DIY LTCC or similar system for LT1s

[spfautsch]

Another update - I'm still alive and well, and still employed. I hope everyone here is enjoying the same.

Made some significant progress today running fully interrupt driven dwell (i.e. no hazard of routine starvation). In the process of locating the issue I implemented some real-time profiling that helped me identify how dwell wasn't being initiated by the controller around the transition points in certain conditions. This is where dwell needs to extend beyond one cylinder firing event or simply put more than 90 / 180 / 270 degrees of crank rotation. After sweep testing a few more routines I found a couple possible issues that were easily fixed.

The result is something the butt-dyno says is damned good, but the profiling numbers say still lacks ever so slightly. Possibly a missed dwell happening during decel events where I lift my foot off the go pedal.

Hoping to put a couple hundred miles on it Monday and will report back. I'm excited! Not only because it's no longer exhibiting the rev-limiter type behavior around the transition points, but I'm also happy to report the microcontroller seems to have more than enough processing power to compute dwell from time to degrees on the fly once per combustion event (using the 16 bit RPM value) as far as I've been able to push it (around 5800 rpm).

[NomakeWan]

That sounds awesome. I'll admit I was on the fence about running any of the new stuff due to the dwell bugs (the older branch seemed more robust, if far less feature-rich), but it seems like those last few remaining issues are going to be ironed out in short order. I'll need to hit you up to swap to the new board. I've had a full set of coils and harnesses laying in my lab for months now, just need to get spark plug wires and figure out a way to mount them in the engine bay. Well, that and buy up a few more connectors so I can try to fab up a "plug and play" harness to let me switch between your controller and the Opti for testing. I prefer to modify as little of the original wiring as possible, and thankfully the majority of the car's connectors are still available on Mouser or Digikey, so yeah.

It may sound like crickets around here but I still watch this thread and the EEHack/Flashhack threads like a hawk reading every update. Glad you're still safe and still employed, and hope it stays that way.

[spfautsch]

For plug wires I generally like making my own - a set of Taylor 46052 straight boots (search amazon) and I generally go to Autozone and buy a universal set of 8mm wires. With headers I needed 90* boots on all but 6 & 8 plugs so I got a Duralast 1601 kit for like $24. Heck they may even have a universal kit with the LT-1 straight boots (the Taylor PN above). Get a plug wire crimper and a can of silicone spray and you're all set.

There are no major functional changes to the new board, just the addition of an ICP header and provisions to add jumpers to disconnect the ground planes for the leds. I PM'd you details - all you need is a .1uf cap for the ECT sense input.

I don't know what possessed me, but I invited my wife to go for a ride today because I thought she'd be helpful in switching a parameter I wanted to test. Got tired of hearing "stop trying to make me puke" real quick. After dropping her off at the house and re-connecting all the USB cables she yanked loose, had a successful test drive. Spun it up slowly to about 6400 in 2nd and couldn't feel any missing even though the profiling metrics showed 2 or 3 missed dwells. Hoping to have a more productive day tomorrow logging my commute.

[spfautsch]

I was hoping the problem was solved, but after a 30 minute drive yesterday morning with zero errors counted in profiling, the wheels fell off.

Initially I started looking for obvious causes such as memory corruption / leaks or a thermally triggered issue. But after additional testing, and then sleeping it over and adding more profiling metrics, it looks like the possible cause is that it's seeing quite a bit more EST falling edge signal counts (i.e. fire the currently dwelling coil) than low res (tdc) counts. This might be the last 1 percent that takes more work than the first 90 percent. But I'll figure it out, or die trying.

If anyone want's to pitch in on the effort PM me and I'll send you the current source. Oh, wait it won't matter b/c no-one has a running setup. :-[

Aside from this one (very) major problem, starting, ECT detection (the PCM switches between two different tables), and all the other minutiae have been working like a clock.

[Joshie225]

I'm sorry. How did the wheels come off? I'm not understanding what's happening.

I'd like to be one of the people with a running setup, but the world as it is gives me scant opportunity to use the car so my motivation to improve it is nil.

[NomakeWan]

I'm sorry. How did the wheels come off? I'm not understanding what's happening.

I'd like to be one of the people with a running setup, but the world as it is gives me scant opportunity to use the car so my motivation to improve it is nil.

It’s a figure of speech. He means everything went to shit.

[Joshie225]

Hmm. How can I say this nicely? ...I can't so...

What I'm interested in knowing is what went wrong.

[spfautsch]

How did the wheels come off? I'm not understanding what's happening.

Let me clarify. The problem I'm having with the current development branch seems to be intermittent / aphasic (is that even a word?) / possibly temperature related. It's also possible the board I'm currently running in the car has had something happen to it - it's the first of the manufactured boards I put into operation and I later changed the coil driver resistors to 100 ohms thinking there was an electrical problem that was purely imaginary (one of the joys of being an un-educated country bumpkin with an uncontrollable penchant for tinkering). If I end up pulling all my hair out with the current development branch I'll temporarily mothball it and verify that 0.9.17 still works as well as I remember it.


The project in general is not doomed.


I've just put too much effort into this very full-featured branch (it senses coolant temp, has 2x 3d tables for dwell with temp and map compensation, calculates dwell from time to degrees on the fly, etc.) to backport all the bug fixes I've added since then (0.9.17).

I think I've devised a way to monitor what I believe the problem is (noise on the EST input) and let me dump the recent controller activity after several misses have been picked up. But instead of driving it I'm posting here because it's threatening to storm and I feel no reason to push the envelope (yes, it's my garage queen and I don't drive her in the rain, or worse).

... the world as it is gives me scant opportunity to use the car so my motivation to improve it is nil.

I suppose I can understand, depending on what your motivating factor is (fear, loss of income, loss of a loved one). Loss of income would wipe anyone out. Loss of a loved one may make one long for the distraction of a project (my father passed away a few weeks before I first drove the car over 100mph with the prototype breadboard duct-taped to the windshield). I don't mean to sound like a complainypants, but my current situation is that my employer handed us a 20% pay cut ten days after laying 40% of the company off. I get it. Luckily gas is almost as cheap as I remember it when I first filled up my '74 Camaro in 1987. If it weren't for the price of gas I probably wouldn't even be driving it, and there would be no progress on testing.

I'm going to stop now before I get into a discourse on political matters. Let me just summarize my thoughts about cornavirus thusly - I never knew I had Tourette's syndrome until this hit. It takes phenomenal restraint each morning to resist throwing my coffee mug at the television, even though I'm fully aware it is an inanimate, soulless object. All I want is the goddamned weather forecast, and when you commute a minimum of 115 miles a day the "app for that" sucks balls.

[Joshie225]

You've coded an ignition system that is far more complex than I thought necessary. The coil needs a certain amount of time to charge fully. If the voltage is low it needs more time. If the MAP is low less spark energy is needed, but why add the complexity? Component life? Before considering dwell vs temperature I would want to see that the coil and driver are significantly temperature sensitive across what -40°C to 120°C? Forgive me if all this was explained and I've missed it. I don't understand how the project arrived here and it's not really my business so don't feel the need to explain.

My motivation for playing with cars was that I found it fun. With driving so little, and not feeling much desire to drive, the car simply gets less of my attention.

I'm sorry you've lost your father. I'm grateful that my father is in good health as he approaches 80 this June.

I wish you luck. Much of my working life a 20% pay cut would have been rather painful. Shit, it was painful. I once was laid off and then worked for 70% of what I'd been earning. It was precisely enough to pay the bills with nothing left over. You might call it economic social distancing as there just wasn't any money for going out.

[spfautsch]

Before considering dwell vs temperature I would want to see that the coil and driver are significantly temperature sensitive across what -40°C to 120°C? Forgive me if all this was explained and I've missed it.

See the spreadsheet in this post. <link>

I know there's a lot of off-topic discussion here so to summarize, I found that different coils react differently to both temperature and voltage. Implementing a voltage vs temperature vs dwell time table that can be specific to each part # seemed to be the most intelligent way to accommodate anything and everything ppl might want to use this for.

I don't understand how the project arrived here

It's all here, and there's a lot of it. If you choose not to read, what purpose does complaining serve?

[Joshie225]

Thanks for the spreadsheet. Indeed there is much more dwell variation with voltage than with temperature.

Expressing not understanding isn't complaining. I better understand now. And with the understanding I'm dropping out.

[spfautsch]

Thanks ... Indeed there is much more dwell variation with voltage than with temperature.

Had you read any of the surrounding data you might have noticed where kur4o and I discussed how the D585 (aka hot truck coils that everybody wants to run) are significantly more non-linear with temperature. Also, and a whole lot more important is they lack "soft" dwell limiting circuitry. This means when you give them too much dwell time, they fire the plug with full energy regardless of what the PCM is commanding. This poses a huge problem, and also explains why the Bailey LTCC setup that uses fixed dwell without temperature compensation (or the ability to customize the dwell target / table) was not recommended for use with these coils. Google: "bailey ltcc kickback".

Expressing not understanding isn't complaining. I better understand now.

I've been married to a half Phillipino, half Irish woman for going on 27 years. I'm pretty sure I can identify complaining. It's sometimes what we do as foreplay. I respect people that complain - means they're thinking for themselves. Seems a lost art these days.

And with the understanding I'm dropping out.

That's your choice, but I hope you will reconsider once sanity returns to the general population.

Testing yesterday with some rudimentary misfire detection revealed an ignition event where subsequent to a massive "hiccup" at ~2200 rpm, a coil was "fired" 3 times, starting at 39 degrees, then again at 31, and then finally at 30 - where the spark advance tracker and eehack were showing spark advance was at the time. All within the same combustion event on #3. I'm extremely happy to see logging data to show me this because it's indicative of a noise situation on the EST circuit. Minimally, this gives me an idea of what area the problem lies.

I have a more robust debouncing algorithm for the EST line ready for road testing tomorrow, as soon as the rain, and the massive pond in my back yard dissipate.

[spfautsch]

Well I'm back to replying to myself here... but, couldn't help but share my enthusiasm.

Sorry it's taken so long but life events and weather have been less than conducive to data collection efforts.

The multiple firing event thing seems to have been easily fixed by oversampling the EST input on the falling edge. This explains all the seemingly random detonation I was seeing at high ambient temps while cruising in 6th gear @ 75mph.

Unfortunately this still wasn't the last remaining bug. After more work profiling the program execution I was able to clearly identify there were dwell charging events being mysteriously missed. It took several days of testing and lots of pulled out hair to discover what I believe was a race condition between calculating dwell and initiating dwell. Building a buffering / queuing mechanism seems to have resolved the problem, and concurrently made me wonder why the hell I didn't implement one from the get-go.

The evening commute was almost flawless with the fix in place. The "almost" qualifier deserves the explanation that I haven't fine-tuned IAC vs A/C pressure tables so when the compressor engaged I felt surge. I also felt / heard something strange when I wound it out to 5800rpm but that could have been any number of things.

I still need to test all the functional modes to make sure it works as well with and without logging and / or profiling disabled, but I'm pretty sure this was the last major turd left in the punch bowl. If things go well tomorrow I hope to upload a release.

[NomakeWan]

Congrats on more dragons (seemingly) vanquished! I hope the next few tests go well and look forward to the new code!

[spfautsch]

I was unable to test as thoroughly as I'd like. Last night I was greeted with a thunderous racket after closing the garage door. Seems the plastic wall brackets for my wire shelving decided to let go, showering my test mule with both of my biggest beer coolers. Thankfully one bracket didn't give up, allowing the washer boxes to retain their purchase and not end up bouncing off of (and likely cracking) the plastic targa top lense. No harm but some scuffs, but I spent most of the day cleaning up from that. Rest assured if I encounter any problems in the upcoming days I'll sound the alarm loudly.

Here's the 0.9.33 release. As mentioned previously during my 60 mile ride home last night it ran as well as I can recall since the 0.9.17 release that used essentially fixed dwell and was incredibly rudimentary by comparison. I smelled quite a bit of burning clutch and a few instances of butthole puckering understeer were encountered. If it missed any dwell events it was around 4700 rpm and was completely unnoticeable (but easily fixed by using an array based queuing mechanism).

I believe there's a minor bug in the logging "repeat" function but on a scale of 10 I would characterize my concern with that at about 0.2.

Since my employer secured a PPP loan and temporarily restored our wages I'm going to reinstate the challenge gift card, at a value of $100. I know that's not a lot for all the trouble involved, but it's what I can do. I'd really love to know that this works on another LT-1 before I lose interest.

Edit: It doesn't have to be an Amazon gift card, I can also paypal the prize, or mail a check. Winner's choice.

In thinking about the state of the project it dawned on me that some research may be needed to extract the cranking spark advance table from the eeprom based LT-1s and possibly adapt to this application. If anyone wants to pitch in with that I'd really appreciate it.