[metcap]

Ok guys I'm not afraid to be proven wrong, I'm no expert and I'm always willing to learn from mistakes. I begged, borrowed a scope from work and had a closer look at the VR and spark signals. First off, working off information from this forum and elsewhere this is my VR signal pre n post conditioner. Looks about right, hahah hope it's right.

I'm using a MAX9924 based conditioner on 202207.1 code and all test was done on bench.
Spent few days on this, initially comparing the spark signal output from the board vs the spark event at the spark plug but found almost no delay there and was convinced that all the delay I'm seeing is caused by the timing light. Packed up all the bench test stuff and was done. But somehow sleep on it and just thought why not test spark event vs ignition timing signal. This is when I found something interesting, tests here done with these settings trigger angle at 180deg, fixed timing at 0 deg and fixed dwell at 3ms.


First n second image shows test done at 1k rpm. Not sure why but I'm seeing 280uSec (sometimes up to320uSec) error between vr output signal into ecu vs spark event.


These next image is test done at 5k rpm and now the error comes down to 60uSec.



When I tried trigger angle at 0 deg I'm seeing almost no error at 1k rpm right up to 5k rpm and the trend is increase in error as I go higher trigger angle right up to 180 deg. The puzzling thing to me is if there's wrong setup in vr setting or setup, why does the error only increases off 0 deg?
This dun seem normal but I'm just unsure what to make of it, I've checked vr polarity seems correct. So in my case I've trigger angle at 161deg and I'm guessing I'm starting out at an advanced timing and I'm seeing it retard as I go up in speed?

[jonbill]

Isn't this just measuring the time taken for "trigger angle" degrees of rotation?
By my calcs, 180° at 1000rpm takes about 30ms and only 6ms at 5000rpm.
The units are out, but....

[miker]

can you post a high speed trigger log? That will show if you've got the cables the right way round to the VR conditioner.

(what you're seeing does look strange & with my knowledge of the coding i can't explain it.)

[metcap]

If you mean the TS tooth/composite logger then I dun've it coz I'm on an unregistered version.

[PSIG]

There's a small lack of info with the images. What I'm seeing in the first image is "normal" for a proper polarity VR sensor signal, into a MAX chip, and output inverted - IF it is being probed on the proper ± locations. Note the edge settings must be set to RISING. I also see a large amount of hysteresis, so I'd look at how the MAX is set-up and in which mode. Most MAX setups typically use A2 mode here, but that has its own drawbacks. If you don't want any hysteresis or too-smart features, drop to zero-crossing mode outlined in the datasheets, which would be good anyway just to remove that variable.

The last images show (to me) simple driver lag. The time between the blue output (Speeduino) and the yellow firing (coil) is entirely on the driver you added between them, whatever it is that you're using. In either case, it can be tuned without special considerations. Give the engine what it wants. As the info is limited, tell me if I am not interpreting them properly. I had to make some assumptions, and I don't like to do that of course.

[metcap]

I'm not to trying to prove that what I'm showing or saying is correct, I'm just sharing what I'm finding out. Could be my setup is wrong or whatever and I'd be happy if I get corrected and pointed in the right direction.
As I said earlier I did test the output from the Tc4424 chip vs spark event and saw minimal errror. About 20 to 60uSec is what I'm seeing and I dun think this scope has the resolution to do any better.
Besides what you're proposing that it's driver error, I'd think it'd be the same with trigger angle at 180deg vs 0 deg?

[metcap]

My max9924 conditioner is built based off this, which is the Speeduino official schematic hahahh


So if you're advising to just use zero crossing without adaptive is it I need to set it up as mode B?

[JHolland]

Have you got C4 and C4 fitted? you should not have them unless you have a noise issue.

[metcap]

Yes I do hav C4, MAX9924 is a single input version of the max chip BTW.

[PSIG]

I'm not to trying to prove that what I'm showing or saying is correct, I'm just sharing what I'm finding out.

That's good, and allows us all to look at your data. That's why I commented on what I saw.

Besides what you're proposing that it's driver error, I'd think it'd be the same with trigger angle at 180deg vs 0 deg?

The coil driver, the external one between Speeduino and your coil. Or, the one in your coil if it's an active (smart) coil. That driver, not the signal driver (TC442x, IXDN6xx, etc), although it and it's capacitor could be a part of it, relative to the external driver's gate capacitance, which would affect switching speed of the external coil driver.

Hmm. It's possible certain external drivers need greater current to switch quickly (larger TC/IXDN capacitor or reduced output ohms?). We see the older Bosch technology requiring substantially greater gate current, e.g. 106B coil pack drivers.

I assume (don't like to assume) your probing was on the Speeduino ECM signal output, and the coil (-) terminal? If so, the time between those two signals (and potential delays) is primarily on the external driver's response. Just saying what I'm seeing, assuming some things. A coil driver delay of x-µs would cause an increasing timing drift as rpm increase, though should not affect trigger offset timing, and I would suggest 'scoping the direct effects of only trigger angle change. It could be a random issue based on components used, or a certain/specific combination of components or traits?

This is an investigation, and that's what we're doing. Good stuff. But, I still have an unanswered question, which is — what is the specific problem this issue is creating? Exactly why would we care? Simple question asking for a simple answer. "It will destroy the engine" (and why) or "It just bugs me" (and why) are both valid answers, and sets the foundation of how we address it.