DIYPNP smooth idle help!
 

Hey all, hoping to get some tuning advice from you guys here. I'm a total mechanical/car noob, and have been learning the ropes with this miata I've picked up.

I have a stock 94 R pack miata with a DIYPNP I've put together.

The DIYPNP is set up for sequential injection, and I have a MTX-L wideband hooked up (plugged into the stock O2 bung). Using the GM IAT. Everything else is bone stock. I purchased TunerstudioMS and MLV. Loaded the recommended DIYPNP base map.

At idle, the car runs rough. Revving it up past 1500rpm, it begins to smooth out, but I want to work on my idle first. I've modified the VE table a bit so it's not running super rich in CL idle (seems to sit around 14-15AFR).

Setting up my fixed 10 deg. timing:

The timing mark was very smooth and dead on until I raised the timing offset past 3.7 deg. The white timing mark started to jump all over the place as soon as i bumped it to 3.8 deg. Why is this happening? I can get it to about 9 deg before it starts to jitter.

I ordered new plugs and wires as I don't know when they were last replaced (the o2 looked OEM at 130k mi).

MTX-L wideband offset:

AFR in the tuning software reads about .4 high. I have ground connected to ground and 12v connected to 12v via the DB15 connector. I connected the wideband to the DB15 with a homemade CAT5e cable + 4-pin computer molex connector. I've compensated for the extra resistance by modifying the linear voltage mapping (.14v to 5.14v instead of 0v to .5v) as I understand it's a ground issue. Kosher?

Any thoughts/advice are appreciated. Thanks in advance.

Played around some more and got a closed loop idle capture.

I noticed my battery voltage was fluctuating between 12.9 and 13.2 while idling. Enabled EGO, engine still shakes quite a bit under 1800 RPM. Could someone take a look at my MSQ file to make sure I don't have anything totally screwed up?

Wideband seems to be affecting the battery voltage, fluctuates rapidly between 11.7 and 12.1 with the car off (smoothing lag factor 100). With the WB disconnected, it has a slower range of .1V It takes a smoothing lag factor of 30 for it to be stable with the WB. I suspect this is was not helping my idle, injector MS/V was set to .2ms/V Should I drop a big capacitor in between 12V/GND in the DIYPNP?

About 6 ft of cable between WB and DIYPNP. 5A fuse inline with 12V.

Another MSL attached, the big dip is from the headlights. WB is attached, smoothing factor 100. Car is off.

Thoughts?

Where is the wideband getting power and ground?

Try turning closed loop idle off; it's not very good at catching a hunting idle. You want to stabilize the idle through fuel and spark tuning before you use closed loop.

DB 15 pins 1 and 6 are gnd/12v respectively.
https://www.miataturbo.net/attachmen...ine=1364567813

Thought I'd pull the plugs, one of these things is not like the other.. Closest to firewall on the left. Now I'm scared to continue.

https://www.miataturbo.net/attachmen...ine=1364567813

Always start with new plugs.

I replaced plugs and wires yesterday, played with a few settings and now I'm happy enough with idle RPM, for now.

Idle below 950 RPM and around 1600-1800 rpm is still far rougher than stock. Sacked up and decided to see how she runs now.

Half my horses were put out to pasture. Exhaust sounded lazy and muted. Ran MLV VEA 3 times, but it did not help the larger issue.

https://www.miataturbo.net/attachmen...ine=1364690149

https://www.miataturbo.net/attachmen...ine=1364690049https://www.miataturbo.net/attachmen...ine=1364689982https://www.miataturbo.net/attachmen...ine=1364689982https://www.miataturbo.net/attachmen...ine=1364689982https://www.miataturbo.net/attachmen...ine=1364689982https://www.miataturbo.net/attachmen...ine=1364689982

Is this just a matter of tuning spark/VE or am I missing something serious?

Thanks for looking!

Two thoughts:

1) Trigger offset angle oddity


Start -> 12sec: 4deg trigger offset
12sec -> finish: 3deg trigger offset

Why the huge difference? Stock ECU was never this stable (GND-TEN)

2) Good call on the plugs. New plugs on the bottom, old on the top. Old were run with the DIYPNP briefly. (4-3-2-1):

https://www.miataturbo.net/attachmen...ine=1364831166

Scary. Can I troubleshoot the DIYPNP on the bench? I assume this is the cause of the shakes and power loss. Hesitant to start the car in this state.

Could you post the actual MSQ and a data log of the trouble running?

Latest MSQ is attached in post 7. Easy driving log in 7 as well, but I don't have an up to date idle log. Thanks for looking.

Ran compression:
Cyl: 1-2-3-4
Result:180-150-180-180

It was 180-155-180-180 a couple months ago

You might want to rewire the wideband to have separate power and ground. That will cut down on noise as well.

Problem solved:

https://i.imgur.com/6J5DEzk.jpg

3rd pin from the top right on one side looks to be marked as 4Y to me. On the other side, it is correctly marked as 4U. Anyone else have this problem on their board?

INJ2 should have been wired to 4Y and ended up at 4U.
INJ1 (cyl3, problem cyl) should have been wired to 4U and ended up at 4O (which looks like 4U when cut off) because I read "The injector at cylinder #1 is 4U" and thought INJ1 was 4U.

It was a perfect storm of events that made sense at the time. I can't believe it ran as well as it did. INJ1 must have been open or partially open and running super rich. I'm surprised the other cylinders aren't hosed from running so lean...

I found this problem by using the DIYPNP test mode with a multimeter (UniTrend UT61E) in duty cycle mode.

The timing mark no longer jumps around either. I think I will rewire the wideband since VBatt still wiggles when it's connected.

Made a quick run tonight and it was great, can't wait to dive into the details.

Thank you all for following.