High load AFR unstable :(
 

Hi all

Still having issues with AFR being shitty sometimes and awesome others. I was happy enough if it was semi stable at high loads but its not always. Here are 2 pics of 2 differnt days but with the same tune. First is great second is IMO horrible and unusable. Any opinions on wtf could be happening would be welcome. Car doesnt feel / seem different regardless of what AFR says is happening. Can chop up the logs and post if that might help.

https://www.miataturbo.net/attachmen...ine=1353482121

max = 13.0 min = 12.6 Target = 12.9

https://www.miataturbo.net/attachmen...ine=1353482121

max = 13.6 min = 12.0 Target = 12.7

Do you have like 2000cc injectors installed?

Why are yo uusing the default range from the LC-1? Most of the time we use 10-20AFR, not 7.5-22.3. How are the ground run? Any chance one log was taken with the lights or the AC on, and the other without?

why do you reduce the resolution by ~30% if you dont have to?


his issue is the injectors.




































In before:




BUT BUT BUT ID1000cc injectors are the best thing in the world.

https://www.miataturbo.net/attachmen...ine=1353507112

and now we postin in a troll thread.

I'm sorry if you cant handle harsh reality.

I didn't have this problem with 750cc IDs on a 140hp 1.6.

and his injectors are 30% larger. look at his PW between the two logs, but the again the load is hardly different. With Injectors this large and horsepower this minute, you need to be able to control these things by the thousandth of a second to get good AFRs...so perfectly inputting deadtimes and voltage corrections curves is vitial.


Oni, can you post the logs instead of screenshots? I'd like to look at other things in the log.

Just getting Office to get the logs to a smaller size, they are all taken while driving to work.

Grounding ive tried a few systems

-Power from blue connector behind headlight and grounded to head where ecu grounds
-Currently power from MS harness and ground to MS harness ( Rev MS unit comes with these to use for wideband ). currently crimped, have tried solders and also twist and tape without any help.

I hope its not my injectors :( i got them to fix this issue based on the recomendations from here. ah well :). Ive been using the deadtimes / voltage correction from the provided data but that could be incorrect for my car i guess.

I got the ID1000 because i have limited access to E85 but want to get everything solid before swapping. Currently only one pump in the state but its semi close to work.

I was worried about low PW stability but i saw a vid of ID2000's idling on a small engine, granted it was super rich.

Also im just using defaults becuase they match on my gadge and in TS so didnt see a reason to change it, happy to try a change though if it might help.

Cheers
Scott

OK log time, 3 in total, good day is from the ss above and so is bad day. the third is one that shows me driving for about 10min total pulling over 2 times to power cycle the car, to show how differnt the stability can be by restarting, worth a look at :). somedays its awesome and i dont have to restart at all, other days i can restart 5+ times and its still bad.

Cheers
Scott

Are you using mulitply map?


honestly I'm looking at the bad day log and trying to figure out how it could be any better...

The injectors are cycling in a range of 3.69xms and 3.63xms and giving you those AFR fluxations...at WOT for over 1000RPMs. that's it; what more can you expect out of them? You want better fuel control between 0.06ms of time...on batch fuel.

Looking at the good day log, I dont see much difference other than a longer log and helps make the peaks and valley filter out. Zoom out of the good day and afr stability log, and it looks just as "bad"

I still cant fathom why people are using 1000cc injectors and utilizing only 3% of capacity.

Greetings,

I dont have a Miata but I have a Protege with the same BP 1.8 engine as the Miatas with ID1000 and works great idles perfect almost as stock with 14.7 AFR.

MS3 with a MS3X board ECU running sequential.

But in order to for this injectors to work it is absolutely critical that you set the dead time properly and the Req Fuel or you will have problems.

I can share my tune if you want.

I've noticed that too over the past 2 years more and more people are buying unnecessarily big injectors. I'm guessing because the market now has big injectors like that people just keep buying them. Remember before when rx7 injectors were considered big? Oh the days :crx:

Maybe people buy them that big just in case they decide to run a borg 17 years from now?

Because nobody makes an EV14 injector in the <500cc/min range. I has a sad

DW makes 600cc injectors. Personally i thought those would drown the market.

Lolwut? I has 445cc FIC EV14 injectors.

mother effin this.

Thanks for the comments im at work ATM but ill grab my injctor settings tonight for people to look over.

Found the multiply map setting and it is set to "Multiply", should i have that on or off? :).

Its intersting if you look over the good day log at high load and just press the arrow key you only see changes in steps of 0.1-0.2 where in the bad day you see 0.1-1.0, do you think this is just beucase its not sampling the same?. would it help if i made 2 logs that are over the same time period? i cant help but still think its differnt as when i watch my afr on a good day it will sit within a small window and not change much compared to the numbers dancing on the dial.

I think people are going with ID1000's becuase were being told there is no reason not to and for some extra head room. Im planning 250rwhp on E85 down the track so 1000's seemed like a good idea :). but for now i just want to get a NA tune squared away.

If anyone has a SS of there injector settings for there ID1000's id love to see them :). i did post mine before ( cant find them now ) and no one poked holes in them though.

Cheers
Scott

Include a set with my ms3x. kthnxbai :party:

its okay for you, since it seems your fuel map is perfectly tuned.


try logging the same amount time like you suggested, the bad log seems to have a more noisy battery log.