I have been trying to get the car ready to take Red Rocks and Okteenerfest. So far the microsquirt system has been pretty great but has some issues to get worked out. From day one the Air Fuel Gauge for the O2 sensor had an issue. It would mostly just flash 7.4 but would seem to give some readings until it warmed up and then just flash 7.4. However, it was able to be tuned in tuner studio.

We also had an issue with the ECU losing connection with the crank trigger/hall sensor and restarting. We believe that it was in the wiring the for the resistor on the hall sensor that was causing this. What would happen is that it would run great until it got hot and then it was like someone was flipping a light switch on and off very quickly. each time it would come back on, the ECU would think it was starting the car and adjust fuel and dwell ect and the car would just bog out and not run. Then it would eventually kill the wasted spark. I could replace it and it would start back up but would repeat the problem.

So The day started off well yesterday. We replaced the O2 sensor and finally got the gauge to work. We (Cary) had already rewired the hall sensor with a new resistor and the car had been starting and running great but had not been road tested yet. I was hoping that all issues were now resolved and I could schedule some time on the Dyno to get the fine-tuning done.

I took the car out to try and recreate the conditions where things were previously failing. I made it all of about 7 miles, with the car running better than it ever had before it just stopped running.

I looked at the gauges and the Coolant temp gauge was red and showing 240 degrees. The "Coolant temp" is really the air temp measured inside of the throttle body. The temp did not feel that hot and I have never had it go that hot that fast. I did not hear any backfiring. I took the rain hat off of the Throttle body and it did not feel that warm. I had some air in a can (for cleaning keyboards) and I blew it on the coolant sensor and did not see a drastic change like I was expecting. It slowly came back down to where I previously may have been able to restart it. But I had no spark again.

We got it back to the shop and I replaced the wasted spark for the 3rd time now. It fired right up.

I was data logging and captured the event but I am still not very good at reading the logs. I didn't,t see the "flicking on and off" like we did before when the wasted spark was getting cooked, but it does look like something is happening with the dwell.
I have the log in a drop box folder here: https://www.dropbox.com/sh/3osmebnfu61632k/...sSkTj-tlFa?dl=0

Can you take a look at it and let me know where to go from here?

I would post this on the Megasquirt Facebook group. Lots of tuners on there.

https://m.facebook.com/groups/514059662014654?ref=bookmarks

Sounds like the system reads your temps wrong and is causing your tune to go into some limp mode. Maybe you have something set wrong.

First thing first, consider buying the head temp gauge from McMark, and rewiring your ECU so that the Coolant is actually measuring head temp. I have this setup, and it has been very dependable. Using air temp as Coolant is going to cause you a LOT of issues with warmup and idle. Air intake temps just vary too much.

Second, make sure your wire running from the crank hall sensor is shielded. Your setup looks like a simple 2 wire. Replace it with something shielded. Otherwise you are going to pick up a lot of noise from the ignition, which may be part of your issue (in fact, I'd put money on it). also, route the crank fire wire as far from any ignition components as you can.

Thank you Zach! We will give that a try. Did you see anything in the datalog?

So one thing I'm unclear about is when we first got it running it ran great and I put about 1700 miles on it with no issues to speak of and then it developed these issues. I was thinking it was because of the cold solder on the resistor for the hall sensor / crank trigger finally getting hot enough to have an issue. Why would something like this develop it seems like if it was getting interference it would have been getting it from the start?

I’m not experienced enough yet to really interpret data logs yet. And am pretty noob on all the EFI stuff. Learning though. And when you get it working it’s really worth it.

The basics though is that ignition throws a TON of noise, and this can cause issues. Doing anything you can to fight that noise helps. Your crank position kinda has to route through your ignition wires, so shield it up anyway you can.

The other thing. SPEND THE MONEY AND GET IT DYNO TUNED! Make sure your tuner guy knows Aircooled cars and knows to tune it on the fat side. I spent about $600 for my time on the dyno and it was some of the best money I spent on this car.

I’m still tuning idle to get my IAC down, but when I drive the car it’s just smooth through every RPM all the way to redline.

Zach

intake temp and engine temp are two different things. Really surprised you're seeing intake temps that high. I don't think I've ever seen intake temps double of the outside temp.

I am ready to get it Dyno Tuned and I even have it coordinated with a shop and with an air-cooled specialist tuner I just need to get this part sorted out first. Given that it has not had this issue before I think the harness wrap that we put together for it is working well and we taken out where it was having an issue with the solder on the resistor so this point I'm trying to find what kind of safety mode might have been activated once it read the temp so high

I think I may just need to recalibrate that I'm going to try it tomorrow and I'm trying to figure out what kind of safety mode it might have gone into once it read that temp. When I was having the issues with it getting so hot that the solder was my suspected issue, the temps would get to about 200 after running it hard and letting it sit without air cooling it. This 240 reading I got was after driving for about 10 minutes and about five minutes of warm-up time before that. So I'm thinking it needs to be calibrated or I may be having an issue with that sensor. Putting my hand in there when it was hotter I could definitely feel what 200 would have felt like and it was nowhere near that. I just need to watch some more videos and read some more instruction manuals I was just hoping for somebody point me in the right direction since there's so much information to go through. I do have one of mcmark cylinder head temp sensors but it's not connected anything at this point

for what its worth, and bearing in mind that I am a neophyte and, still have not completed my MS installation, here's my 2 cents:

1) I noticed your coil pack is on the engine, and there might be a heat problem, causing an intermittent or systemic, slow-motion electrical breakdown... (basis is that I had some V12 E-Types and the stock ignition systems were mounted on top of the engine, causing very odd problems, not dissimilar from problems you are describing) the solution would be to move it to the firewall/bulkhead/ somewhere without heat soak.

2) check for ground faults, intermediate grounds (See#1), and feedback loops. Is the engine properly grounded to the chassis?

Good Luck ! I'm nearly finished with my installation and trying to learn from everybody

What he said, plus put the coil on a bracket you don't want the heat transfer from the engine. Heat soak can be a problem for a lot of electronics.
240 is not an accurate intake temp, maybe 20 over ambient. I measured cooling air off a heater core at 150 max.

I miss spoke, I added a guage for the air intake temp and it is reading accurately. The "coolant temp" is connected to the cylinder head but by a bolt so it is a little different than a CHT. Before starting the engine it was reading at where the ambient temperature was at. However I am not getting the same reading with a laser temp sensor that I am getting on the gauge as it heats up.

I ran the car today in the shop to try and isolate what the issue may be. Once again the car runs great until a certain point when the dwell starts spiking the ARF Drops off and the engine starts cutting out. The temp did not get above 220 according to the guage(see previous post) I was watching all of the guage including ign load, warm up enrichment, lost sync ect... the only ones showing issues that I noticed were the Dwell and AFR.

I monitored the temp of the wasted spark and according to my hand sensor it got up to about 167.5 at the time of the issue.

Just a WAG, check your battery voltage when the trouble begins, low voltage will cause a problem like that.

I have uploaded more in a dropbox folder including a video of the gauges as it was happening. It was at 13.8 so that was fine.

Got an email back from Mario:
" It's still a sync loss, each time it loses sync the computer resets and the coil gets more dwell because it thinks the car is restarting. Fix the sync loss and the coil problem is fixed. The 02 will also restart since its attached to the fuel pump relay which shuts off during a sync loss.

220-250 is normal sensor temp and nothing to be worried about. Air cooled heads normally operate at 325-350, the sensor not being a thermocoule and only getting residual heat didn't get upto actual had temp. On a water cooled car the coolant temp of 240 would be trouble and why the gauge turns colors, you can change when that happens by right clicking and adjusting the warning and danger temps."

So now how do I find the sync loss......

First of all, check the wire and connectors supplying +12v to the ecu. A glitch in the Power supply will show the same sympoms as a sync loss. (dont ask me how I know, it took me a year to find mine)

I run Marios trigger on my car too and I have no issues with the short unshielded cables from the trigger (hall) sensor but then mine connects to the ECU (not MS) supplied cables which are shielded all the way back to my ECU. I'd unbolt the coil from the engine and run the engine till the issues start and then try moving the coil to see if that helps.

What sparkplugs are you using? Resistor sparkplugs is almost a must when running modern electronic injection.

I dont know anything about microsquirt , but on my megasquirt at one time I was getting a sync loss alarm ,
I was able to trace it back to my hall sensor , I tighten up
the gap between the hall sensor & the wheel and it went away , ( I also run mario's
hall sensor kit ) I hope this helps Bill D.

Subscribed ..............

We want 3mm or less of clearance , I'm fairly certain I set it at 3mm.

It is true that is a possible heat issue. I have run that exact coil in the same location and not had any issues. I have mine with a small air gap of about 1 inch.

I remember adjusting mine closer , I checked DIYautotune
www.diyautotune.com/product/hall-effect-crankshaft-position-sensor/

Maximum air gap: 5 mm
Note that this is an absolute maximum. For most trigger wheels, operation is most reliable at a 1.0 to 1.5 mm air gap.
The trigger wheel must be at least 2 mm wide, the teeth and gaps at least 2 mm wide, and the gaps at least 2 mm deep.


Bill D.

Mike here's a shot while we were building the coil mount.
Click to view attachment

The yellow arrow shows you where Matt and I discussed adding an air gap and or heat sink fins. Right now the coil is sitting flush on the mount we created.
Click to view attachment

I bet that helps a lot. It is a amazing how much a bit of space stops heat transfer.
I just added the DIY Autotune Quadspark and performance coil. Real improvement on my engine but the VW coil ran my 2056 just fine.
This gets the igniters out of the heat which seems to work better.

I like that coil design. Clean and simple.

McMark's kit uses 4 coil packs located right next to the park plugs, with very short leads. Its way trick, but you need small hands to get all the fasteners (or meat hooks like I've got and deal with bloody knuckles and some cussing.)

Marks design, though, does give a nice clean run for the crank sensor wire more or less free from spark plug noise. Even so, we used a shielded wire to ensure the signal made the run without dropping.

Zach

Love to see some pictures of that set up. Sounds cool and a lot cleaner on at least a few levels.

I think he uses LS1 coils. It would be even cleaner to find a coil on plug but the bracket might get interesting if the grommet doesn't hold it secure.
That being said, I don't think it's related to your problem. Longshot but Dad would always check for bad soldering on the board, as it could show those symptoms as circuits heat and cool while working. I've actually seen it in action on an E36 with intermittent blower function as the factory fuses are 3 piece and lost solid internal contact.

Click to view attachment

Here's how I mounted my LS2 coils. Nutserts into the rear firewall and then male/female threaded rubber mounts.

Do you have any oscilloscope to scope it with?

My experience has been that Microsquirt has extremely poor signal conditioning and has a really hard time with less-than-perfect VR or even Hall signals**.

Hall effect sensors can also have this unfortunate trait of switching the trigger side and getting off phase; you might have that issue resulting in some trigger confusion. This is actually why people often recommend VR sensors on high end systems with good filtering capability.

**I fully realize that sounds stupid since a hall sensor should be outputting a square wave but I've never really been happy with uS.

I was able to get lots of advice from Mario and others on some of the Megasquirt and tuner facebook groups. At this point, I have ordered another hall sensor and we will be pulling the motor to replace it if needed. The thought is that it may have become loose and is too far or too close to the wheel. At that time we will also shield the wires to try and protect against any interference that could be contributing to the problem (Now or in the future) What we understood was that the hall sensor should be mounted 3mm from the trigger wheel and it should be 1mm. One of the people that responded on the facebook group said I just went through the exact same thing with a clients car at the shop. Move it to 1mm and problem solved. So I am hopeful that this will do the trick. It also allows us to address a pesky oil leak while the motor is out.
So cross your fingers for me, everyone. Hopefully, we will be running solid again and ready for Dyno tuning soon

While its easier to fix with the engine out, you should be able to do with the engine still in the car. I've lowered the engine support bar and loosened but not removed the trans mounts to allow the engine to pivot. From that point its just a process of finding a comfortable way to hang out under the car and do the work. Easy if you have a lift, less easy if you are using jackstands and a creeper like me. But it DOES save a lot of time. I replaced an oil pump that way once (or rather retrieved the dizzy screw the feel through the tiny hole in the dizzy well and landed behind the oil pump. Fun times.)

If I had pulled the engine it would have taken at least 2 days. w/o pulling the engine it only took a few hours.
Zach