The Bean

In reply to pres589 (djronnebaum) :

tell me more

This is just the audience for such "boring" stuff

The spark plugs are gapped to .035". A note for future me.

Shavarsh said:

In reply to pres589 (djronnebaum) :

That's very interesting. Most motorsport writeups I have seen tout the benefits of star point grounding at the cylinder head. If 3 grounding blocks are used, how are each terminated to the battery?

Remember, I'm an aircraft electrical engineer, not automotive.  That said, the theory is all the same.  Aircraft procedures ask for a bond check (making sure the housing is well grounded) for all kinds of procedures. 

The battery is terminated at the frame of the aircraft like at a frame rib on the negative side and at a battery disconnect contactor on the positive side so the battery can be easily isolated and effectively "turned off".  The ground points for case, power, and shield are all well bonded to the aircraft as well.  Just like a car.  You just separate them physically.

What's interesting for me lately is working on another brand's aircraft due to a corporate merger about a decade ago, and they didn't do this kind of thing.  I'm not going to say this causes problems, because I've done very little sustaining/troubleshooting work, but it's not great to folks that came from the area I did who are used to seeing these seperated.

Nukem said:

In reply to pres589 (djronnebaum) :

tell me more

This is just the audience for such "boring" stuff

Ha, maybe a different thread for this?  Are there automotive electrical engineers/specialists/etc here?  I don't know of any... would be better to have different points of view, I think.

Success! 

After swapping the plugs, I was disappointed to find the problem actually worse than with the other set. I could get the engine started, but it was having an rpm spike on most revolutions.

This got me thinking a bit... I decided to retry rising edge instead of falling edge on the crank trigger, and bingo. Running smoothed right out and the engine took no offense to throttle input, unlike before. I threw a conservative timing map on it, and took it out for a longer drive, stopping occasionally to adjust the VE table. While it's still very lean under load in most areas, I'm very pleased to be on the right track again. Even after setting the trigger filter back to "weak" I didn't have any drivability issues. I'll have to go back over the log to make sure nothing showed up in the data.

In reply to pres589 (djronnebaum) :

Good idea, I'm sure there is lots of interest on the board on a grounding best practices thread. 

In reply to pres589 (djronnebaum) :

The fact that you are aeronautical does not invalidate your best practice. I have restored and prepared many road racing cars, and aircraft best practice is our general guidance everywhere.

In reply to Shavarsh :

Yes!

Time to put some miles on it, to fix the maps and see the differences with the carb in drivability...

In reply to Shavarsh :

For my own notes: What polarity are you using on the crank sensor? I think which of Falling/ Rising is correct is going to depend on that.

In reply to Nukem :

Good question, I will check and get back to you.

Confidence is important when driving around with a sketchy tune and making adjustments. One thing I do not have confidence in is the brakes. I've chased a soft pedal ever since adding the booster and swapping spindles/calipers in the front. While there are some inherent flaws with the current system, such as the manual pedal ratio, I still think the current setup should be better. The first step in systematically chasing down this issue was a complete bleed with the motive bleeder. Having done that awhile ago, without improvement, it is time to replace the rotors/pads. The set on the Bean currently are... the set off the junked miata that the suspension is from. There was definite evidence that the car had seen track time, and I hope that the pads are just badly tapered. So, out with the old, in with the new:

Oh, and obvious PSA of the day: don't brace your pry bar on the pilot hub of your cheapo spacers.

Oh, and what do we have here? After thinking about it for awhile, I realized that I didn't tighten any of the rear swaybar hardware when replacing parts a few months ago (with the intention of tightening with the car on the ground). Turns out, when all the hardware is loose, things try to disassemble themselves.

I popped the link in the press to reassemble, then tightened everything with the wheels in the air. I could use a bit of rear spring rate anyway.

In reply to Shavarsh :

That link needs a captive washer. Loose hardware is not the only reason why the rubber popped out, and it is sure to continue now that it knows the way.

Good call Turner.

I've been sneaking up on the ve table with the afr target set to 13.0 everywhere. The car is feeling great, and I can run it out to red line without worry. The timing is conservative as well. Just for kicks, here are the VE and Spark tables (DO NOT USE THESE FOR YOUR TUNE!):

Obviously there are some issues here and more work to be done. The first issue is the 180% efficiency values... injector and engine size seem to be correct in tunerstudio, so I have a fuel pressure sensor on the way to check the pressure, and make sure it is consistent across the board. Also, the area of the map that governs the second half of the clutch engagement is going lean, so taking off from a stop is best done over 1500rpm. The idle zone is pretty blocky so I can play with the PID IAC tuning.

Coming home from a surf this morning the temperatures were a bit higher than normal, and climbing in traffic. When I got home I filled up the overflow tank just in case, and it promptly drained out onto the ground... well there's you're problem:

Looks like the hose got caught in the power steering belt and became 2 pieces. Will need to secure the replacement a bit better.

PS: Always ready to learn if there are some obvious tuning elements I am missing!

Hopefully the temperature stayed low enough not to have bad consequences.

PS. Even though I wouldn't be of any help even if I did, I can't see the images with the maps, just the photo of the engine bay.

I can't open a PNG either. Not sure why, new computer. They don't open with old graphics programs either, but they appear on the screen just like a JPG on the old one.

Let's try again with a jpeg:

While charting the RPM vs IAC Duty % the car started getting a bit hot again. I decided to plug in the second fan relay so all 3 fans could run while sitting there idling at high RPM. Unfortunately, when I plugged in that relay, it took out both fan circuits (without blowing the fan fuses), and seems to have effected the accessory circuit in the stock Pinto wiring. After a bit of troubleshooting on the side of the road, I jumped the first set of fans on at the auxiliary fusebox and drove home. Hopefully there is a good explanation of what went wrong when I have time to dig in this weekend. This fan has caused problems before, and I usually don't have it powered.

In other news, I've been using Andy Whittle's videos on youtube to work through the idle tuning systematically. His videos are very clear and concise. I can't recommend his methods to others until I achieve a good result myself, but its nice to have a logical guide to follow.

Small update: Initial troubleshooting shows the wiring looks fine on the pinto side. However the fuel pump now runs continuously with key on (instead of just the prime duration) even when the LED on the board shuts off. Also, the fan relays are not receiving ground from the ecu, even though tunerstudio thinks they are... I am hoping that I haven't fried part of the ecu...

In reply to Shavarsh :

I'd pop the ECU open and give it a look at smell. If anything let the smoke out you will probably know 

In reply to Nukem :

I'll ad that when I had a diode fail in my Megasquirt years ago the smell was vile and massive.

Thanks for the comments. After speaking with the ecu manufacturer, he told me to check chip ULN2803A. After some digging I found it:

And a closer look:

Despite my cameras inability to focus on small things, you can see a bump on top of the chip that shouldn't be there, accompanied by the smell of burning electronics. New chip has been ordered. Fingers crossed that is the only thing effected. 

Due to the fact that this happened after plugging in a relay to the second fan, I have to assume that the relay is the problem. How else would a 12V spike find its way into this chip? I don't think the fan has direct line of sight to this chip, even if there is something wrong with the fan. 

Which leads me to the question... I have 7 identical relays, how do I find the bad one? I'll start with some simple continuity tests, but any advice is appreciated.

The control circuit of the relay is likely shorted to power. This can easily be in the receptacle, not the relay if the hot has been transposed into position for the grounded by the ECU terminal. The correct wire also needs a specific position, but I have done this error with less serious results. The ECU gets the volts when you plug any good relay into the open receptacle. May simultaneously hurt the relay. So check all of your relays, if they are all the same it is not the relay, but the receptacle.

In reply to TurnerX19 :

I've been bitten by cheap relay receptacles before.

So, I pulled the trigger wire from the receptacle, and replaced it with a wire to chassis ground. I put a 5A fuse in line, and tried each relay. Unfortunately, they all powered the fan without issue, and without popping the fuse. Next I'll wire in the other fan on the same ground to fully replicate the conditions.

All of the relays test identically for resistance and continuity on the bench.

Edit:

Even after hooking up all the wiring as before, and hooking up my fused ground wire to the pin on the ecu plug, all the relays tested fine. The ecu signal pin on the offending receptacle does like to push out the back sometimes though...I wonder if the coil chatter could have been enough to damage the chip. It doesn't look like there are any uninsulated power wires near the terminal that could have made contact.

Since the fan never worked correctly the coil RF has increased in possibility.