Real Time logging of Ignition Timing - Accomplished

[Jack B]

It took a while but the goal has been accomplished, we can now log the real timing curve of the viper. We have not done any wot runs yet, but that will come as soon as the rain stops. The method uses the Innovate LMA-3, although, it can also be done with their DL32 or the new SS series unit.

I think Innovate learned as much as we did during the project. The timing function did not work when they brought it out last year and only thru listening to their customers did they get the firmware/software to work. For anyone that has a functioning Innovate system the install for timing hardware couldn't take more than four hours, on the other hand Joe Dell could do it 30 minutes.

Bascially an optical sensor picks up tdc and inturn is referenced to #1 coil pack. The timing that is logged is precise. That is the simple part, it took a whole bunch of filters and voltage dividers to get it to work.

Here are some of the uses where it might be useful:

1. To set real timing on any piggyback pcm controller, everyone with a piggy back controller is now guessing to a certain point.

2. SC, turbo and NOS cars can precisely graph the timing relative to peak torque thus pulling and adding timing exactly where needed. This would apply to NA cars, but, not as critical.

3. Does the PCM add/pull different amounts of timing for different temperatures - we will find out. If that is the case, we are going to know precisely how to modify the timing for the temperature.

4. Even those of you with a full controller like the AEM are still guessing on the timing. Perhaps the curve is ragged as a result of a faulty algorithm. Fix the curve and possibily pick up some hp under the peak.

5. Dozens (maybe more) have had or still have high speed misses in there cars, those misses are probably related to some point on the timing curve, we are going to find out and then possibly modify the timing curve to bypass the miss.

There are probably a bunch more, but, that is a start. I will compare actual timing with the VEC output and see if the changes are linear as expected. I'll post some of that info next week. What would be intertesting is get one of those cars with the high speed miss and see if the issue is associated with the timing.

[ViperJohn]

Great info, keep us posted.

[Joseph Dell]

35, Jack. I had to stop to pee.

Great work, though. can't wait to see the results!

JD

[Disturbed]

Any updates on this? I'm working on this AEM install on a N/A Viper and between funky O2 sensor readings and not really knowing what the base timming is, it's getting to be a real PITA.


What do you use for the O2's. The AEM O2 sensors suck...at best.

[dave6666]

Where does that optical TDC sensor install?

[EllowViper]

So if I'm hearing this right, the PCM timing (open and close loop) that I'm logging via the ODB-II data stream is not accurate enough? I know I have to match the PCM timing at a specific RPM/MAP with what the VEC is adjusting to get the actual delivered spark. It would be good to see the net delivered spark (VEC plus PCM calculated) vice having to do the math...Is this what you are accomplishing??

[DAMN YANKEE]

JACK,
Remarkable news, can't wait the hear how you did it. I recommend we code word the process "install wipers", no wait a minute, too technical...lets use "open gas cap" as I have learned as of late that it is a very bad idea to actually tune your own car.

[Jack B]

Originally Posted by EllowViper

So if I'm hearing this right, the PCM timing (open and close loop) that I'm logging via the ODB-II data stream is not accurate enough? I know I have to match the PCM timing at a specific RPM/MAP with what the VEC is adjusting to get the actual delivered spark. It would be good to see the net delivered spark (VEC plus PCM calculated) vice having to do the math...Is this what you are accomplishing??

The PCM is way too slow, when you try to log rpm, throttle position, timing and map it is crude at best. If you get a smooth curve the logging program is adding smoothing (behind the scenes) which changes the exact values. In short, at WOT the obd output is worthless with the operating speed of our PCM's. On the other hand, I am logging on a new Solstice GXP, that PCM is fast enough that you can log 16 channels and still produce very accurate logs. That is a Bosch high speed controller.

The other issue you hit is that the true delivered advance is not available. The optical sensor was mounted on a bracket I fabricated and placed on the front cover. It picks up off of a reflective stripe put on the balancer.

[EllowViper]

That's what I thought. I'm playing catch-up "after the fact" and you are tapped into the actual delivered spark timing in reference to the crank position. I'd like to see what the actuals vice computed values are.

[PowerKraus]

Originally Posted by Jack B

The PCM is way too slow, when you try to log rpm, throttle position, timing and map it is crude at best. If you get a smooth curve the logging program is adding smoothing (behind the scenes) which changes the exact values. In short, at WOT the obd output is worthless with the operating speed of our PCM's. On the other hand, I am logging on a new Solstice GXP, that PCM is fast enough that you can log 16 channels and still produce very accurate logs. That is a Bosch high speed controller.

The other issue you hit is that the true delivered advance is not available. The optical sensor was mounted on a bracket I fabricated and placed on the front cover. It picks up off of a reflective stripe put on the balancer.

Will you share the timing you see during a WOT run, in 500 rpm incrediments?

[PowerKraus]

Great job of investigation; which leads me to a few comments or questions. And maybe better understand what you are saying. Having gadiated from Texas A&M, I am happy to be able to use this typewriter, but it is much more challenging to read and answer world problems

Originally Posted by Jack B

It took a while but the goal has been accomplished, we can now log the real timing curve of the viper. We have not done any wot runs yet, but that will come as soon as the rain stops. The method uses the Innovate LMA-3, although, it can also be done with their DL32 or the new SS series unit.

I think Innovate learned as much as we did during the project. The timing function did not work when they brought it out last year and only thru listening to their customers did they get the firmware/software to work. For anyone that has a functioning Innovate system the install for timing hardware couldn't take more than four hours, on the other hand Joe Dell could do it 30 minutes.

Bascially an optical sensor picks up tdc and inturn is referenced to #1 coil pack. The timing that is logged is precise. That is the simple part, it took a whole bunch of filters and voltage dividers to get it to work.

Here are some of the uses where it might be useful:

1. To set real timing on any piggyback pcm controller, everyone with a piggy back controller is now guessing to a certain point. I don't think we are guessing when it comes to the base timing used for WOT, or is there a question as to the accuracy of timing provided by Sean? Certainly a guess at every other load/rpm.

2. SC, turbo and NOS cars can precisely graph the timing relative to peak torque thus pulling and adding timing exactly where needed. This would apply to NA cars, but, not as critical. OK

3. Does the PCM add/pull different amounts of timing for different temperatures - we will find out. If that is the case, we are going to know precisely how to modify the timing for the temperature. You can bet it does, I've never seen a PCM that ignored temperature.

4. Even those of you with a full controller like the AEM are still guessing on the timing. Perhaps the curve is ragged as a result of a faulty algorithm. Fix the curve and possibily pick up some hp under the peak. OK

5. Dozens (maybe more) have had or still have high speed misses in there cars, those misses are probably related to some point on the timing curve, we are going to find out and then possibly modify the timing curve to bypass the miss. Yes ! that would be great!

There are probably a bunch more, but, that is a start. I will compare actual timing with the VEC output and see if the changes are linear as expected. I'll post some of that info next week. What would be intertesting is get one of those cars with the high speed miss and see if the issue is associated with the timing.

I've logged a few high rpm misses with the VEC. The timing delta as entered in the VEC should show up on the log; but it does not. It actually appears there is a moment where/when a timing signal is either not sent/delivered, and is interupted. Which is strange, I beleive this DIS system is similiar to GM's that was used on the Grand Nationals. Where your crank sensor in combination with the cam sensor, is only needed when starting the car. For example, if you start your car, you could unplug the crank sensor and it will run fine. The crank sensor used in conjuction with the cam sensor is simply used to establish the #1 cylinder and initiation of spark to fire. However, the cam sensor is needed on a running basis.

I appreciate what you've uncovered and look forward to more detail.

[Jack B]

Originally Posted by PowerKraus

Great job of investigation; which leads me to a few comments or questions. And maybe better understand what you are saying. Having gadiated from Texas A&M, I am happy to be able to use this typewriter, but it is much more challenging to read and answer world problems



I've logged a few high rpm misses with the VEC. The timing delta as entered in the VEC should show up on the log; but it does not. It actually appears there is a moment where/when a timing signal is either not sent/delivered, and is interupted. Which is strange, I beleive this DIS system is similiar to GM's that was used on the Grand Nationals. Where your crank sensor in combination with the cam sensor, is only needed when starting the car. For example, if you start your car, you could unplug the crank sensor and it will run fine. The crank sensor used in conjuction with the cam sensor is simply used to establish the #1 cylinder and initiation of spark to fire. However, the cam sensor is needed on a running basis.

I appreciate what you've uncovered and look forward to more detail.

If you want some more background, download the Innovate DL32 manual, it shows the circuit, Basically you are juxtaposing top dead center and actual spark. I can tell you this, the the coil pack and the crank sensors have a very unusual waveform. You actually have to filter them to use them because they are more of an oscillating pulse, not a single pulse. I have shots of both taken with an oscilloscope and they are unique.