evo8dad

iTrader: (15)

I'm dealing with a car now that doesn't want to idle (272 Kelfords, BC valvetrain, +1mm valves, street head port, BBK Full, 3" hard pipe with Perrin filter, JMF exhaust/intake manis). Hz are staying in the high 50s to low 60s, and the LTFT low doesn't seem to be adjusting (the car is a 96530006 convert currently running Tephra's v7 96530706).

Regarding the trims, I'm going to try and back ROM the ECU until I can get the trims to read, as on the 96530006 ROM the transition point from mid to low is 62.50hz and the low to mid is 75ish hz, so although the idle Hz are a little high it still should be within the low LTFT window to allow it to adjust. the O2 feedback is adjusting but, no LTFT low.

Another note is that when the car is warming up (open loop) its fine, no idle issues at all, although a little high (1500 rpm or so) it never stales out. Once the car transitions into closed loop and idle and coolant temp has risen above 188 degrees the car just simply doesn't want to idle.

I've tried adjusting the BISS, although without activating the SAS mode as once its activated the car just dies. Also I've only been able to adjust the BISS during warm-up as once the engine is up to operating temp it just stales. The adjustments to the BISS seem to have no effect.

I've tried adjusting the ISCV values but, have gotten no where there either. Although I may be adjusting the wrong areas of the map, however, I did make a wide range of adjustments that should have covered the idle area. I've been following mrfred's ISCV disassembly and have tried using some of those maps but, still not quite sure what effects what. I have logged ISCV demand percent.

Attached is a log of the car at start-up through to when it transitions into closed loop and subsequently stales.

Any suggestions would be greatly appreciated.

l2r99gst

iTrader: (2)

The beginning of the log shows -25% for the stft, so it is maxed out. Looks like you need to work on the injector scaling and/or latency a bit. That's probably why the long term trims don't have a chance to adjust yet. Once the switch from open to closed loop happens, I'm guessing the maxed stft and/or lack or LTFT trims are causing your issue.

As far as the biss, you should adjust that, so the car won't die when the actuator is enabled. But your isc steps of 16 would suggest that you have it adjusted out enough.

One other weird thing I noticed is that the MAT goes way out of whack once the stumble happens, but that could just be the reading changing due to the stumble and voltage decline.

What I would do is either work with the maf scaling and or the injector scaling/latency to try to get the trims to begin adjusting and/or the stft away from being maxed out.

evo8dad

iTrader: (15)

Thanks for the input as always, its very much appreciated

I noticed the STFT showing maxed (which is logged right at cold start up by the way) out but, thought that it was odd it adjusted itself throughout the rest of the log but, the readings were all over the place and never settled down to +/1 5%.

I had thought about adjusting the MAF scaling but, thought that even though its idling in the high 50s to low 60 Hz range it was still within the low to mid and mid to low trim window.

I presume though since the STFT is initially maxed out and never really settles down that the latencies still need to be adjusted to compensate or through the MAF scaling.

Now since my LTFT low isn't registering but, my STFT is I assume that I should simply reduce the MAF scaling in the 75 hz and below at 10% increments or so until I find the Hz at idle beginning to lower but, also the STFT to come off of being maxed out.

At what point than do I worry about adjusting the ISCV tables?

Appauldd

iTrader: (22)

Here is what I did to get my Revolver cams and supporting mods to idle.

Try flipping over the MAF and removing the silicon circle. Under it is a phillips plug. If you look through the MAF you will see that the plug actually blocks a bit of the lower portion of the MAF. Adjust out the plug until it is flush with the inside. This will open up the MAF a tad and lower the Hz at idle because more air will bypass the sensor, thus the MAF will read a tad less.

Increase timing in the cells 1000 RPM and below and 60 Hz and below to at least 8*.

Next, up the idle sensitivity to around 146 from the stock 128. This will make the ECU react a bit quicker and hold idle a bit better.

In the MAF scaling table up the values in the 75Hz and below by 15%. Then smooth the graph to look like a stock one.

Set the idle RPM values at about 1070 RPM.

Since your LTFTs are not setting you will never see the STFT settle. You need to get the LTFT Low to adjust.

I assume you are working on VIII so after you make these changes....disconnect the batter for 10 mins to clear the residual memory.

Start it and let it idle for 20 mins, then adjust as necessary.

evo8dad

iTrader: (15)

I'll try this. I had thought about it as it is something I've done on larger turbos with a 4" suction pipe but, didn't think about doing this on this setup, however, now that you've mentioned it it makes sense to try it given the high idle Hz.

Did this one already.

Not quite sure what you mean here. Where is this adjustment made (I couldn't find anything in the ROM related to what you've mentioned).


Why would I increase the Hz values if I'm already showing a high Hz at idle currently? I presume the Hz values will drop once I back out the MAF plug but, I would think I should reduce the Hz values in the ROM table?

Interesting you chose 1070 rpm, I used 1000 rpm but, will try 1070.

I presume after trying your suggestions I should hopefully start to see some LTFT low action at idle?

Yes it is an VIII. Thanks for the input and I'll try this and report back.

evo8dad

iTrader: (15)

Thanks for making this note. I went back through the ROM and I believe Tephra's v7 patch has the incorrect MAT scaling so I added mrfred's MAT patch and it seemed idle much better.

I left the injector scaling at 650 and the latencies alone and adjusted the MAF scaling by reducing the Hz values from 75 and below by 25%, which seemed to help but, the engine still has an erratic idle and the LTFT low still does not register.

Attached is another log after making these adjustments.

oldevodude

iTrader: (3)

Are these new injectors? As stated before I would focus on the scaling of injectors.

What about a vacuum or leak somewhere?

l2r99gst

iTrader: (2)

You changed the Hz values in the first column or the values in the second column?

Chaning the Hz values in the first column won't do anything. You change the values in the second column, which tells the ECU how many L/Hz or air is flowing per Hz.

evo8dad

iTrader: (15)

FIC 750s, however, I've scaled countless injectors in the past and have a pretty wealthy database to pull from but, none of that seems to be helping.

I've done vacuum check and everything was ok. 7137644

Yeah I changed the second column values

Appauldd

iTrader: (22)

The Idle stability control is this.....

<table name="Idle Stability Timing Control" category="Timing" address="174e" type="2D" level="2" scaling="uint16">
<table name="Settings" type="Static X Axis" elements="3">
<data>Sensitivity</data>
<data>Timing Limit</data>
<data>Untested</data>
</table>
</table>

Add this to your 94170015 xml. It will allow you to increase the sensitivity for idle. Helps to keep the car from hunting at idle. Stock value is 128....raise it to about 146.

Second question....I didn't mean raise the Hz values, I meant raise the scaling values. Sorry about that.

Third question.....Yes you should start to see some LTFT adjustments.

evo8dad

iTrader: (15)

Do you have the address for 96530006?

I would suspect that raising the scaling value for the injector won't do much since it effects cruise trims more so than idle trims. If anything I should try (which I have) adjusting injector latencies. Thoughts?

I was hoping to try this out today but, didn't get to it. Tomorrow I plan on trying this stuff.

kanyonkid

iTrader: (33)

i dont think ull get those kelfords to idle PERFECT... my idle is at 1070 with 7* timing... my injectors were scaled perfect before the cams, and right now afr's range from 13.9-17.0.. it will chill at 14.7 for a few "lopes" than go lean and then back to rich then hit 14.7 for a few lopes, and on and on and on... unless my motor is HOTTTT than it idles perfect

evo8dad

iTrader: (15)

Believe it or not I've gotten Kelfords to idle pretty nicely in the past on a bunch of other Evos. This Evo has a few other mods other than the cams (head work being the biggest difference).

oldevodude

iTrader: (3)

Are the injectors new? Maybe you simply got a bad set? It just sounds like so much adjustment when as you said you have done many others before. I ended up having 2 of the 4 injectors flow different from the good 2 These were brand new and after pulling my hair out I finally got them flow tested and one way off and the other was just off a little . I swapped injectors and life is grand.

Good Luck

oldevodude

iTrader: (3)

Are the injectors new? Maybe you simply got a bad set? It just sounds like so much adjustment when as you said you have done many others before. I ended up having 2 of the 4 injectors flow different from the good 2 These were brand new and after pulling my hair out I finally got them flow tested and one way off and the other was just off a little . I swapped injectors and life is grand.

Good Luck