Page 3 of 7

Posted: Thu Feb 09, 2006 3:21 am
by mikesb
Yeah, I can see why you're saying that. It was the first thing I tried, but I discovered that high map low rpm section is also used during cold running and initial acceleration. I'd tweaked up those bins to as much as 2x what I have now, the lean peak was gone. But the next day I couldn't start the car until I switched back to my alpha-n table. Warmed up, back to the speed-density table - the car would not pull from idle - when the throttle opened it would jump to the high map / low rpm area, run very rich (even after the AE cut out) with no power. I couldn't pull away in gear unless I got it up to 2500+ rpms first. With lower bins there, I pull away quite well after the initial "pop".

The weekend is looking promising, a little snow but only -2C. Looks like AE and the high map / low RPM bins are interrelated and it'll take some tweaking for me to sort them. I've found a construction area not too far from my house, Sunday's it's deserted, a perfect place to play.

any tips to tune out an off idle stumble?

Posted: Thu Feb 09, 2006 3:48 am
by ptegler
cold vs. hot start.... has anyone considered the dual table settings to set two different AE/map table configs?
Perhaps using a thermo/temp switch to change tables after warmup?
 
Paul Tegler
ptegler@cablespeed.com (ptegler@cablespeed.com)
www.teglerizer.com
----- Original Message -----
From: mikesb (m.polan@rogers.com)
To: megasquirt-tuning@msefi.com (megasquirt-tuning@msefi.com)
Sent: Thursday, February 09, 2006 6:21 AM
Subject: any tips to tune out an off idle stumble?



Yeah, I can see why you're saying that. It was the first thing I tried, but I discovered that high map low rpm section is also used during cold running and initial acceleration. I'd tweaked up those bins to as much as 2x what I have now, the lean peak was gone. But the next day I couldn't start the car until I switched back to my alpha-n table. Warmed up, back to the speed-density table - the car would not pull from idle - when the throttle opened it would jump to the high map / low rpm area, run very rich (even after the AE cut out) with no power. I couldn't pull away in gear unless I got it up to 2500+ rpms first. With lower bins there, I pull away quite well after the initial "pop".

The weekend is looking promising, a little snow but only -2C. Looks like AE and the high map / low RPM bins are interrelated and it'll take some tweaking for me to sort them. I've found a construction area not too far from my house, Sunday's it's deserted, a perfect place to play.



Mike


This post is at: http://www.msefi.com/viewtopic.php?p=100847#100847
Posted by email.

Posted: Sun Feb 26, 2006 3:25 pm
by mikesb
Trouble with dual tables would be that I couldn't drive the car till it was warmed up and off the leaner tables.

Here's an update: I've switched to squirt-n-spark extra (still fuel only). I changed to MAP-based accel and decel, and there seems to be some improvement. As far as I can tell, decel doesn't matter if I've got the enrishment set to 100%, right?

I still occasionally get a lean pop when I blip the throttle. I can't raise the AE value any further, as you can see the pop is followed by a rich stumble, which I assume is the accel squirt being burned off. The image shows about 1.5s start to finish.

Anyone have any suggestions on what I should try next? I can't enrichen the 84map/21rpm bin enough to overcome the lean range, as when the engine is running there normally (usually when the engine is cold) it runs way to rich.

Posted: Mon Feb 27, 2006 12:29 am
by arnold
Aren't you experiencing fuel gathering in the manifold?

Posted: Mon Feb 27, 2006 4:56 am
by mikesb
Yeah, I could be. Not sure what to do about it... I think it's these d*mn siamesed ports - when the flow changes direction at low speeds I could be getting a pulse through the log manifold that stalls the fuel. At higher air speeds it's not a problem, just at off idle.

Maybe I should idle at 1200-1500, that seems to help. If I could get some tight mandrel bends I could try S shaped runners 2 into 1. Not much room to fit that though :(.

Posted: Mon Feb 27, 2006 5:00 am
by arnold
Maybe warm up the inlet manifold a bit (coolant), or put it under a slight angle? I experience the same, even with a blower pressing the the air in....

Posted: Mon Feb 27, 2006 10:03 am
by mikesb
Ugh, plumbing for coolant through my home made manifold, I think I'll try that last...

Looking at my logs, my lean spike shows up almost 0.5s after the throttle first moves, so I assumed it was sluggish fuel too. At 600 rpm that's 2-3 ignition events plus another 5 to the rich point. 1 event every 100-200ms at idle.

Thinking out loud now - I only see 1/2 the events 'cause my O2 sensor is on the middle branch of a 5 port head - exhaust signal from cylinder 2 & 3. The response time of my sensor is < 100ms, but add to that the latency of the unit, the logger and the time between events, and it may be only a cylinder or two lean event just off idle.

Mops may have been on the right track earlier - I need a very short, very rich AE injection for one or two cycles. I'm not sure how to do this for MSNS, I think I'll ask over there.

Posted: Tue Feb 28, 2006 11:50 am
by teg
I have exactly same kind of problem in my engine (also siamese port). I haven't found any solution yet - have you?

Posted: Tue Feb 28, 2006 2:26 pm
by mikesb
Still working on it Teg, but rest assured I'll post here if I solve it.

AW asked for more info, so I've added my settings for Extra024s13c. I haven't loaded 26h4 into my MS yet, but it looks like my settings are (mostly) preserved when I tried to look at them using megatune.

My reqfuel is 9, w/ a PW of about 1.8ms at idle. I'm tuning closer to 12:1 as the 5 port engine seems to idle best there so far, and I use that for full throttle too. At cruise I'm running it at 14.7:1 and it does seem to run well there.

I'm using normal AE (no decay) with mapdot accel. MAP bins of 1.0/50, 2.5/100, 5.5/200, 8.0/300 and a mapdot threshold of 30. Accel time of 0.2s.

Posted: Tue Feb 28, 2006 2:48 pm
by whittlebeast
PM me with your email I found the issue

AW