DIY LTCC or similar system for LT1s

[vilefly]

Hey spfautsch,
I noticed a youtube post where a guy did an EFI connection conversion, but had problems above 5000rpm with the crank sensor. His pcm read a much lower RPM when he hit 5200 rpm. So I am wondering if this might match your symptoms you have been having. I suspect there is a resonant frequency problem going on. Perhaps a shielded input or filter should be installed inline to prevent this.

Another thing, free LS pcm tuning is now available through PCMhammer and LS_droid. I think people will beat a path to your door if you make a camshaft output that is synchronous with the LS motors in terms of degrees. Despite all these changes......THIS PROJECT IS NOT OBSOLETE! I will be digging up some LS data so that you may know the timing for this.
LS1_timing_chart2.jpg

Looks like a simple signal. Goes high at 360* and low at 720*.
People have been running the ls pcm with the 4x crankshaft reluctor found in the 96-up vortech and lt1 engines with success. I bet if someone grabbed a pcm from an LS motor, and programmed it to a vortech crank input, it should work with coil on plug operation.

[spfautsch]

... a guy did an EFI connection conversion, but had problems above 5000rpm with the crank sensor. His pcm read a much lower RPM when he hit 5200 rpm. So I am wondering if this might match your symptoms you have been having. ... Perhaps a shielded input or filter should be installed inline to prevent this.

I doubt that would be at all related. I'm fairly confident I identified the source of the stumble as an oversight I made in transitioning correctly at speeds where coils needed to be dwelled right around the 90 degree multiples (i.e. more than one TDC event ahead).

Bear in mind that the opti has open collector outputs that are quite different from the outputs of un-amplified hall effect sensors in terms of noise immunity. I've no first-hand experience in the field but I believe most factory harnesses use either twisting or shielding around hall sensor pairs due to the nature of the signal. I noticed no such care taken with the opti lines when I spliced into my harness.

Another thing, free LS pcm tuning is now available through PCMhammer and LS_droid. I think people will beat a path to your door if you make a camshaft output that is synchronous with the LS motors in terms of degrees. ... I will be digging up some LS data so that you may know the timing for this.

That would be very helpful (more timing info). Assuming the signal transition coincides with #1 and #6 TDC, a synthetic CMP signal would likely be so trivial to implement I'd do it just for fun. The only drawback would be the controller would need to detect sequence before the synthetic CMP signal could be properly timed, and then the PCM would finally be able to detect sequence - possibly 360 degrees later. This means start time could be higher than with a native CMP signal, and it could potentially trigger a CKP or CMP code in the LS PCM.

I've been casually following the progress of the LS_droid stuff. It's a very significant milestone, but to be honest PCM swaps for a "nostalgia" gen 2 LT-1 wasn't the first thing I thought would result from it. My first thought was it will just make LSx engine swaps that much more ubiquitous (if that's possible).

Whatever the case, thanks for bumping the thread. I just got the tracking # on my reciprocating assembly yesterday afternoon so I'm finally going to be able to start building in a week or so.

[vilefly]

It requires 1.5 seconds of cranking without a cam signal to set a code p0342 or p0343. If it syncs within this time......no code will be set. Codes can be disabled in the pcm with a proper flash programming. Average crank time is about 150-230rpm. So, by that reckoning, roughly 4 crank rotations without a cam signal will set the code.
So, I think we are good.

And now....a more detailed waveform for the crank wheel. Per degree of rotation.

ls cam-24x crank signals.jpg

[lionelhutz]

The truck 0411 applications with a 4x crank sensor still use a distributor. You're not going to get CNP from that setup. It would require a 24X wheel under the timing cover.

[spfautsch]

And now....a more detailed waveform for the crank wheel.

I'm not sure that's of much use.

Before I risk putting my foot further in my mouth, can you clarify your intentions? Are you saying it's possible to have the PCM handle sequencing of the ignition coils with a 4x crank reluctor?

Edit: Sorry - after re-reading what you said here...

People have been running the ls pcm with the 4x crankshaft reluctor found in the 96-up vortech and lt1 engines with success. I bet if someone grabbed a pcm from an LS motor, and programmed it to a vortech crank input, it should work with coil on plug operation.

I'd say I'd bet not.

From a simple search I'm seeing the 4x wheel is non-indexed, so the CKP/CMP waveforms would look like this?

vortec-CKP-CMP-waveform.jpg

Or, were you thinking of using the diy-ltcc to handle sequencing the coils and just picking the EST signal from the PCM for timing?

On the vortec setups the engine runs with injectors in batch mode until it has picked up sequence from the CMP. Having the PCM control the coils in such a setup would be wholly different because there's no margin for error with sequence. This is why the 24x system has an indexed crank reluctor.

Honestly the only variable I would need to know is if the CMP signal goes high on #1 or #6. I think it's safe to assume #6, but I'll wait until someone comes forward wanting to try.

[kur4o]

Ignition with 8 coils is only possible with 24x crank and 1x cam signal. A signal translator from opti signal to 24x,1x signals will allow the use of coil on plug and ls1 pcm.
High res opti signal can be used for very accurate generator signal trigger upto a degree. The ls1 wheel is 360 degree and the 1x cam sensor in ls1 shows if the #1 or #6 is at TDC.

A high res degree based generator will also fix the change in frequency, the higher the rpm the shorter the signal`s duration, which will be very hard to encode with time based generator.

So for sync you need to find #1 or #6 TDC and start the 24x output from there in repeating pattern each 360 high res counts and changing the state of 1x signal from high to low each 360 degree also.

I think you will need 2 low res slots to determine crank position and just wait for 360 degree trigger to output the 24x signal.

I do have a busted crank with the wheel laying around so can measure the degrees of the slots manually.

[vilefly]

Well, I don't know enough about programming the 0411 pcm to know if it won't work. Retroprogramming, if you will. I figure somewhere in the memory, the math for crank velocity can be accessed from a vortech 0411, and spliced into a ls1 binary, and possibly work. This is idealistic thinking, of course. I expect a loss of resolution, but wonder if it can be managed.

My original thought, was to just use the pcm to handle timing and misfire detection, while letting the the diy-ltcc do its thing. Then I started imagining the above scenario. I have no immediate plans to do this just yet, because I need to learn a whole lot more about 0411 binaries.

If everyone says it can't be done, then so be it.......I will stop pestering. Kur40's post beat my reply before I could press enter.

[spfautsch]

Ignition with 8 coils is only possible with 24x crank and 1x cam signal. A signal translator from opti signal to 24x,1x signals will allow the use of coil on plug and ls1 pcm.

Assuming this is accurate (I tend to think it is), a hybrid setup where the diy-ltcc handled coil sequencing & dwell, generated a synthetic 1x cam signal, and picked off the EST signal for spark advance would be incredibly simple with a 4x crank reluctor.

High res opti signal can be used for very accurate generator signal trigger upto a degree.

A high res degree based generator will also fix the change in frequency, the higher the rpm the shorter the signal`s duration, which will be very hard to encode with time based generator.

I thought about addressing the idea of synthesizing a 24x+1x signal in my previous post, but decided against it. My concerns are with the "lossy" nature of the high res opti signal - I've logged a lot of instances where 1 or 2 degrees are lost or gained per TDC signal. I fear synthesizing a 24x signal from this info could falsely trigger misfire detection in the LS PCM. But I'd be willing to try if the right person came forward with a test mule.

I think you will need 2 low res slots to determine crank position and just wait for 360 degree trigger to output the 24x signal.

Maybe not. If we can control both the CKP and CMP signals going to the PCM then it would probably work to just wait until sequence is detected to begin feeding both signals. This would (theoretically) add a maximum of 90 degrees of crankshaft rotation to the start sequence.

I do have a busted crank with the wheel laying around so can measure the degrees of the slots manually.

vilefly's image of the 24x wheel's waveform in post #415 should give adequate information to work from. It's just a variation of 3 and 12 degree sequences. The beauty of it is in it's "native" form the PCM would never need to see more than 90 degrees of crank rotation to determine sequence. Ingenious.

Doing a opti to 24x signal converter would be a relatively simple fork of the current code. As I mentioned, I would be willing to try it if the right person came along with a test mule they wouldn't mind doing some trial-and-error wiring on. It's not a conversion I'd anticipate doing myself - at least not in my current Y body where I'd have to get the CCM and the ABS/ASR modules talking, etc. I don't see that the gains to be had with the newer PCM would be worth the sacrifices in overall vehicle function and originality. Not to mention I'd really like to get my car running so I can complete this controller project!