Mitchell's Megasquirt finally runs!

[S_Mitchell]

After I cleared the difference report, it did say there was a comm port issue. I had forgotten about that. I have a Windows 7 PC, serial cable ends just under the rear seat support rail, and a 10' usb from there, FW is 029y4a. I'll be working on comm ports today.

[Piledriver]

You are far better off with a long serial cable and short USB cable in my experience.

Also make sure to turn your FIFO buffers to 1 if the driver gives you that choice.

The latest alpha FW and TS betas probably fix some serial issues...

[S_Mitchell]

I've wanted to pull my hair out today! Finally gave up and decided to quit working on the car until Monday. For some reason I decided to look at the short datalog I made. Even though I don't really know what I' m looking at, I know it doesn't look right at all. Think I have noise. The Duty Cycle and Pulse Width are everywhere, but appear stable on the TS dashboard.

[westypoo]

my comms issue i had was cause i had the usb/serial adapter at the ms unit, and the same 10ft usb cable running to my laptop. put the adapter at the labtop, and run the serial cable to the ms unit. it solved my issue with comms. i now am running the gigaware usb/serial adapter cable that you can get from radio shack, and ive had great luck with it.

[Piledriver]

The only setup I have had work well with the USB adapter at the MS1, 2 or 3 is with MS3...that has an FTDI adapter built into the CPU board. I have a BT adapter on the std RS232 port, (turn it off to use the USB or the adapters fight each other, all you get is junk)

With 3 different adapters, plugging the adapter into the Laptop and then make the long run with a serial cable worked FAR better. (my USB adapters all have really short cables, extending the USB did not work for jack.)

Also, I treat the mounting plate for the MS as the "engine" as far as all the grounds are run, and have external coil drivers (LS2 coils have the drivers built in, so ~no current logic level signals only)

I keep the nice, fat power and ground cables directly to the battery as short as possible, everything lives under the back seat.

[S_Mitchell]

I've wired this thing every way Pile has said not to. Wasn't my original plan, but when I found giant holes already drilled where the the air conditioner used to be, I couldn't help myself. Everything except the O2 is bundled together, and they all enter the car at the same place, 3" or so from the relay box. I don't expect it to be alt noise, but I'll check that today and try to track down some shielded wire to reroute my VR.

Also, I treat the mounting plate for the MS as the "engine" as far as all the grounds are run, and have external coil drivers (LS2 coils have the drivers built in, so ~no current logic level signals only).

I do have coil drivers in the box also, they wouldn't be a big deal at all to move. Can you explain using the mounting plate as the engine?
All my wires go to the relay box and I use the ground, by the main power and 12v switched, to go strait to the battery. Do I need another cable from the neg battery post to the cases of the relay board and MS?

Also should I route my coil driver output wires with the sensors or with the VR, and would it be helpful if they were shielded too?

[Piledriver]

In the MS installation guide, you run all the grounds to the engine block.

As the guides are generally made for "normal" cars with the battery next to the engine and the ECU under the dash, that is a reasonably good way to go.

However, on an ACVW, you have your main power source/ground under the back seat, nowhere near the engine, physically or electrically, and the "dash" is about as far away from the engine or battery as you can get.
(VW busses are a possible exception, but bear with me)

I installed my setups under the back seat passenger side, on an aluminum mounting plate with the fuse panel and all the relays etc, so I can pull everything as a unit in seconds.

All the grounds that the guides say to attach to the engine block I attach to the ECU mount plate (~6 inches long, max), and have a pair of as short-as-possible 12 ga jumpers for power and same for ground straps to the battery terminals.

The only grounds I have at the engine are the power circuit grounds for the LS2 coils.
These get their power from their own pair of fuses powered by another relay, with it's own 12ga power wire from the battery, with a ferrite core with as many winds as i could fit on a pair of rat shack split ferrite cores to avoid feeding back ignition noise into the electrical system..

On my MS2 setup, when I initially used "normal" coils (from a HD twincam) I installed the PWM injector drivers and coil drivers in my original MS case, and had the MS+ *error board in an MS3 case, making up shielded cable in between.
(I added 1K 5v pullup resistors to the coil drivers to control the LS2 coils later)

The "high power" (injector/coil driver) harness from the drive case was routed back to the engine on the other side of the car from the "sensor" harness.
(was using a v1.01 board set up for full sequential, was concerned about noise, so i went with best practices I know from being an electronic tech, was an ECM troop in the USAF back in the day)

The battery is the best (and probably only) effective noise filter you have. Use it.

Pulling 12v for power from the "main" harness (as in an in-frunk install) with the chassis as the ground is an absolutely fabulous way to see all your alternators electrical funk greatly amplified on all your inputs.

Overkill? I have zero noise issues.
When I initially decided to "try out" the MS I had a wiring mess with long pwr/ground lines and a noisy 40 year old alternator, and it would_not_work_for_crap.

After doing a proper install as described, all the electrical noise related issues went away.
(had to buy a serial cable though, as the USB extension cable setup didn't fly)

[S_Mitchell]

I guess something in my corrupted haywire settings really messed my engine up. I can no longer turn it by hand. I have a high torque starter, and it will turn over and run. I noticed last week that when I turn the ignition off, the engine abruptly dies, there is no wind down of any kind. I don't even have it in me to do a tear down right now. It will be a kick in the junk to go back to a 1600SP from my 2017.

[MarioVelotta]

That sounds like a bearing failure to me, I've had it happen a few times now. I have switched cases because it was the rear main each time.

[S_Mitchell]

Pulled my engine Friday and tore it down yesterday. #2 was the seized one, 4 was pretty bad.

[S_Mitchell]

I'm back in business! Took me long enough, but I finally got my engine back together and broke in the cam yesterday. Planning on running ignition only and carbs for about 500 miles then switching back to injection.

[S_Mitchell]

My car is having the exact same problems running ignition only as it did with full EFI. It stumbles between 1800-2300 RPM, has no power when leaving from a standstill, and bucks under cruise. It performs great under WOT and is smooth when accelerating just slightly. I have adjusted dwell, spark duration, and my spark map all to no avail, I don't know what else to do.

[S_Mitchell]

Thought I'd post an update. I sent my ECU to Mario to check out and it works great now. I had more engine problems and I'm back to the trusty old 1600SP. It's running the crank fire and its performing flawlessly now, just not as fun to drive as the 2017.

[andy198712]

Thought I'd post an update. I sent my ECU to Mario to check out and it works great now. I had more engine problems and I'm back to the trusty old 1600SP. It's running the crank fire and its performing flawlessly now, just not as fun to drive as the 2017.

at least its running smooth now though!