Well, I can guarantee that the underhood components aren't 55 or 60. The heat soak is what can kill your rail's liquid contents and turn it into vapor. Just for fun, when you're in hot start situation, squirt some ice water on the rail and injectors (carefully of course), let it sit a minute and give it a shot. Some people thought I was nuts with the fouled plug scenario, this is similar. BTDT.

-Mike

 

Ohhh...crap. Profuse apologies muythai.
I checked my wiring last night and all my notes pointed at the wiring being correct, I couldn't find error's page on the pinout, and my old notes, as I read them had them in the correct positions. I just checked the pdf that Glen posted to confirm my notes and sure enough, it seems I have Ne+ and G+ mixed up.

Damn, man.. That is totally my bad I know you mentioned the wiring a long time ago. I didn't check because the car ran fine, I was pretty sure I had everything right too because I checked it against my notes the first time I had trouble starting a month or so ago. I just thought it was something else.
I don't remember where the mixup could have happened, I was so sure...

Anyways, I will reverse them when I get back to my car this Friday.

EDIT: Actually, I found error's notes about the pinouts for the daughtercard, which is what I based my notes off of. He has the "base wheel" as the 2pin wheel, and the "reset wheel" as the 24pin and I thought the terminology was vice-versa. I'm not saying it's error's fault. I'm just saying that's probably where I got mixed up. I definitely should have looked at that more closely.

 

Well, I never thought you were crazy, MIke. I had the same problem with plug fouling before.

The only reason I didn't think it was vapor lock was because the car would start with Time Based. If it was vapor lock I figured it wouldn't start then either.

 

do you mean you have the polarities reversed or the Ne and G signals reversed... like are the +'s and -'s reversed or are the Ne and G signals reversed...

The reason I ask is that you shouldn't have been able to start the car at all if you had the Ne and G signals mixed up.

 

No, the polarities are fine.

The wiring mistake I made was getting the G and Ne pairs mixed up, but since the daughterboard grounds both signals together the negatives shouldn't matter.

I dunno how it works either. Maybe I got the outputs backwards also. I will have to recheck it. So far as I can remember, getting Ne+ and G+ backwards was my only mistake, but I was sure about the other wiring too, so I will need to go back and recheck everything now.

 

yeah, if you had Ne+ and G+ backwards, then the car would NOT start.... ever... Most likely you have those right.

What I was talking about wasn't internal on the MS, but are you sure which are Ne+/- and which are G+/- on the CAS. You'd still be able to run if the polarities were off, but you could potentially have problems like the ones you describe.

Ken

 

I am 99.9% sure that the white and white/black wires from the CAS are going to T5 and T6 on the daughterboard. Which are the grounds for the VR circuit, IIRC. I can double check it again, but when I wired it all up I know I used the wrong pairs..

This is what I believe it should be:
http://www.mindspring.com/~wilkinsjr...%20Wiring.html

Before, I had it backwards and I have since corrected the above webpage.. As far as I know the above link has the correct wiring, so correct me if I'm wrong.

 

your G+/- wires should be going to the "top" lm1815 circuit (assuming that has the 82k resistor).

Your Ne+/- wires should be going to the "bottom" lm1815 circuit, assuming that has the 56k or 39k resistor.

It looks like according to the link, that's how it should be... but it depends on T3 and T4 are wired.

T3 should be wired to pin 11 of the CPU through a 1k resistor, while T4 should be wired to u4 (opto-isolator) pin 6... This is assuming a v2.2 board.

If your car was starting and running the way you had it set up, then T3 and T4 were most likely reversed as well... as the megasquirt won't even get the correct rpm reading if the two are reversed....

 

If I got the 2 and 24 tooth circuits mixed up, then it's likely I got the outputs mixed up as well.
Think I'll blame it on bad intel... heh

I hate having to wait to work on my car.. Oh, well. I will know Friday exactly what's up with it.

 

Im pretty sure the top lm1815 on the D board is the 2 tooth ckt. If your using a v3 board you only need to populate and use the top ckt. I checked it with a good quality dmm and the top ckt has the 82k resistor. My board had a accident so I cant look at it again,Shawn

 

I am using a v2.2 board with the error add-on card and totally mixed up the circuits somehow... :-/

 

Just got everything lined out with my spare ECU.
It has idle spikes, however. Can't figure out why.
Only at idle and below 1200 or so, it'll spike by 1k RPM or 1500rpm.. If it goes above that, no problems, or at least none that I can tell.
Any ideas?
http://www.mindspring.com/~wilkinsjr...e%20spikes.xls

BTW, yes the circuits were backwards. Had the 2 and 24 tooth circuits mixed up.

 

usually spikes are caused by noise on the "g" circuit... you don't have noise any higher because the lm1815 filters noise... and the more "real" pulses it gets, the easier it is for the lm1815 to filter the noise....

To get rid of the noise, I suggest turning up the load resistor (potentiometer) on that circuit... turning it clockwise a few turns at a time, then checking to see if the spikes still occur. It might take a lot of turns... on one of my cars that had a ground loop problem, I had to turn it to like 18kohms of resistance... almost its max.

 

All right. I'll dial in some resistance on the 2 tooth circuit. Hopefully that's all it needs. It's definitely the circuit as my other ECU works fine.
I did some testing. Trigger mode works real well now at cold, but there are some hot start issues still. My alt. is having some trouble, I think, or the new battery I got isn't holding a charge.

However, I do want to ask something. I did think that trigger return was a "safer" mode than time based, but if there are spikes at startup with trigger return, won't that cause problems as well?

 

I had RPM spikes, and they were caused by reversed polarity on the CAS signals, and were aggrevated by proximity of a spark plug cable to the CAS cable. With the polarities fixed and the plug cables rerouted the spikes are gone.

 

If there is a spike it's bad no matter what.. time based or trigger return you'll have problems with spikes while cranking.

 

Update!
After resolving my wiring problem, trigger return had the same symptoms as before. After a bit of troubleshooting, I believe I have found a big piece of the puzzle.
Apparently, the problem has to do with the idle advance settings. With a wait time of 0, idle advance of -5, and while using map settings on the spark menu, it just will not start with trigger return. Using a fixed angle solved the problem, I noticed.. So, I added a wait time on the idle advance. Trigger return starts the car without any issues now.
Is anyone using a wait time of 0 and NOT having an issue?

I may try a trigger angle of 65 and see how that does with a 0 wait time.

 

BTW, I also noticed that the idle adv. timer reactivates anytime you come off idle. Is that intended, or is it not a big deal, or is it something that should be fixed?

 

the timer is supposed to restart every time you do change any one of the conditions that turn on idle.

So step on the throttle, the timer restarts, go above the idle rpm, the timer restarts, etc....

Then when all the conditions are met again, the idle advance will wait the specified number of seconds before coming on again.

I'll look into the idle advance causing hard cranking... it shouldn't make a difference since none of the idle advance settings are used during crank.... but I'll take a look anyway.

 

IT'S DOING IT AGAIN!
Earlier, I went back and forth between the settings to verify the problem. It behaved like I described. I'm sure I went back at least 5 times so I could ensure that was the cause of my problem. A few hours later and trigger return ceases to work.

I give up. Time based works everytime. Trigger return sometimes decides to work, usually when hot it won't start. I never have issues with time based, so I will use it from now on.

 

I was thinking that at 0 wait time, the idle advance may be causing an issue with the cranking advance trigger return uses. Not sure about that now, though..
I just can't believe that I am the only one who has this problem! It's pretty frustrating.
I snagged a datalog of the start attempt. I'll try and post it later.

 

maybe the heat from your engine is doing something to your wiring? I haven't had any problems like this on any of the 3 rx7's I've run the megasquirt on so far... but it sounds intermittant and since you're saying it only does it hot... maybe the heat is affecting your wiring...

 

I don't have trouble driving around or anything. I can try and find the datasheet on the wire I am using for the CAS, but I am pretty sure its temp ranges are within spec, but as far as I'm concerned, it could be anything at this point...

Right now I have two v2.2 boards built the same way. The main one I have in the car works fine, but the 2nd one I built developed a idle spike problem that I haven't been able to fix yet. So, I don't know if it could be something in the MS or not, but it only developed this problem recently, which is weird. Before, it worked fine. Outside the trigger return problem, anyways.

 

Here's the datasheet for the cable I'm using for both the CAS and the ECU to coilpack wiring. I had extra so it was convenient to use it for the coilpacks.

http://bwccat.belden.com/ecat/pdf/9418.pdf

Latest MSQ and log files here:
http://www.mindspring.com/~wilkinsjr/MS%20Files/

Any ideas?

 

did you ground the shield for the coilpack wiring? IF so, try ungrounding it...

Also, This wire is non-twisted-pair... Twisted pair is preferable for the CAS wiring (although I don't thing the stock CAS is twisted pair, and I don't think that's causing you a problem).