cam and crank trigger combo inside a distributor (megasquirt

[Piledriver]

I ran 6 on the crank/1 sync for years, it worked fine, at least for more or less stock engine.
On a higher spec/faster accellerating engine I agree the more resolution the merrier.
It would be an interesting test to see how much it really matters.

[juki48]

HI Guys, just REMEMBER 12-1 "gear down" is ONLY 6 teeth on the Crankshaft.

This is a bit low for position prediction (6-8 degrees) guess@2500/second.

Lance

I realize this and will play with more teeth once I get it running. I will make a 24-1 in the future and see how it goes. I have a 24 position index head and a 360 position index head. so 12 was a nice easy one to make and I'm confident it will work. my engine is a 2276 turbo running 5psi for now. it is quick but it is a driver, not a drag car. I've never had it over 6k rpm, usually shifting at 4500. don't need to wind it out with all that torque.

[Devastator]

Nice looking wheel Juki.

[Paul H]

I've ran some tests on a 12 tooth in the distributor v 36-1 on the crank up to 7800 rpm on a 2180cc with no difference in power output in fact it placed one power curve right on top of each other also no difference in timing at any rpm. Another test I ran on a stock 1600 gave a slight increase in power with the distributor trigger but that was down to the sequential fuel. We ran the 2180 in two stroke mode to negate any difference between sequential and batch.I could have run the test using the 36-1 crank wheel and the cam sensor to give exact same fueling which would have saved time faffing with the fuel map but as the result showed no difference it was pointless.

[Piledriver]

Paul, are you set up to do any acceleration timing prediction error tests, say between the 12-1 "gear down" and the 36-1?

[Paul H]

Paul, are you set up to do any acceleration timing prediction error tests, say between the 12-1 "gear down" and the 36-1?

I did that mathematically in a previous post -I suppose i'll have to go and find it and re post.
also when you have a multitude of engines fitted with 3 triggering patterns-flywheel,pulley,dizzy and a digital dyno you can get to the real information and results rather than dwelling on irrelevant theories -"the proof of the pudding---"
Like I just said in my last post

also no difference in timing at any rpm

check the timing at different rpm's and make sure the reading concur with the timing number displayed in the GUI

[Paul H]

Assuming the ecu calculates rpm from the time that passes between each tooth
By my calcs the error in rpm readings between a 60 tooth and a 12 tooth at 2000 rpm with accel rate of 2500rpm/sec is 5 rpm.

60 tooth = 0.0005 sec per tooth
12 tooth = 0.0025 sec per tooth

So what happens in 0.002 seconds at an accel rate of 2500rpm per second - a change of
2500 x 0.002 = 5 rpm

I checked my calcs a couple times but they may be wrong

I always wondered why a 60 tooth wheel was better than a 36 or less tooth wheel

The previous post -I guess if you have a problem with a potential 5 rpm difference then you need more teeth

[juki48]

I've ran some tests on a 12 tooth in the distributor v 36-1 on the crank up to 7800 rpm on a 2180cc with no difference in power output in fact it placed one power curve right on top of each other also no difference in timing at any rpm. Another test I ran on a stock 1600 gave a slight increase in power with the distributor trigger but that was down to the sequential fuel. We ran the 2180 in two stroke mode to negate any difference between sequential and batch.I could have run the test using the 36-1 crank wheel and the cam sensor to give exact same fueling which would have saved time faffing with the fuel map but as the result showed no difference it was pointless.

Paul, Thanks for sharing that. It's always nice when someone has real data related to your project. Thanks again!

[Piledriver]

Groovy

[Redline Weber]

YES, math works though you should finalize your report in DEGREES.
ANOTHER FACT, do NOT limit this thread to MS.
MY ECU-882, MS-3, some other EMS can do the work, MANY "big name" EMS can NOT.

NOW, a simple "timing lag" test method to prove the math.

The Spark Instant can be SET to a SPECIFIC value with most EMS, only ONE amount.

Thus doing so with a running engine one may OBSERVE that value on the crankshaft pulley.

This observation should exhibit a clean "flicker free" mark with NO drift.
Once finished, the next step is to "blip" your running engine with additional observation.
No drift should be seen with respect the wheel limits/2500 RPM per second change being within TW tooth count quantity.

Lance

[Piledriver]

Theoretical RPM delta doesn't matter, the actual timing variance vs predicted is what is important, and that is going to be ECU dependent.

I have done simple tests set at 10btdc on ms2 ans ms3, with the 6/1 setup there was a ~degree or so of jitter at the blip, but that was a stock motor/flywheel combo. The 36-1/1 wasn't much better except near cranking speed.

MS2 and MS3 can log a running %timing error, where it can evaluate what its predicting vs. what it measures.
(Ms3 is logging over serial at ~50hz using jSSC drivers now, and ~400hz on a somewhat limited dataset to SDCard

I have made up some new sensors to play with, and I can pop the 16-1 gear down (on the Mallory) and compare it to the 36-2 I just made and see how much they differ.

I'm trying to find a larger gear to swap into my setup, at the same pitch it looks like I could go with a 60 or even 72 tooth wheel and still fit in the 009, and much larger on the Mallory with a different sensor.

Beyond a certain tooth count it probably doesn't matter, though.

[Paul H]

I think the answer in degrees potential lag is 0.01 degrees.
Has anyone got a timing light with an accuracy of better than +/- 2 degrees ? send it over so I can carry out some tests

[Redline Weber]

HI Paul, your math is NOT completely correct, or finished, though you did get one numeral correct the "1".
NOW add some "zeros" or 10 degrees for a 4 tooth (90 degree) TW, still better than a 3 tooth TW.
OK, sure, some EMS work worse than the (your) MS standard. (Pile)
The factor of CRIP value adds to lower the error in position.

The fact the ANY "jitter" exists in Steady State is VERY bad.

THINK of a 3 (three) tooth TW, NOT evenly divisible using the four cylinder engine.
This Trigger Count is uneven AND will cause a different "lag" between cylinders depending on CRIP.

These rules are Gods law, the law of mathematics, not a NIST law.

My Blue Point timing light "reads" clear like a Laptop Display.
The chance for error was also tested by using a Military Strobe Light.
YOUR "overseas" problem for the timing light may be easily solved as i did when I bought TWO "mill-spec" strobe lights for $10.00.
They were bought at a Surplus House.

Lance

[Paul H]

Well lets get back to reality and away from the theories.
I've tested many triggering devices and patterns with three different manufacturers ECU's and had no issues
with timing drift or errors. The only problems I can see occurring in this thread is trying to cram a system into the small
space of an 009 and the use of the associated small pickups and tiny teeth. All devices I have used that contain these pickups have failed MSD distributors, Petronix, Mallory even Honda had problems with their distributor based sensors with "cross talking" and interference

[Redline Weber]

We fit the Honda VR sensor (our) inside our 009 VW distributor/CAS used with a 24-1 TW.

Lance