 is 1.1mm BTDC

Thank you for the reply. Realy looking for graph of advance and retard details.
Can't find any official info from Yamaha.
I understood that these were the first to have a retard curve on a production racing motorcycle...

[DoldGuy]

Currently I am trying to read a cylinder pressure pulse just before exhaust port opening as a way of determining if the motor has fired or not. The timing of the pulse and pressure will hopefully be more consistent.[/QUOTE]


TZ,

I know NOTHING about injection on two strokes (but have tried to follow/understand your venture) and curious of the pressure location just before exhaust opening. Would there be greater differintial just after TDC to determine combustion did/did not occur?

Doldguy

[TZ350]

TZ, I know NOTHING about injection on two strokes (but have tried to follow/understand your venture) and curious of the pressure location just before exhaust opening. Would there be greater differential just after TDC to determine combustion did/did not occur? Doldguy

Thanks for taking an interest in my EFI project. I am not sure I know much about 2T injection either but I am finding it a fun and challenging learning curve.

I agree, just after TDC would give a pretty identifiable pulse and was where I wanted to go at first. After thinking about the 1,000-2,000 psi combustion pressure and taking advice about the damage that detonation would quickly inflict on the pressure sensor I went with Clint's (from www.tfxengine.com) suggestion to look at making a pressure tap 10 or so degrees before exhaust port opening. I figured that would be about 100-200 psi and a much lower gas temperature, so much easier to work with.

Actually drilling a small hole for the pressure tap proved challenging. On the NSR cylinder there is not much useful material for drilling holes through that bridges the gap between the cylinder and the water jacket. It was a real challenge to drill a 0.75mm through hole without breaking out into fresh air.

Tonight I ran the bike up and put my finger over the pressure sensor hole in the cylinder. At idle there was plenty of hot gas coming out but as I revved the engine up the volume and temperature of expelled gas significantly reduced. Wasn't expecting that but it is in line with the reduced signal I have seen from the pressure sensors I have tried.

Again it looks to be all about "Time". The response time of the sensor and the the time it takes to expel enough gas to fill all the cavities and start applying pressure to the sensor. As the revvs go up there is less and less time so the signal gets smaller and smaller.

[EssexNick]

TZ. I've also been following your progress with great interest. It seems to me, (and what do I know?), you will always struggle to find a pressure sensor that responds fast enough. Now if you only want to determine whether or not combustion has taken place, would it be possible to use a contact microphone on the outside of the head and just use electronics to differentiate between the sounds of combustion and non-combustion?

[TZ350]

Yes it might have to be where I look next.

[husaberg]

Yes it might have to be where I look next.

Another place to look assuming you cant get the mic or sensor to work would be piston acceleration

During a normal combustion event, the piston speed is lowest at top dead center on the compression stroke, and is the highest with the piston moving downwards during the power stroke. The average engine speed for a single engine revolution would lie in between the lowest and the highest instantaneous values. The difference between the lowest and highest engine speeds is significant enough that any deviation from the values expected during normal combustion can be used to detect a misfire event. If misfire occurs, the crank speed during the power stroke will be significantly lower as there is no downward force being applied on the piston head from combustion of the air fuel mixture. the variation of instantaneous engine speed with crank angle for normal combustion and misfire conditions.

https://www.pi-innovo.com/misfire-de...-with-openecu/

http://www.freepatentsonline.com/6885932.html

[Frits Overmars]

Tonight I ran the bike up and put my finger over the pressure sensor hole in the cylinder. At idle there was plenty of hot gas coming out but as I revved the engine up the volume and temperature of expelled gas significantly reduced. Wasn't expecting that but it is in line with the reduced signal I have seen from the pressure sensors I have tried.
Again it looks to be all about "Time". The response time of the sensor and the time it takes to expel enough gas to fill all the cavities and start applying pressure to the sensor. As the revs go up there is less and less time so the signal gets smaller and smaller.

You nailed it TeeZee. Sure, the sensor may be responsible, but even the fastest sensor can only register the delta_pressure/delta_time-variation in the system consisting of engine cylinder, connecting bore and volume in front of the sensor, the worst situation being a long, narrow connecting bore and a large volume in front of the sensor.

It's a bit comparable to an exhaust duct, where the return pulse from the exhaust pipe has to raise the pressure in the duct before it can do any good for the cylinder.

Assuming the sensor itself is up to the job, the bottlebeck in your case will be the 0,75 mm connecting bore. And since that is already borderline, you might concentrate on shortening it if possible, and reducing the cavity volume to absolutely zero.

[jbiplane]

It's a bit comparable to an exhaust duct, where the return pulse from the exhaust pipe has to raise the pressure in the duct before it can do any good for the cylinder.

Is there any idea how make optimal Ligth weight exhaust.
Says for 66 bore 40mm stroke 2 cylinder engine to have good performance at 6800-7200 rpm.
Exhaust temperature 600C. Exhaust duration 190. Crancase compression 1,6

Proposed shapes are





Classical variant of 2 scooter muffler could be bulky (and they designed for 9000 rpm)

[Frits Overmars]

Another place to look assuming you cant get the mic or sensor to work would be piston acceleration

During a normal combustion event, the piston speed is lowest at top dead center on the compression stroke, and is the highest with the piston moving downwards during the power stroke. The average engine speed for a single engine revolution would lie in between the lowest and the highest instantaneous values. The difference between the lowest and highest engine speeds is significant enough that any deviation from the values expected during normal combustion can be used to detect a misfire event. If misfire occurs, the crank speed during the power stroke will be significantly lower as there is no downward force being applied on the piston head from combustion of the air fuel mixture.

https://www.pi-innovo.com/misfire-de...-with-openecu/
http://www.freepatentsonline.com/6885932.html

My evil mind can't help noticing "the piston speed is lowest at top dead center on the compression stroke". Lovely. I'd add that the piston speed is zero at TDC,
and not just on the compression stroke but at every TDC.
Next, they juggle with the term engine speed. One moment it stands for piston speed, the next moment it's crankshaft rpm, which is quite a different animal.
Ultimately, it comes down to the fluctuations in crankshaft rpm because that's the only thing that can be measured with the encoder they are using. It has to be reliably connected to the crankshaft and it has to be an expensive encoder too, because those 'significant' fluctuations are maybe 1 percent of the nominal crankshaft rpm. Moreover, the cited papers discuss a four-stroke engine. The rpm fluctuations in a two-stroke engine are maybe half of what they are in a four-stroke, which makes it even more difficult to distinguish misfiring.
Compare that to measuring cylinder pressures: the pressure ratio between a firing and a non-firing engine just before exhaust opening can be 500 %.
That's a hell of a lot easier to detect.

Last but not least, if you insist on installing an expensive encoder and measuring the crankshaft speed, what exactly are you seeing? Fluctuations caused by cylinder pressure? Or fluctuations because of a slipping/grabbing clutch? Or a whipping chain? Or because of momentarily contact loss between wheel and road? Just listen to the rising and falling engine rpm on any Isle of Man-video.

[TerraRoot]

TZ, can i point out that DI would solve your part throttle problems? while TBI/port injection solves DI's "not enough time to inject all the go juice at revs" problem

combine the two systems and drop trying to detect misfires as they happen?

[Frits Overmars]

TZ, can i point out that DI would solve your part throttle problems? while TBI/port injection solves DI's "not enough time to inject all the go juice at revs" problem.
combine the two systems and drop trying to detect misfires as they happen?

It's not just about misfires, Terraroot. It's mainly about coming on the throttle (or should I say on the pipe?) after a throttled-down deliberate non-firing period.

[TerraRoot]

non-firing, much better word.
what i was getting at, i'm equating the response trouble TZ is having with >
carb:
fourstrokes on part throttle, each non-fire the vaccum is less so less fuel, engine recovers.
FI:
fourstrokes on part throttle, each non-fire the FI-ECU does not know about drowns the engine with fuel making each non fire worse until engine speed drops and ECU notices.

Solve the fourstroking, solve the throttle response. Every DI two-stroke i've heard has a great solid smooth idle/part throttle.