I'm in the process of upgrading my school's 7 which was running just just fuel on MS to both fuel and spark. We are using the duel VR circuit with a 2nd gen cas. I can pick up RPM with the megasquirt, and I have verified that the processor is successfully receiving the outputs of the VR conditioners (whose output also match their inputs). We are able to get spark when we spin the cas quickly (megatune reads approx 1000 rpm), but at lower rpms (read cranking speed =150 rpm) the coil doesn't fire. We still read rpm in megatune and the fuel injectors click away.

At this point, I'm sure that my spark output circuit is good and so are my VR conditioner's, so it would seem the issue lies in the firmware/tuning.

From what I've garnered from the poorly labeled wheel decoder settings in the MS1extraFAQ, I want 12 tooth, 2nd trigger enable, rising edge, no missing, and 1 3 7 9.

When we try to crank with it in, it will fire 2 sparks initially and then completely stop sparking.

Do ya'll have any ideas?

 

Two sparks initially indicates that it is not reading the 2 tooth wheel.

How is the shielded cable to the CAS grounded? It should only be grounded at one end (the ECU). The engine also needs a good solid ground cable between it and the car.

 

I agree with aaron. What is poorly labled about the settings in the faq?

Ken

 

Sorry, if I sounded bitchy last time. I was just frustrated with the car. Ya in the FAQ the screen shots of the wheel decoder setting aren't really labled as to what they are for.

I'm actually a computer engineer myself, and I suspected poor grounds too. We added grounding straps, but that didn't fix the problem. Aaron you were right about the 2 sparks thing. We had the two VR circuits switched. We now get spark consistently, and have gotten it to kick over.

By no means is it running though. We were able to get it to keep itself running but it wasn't smooth at all. If we had it at WOT and cranked it, it would fire up then falter then fire then falter. Looking in Megatune when it was firing it had a pulse width > 0, but when it was faltering the fuel cut out (pulse width=0). I scoped the inputs and outputs of both VR circuits and they look absolutely beautiful. I also added a shield grounded at the ECU as suggested to the VR sensor wires. If I set static timing and check it with a light, it looks dead on. Do you guys have any ideas?

Aaron, I like your website. I discovered it a while back. Its good to see another electrical type guy, working on cars.

 

what method are you using to set timing you have to make sure you set it -5 in megatune spark setting fixed angle and then when finished return it to -10. You should double check your coil connection are correct. Last time i put my engine I had a similar problem. I had to turn of trailing to get the car to run. I still can't get it working again.

 

I'm working on a step by step Megasquirt install as we speak. It will fully cover all of the confusing and poorly documented steps. But it is a huge job and probably won't be out for months.

Sweet.

Make sure that the TPS is calibrated.

With the timing locked at some value (-5 is easiest) the engine will run very poorly especially if the VE table isn't tuned. As long as you have your timing zero'ed out and you are sure it is firing correctly, then give it about 15 degrees at it's current idle bin and it will pick up considerably. Of course this assumes that it's not running 10:1 AFRs.

It's funny, but it was the Megasquirt mostly that got be back into electronics.

 

Aaron, I'd quite like it if you could use my pictures. I haven't written any real explanation.

https://www.rx7club.com/showthread.p...t=contribution

If you want jpg/png versions of them please just PM me your email address!

 

I wrote those steps for the more experienced user anyway, but I still don't see how they are poorly documented. If I can have a specific example of a step that is poorly documented and how, I'll go fix it.

Ken

 

Sorry, I think "poorly documented" was the wrong choice of words. What I meant was that the current FAQ (for MSI anyway) is a big jumbled and can be hard to understand for the new user. I have honestly not looked at the FAQ for MSII.

My writeup is a step by step install covering both MSI and MSII, not really a FAQ.

 

So I finally got it running!

I'm not sure if I mentioned it before, but I'm a miata guy. Most of us Miata guys are running high-res, well as it turns out high-res doesn't work with rotarys. Yesterday we flashed down to 29y4 and the car fired up the first time with the same settings we were using with high-res.

This might be something you want to include in any guides or how to's.


Now I have another problem though. This one is a tricky one though. So after we got it running, we went and thrashed the car around and it behaved beautifully. Well we parked it, turned it off, and got some beverages to celebrate. When we came back 5 mins later to let someone else drive the car, it wouldn't start! After hooking up with a laptop, all our gauges were off (198kPa, 6 deg MAT and CLT, no rpm). After download the msq again nothing changed. Finally we reflashed and reloaded the msq and the car started.

It seems like every time the MS loses its power its completely, not only do the settings get erased but also parts of the firmware. Has anyone seen anything like this before? Or have any ideas about what could be wrong?

 

I've only seen this when there are voltage spikes on the +12v when you turn the unit off.

Just as a test, can you access the ECU easily? IF so, I'd like to see if unplugging the MS before turning off power to it gets rid of that problem for you. That would help avoid any voltage spikes or anything like that on power down.

So just unplug it from the harness, then shut off the car. Then plug it back in and turn on the car, and see if it has the same issue.

Also, do you have a stim? Does this happen on the stim?

Ken

 

Checked a bunch of stuff. Turns out it only happened when we programmed it with our MSQ. If we just flashed and cycled power it was fine. So at that point we new it was something in the MSQ. After some mucking around with wheel and codbase options, we found that we had both the distributor and generic wheel turned on (don't ask me why, my friend was piloting the laptop during all this). After fixing that it worked just fine.

As I stated before I'm more of a piston engine guy. As such I understand advance in a piston engine context, but I don't have a good idea of what is reasonable on a rotary. Does some one with a reasonably stock 13B have a spark table that I could start with.

 

There was a thread in this section not too long ago that had several base maps that were reasonable.

If nobody has posted anything I'll see if I can find my NA and Turbo maps. None of them have been tuned on a dyno, so I'd expect them to be rough, but they'll run the engine well enough to drive around.

Ken