Trigger Wheel Decoder / Coil driver

[antus]

Cool.

In my case there is no adjustability, and I tend to come up at up to about 3 degrees out building it by hand. So I'd be looking for fine resolution over small range, maybe a couple of steps and up to 5 degrees. Your right about it being good enough to start and could then handle it in the delco, but it would be nice to be able handle it upstream of the delco, so that I can put the delco in diag mode, put the light on it and go "click click click, bang on 10 degrees".

I think sending the signal to the delco to do the retard would be better, because then it'd be visible in the logged data stream without anything extra custom or another piece of data to take in to account, and you'd get knock count logging if you wanted to use that too.

[BennVenn]

Good point about the 10deg in diag mode, as that is the only timing mark on my pulley and I guess yours too? I should have a few timers spare in the micro for degree resolution but reducing down to +/- 8 deg offset isn't all that much considering a tooth is 10deg so there's still the chance of being out 2deg?

How fast is the delco at pulling timing? 1mhz ADC sampling and 92mhz core freq would guarantee timing is pulled before the next spark event, even over 20,000RPM. Also is delco knock 'count' really a count? From testing with a sig gen, the value tunerpro reports isn't a count but rather the pulse width on the knock pin on the memcal.

Do you know if there is a post with the waveform captures of EST/Bypass/REF ?

[antus]

Im planning on sticking with the GM 6+1 pattern, so notches every 6 Notches, every 60 degrees that are 10 degrees wide. Plus one to signal home at 10 degrees after TDC. So in my case its all in, on getting it right. I put the second 60 degree notch to align with the timing mark on the block and use that as a reference. Meanwhile the trigger is 60 degrees around on a custom mount lining up with the rising edge that is about to be followed by the home mark.

Then I'd run it up and use the fine tune to dial in 10 degrees base timing.

I dont know about the how fast on pulling timing.

The knock counts I am referring to is additional code in 12p that logs counts to a table in NVRAM so that you can road drive and see if you are knocking in an area to reduce the timing a little there. Its not often used as the original system isnt that great, but if someone was to run this saab system I think it'd be great for road tuning and fixing knock before destroying your engine. You are correct its not a count of individual events (at least I dont believe thats possible the way it works) but you can see if you are knocking, and where, after you have been for a drive and without having a laptop connected.

The tables are ESC - Knock Count Logging 20-100kPa and ESC - Boost Knock Count Logging and the feature needs to be enabled and the ecm needs to be running in nvram as literally using those tables as ram, the tables are larger than the size of the native ram in the ecm.

im not aware of one, but I believe EST and REF are 5 volt square wave, Bypass is a 5v local level. There might be some useful things to know in VL400's thread here, including that controlling dwell from the module made a significant difference. This module was very good, but is no longer in production, and possibly overkill with a high price. Being implemented with less hardware and lower costs, but still being able to configure trigger pattern, outputs, and some method of dwell control, even its just an adder for more dwell for bigger coils, not a fully programmable table would be perfect for most people. viewtopic.php?f=11&t=2913

The hard part with the dwell is guessing the next event timing and starting the dwell early before the delco asks for it.

[marantz2600]

I just captured a waveform from my Cavalier which seems to have the same 6+1 type of system.

My system is coil driven. It is not a Hall effect sensor.

I know the thread is old, but here it is for reference.

A0001DS.png

In trying to find out what the reluctor looks like I found some pictures. Seems to be just a wheel with notches in it.

https://www.bkracing.shop/product/6-plu ... se&q=false
https://www.cobaltss.net/forums/attachm ... 113edd.jpg

[antus]

Cool, thanks

[Poida]

I just captured a waveform from my Cavalier which seems to have the same 6+1 type of system.

My system is coil driven. It is not a Hall effect sensor.

SNIP

In trying to find out what the reluctor looks like I found some pictures. Seems to be just a wheel with notches in it.

SNIP

I've been exploring with the Cavalier 4-cylinder wasted spark ignition system.

From everything I have found, the engine reluctor 6-1 wheel is actually part of the crankshaft casting in the form of a thin slotted disk formed between cylinders 2 and 3 and the reluctor sensor mounts into the block.

Not having an engine to see firsthand, I have manuals that show the 2.5 litre engine versions had an integrated sensor mounted directly to the base of the module, but I'm not sure what vehicles they were fitted into. The rest of the 4 cylinder engines have a separate sensor.

The 1989 'GM DISTRIBUTORLESS IGNITION SYSTEMS' workshop manual (16020.05-2) has valuable info to guide through the peculiarities of the various GM systems that did away with the distributor.

Trying to add 30Mb pdf file but the system is rejecting it, saying it's too large. (Maximum filesize per attachment: 52 MiB.)

[antus]

Thats wierd, send me a pm to a location with the file and i'll do some testing and get it sorted if its all good to post.