thetech

iTrader: (1)

This is on an SCCA EP racecar, running an MS for fuel only using an 85 distributor for spark.

I'll summarize the timeline of events: bought the car as a shell with the wiring already done, after building the car started it, had a problem with the primary injectors, resoldered and checked all the wiring, updated to msns-extra 029t with staged injection and everything was fine. Did approx 35 dyno runs without any hickups and everything running beautifully making very good power with staged injection around 4000rpm.

One day, I am moving the car into my trailer and it won't start. After about an hour of fiddling, I try turning the staged injection off and running with all 4 injectors firing; at this point it starts, but running very rich (as expected). After fiddling with the main DB37 connector, switching back to staged injection mode, and checking wiring the car miraculously starts. We do another 10+ dyno pulls.

The next time I start the car is at California Speedway at a race event - same problem occurs, where the car won't start unless staged injection is turned off. Eventually, I am able to get the car running fine after some more fiddling with the connector. Everything seems great, I get out on track, car dies immediately on first corner. After being towed back in, fiddle with the wiring some more, and it runs fine again.

At this point the obvious conclusion is that the wiring harness is bad or some element of the wiring is faulty - I order replacement MS harness from DIY Autotune (very nice piece btw), new injector plugs, and rewire the whole car myself with a Painless racecar harness. Every circuit has its own relay, and I gave the injectors their own direct relay rather than using the same power source as the ECU itself. Fast forward to today, car still wont start after all this rewiring - I upgrade to 029v firmware and the car starts with staged injection turned on, but sounds like its running on one rotor and dies after 20-30 secs.

All wiring has been checked, double checked, soldered, heat shrunk, etc. The injectors are confirmed to be in working order as well and there is spark. The fuel pump is directly to a switch, is running, and there is good fuel pressure. All sensors are reading correctly (IAT, CLT, TPS, MAP using on-board 2.2 MAP sensor and no O2 sensor).The only thing left at this point is the MS unit itself (its a 2.2 and unknown if it was put together by previous owner or purchased preassembled).

Any ideas?

EDIT: Attached .msq

muythaibxr

I would upgrade to 029w, but that is unlikely to be the cause of the problems.

As far as components in the MS, I would suspect the Injector FETS, or the FET driver.

Have you checked for the "flashing fuel pump" problem?

If you get that problem, it means you have a misconfiguration. You can tell what is misconfigured by closing megatune, turning off the MS, then opening hyperterminal. Set it up for 9600 baud, 8, n, 1 and no flow control, then turn the MS on, and it'll tell you what is wrong.

Ken

thetech

iTrader: (1)

Thanks Ken. The FP indicator is NOT flashing, but I connected with Hyperterm anyway and it connects fine and displays nothing.

Just to reconfirm - my IAT, CLT, TPS and MAP are all working fine in Megatune - these are the only sensors that are needed for correct operation, right?

T2GTUS

I spotted a 1.0ms pulse at 160+ in your msq ... does this only seem to happen when the car is warm?(1.0 seems low to me at that temp)

thetech

iTrader: (1)

No, I have only attempted to start the car when cold with the new car and megasquirt wiring harness completed. I think some of the cranking pulsewidths are probably screwy as I was playing around with them at the track seeing if I could get any different outcome to the problem.

T2GTUS

ok, was curious,since its always a cold start, assuming the board/wiring/components are good, I'd try adding a couple ms to crank pw, (I see that 4.0ms at 60F, musta been a track attempt) and I'd try to up that ASE table a bit ... and as Ken said, flashing 029W wouldn't be a bad idea.

thetech

iTrader: (1)

Thanks for the reply. I think this is a different issue, as the car starts fairly quickly but is clearly only running one one rotor - the cranking PW obviously dont matter once the car is running so I'm not sure that is related (although I should fix my cranking pw table for sure!).

thetech

iTrader: (1)

Here is a datalog from a quick crank attempt; as you can see I am out of battery (need to pickup a charger today) so its tough to tell what effect the low battery voltage can be having.

muythaibxr

If it's only running on 1 rotor, the MS is probably not the culprit. Usually I've seen bad plug wires, bad wiring on one primary injector, 1 bad sparkplug, etc... cause things like that.

The MS shouldn't be the cause of only running on one rotor; if wired correctly, both primaries squirt at the same time... and due to the way the leading coil is set up (wasted spark), it should be firing into both rotors at the same time as well. If one rotor works the other should be working, and if it's not, it's not likely to be the megasquirt that's causing it because if the injector FET is squirting one injector, there's no reason why it shouldn't be squirting on both.

Ken

thetech

iTrader: (1)

I agree, and unless I had just rewired the entire car and spent extra time ensuring that the injector wiring was perfect, I would be going over it all again (and I will if I can't figure it out). Running on 1 rotor is just a guess as it generally sounds 'wrong'.

Here are some datalogs with a fully charged battery where the car ran for a few seconds before dying. Please take a look and let me know if anything looks abnormal. Thanks so much!

muythaibxr

You're getting some VERY odd rpm spikes, your pulse-width seems a bit high (I guess b/c you're never building much vacuum.

I think you may also have a wiring problem. Every time you crank the engine, your MAT drops a few degrees, but CLT and TPS stay the same.

Even though it's not really showing up in the log, I think your Accel enrichment may also be triggering sometimes when you don't want it to.

The threshold for TPS accel is set fairly low. If I were you, I'd temporarily set it to 1v or higher just to see if that's part of the problem.

Also, I'd run it in flood clear for a while to make sure the engine isn't just flooded.

Ken

thetech

iTrader: (1)

Thanks Ken, will try those and report back. I can't imagine what wiring issues I could have - the MAT for example is just the 1 wire back to the ecu (via the DIY autotune harness) and then grounded on the block. I will double check the engine is grounded sufficiently (the MAT is the only sensor grounded on the engine block).

thetech

iTrader: (1)

Well I fixed the MAT issue (I replaced the engine ground strap with a 4 gauge battery piece), adjusted the TPS accel, but still nothing. The car will idle really lumpy and die as soon as I give it any gas.

I'm pretty much stumped here; I rechecked all the injector and coil/distributor wiring, all the sensors are reading fine as per the logs but it's still not running right. What's most bizarre about all of this is that we did so many dyno runs with this exact configuration without a single hickup!

muythaibxr

Where do you have the megasquirt grounded?

thetech

iTrader: (1)

To the chassis, pins 7-9 and 17-19.

muythaibxr

that could be part of the problem... Every time I've seen a megasquirt grounded to the chassis, with no ground to the engine or battery directly, I've seen all kinds of weird problems... including temperatures changing just from touching the brakes, accel enrichment that triggers randomly, or being very sensitive, tach spikes and dropouts, etc...

When you had your engine running, those datalogs showed many of these symptoms.

My suggestion is to ground the engine to the battery, and the MS to the engine...

You will have all kinds of problems if the grounds are not done right. On tofuball's S5 rx7 we measured a voltage drop of 2 volts or more when measuring from the chassis where the stock ECU was mounted to the negative terminal of the battery.

Ken

twister7

iTrader: (3)

try the 029w firmware ,,the 029v did some thing like that on mine ""like running on one rotor""!

http://megasquirt.sourceforge.net/e.../alpha/029w.zip

muythaibxr

He was running 029t, which didn't have the changes that 029v did that made it run funny, I don't think that the firmware revision is the problem. Also, he's running fuel only, which means even with 029v he wouldn't have seen these problems.

thetech

iTrader: (1)

I seem to have isolated the problem to 'wiggling the DB37 connector' - if I wiggle and hold it in a certain spot the car runs fine. It was suggested that I get a multimeter and do some testing, but being too lazy for that, I ordered a new V3 pre-asembled from DIY Autotune.

Will keep you posted, thank you all for the help so far.

muythaibxr

I would strongly suggest regrounding the MS to the engine, and the engine to the battery. I've had many odd problems with grounding to the chassis.

thetech

iTrader: (1)

Absolutely will do; currently I have battery -> chassis, engine -> chassis, MS -> chassis. I will add another strap from the engine -> battery and then change MS -> engine as you suggested.

thetech

iTrader: (1)

New problem: I just bought a new V3 assembly to replace my somewhat-broken 2.2 setup. I have no idea what IAT and CLT sensors are on my engine (the 2.2 was all setup when I bought the car) - I've tried the GM, Bosch and RX7 defaults from within EasyTherm but no luck - MegaTune on the V3 reports the IAT as -40*, where the 2.2 is reading it (correctly) as 60*.

The good news is that the old 2.2 board is setup correctly with the appropriate sensors - how can I retrieve the appropriate s19 files from the 2.2 board and upload them to the V3?

Help appreciated!

muythaibxr

You can't retrieve them from the ms1... you can just take the ms1 chip out of the 2.2 and put it in the v3 though.

Ken

thetech

iTrader: (1)

I'm an idiot; needed to install the 40k resistor on R4 for the stock IAT. Car runs great, all problems solved...must have been a problem with the old 2.2 board.

Thanks for all your help!

muythaibxr

haha, hadn't thought that you might've missed a resistor!

Glad your problem is solved!