DIY LTCC or similar system for LT1s

[kur4o]

I got everything hooked up in the engine with ign on, while spinning the opti, so the est line should be fully functioning.
By speaking to the est how it shoud be : connect it to the icm module all the time, disconnect it completely from the icm module. Can I run the engine for testing with stock icm and coil and connect cntroller to monitoring proper operation.

I am sure if the opti is spinned through the engine cranking there will be no problem at all.

The worst case scenario will be some very brief crank time that enable this condition and ign is left on. Than some coil can be left in constant charge state.
I think the easiest way to prevent this is to turn low all outputs in the moment the low res signal is lost.
A deeper analysis migth be needed to figure what might be causing it. Is it loss of sequence loss of est signal.
I think it is the start of charge happened for a coil but the est line never got triggered. So it hangs in semi triggered state. The firing sequence didn`t completed by est but the coil have started to charge.

I will keep thinking about it and assemble everything for final testing. Still waiting for that china crap connectors. WIll have to figure a coil mounting too. driver side is piece of cake but passenger side is really cramped out.

Edit:
I make a boil comparison of the 71* and 77* thermostat I have. The 71* supposed to be hypertech 160 degree is also made by motorad and have wider opening [34mm] and stiffer spring, so it doesn`t open too deep, while the 77* have slightly smaller 31mm opening but opens much deeper than the 71* due to weaker spring.

[spfautsch]

I got everything hooked up in the engine with ign on, while spinning the opti, so the est line should be fully functioning.

Opti on the engine, or loose?

Have you had it running since your optical pickup failed? If not, you really must do that first to make this a fair comparison.

... the est how it shoud be : connect it to the icm module all the time

You'll want the icm module disconnected when testing the controller. The additional load it presents to the EST circuit could cause an impedance mismatch.

I make a boil comparison of the 71* and 77* thermostat I have. The 71* supposed to be hypertech 160 degree is also made by motorad and have wider opening [34mm] and stiffer spring, so it doesn`t open too deep, while the 77* have slightly smaller 31mm opening but opens much deeper than the 71* due to weaker spring.

I have zero concerns with the 77c motorad thermostat. I was sitting in a line of 12 cars with their cooling fans all running on high, in the mid-day sun, in oppressive humidity (i.e. massive a/c load) and 34c ambient temps and moved less than 45 meters over nearly 30 minutes. The gritting of teeth was in hoping the ignition controller didn't overheat and cause me to have to push the car out of the queue (in the 93f heat + 90% humidity, not to mention embarassment - this Wendys is less than a 3/10s of a mile from the St Louis Lamborghini / Maserati dealer).

[kur4o]

A little update.

I finally have running engine. I connected in parallel controller only with stock ignition and the est line signal is good enough for the ligth show. I guess hand rotating opti is not a good idea for testing anything. Will do more brief crank testing to try to similute hand rotating opti condition. Hope it never shows up again.

I made some custom opti venting and plan to use the egr valve to control the flow through the opti. Still thinknig what will be the best approach and conditions for running the vent. Time based from engine start, IAT based, Coolant based, Map based or mix of all.

China crap connectors arrived Friday and don`t look that bad, A little flumsy but not that horrible. Adding wires for coils and making some coil mounting is the last in the list.

Looking for much better idle quality and at least 20hp in the high rpm increase.

A note on the controller grounding. I think that opti low res ground path will be enough for the controller. Since it have common ground accross the board if I add a second ground to chasis or battery it will lose ground potential within the pcm. I will trace the ground within the pcm to confirm it is sufficient for extra 1.5 amp load.
A future improvement will be to add a constant power supply with sleep mode with separate ign sense input that read coils voltage. I am not sure how hard will be to divide the two.

[spfautsch]

Sounds great, I'm glad you're getting closer!

A future improvement will be to add a constant power supply with sleep mode with separate ign sense input that read coils voltage.

I put an immense amount of thought into going to an always powered controller versus having power switched by the IGN circuit. I'm still open to discussion, but I'm pretty convinced there's zero benefit to trying to make the controller stay powered up all the time. Primary being that (at least with manual transmission cars) the engine can be turned backwards by vehicular inertia in certain stall conditions. If you stall it, let it re-discover sequence. I'm a bit too impaired to go into extreme detail here, but trust me, it's been considered extensively and I still find no reason to dig it's corpse back out of the dirt.

[kur4o]

The separation of Ign on with Voltage sensing will simplify wiring and give some burned fuse indicator.

I guess there will be no sleep idle mode that can be used with ign on.
A reboot is mandatory when ign on is sensed.


The next idea I have is to make some smaller form factor board with smt components and integrate it within the pcm. Use spare pcm pins for coil output. We need only 8 of them. All the wiring will be to the pcm pins so there is no need for anything else. The pcm have 5v output already so there will be no need to add voltage regulator.

I found some really cheap board manufacturer, You can take a look at them
here

I hope you like the idea and give it a try or think about it.

[spfautsch]

The separation of Ign on with Voltage sensing will simplify wiring and give some burned fuse indicator.

In what way? Why is a blown fuse indicator needed?

I guess there will be no sleep idle mode that can be used with ign on.
A reboot is mandatory when ign on is sensed.

It's not a "reboot". The bootloader jumps directly to the setup() routine when first powered on.

Again, I'm open to debate the merits of an always powered controller, but you're going to have to have a lot better arguments than what you've brought thus far.

The next idea I have is to make some smaller form factor board with smt components and integrate it within the pcm. Use spare pcm pins for coil output. We need only 8 of them. All the wiring will be to the pcm pins so there is no need for anything else. The pcm have 5v output already so there will be no need to add voltage regulator.

I hope you like the idea and give it a try or think about it.

I considered going to surface mount tech. The first thing you need to consider is the minimum quantity you need to commit to in order to have a SMT product manufactured by a third party at a cost that makes it feasible. Additionally, I'm turning 50 soon, and I needed bifocals four years ago. So it was a very short debate. I have no intention of doing anything like that. The through hole board could be reduced in size somewhat, but I fail to see a compelling reason. If it functions identically who cares if it's the size of a slice of cheese or the size of two postage stamps? And it doesn't require a reflow oven to assemble.

I honestly respect and appreciate your suggestions. They just don't align well with a do it yourself project in my opinion.

[NomakeWan]

I hope you like the idea and give it a try or think about it.

Apologies for being blunt, but considering this is an open-source project...why not do it yourself?

I designed an SMD board for a personal project, went through the growing pains of learning how to handle PCB design and optimizing components, and had several prototypes before settling on a design. So I'm well aware of what goes into that sort of development, having done it myself.

And as such, I just gotta say...give it a shot for yourself if you're passionate about having a design that conforms to your personal specifications.

[spfautsch]

I have to bump the thread in hopes someone somewhere is still thinking about / planning to test it on their car.

I've ramped back on daily driving the test mule. The odometer is reading 1733xx and I think it was around 169xxx when the last major code changes were made. The tune is at a place where I would feel comfortable letting my wife drive it to work. The only change I intend to make any time soon related to the ignition controller is to build a new board using the original design 470 ohm resistors in the igniter r-c network. The board I've been running has 100 ohm resistors here.

The start counter reads 199 tonight, so I think I'm going to take it out of the car and stop worrying about this statistic. Of those 199 starts there were about 5 failed start attempts. Of those only one wasn't directly caused by tuning changes and / or my use of kur4o's experimental eehack build (I was trying to perfect cylinder balance). And the one that couldn't be attributed to tune changes might have been a simple flood condition. Since that one occurred I've extrapolated more resolution on my injector flow constant, and raised it from 41.5 to 42.9 lb/hr along with the requisite changes to the prime pulse table. This amount of error in startup fueling along with wild changes in weather could account for the flooded start attempt here.

I'll probably limit my driving to one day a week from here on, and only when weather is favorable. I'm feeling good about the overall reliability - over the past several months it's spent at least 3 hours idling at a complete stop with the a/c on max in Missouri's 90 degree F summer with dew points that make the "feels like" temp > 100. I think if there were a glaring design fault it would have surfaced by now.

The only issues I have remaining with the tune / driveability is a closed TPS "parking lot buck", and what I believe to be a closed loop lean miss that happens in somewhat rare conditions when the IAC closes prior to the A/C clutch disengaging. I can find no reason to even remotely consider this is being caused by the ignition system.

Barring any other test volunteers I'm considering looking for a test mule that can be had for cheap. I've seriously considered approaching youtube's Cleetus to see if he still has / would sell the ragged out f body he bought for $1k and then bottle fed it. Unfortunately Bradenton is 1100 miles away. Not to mention my wife may kill me in my sleep if I buy another car, but it's a risk I'm willing to take...

[kur4o]

I am on the final steps of controller assembly and I am looking for at least quick idle test before doing some coil mounting. Summer time put on hold lots of projects, and the fact that the chinese crap connectors came with some weird useless female terminals.

I didn`t have the time even to do some fine tuning at least for the last 2 weeks.

[spfautsch]

Don't let me rush you - life happens and I completely understand. Today I spent a solid 3 hours dripping sweat while pulling out a crabgrass infestation that I've put off addressing the entire summer because I've been focusing on finalizing my tune. My patience muscles might be wearing out - this thread will be 3 years old in a few weeks and I'd love to finally know whether the effort was worth it or not. But I'll crack open another can of dead bull (works the opposite of caffiene) and work on rebuilding said muscles.

Other than building another controller board, I have no intentions of expending any more effort on the project until I know it hasn't been a huge waste of time and only works on my car. Well, I might try adding an auto ASR disable function. But nothing that's going to impact basic functionality.

I'm also hoping to hear back from the McFarland bunch - maybe they'll be willing to be my guinea pig. But I'm not holding my breath. I'm thoroughly amused by the stuff he does, but I'm also the antithesis of a youtube celeb.

[NomakeWan]

I like Cleetus but he doesn’t seem like a DIY-EFI kinda guy. But hey, I could be wrong. Maybe he’ll go for it.

Anyway, I’m still dogged down by another project that requires my time so I haven’t built my controller just yet. I did however at least grab a chip for it and burn the latest revision of the code to it, so there’s that. Last month I finally did get my car repaired and have been dailying it again since. It’s been 100F these past weeks and as long as I’m moving it’ll stay in the 210F coolant range...but if I just sit idling it with the A/C on it’ll creep up well into 235F+ territory. I’m hoping it’s just crap stuck to the radiator.

Anyway, as always, if I make progress I’ll let ya know.

[spfautsch]

I did however at least grab a chip for it and burn the latest revision of the code to it, so there’s that.

Did I forget to send you the microcontroller(s)?

Last month I finally did get my car repaired and have been dailying it again since.

Right on! They're made to drive, not to look at. I'm having a hard time living with my decision to go back to primarily dailying my 493k mile diesel beater. Since I've stopped logging every drive in the test mule, starts have popped off perfectly every single time and if I drive it without trying to set speed records it's getting 23-24 mpg.

It’s been 100F these past weeks and as long as I’m moving it’ll stay in the 210F coolant range...but if I just sit idling it with the A/C on it’ll creep up well into 235F+ territory. I’m hoping it’s just crap stuck to the radiator.

Having just pulled the cover off the radiator shell to clean the condenser core out, I can say there's a pretty big space between the radiator and the condenser, and I was shocked at the volume of leaves, pine needles, cigarette butts and the like that had built up between the two. None of which are very conducive to air flow.

Also, it doesn't hurt to make sure the hard line coming off the back of the heads isn't clogged with crud / bars-leak. When I cleaned mine out I got a gallon milk jug full of dark chocolate colored "stuff" out of it. If this line isn't flowing coolant freely air will get trapped in the heads, causing pockets of boiling coolant around the rear exhaust valves.

... if I make progress I’ll let ya know.

I trust you will. I'm just getting ready to dive into another less technical project, and am afraid I'll be waist deep in building my workshop (aka the diy-ltcc global headquarters) when someone gets one going and needs technical help and I won't be able to give it the attention it deserves because I'm worn out from wearing my construction hat.

[NomakeWan]

Hey, thanks for the tips! I hadn't thought to check the crossover tube. I'll have to add that to my list of stuff to check.

And yeah, there wasn't a microcontroller in the pack, but it wasn't a big deal since I had to order a bunch of random parts from Mouser anyway, so I just added one to my list. Don't even worry about it. ;)

[kur4o]

I have some good and some bad news.

I finally put all the parts together and made some idle test run with the controller. Started on the second try which is not unusual on my setup. Everything went smooth without issues. Running was like 5-10 minutes upto full warmup with some light throttle to 3-4k rpm.
One thing I noticed was exhaust headers got too hot, not quiet sure about it. Still have to verify the reason about it.


Now the bad part. After letting it cool down for couple an hours I decided to start it again and bump the advance to see if it will help cooling the headers. I couldn`t start the engine anymore and all kind of bad things happened. I was experiencing the same condition as spinning the opti by hand. First it started on couple of cylinders and run like crap. The leds also show that it was running on some of the cylinders. Than it was complete failure to start with some coils` leds freezing on all the time while cranking. Some leds freezing on when I quit cranking and leave ign on. I hope the coils survive the torture. Unplug coils and tried to clear flood the engine, suspecting flooding and to monitor the leds. It was again very weird sequence of firing sometimes some stay on, sometimes all went off during cranking.
Checked pcm log and all opti signal are good and strong

After some in depth analysis what went wrong the only difference was successful try - car was in neutral, ON failure it was in park. Also one of the coils was lightly repositioned to prevent spark cable from melting and was right above the opti harness. Moving the coil to original position didn`t improve anything.

I suspect the slower cranking speeds don`t play well with the current code and complete rewrite will be needed to address the issue. And some safety feature for leaving coil output on, all the time, when it fails to start. I guess on loss of low res signal, wait some time to make sure the engine is not spinning and discharge all coils to off position will cure it.

[spfautsch]

And yeah, there wasn't a microcontroller in the pack, but it wasn't a big deal since I had to order a bunch of random parts from Mouser anyway, so I just added one to my list. Don't even worry about it. ;)

$#!1!!! I'm sorry! Did I short you any other parts?

Started on the second try which is not unusual on my setup.

This bothers me. My first, last and only desire here was for someone to test this on a setup that already runs well. With a relatively decent tune and healthy opti pickup these things fire off almost too quickly.

Running was like 5-10 minutes upto full warmup with some light throttle to 3-4k rpm.
One thing I noticed was exhaust headers got too hot, not quiet sure about it.

This is concerning. Headers always get really hot. Are you implying they were glowing, or ???

If you have a video of this I'd like to see it. Then you'll have 100 USD for your troubles.

... decided to start it again and bump the advance

Can you be more specific with how you bumped the advance, and how much?

I was experiencing the same condition as spinning the opti by hand. First it started on couple of cylinders and run like crap. The leds also show that it was running on some of the cylinders.

Sounds like some pretty serious noise issues. Question, do you by chance still have the stock ignition module and coil connected?

Is there any possibility you have the igniters on your coils wired backwards? I've never tested this possibility, but the coils may function this way but everything is going to be way off (especially spark advance).

I suspect the slower cranking speeds don`t play well with the current code and complete rewrite will be needed to address the issue. And some safety feature for leaving coil output on, all the time, when it fails to start. I guess on loss of low res signal, wait some time to make sure the engine is not spinning and discharge all coils to off position will cure it.

Let's do some systematic troubleshooting before we jump to wild conclusions.

I'm swamped with stuff for the next few days but I will probably have time to PM you some instructions next week. Meanwhile, if you can I'd like to see pictures of both sides of the circuit board and the code it's running. If you don't want to post here PM me and we can figure something out.

I know connecting plug wires to plugs in an f body is a pain with headers, and this is a cruel request. But my suggestion is to step back, disconnect the system and make your car start on the first attempt and run perfectly with the distributor. I've wasted an immense amount of time on my own setup trying to solve problems that I wasn't sure existed prior to the ignition system because I changed too many variables at once (engine rebuild, unknown fuel injectors, imperfect tune, etc.). I have no desire to do that again.