Sequential Injection - Timing Table

[richyvrlimited]

I'm having the below issue....

See how the AFR dips and the returns to normal? I've tried tuning it out by boxing that area in on the VE table but it's not worked very well, I've just ended up with BIN's closely stacked together and a smaller spike in AFR's but it's still there - and annoying.

(original thread on msextra.com http://http://www.msextra.com/forums...=34600&start=0)

At 1894-ish (2nd manual mark) I go to WOT, AE, (which needs tuning!) kicks in and the AFR's drop as expected, where they shoot up to a solid 14:1 - which is way too lean for my liking, but I'm getting there... suddenly they drop to 10:1 for no obvious reason. Then pop back up to 14:1.

From what's been said on MSExtra.com and from my tuningexperementation/as I understand it in my feeble brain, it's possibly an injection timing issue (I think that at this point the fuel is injected at the perfect moment the valves open/are opened so all the fuel goes in the cylinder and I get a very rich period). It only happenes with sequential injection, with Batch it doesn't occour. It wouldn't bother me if it couldn't be felt when driving, but the car half-missfires/bucks slightly in this period so it's irritating.

it's definately NOT a wideband issue, nor is it noise on the injectors causing them to spike at that RPM. I've scoped the injectors and they're fine.

It happened on my old and busted engine and my new just swapped in this weekend engine

I cribbed the sequential injection settings from DIYAutotune's DIYPnP basemap, this uses a fixed injection timing of -350deg. I'm hoping (nay praying) that I can resolve this by enabling the injection timing table and in the 2-3k zone change the timing from -350 to something which stops the above.

I've not a clue on where to start though, should I start at -300 or is that too big a jump? or should I roll over in the other direction and start at -360 -> 50deg? Initially I thought at setting the entire table to -350 and in the 90-10kpa rows drop that down, (up?) to 360.

Opinions? will this make my car fall on it's arse, might it help? Am I barking up the wrong tree etc

Help!

EDIT: Just adding my spec!

1996 MX5, with a 4G63 CAS, MSII 3.0.3r & Sequential using Jean's 4 channel driver board

[jbelanger]

I'll have a look to see if I can reproduce it on the bench and, if so, find what is causing it.

Jean

[richyvrlimited]

Cheers Jean. I tried to rcreate it on k JimStim but got no-where. The actual commanded and actioned outputs to he injctors are correct, it's just the engines response.

[jbelanger]

Were you looking at the injector output on a scope? Because that looks really strange and a very abrupt change in AFR especially since there's very little change in RPM and MAP and no change in PW. However, have you looked to make sure that PW2,3,4 are also not changing? What about injection timing? Any big change in that region?

Jean

[richyvrlimited]

Yeah I scoped the injector output, it doesn't do anything weird, all PW line up with each other, and currently injection timing is fixed at -350

I thought it might be that with a fixed table that problem area is actually the perfect injection angle, long term it'd be amazing to get the perfect angle all the time as i'd save a ton of fuel! Short term I'd hoped to make it 'wrong' but have a smooth AFR and a smooth engine

any suggestions on where to start with using the injection tming table? I thought about blankng the lot to -350 but then altering the problem area, but to what?!

[jbelanger]

I think that what you suggested before of changing it by 60 degrees each side is reasonable. You'll see if there is a change for the better and then try more to see what happens. If that doesn't help, try less (30 degrees).

Jean

[richyvrlimited]

I think that what you suggested before of changing it by 60 degrees each side is reasonable. You'll see if there is a change for the better and then try more to see what happens. If that doesn't help, try less (30 degrees).

Jean

Hrm, well I set the table to a blanket -350 and the top row from 0rpm to 4k to-340, (taking small steps )

Didn't see much of a change, BUT I now have an issue where lifting the throttle from either a part throttle position or from WOT suddenly causes the car to surge and jerk, Itentical in feel to how a 'burn' feels when driving.

There's no resets/sync losses etc. and it'll stop as soon as i go back to a fixed timing table... It's not the actual timing figures at fault as far as I can tell as it'll happen even when in a -350deg 'zone'

[jbelanger]

This sounds like a decel event and you probably need to either change your thresholds or tune your values.

Jean

[richyvrlimited]

Completely agree, however it doesn't do it when using a fixed injection angle (at the exact angle I have in the table)

Weird! I'll keep messing anyway

Cheers for the help!

[jbelanger]

Ok. I'll have to have a look at what happens with a quickly changing RPM and changing timing to see if there is something weird done in the code. Can you post your latest msq so that I can have the same setup for my tests?

Jean