sparkstack

Guys, this is my first time with any aftermarket ECU, so go easy on me if you see anything dumb and if you want me to produce any logs or other files please let me know.

Here's what i have:

1990 1.6 Miata with FM Voodoo II kit. Tan Top injectors. 5psi boost.
AFM Delete, GM IAT. Stock TPS (unplugged)
Electronic Boost Control (mounted & wired, not plumbed in or activated)
AEM-30-4100 wideband, Powered by the DIYPNP and connected to the stock 02 input.
DIYPNP unit, wired and configured as per the instructions here My only change being the DIYPNP is running MS2/Extra 3.1.1 release firmware.

Base tune from the link above.

Set CLT sensor to values shown on the page above. However I've noticed the CLT sensor readings are really high. the DIYPNP and therfore tunerstudio reporting my temps in the 240f range. even though the dumb guage is to the left of vertical. I've never had an issue with overheating on this car, so i'm inlined to belive the DIYPNP/tunerstudio is reading high. This may be effecting what i'm describing below?

Initial start from the 240 base map was sluggish, even after i adusted the req fuel for the 265 tan tops. Was running really lean, numbers in the high 18's. and ocilating quite badly. Eventually was able to get a stable enough idle so i could check base timing. Locked timing to 10 and checked against the crank with a timing light. My offset comes to 2.

To get the idle to smooth out i adjusted the air bypass screw. Now i don't know what the initial (stable, stock ECU, 850rpm) idle was because stupidly i didnt count the number of turns i made of the screw
Is there a way i can set the car to a known good idle like the old GND & 10 procedure without having to plug the stock ECU back in?

I have a problem getting a good idle, especially with the AC on.

From cold start, in garage and not driving, it idle's above 1500 and then as the engine warms it slows down to just below 1000rpm. Fair enough. Putting the headlights on makes it occilate a little, but it settles down again to just below 1000rpm. However, putting the AC on makes it bog horribly, car sounds like a diesel and idles around 500rpm. Not good.

I've got the car running pretty well during normal driving, crusing and accellerating. There is obviously more work to do and Autotune has done the most work so far, i just adjusted the AFR table. It's leaps and bounds smoother than using the Voodoo box with the stock ECU.

The interesting thing is while driving the car, the idle hovers around 1200-1300rpm. If i put the AC on the idle drops down to 1000 but the engine doesn't knock. This is repeatable while driving, however if i then stop the car and restart it, idle resumes at 1000rpm and cannot be made to deviate from that unless i take it for a drive again.

Matt Cramer

Fix your temperature reading first. What sensor calibration data did you send it under Tools -> Calibrate Thermistor Tables?

sparkstack

Thanks for the reply Matt. I set it to this - taken from the 1990-1993 Application Doc on your site.

For the CLT, use the following table with a bias resistor setting of 2490 ohms:

Temperature F / C Resistance In Ohms
-4 / -20 16150
104 / 40 1150
176 / 80 330

Matt Cramer

Could you check the voltage on the CLT input wire on the main board when the car is cold and when it gets up to temp?

sparkstack

I'll give you cold and fully warm in the AM but right now, after the car has been sitting for around 4 hours, TunerStudio says 107F and the voltage on 1Q is 1.499

sparkstack

COLD 2.419v TunerStudio says that's 69F.
It's also telling me the Air Temp is 70F. Which is about right. Car is in the garage.

As an aside, can't get the car started this morning. Will swap in the stock ECU and get it running & the timing and idle setup then put the DIYPNP back in. Hopefully i can get it to fire up.

sparkstack

HOT (after running with the OEM ECU till the fans come on) is
0.010v TunerStudio says 208F

sparkstack

Also, and this may/maynot be helpful. If i stick my fluke thermocouple in the head outlet back to the radiator i see a steady 194F at idle.

sparkstack

So it runs.

Fired it up with the oem ECU and got it to give me a basic idle. Adjusted the idle screw accordingly and will not touch it again!

Plugged the diypnp back in and it fired up, though was very reluctant to idle. over the course of a few hours i've got it to idle quite nicely even with the AC on, courtesy of the latest beta firmware and the AC up idle code. With the general release code it couldn't get both.

Driving the car and it's very smooth. Still running auto tune to get it close to spot on. One thing though is it seems a little too smooth, if possible. seems like the engine wants to give me more but is being held back, especially in first gear. Most likely the spark table as i'm using a very lightly modified version of the diypnp default tune.

Still need to figure out why my coolant temp is reading so high, but msq and log files attached. Comments and suggestions appreciated.

sparkstack

Matt? where'd you go?

Matt Cramer

That appears to be a bit too low - maybe there's a partial short between the coolant and ground. I'd check for stray flux and such on the adapter board.

Seefo

iTrader: (5)

how do you know you aren't actually running that hot? My fans don't turn on till about 200-205F reading on the scangauge and a bit of movement on my modified water temp. I have easily hit 208F while sitting in traffic before my fans really started to change the temp. I think your readings seem normal. and the 194*F reading is spot on for the stock temp t-stat (192*F on the NB, can't imagine the NA is far off from that).

Plus, your temp gauge is a dummy gauge. it has a dead spot from 165-225ish *F where it stays in the middle. I would shy away from using that as a reference for anything other than overheating.

sparkstack

Because like i said i stuck my thermocouple in the inlet hose and got 194, MS thinks it's 208. 14 degrees difference seems a bit much, even though the temps are being measured in different places. if i'm wrong, fine, i just want to know.

So do i. I was worried at first because tuner studio kept sticking the gauge way over into the red. Seems i missed the part that told me to turn on Expanded_CLT_TEMP. Now it's not freaking me out as much, but again, i'd like to be sure it's getting the correct data.

I'm aware of this. Hence the fluke in the filler neck.

Seefo

iTrader: (5)

What I am saying is, if your fans are on and its reading 208*F then its probably reading correctly. fluke or not, I don't know, did you check them at the same time? your posts are a few hours off so I am not sure there. but it seems like you got a good idea of how to get this fixed.

sparkstack

Fans are on. But then that is controlled by the temperature i set in the MSQ config for the fan output. I may be running a little hot (it's a 20 year old radiator for starters) but i'm not overheating.