Hi,

there is this thread running at MSextra, but none of the MS guys seems to really care about the problem.

I know I am not the only one (especially of the 99/00 Miata guys - Reverant etc.) who experiences these hickups at light throttle conditions.

I made a lot of logs and screenshots...

Here are some screens -

Again - you can see the changes in AFR, when nothing else really changes. Not the TP, not the TPSdot, no change in any of the PWs.
I can reproduce it mostly in slight uphill drive with steady throttle at ~15-20% and MAP being just below boost (~90kPa)









Last week I might have made a new step to see what causes the hickups...

I had used Tuner Studio with an Laptop before with an serial to usb adapter. There were some problems with that in the past so the data rate was set to 10 frames per second. (And I forgot about that...)
As I am using an In-Car PC now with an direct serial connection I set the data rate to max now and did some new logs.

Here is a screenshot of the hickup with the new data rate:



As you can see in the screenshot of the log - the RPM starts to fluctuate wildly just before the hickup-event.
(Which you can see in the AFR line).

So what could cause this?

I have no sync losses. Which I would suspect to see if the CAS or CAM input would drop out...

My RPM lag is set fairly high(low?) at 85. Should I try a lag factor closer to 50 or 60?

Thanks for your input. I hope to get rid of this annoying stumbles....

And I think the whole 99/00 crowd would appreciate it...

 

This problem exists on all 90-05 models, and I'm pretty sure it is tied to sequential injection.

 

This is why I suspected it would be better off on MSextra, but there seems to be no care for MS2 anymore...

 

Yeah, I may have to resort getting it fixed myself. I don't know if I am that good with the MS2 code, but the latest development I have done has given me enough confidence to try it out.

 

Please do Reverant. It's something which does my head in and despite bumping that thread whenever I can on MSExtra the devs just seem to ignore it.

 

Have you tried playing around with the injection timing to see if it impacts when the problem presents?

 

My DIYPN suffers from this as well. Glad to know I am not the only one!

 

It is not due to injection timing. My car has fixed injection timing and the problem appears randomly.

 

Sure, but move the timing to see if it changes the symptom, if it doesn't no harm, if it does then it tells you it is a timing vs spark related thing. Batch is very different timing to sequential (from a fueling vs spark perspective) - gives you something to try rather than doing nothing waiting for someone to tell you different.

 

I've tried moving the injection timing around, makes zero difference.

What I think's happening is a squirt is being missed, just a solitary one under certain conditions which creates this stumble. But getting the MS Devs to notice isn't happening.

 

Would this be similar to "Lean Tip-in" that us guys with a stock ecu fix with one of Olderguy's O2 clamps? Maybe your wideband's output is out of step with what is going in in light throttle.

(I made a O2 clamp (similar to the Olderguy one) that can be adjusted to trip in vacuum. One of my Mazdaspeed owning buddies is testing it with his otherwise stock-ish MSM. If anyone wants the component part numbers I can find them and post up here. Note the components cost more than Olderguy's unit does shipped).

 

it's not tip in either i.e. Acell Enrichemnts being off, you can see that from the logs Zaphod posted.

 

James is looking (at MSextra) for someone near Birmingham to try to find the culprit...

 

Which I am but my car is immobile currently. I've said I'll ask around on Nutz so see is anyone is in the midlands with MSII

 

Should be an DIYPNP or MS2 with sequential mod... and 3.1.1 or 3.1.2 code running...

 

I knows this :P

 

I knows that U knows

 

I had to have another look at my logs to see what happens, when the misfires start:

I had a look at every single value available in MLV - now the only thing (but certainly not unimportant) is this:

Always just before the misfires beginn the timing err% starts to swing wildly.
I have no idea what this means, but maybe one of the knowledgeable guys here can explain me, what this means and what could cause this...

Here is the same log as above with the timing err% as light blue line: (AFR is the light green line)

 

Richy I'm running MS2 with Sequential built by Reverant, maybe I can help as I also experience this?

 

Timing err% means nothing other than what tooth was expected when vs when it was. Considering a falter just occurred, it would be normal to see this just before the AFR dip/spike. It is an indicator, not a cause.