AE Help

[kev_rm]

MS-II v3 running latest beta code .684.

Can't get the off-idle AE right. Goes lean, then goes rich afterwards. tried lots of combinations. This engine has a large throttle body to displacement ratio (65mm:2.6L) and an agressive manifold.

It was a lot easier to get right with MSnE!

Would love any pointers.

http://home.comcast.net/~kevin_connell/ ... 241232.xls

[Bernard Fife]

kev_rm,

Is this with X-Tau, the standard accel entrichments, or both?

If the standard (or both), are you using MAPdot, TPSdot, or a combination?

Are you using AE rpm scaling? If so, is the low rpm high enough?

Does this occur mostly when cold, or warm, or both. If cold, note that the cold accel mult needs to be increased in the 2.684 code, mine went from 130% to 155%.

Finally, what is you TPSdot and MAP dot thesholds, if applicable?

Lance.

[kev_rm]

HI lance, I linked in a screen shot of the AE page, but to answer your questions:

1) no X-Tau.
2) combination MAP/TPS due to large tb diameter. 50/50 seems to work best so far.
3) Not sure what AE rpm scaling is, sounds interesting, most of the problem is at off-idle!
4) warm
5) 40kpa/Sec, 5%/sec. little worried about 40Kpa/sec because of turbo boost, but we'll solve that problem when we come to it.

[kev_rm]

I read up on scaling. Shouldn't be a factor yet. Want to get off-idle working first then deal with rich accell at higher rpm later!

EDIT: 2500/5000 is the setting.

[Bernard Fife]

kev_rm,

Ah, when you post a link to an image, if you put {img}{/img} tags around it (but use square brackets instead of curly brackets), the image will show up directly in the forums and emails.

You accel time is very long. I would drop it to 0.2 or 0.3, then if necessary increase the pulse widths values.

Note that though the pulse width under accel goes to 11.2, the engine is actually slowing down when the pulse widths are that high, and it doesn't really start to accelerate until it gets back down to around 5.0, so you may need to decrease the pulse width bins a fair bit.

If this were mine, I would also use a much larger range for the TPS%/sec bins. For the top bin, I would use at least 200%/sec. (Then maybe 100, 50, and 15). I would also use a higher bottom bin, around 10%/sec to 15%/sec. At 10%/sec, it takes ten seconds to go from closed to wide open throttle, and the VE table should be more than capable of handling this.

If the stumble is still there, I'd work on the VE bins just above idle, they may be too lean.

Lance.

[kev_rm]

Hi, I'm confused, why less gas? the engine just goes lean in this log, and any decel when prior to throttle shutting here is lean stumble - i just blipped the throttle here, that's it. The log does not indicate it goes rich before it goes lean if you were thinking rich misfire?

kev_rm,



Note that though the pulse width under accel goes to 11.2, the engine is actually slowing down when the pulse widths are that high, and it doesn't really start to accelerate until it gets back down to around 5.0, so you may need to decrease the pulse width bins a fair bit.

[Bernard Fife]

kev_rm,

A couple of reasons:

- the response time of a wide band sensor is too slow to be used for transients (despite the marketing claims),

- a misfire from a too rich mixture feels a lot like a bog.

Lance.