DIYPNP Starts, But Wants To Die
 

Hi All,
I'm in the process of installing a DIYPNP v1.5 that I recently built for my '90 1.6L JRSC Miata. It starts, but I've got to work the throttle quite a lot to keep it running. RPMs rise but inevitably crash all the way down to 4-500 no matter what I do with the throttle. RPMs are very poorly correlated to throttle input.

Other details:
-all edge pin connections double checked
-PLX wideband O2 sensor currently wired to simulate narrowband (trying not to add variables)
-running DIYAUTOTUNE's 460cc basemap, with injector size and required fuel changed for stock injectors
-RX-8 yellow bodied injectors waiting to go in and Jackson AFPR waiting to get ripped out (trying not to add variables)
-stock TPS switch for now (no VTPS)

I've attached the current tune file and a log of it trying to run. Only two (perhaps irrelevant) things in the log stand out to my inexperienced eye:
-Injector duty cycle and injector pulse width do not seem to be directly related. I'd have expected one to simply be a multiple of the other, but the traces move independently.
-RPMs and pulse width / duty cycle are inversely proportional to one another and all are cycling faster than the narrowband O2 is giving lean/rich feedback.

If anyone can have a look at the attached and make some suggestions, I'd really appreciate your thoughts on how to get the thing to idle. Thanks!

...Looks like my files didn't upload correctly, trying again...

It would be easier if you set the wideband to give a real air/fuel ratio; that way I could tell if your fueling is correct.

Sounds like the idle valve is not tuned yet.

*edit* Are you saying that even keeping a steady throttle angle the engine still droops to 500 RPM?

^i was just going to say what he said

I know a few guys in the Madison area that may be able to help you locally as well, if not I've had pretty good luck troubleshooting my DIYPNP and I'm about an hour from Madison myself.

-Matt Cramer: That makes a ton of sense. Will try to get to that today. Might try the 230cc basemap just to see if it exhibits the same startup symptoms.
-Ryephile: Both assertions are correct. Even at a throttle angle that should net me 3k rpm (or more), it's still oscillating wildly wanting to stall.
-Padlock: Cool! Glad to hear I'm not alone around here! If I can't work this out on my own, I may come begging for help and/or contact info.

Thanks for the input, guys! Will get that wideband rewired and report back with improved logs.

I wired up the wideband as suggested, then tried
-running the 460cc basemap (with required fuel changed for stock sized injectors and wideband set) - attachment 1.
-running the 240cc basemap (with wideband set) - attachment 2.

It's lean-ish everywhere, but very lean at low RPM / high MAP values. Is it likely something that should be dealt with via (large) adjustments to the VE table, or is some more system-wide variable likely to blame?

Thanks!

Starting over with a new Tunerstudio project, in case I messed with a setting I shouldn't have. Will update when I get stumped again. Thanks all for your help so far!

the basemap isnt good at all, even on an N/A 1.6. Assuming you have the JRSC on it, it would make sense that the VE table would need some serious loving to get your AFR's where you want them.

I'd personally try adjusting the VE table some to see how much it takes to get your AFR's where you want

Hi Guys,
I found the cause of the really high idle: though Tunerstudio was set to idle control valve "none", unplugging the IAC dropped the idle to a solid 850. The fuel was still way off (lean), so I did the following:
-removed JRSC FRP
-removed JRSC "black box" that modified the MAF signal
-installed RX8 yellow injectors
-checked/slightly corrected timing

After that, I started back up on the 460cc basemap (derated to 440cc, 0.8ms dead time, 0.1 voltage correction) and it idled very rough and the wideband was pretty much pegged at it's 20:1 max reading. Rather than tweaking the VE table, I adjusted the injector size (and thus "required fuel") until the car would steady state idle a little richer than stoich. I had to claim 250cc injectors...?

So, I'm stuck again. It seems very unlikely to me that the basemap's VE table needs to be increased 75% in the idle region. The problem is not the new injectors, as it was crazy lean on the old stock ones that were known good. Before I go out and try to find a fuel pressure sensor/gauge and slow this process up a couple more days, can you guys think of a setting in Tunerstudio or a DIYPNP hardware issue that is likely causing this? Are there Tunerstudio settings that persist through a new project with a newly loaded tune (ie should I reinstall Tunerstudio)?
Attached is a log of the RX8 injectors idling the car that thinks they're 250s and the current tune. If there are any other logs that would be helpful, I'd merrily go take 'em.

Thanks for reading, and thanks in advance if you have any suggestions.

Little update, for posterity: Turned the injector size back down, bumped up the VE tables 60% across the board and took it for a drive. Not as scary as I'd been imagining...going to start out really rich and just let VEAL work it's magic. Turns out that the basemap was least appropriate in the idle region that I was confining myself to.
Thanks to all that weighed in. Very excited about the tuning to come...

glad to hear all is working out. have fun tuning!