Mazda MX-5 Miata banner
1 - 14 of 14 Posts

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #1 ·
As some of you may know I've had no end of issues with my supercharged car since it was "tuned" by a certain garage. The car is barely used, and due to other commitments I've had really limited time to work on it or fault find. It went somewhere competent and was largely dialled in except for a somewhat random miss fire.

I've never been able to get my idle quite right (mid 13s AFR and fluxes ~0.5 points) and the car does a weird hunt/lean spike during idle, at the weeknd this got much more pronounced.

My question is does anyone what any idea what could be causing such a spike? Injector sticking, spark missing, idle control doing something weird? etc. The revs seem to rise as it spikes, up to about 1400rpm at each lean spike which makes me think it's not a miss fire (would stumble surely). Idle is open loop control, fuel is open loop, injectors are semi-sequential.

I'd like to ask that feedback remains on topic, and that people actually think about the issue rather than blurt out things like "SPARK PLUGS!", I've replaced or swapped a lot of sensors, plugs, coils etc so im confident stuff like that isn't an issue. I'd like to know if anyone has some better insight into potentially more complex causes/

 
  • Like
Reactions: guitaristanime

·
Hiding from the Ayatollah or making Full Metal Jac
Joined
·
1,809 Posts
I suggest you get a log off your ECU.

Ideally a cold start idle as well as a warmed up post-drive idle.

Ideally you've got a 12-13-ish idle cold and 14.7-ish hot idle AFR average.
 

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #3 ·
Yes, warmup idle is 12-13, hot idle is as in the video (car was fully warm)
 

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #6 ·
This appears to be injector related. A set of ev14 640cc seems to have helped a lot
 
  • Like
Reactions: Martin Y and Zed.

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #8 ·
Rx8 yellows, genuine, flow tested and cleaned. I think it's a combination of these and the limit of the ECU. Even with the ev14s my idle still isn't as stable as guitarists me221 and his rx8s
 

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #10 ·
Likely drivers, the precision with which they are controlled, etc. I spoke to Bailey a while back and they don't tune speedunio due to poor ignition control, they've found some units to feature too much latency and some scattering. I will say though that a few shops are now picking up the DIYEFI and have been very impressed. I'm guessing at the root cause really, and it might purely be my injector setup. Iv also not really put much effort into getting a super stable idle with the ev14s, so that might still be possible eliminating the ECU as a factor at all. All I can say is moving to ev14s has helped improve my idle to an acceptable, but not perfect, degree with limited effort.
 

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #11 ·
I do need to double check a reference tune I have to make sure iv not had a different setting. There does seem to be varried info online regarding rx8 dead times too, everything from 0.7-1ms so it might still be tune related. Ultimately I'm at a point where I just want to have the issue resolved now, and if the easiest way to do that is throw the money at new injectors I'll do that now.
 

·
layabout
Joined
·
2,791 Posts
This appears to be injector related. A set of ev14 640cc seems to have helped a lot
larger injectors on a lower duty-cycle are more efficient than smaller injectors flat-out (I believe 🙃 )

Rich.
 

·
Banish a little rust a day people
Joined
·
5,407 Posts
Discussion Starter · #13 ·
larger injectors on a lower duty-cycle are more efficient than smaller injectors flat-out (I believe 🙃 )

Rich.
It's the opposite at idle though, smaller injectors are able to slow down/spray less more easily to meet the lower fuel requirements at idle. Spray pattern is a factor as well obviously and largely where injector technology has developed in recent years. I'd of though the rx8s should be able to handle the idle, as I said, Guitarist has sent me a clip of his idle with rx8s and it sits dead on 14.5- 14.4.
 
  • Like
Reactions: Zed.

·
Happily expanding the To Do list since 1997
Joined
·
10,617 Posts
Yes, large injectors idle with very brief injection durations and can be a problem when the dead time is larger than the squirt duration and the dead time isn't accurately known and it varies with battery voltage anyway.

The error in telling how much of the injection period is actually delivering fuel can get very large.

EV14s are capable of very good performance at short injection times but the ECU is only going to control them properly when it has a good idea what the real dead time is.
 
1 - 14 of 14 Posts
Top