ESE's works engine tuner

[wobbly]

Here is the TM parts page.
There is a newer reed petal set, that makes better power with both bottom backups pointing inward
and the single top one pointing away from the petal.

[peewee]

In a sim the power continues to rise as the reeds are made thinner ( from being too thick ) , due to the petals resonant frequency coinciding with the forcing frequency of the intake tract.
BUT,as you drop the resonant rpm you are also effectively increasing the lift and petal open duration.
This very quickly runs into trouble at peak power and beyond,as the instant the petals hit a stop, or are not stiff enough to exhibit controlled drop back
onto the seat,they go completely spastic.
Easily seen in the sim as wildly varying tip lift ( and a big power drop ) with each sim iteration in turn.
One trace run will show a good result, the next trace the lift is all over the place.
Add a couple of 1/10 of a mm to the petal thickness and the resonant rpm will rise a few hundred, and the effect goes away in the top end.
This is exactly what the sim is for - to point you in the right direction with what to be trying on the dyno.
IMHO there are way to many variables involved to even begin to formulate a workable rule of thumb for reeds, as in reality, suck it and see is the only way.

I have just completed a huge number of sims for the KZ10 with all manner of different reed blocks, the only way to get a meaningful result was to set the petal thickness for every block
such that the 1st harmonic rpm was 10,000 with engines natural peak power at 13200 ( as in reality ).
This then shows the real effect of the reed block geometry on the power - not the differences due to the varying petal thickness - as this is easily changed during the subsequent dyno confirmation process.

theres no rule of thumb for the stiffeners ? I see some are smaller rectangles on top of the main rectangle petals. some were triangle shape. what about the thickness and material in relation to the main petals. has to be some kind of approximation to get started in the right direction

[jfn2]

First, what is magmalyte?
I see that the cage itself has a rounded nose. What would be the cage angle and why is the nose rounded and not more v-shaped? Does this have something to do with the shaping on the inside?

Wobbly:
With turning of the backups on the bottom and turning around the backup on top be in line with trying to get the intake charge to favor the top set of reeds. Does this not also change the stiffness of the top set of reeds? Couldn't you just change to a thinner material on the top? Do you have a pic of the reed and cage assembled? Thanks.

[marsheng]

I've run CHT/EGT on 5 different bikes (all air cooled, 50-160cc) and use 175-225C as the safe range for CHT. This was for endurance rides where the temp is held for long periods of time. I find CHT tends to be highest at lower RPM and downshifting will bring it back in line.

Thanks very much. Useful data for me.

I wonder if it is lean at low RPM and a downshift brings in a lot more cooling air and gas ?

[marsheng]

Temp range for the under plug depends alot on how good the cooling of the insert is ( in water cooled ) and how good
the head design and material type is ( in air cooled ).
The CHT range is around 160 to 260*C

is this for both air and water cooled motors?

[Frits Overmars]

I wonder if it is lean at low RPM and a downshift brings in a lot more cooling air and gas ?

Air and gas are often brought in with the intend to make a fire...

[cotswold]

My old 50 the NSR one,
It's onto it's 3rd owner now and at the weekend it seized, I was asked to have a quick look as he wanted to get back out on it. I took the top off and there was a ring of melted ally around the outside of the piston and the combustion area of the head had been hammered into the squish area.
I should have taken photo's
when I was checking over the head I found the plug was hand tight, could this have been a case of very lean mixture or has anyone got another idea?
I will ask him to take a couple of photo's if anyone is interested

[wobbly]

Yes the whole idea of differing reeds and backups to to bias the flow into the upper ports.
The KZ10 reed block isnt rounded - thats just an artists impression, the angle is 58* and its a normal exit shape.
There are various aero blade inserts made for these, and the biggest/longest horizontal foil section insert works best.
As I said previously you cant simply keep on making the top petal thinner, as you quickly end up with the reeds natural frequency
making it loose control at high rpm and it " flutters " uncontrollably.
I saw this on the flow bench with a set of stock CR125 petals, they have no backups and as soon as the lift/length ratio approaches 1/3
the petals bounce around in a rhythmic pattern - the best petals on the dyno,with backups, went to full lift and sat there.
Notice that the backups only contact the outer sides of the main petals, this seems to be designed to prevent fretting on the outer corners
but obviously they affect the stiffness of the main petals as well.
Maybe it also reduces the turbulent exit flow spilling over the sides of the ports, and forces more flow down the middle.

Re the CHT numbers - yes the range is for water cooled at the bottom end and air cooled up over 200.
Yamaha KT100 engines had a change from sand cast to die cast at some stage, and the different material/ casting technique ( exact same head shape ) caused a drop in CHT
from 230C to 170C.

[Haufen]

That may not break the pocelain but it can very well cause an different problem. An open ended wrench bears on just two points of the hexagon and as the spark plug body is relatively thin-walled, you'll risk pushing the hexagon out of round, causing leakage at the top seal between the hexagon and the porcelain.

Thanks, I will keep that in mind. Although a quick inspection of the broken plugs revealed that all of them were fine in the transition area between the hexagon and the porcelain.
I will now carefully mount the new plug with the proper tooling and report back when I have new findings.

Okay, so I can report back that I can now repeatably destroy NGK R7376-10 plugs by running just a tad too lean when pushing to the limits on the dyno in a high bmep application, due to causing "funny" combustion pehenomena. The toolings used to tighten the plug had no influence on the result, unfortunately.

Enough motivation to stop procrastinating on the installation of a deto sensor. With a diameter of about 20mm at the mating side I am asking myself where to put it ideally (and where not to put it). As I run a paper gasket, I suppose somewhere on the water cooled cylinder would be nice, but there is not enough space available, except "on a stick" on one of the bolting areas for the head cover. Where du you guys put yours?

[marsheng]

Air and gas are often brought in with the intend to make a fire...

Oh yes it does but here is a lesson I learned the other day.

I have a very heavy dyno drum. 700 KG

In running in a motor, I reved the bike up 10K and then closed the throttle. As it slowly coasted down to an idle, the exhaust pipe began to glow red hot. I opened the throttle and the pipes went from red to black again. I repeated the process to confirm and it replicated the result.

I understand that a lean mixture burns slower and the gasses can still be burning when the exhaust valve opens, hence the exhaust pipes get really hot. So in this case, more air and fuel actually cooled the exhaust.

Cheers Wallace

[Frits Overmars]

Okay, so I can report back that I can now repeatably destroy NGK R7376-10 plugs by running just a tad too lean when pushing to the limits on the dyno in a high bmep application, due to causing "funny" combustion pehenomena.

That's not nice to hear, but good to know. There seem to be two kinds of NGK R7376-10 plugs on the market: one is expensive and reliable; the other is neither.
How much was your version?
And of course you made me curious about your combustion phenomena. Care to elaborate?

Enough motivation to stop procrastinating on the installation of a deto sensor. With a diameter of about 20mm at the mating side I am asking myself where to put it ideally (and where not to put it). As I run a paper gasket, I suppose somewhere on the water cooled cylinder would be nice, but there is not enough space available, except "on a stick" on one of the bolting areas for the head cover. Where du you guys put yours?

As you say: on a stick, like on the RSA-picture below. But with limitations: the sensor should be pulled directly onto the head; no nut in between.

In running in a motor, I reved the bike up 10K and then closed the throttle. As it slowly coasted down to an idle, the exhaust pipe began to glow red hot. I opened the throttle and the pipes went from red to black again. I repeated the process to confirm and it replicated the result. I understand that a lean mixture burns slower and the gasses can still be burning when the exhaust valve opens, hence the exhaust pipes get really hot. So in this case, more air and fuel actually cooled the exhaust.

I see what you mean Wallace. In an engine running at 10.000 rpm with a closed throttle there should be no combustion at all. In your case the mixture, although very lean, must still have been rich enough to be igniteable. That's not good. And apparently combustion is so slow that the mixture is still burning when the exhaust opens. That's bad. If it makes the exhaust pipe glow red hod, think what it will do to the exhaust valve. Wait a minute: exhaust valve? Carry on; you'll be fine .

[Haufen]

That's not nice to hear, but good to know. There seem to be two kinds of NGK R7376-10 plugs on the market: one is expensive and reliable; the other is neither.
How much was your version?
And of course you made me curious about your combustion phenomena. Care to elaborate?

As you say: on a stick, like on the RSA-picture below. But with limitations: the sensor should be pulled directly onto the head; no nut in between.

Some time ago someone (you?) posted the part number of the expensive kind - and I can confirm that I have recently destroyed the 4th piece of the expensive kind.
(This then led me into conducting a plug test in which the standard NGK BR10ES gapped @ 0,45mm surprisingly matched the 7376. No surprises from Brisk, Bosch and other fancy NGK stuff, but mostly a tad lower or not comparable due to different combustion chamber volume and / or electrode position)

I would love to approach this combustion phenomena with fully indicated engine and a couple of combustion chambers, as I am sure one could learn a lot from it in a short period of time. But for this hobby application I have to do it the oldschool way, I suppose. I will check if I can find a recording of the engine sound. As the engine approaches maximum power or shortly thereafter, there may be a kind of "poof" if it is slightly too lean, and if its a bit leaner there is no "poof" and the engine just dies as if the kill switch would have been activated. If I do not lean the engine off with the e-PJ, nothing happens (except for that lost horse and a bit of powerband). As this is caused by a too lean condition, I suspect it to be pre-ignition and wrongly timed, as it also kills all the resonance in the engine. Fortunately, besides from dead spark plugs, the rest of the engine seems to be fine.

Unfortunately my head cover and the combustion chamber do not share a bolt position as in the RSA. So I could only get the signal from the cylinder, not directly from the head insert. Do you expect this to cause issues or significantly weaken or change the signal?

[Martin1981]

hi wobbly, i sent you a PM.

[Frits Overmars]

Unfortunately my head cover and the combustion chamber do not share a bolt position as in the RSA. So I could only get the signal from the cylinder, not directly from the head insert. Do you expect this to cause issues or significantly weaken or change the signal?

I can't say without knowing exactly how you are going to mount the sensor. I guess you will just have to find a place in direct solid contact with the combustion chamber, and even then it will be a matter of trial and error.
All parts of an engine vibrate with some frequency, with some amplitude and in one plane or another, and these vibrations may attenuate the detonation-induced vibration to some extent. Conversely, parts resonating in their own root frequency may fool the sensor into reporting detonation when really there isn't any.

[wobbly]

I use the Bosch sensor bolted directly to the head frame stay of a CR125 ( so its at right angles to the cylinder axis ) and it works perfectly.
On the TM KZ10 engines I have one longer head cover stud, with a small tophat spacer under the sensor, this also works perfectly.
The frequency/amplitude of a real deto event is very specific, and the little sensor light box I use I can wind up the sensitivity such that all the lower level led's flicker
when running the engine under normal load - and then only the big red warning light comes on when deto is present.
It also has a 0-5V output, showing the level of det on the data logger.

Re the R7376, apart from the Hp gain I found over all the other plugs I tested, the biggest issue it solved was two fold.
When a race engine started to detonate, the Iridium type plugs would crack the porcelain at the tip and fail quickly afterward by dropping pieces onto the piston.
The other plugs with big section earth electrodes have also killed several engines when this piece has dropped off.
The R7376 is expensive due to the tip porcelain construction, and the thin rare earth laser welded ground strap.
I have run hundreds of hours on these plugs and never had a failure ( inside the engine ).
Surely i would imagine a tuning issue that cracks the upper porcelain on a R7376 would do exactly the same to any plug ( or worse, inside a cylinder as I
discovered ).