Engi-ninja

Shorts in wires can appear to be extremely unpredictable. I've spent weeks chasing erratic valve behavior on a test cell engine, only to find out that it was a short in the wiring harness. If you were able to get it to die by shaking the fuel pump wiring, I would spend a little more time looking closely at that. Are there any tight bends in the harness in that area? Sometimes tight bends of bundled wires can lead to kinking of the innermost wire, which will eventually cause an intermittent connection issue.

poormxdad

Attached is a log I took around town after installing the new main relay. She cuts out at about 238 seconds. I don't see anything that would lead me to a solution.

I was fiddling around with looking at different logged parameters and saw the ECU Temp. My minimum says 238 degrees and max 248. I looked back at older logs from when I first added the PNP Pro and the max I see is around 220, but that was back in April. Is 238-248 too high?

Thanks,

aaronwalker1974

If you are saying your coolant temp sensor at the ecu is being reported at 238-248? Yeah that is way too high. I wouldn't want the temps to go above 220. Especially for street driving.

Do you have an upgraded radiator?
Any ducting for the radiator?
When you installed the kraftwerks kit did you cut up your under tray?

poormxdad

No. One of the parameters I can see with Mega Log Viewer is called ECU Temp. I assume it's the temp of the MegaSquirt's processor, but it's just an assumption.

aaronwalker1974

Ahh...gotcha.

poormxdad

Ever check yours? I have no idea if those numbers are good or bad, or even accurate.

poormxdad

Matt Kramer says those numbers aren't real.

Tico556

If your ECU processor has a temp of 200+ degrees that's a big problem. Does it get really hot when the cars running? Is it getting ventilation?

aidandj

iTrader: (5)

Off topic bitching.

TL;DR.

ECU temp is not valid numbers, get the firmware updated if you want real numbers.

Your tune might need to be updated to go along with that.

poormxdad

Gents,

I posted that question to Matt this morning when I was away from my tuning laptop. Matt told me I needed to check the temp from a cold start--obviously the starting temp should be ambient. He also said if MLV showed something other than ambient, I needed to check what calibration was used. I have not yet been able to do that.

However, on the way home from the track three days ago, I stopped in an empty parking lot and checked everything I could, including the connectors to the MS. The box was not hot. I had my hands and fingers all over it, and it never entered my mind that the box was even warm.

I was searching the log for anything that might lend a clue to my cut out issue and saw ECU temp. I checked ECU Temp on a dozen logs from as far back as March. They all showed 200 degrees or more as the min. I know my desktop processor is much hotter than the outside of the case it is in, so I asked the question.

poormxdad

Gents,

Matt has also asked me to get a voltage reading, VREF, from the MS when it goes Off Line. I thought you all might present a Miata perspective based on the results.

I added a wire to the H port on the back connector and hooked it up to a multimeter. With the car started and running, the voltage from Port H was rock steady at 4.95V. She idled in the driveway for 32 minutes before she died. The voltage was 1.648, key on engine off. CLT was about 200, and the MAT was up in the 120s. I recycled the key and got the full fuel pump prime cycle. She restarted normally. The multimeter read 4.96V and was rock steady till she died again 25 minutes later. I saw the voltage fluctuating and settle out at 1.708V.

Any ideas?

Thanks,

sixshooter

iTrader: (12)

Main relay? Other relays? There has been some discussion of these relays getting hot and losing connection.

poormxdad

She has a new main relay in there now. It did not change the behavior of the problem at all.

poormxdad

Gents,

I wanted to provide an update. Based on the info I mentioned in post #51, DIY sent me an RMA number. They were able to duplicate the problem in their lab. They "replaced the MS3Pro module". I just did a glorious track day at Dominion Raceway without a single hiccup. I did notice a couple of interesting things.

After reloading the 1.4.1 firmware and my current tune, she started right up. But, I had to recalibrate the AFR Table. I had been using the Custom Linear Wideband setting to get the Bosch O2 sensor and TS to agree. They were now way off. Going back to the default Innovate Sensor pretty much put things spot on.

My TPS now calibrates at 0%. It still reads 101.7% on the high end at boost, but at least now when the throttle is closed it reads 0%. Before,no matter how many times I did the calibration, closed throttle could be anywhere from -0.7 to +0.4%.

I was also having issues with Shadow Dash. Sometimes the display would stop functioning even though the car was still running and presumably the MS was still working. It wasn't locked up. I could scroll through screens and MPH still worked from the GPS in the tablet, but I got no other sensor readings. It would also continue data logging all the sensors, but gave no useful display. Shadow Dash is working normally again.

Thanks to Matt and DIY for getting me back on track.

shuiend

iTrader: (24)

​​​​​​​How is the new Dominion Raceway? I have not been out since they moved down to Stafford. When I live up in NoVA I was about 8 minutes from their old track in Manassas. Some of my friends would rent it on Thursday's for way cheap and we would go out and have fun.

poormxdad

I love it. It's a great Miata track. There are lots of folks who don't like it due to the proximity of the jersey barriers and lack of run off. It was also poorly paved due to a lack of funds. There are some pretty good bumps as you go under the bridge. I would go there even if it sucked since it's only 100 miles from my house. VIR and Summit Point are both almost four hours each way for me.