Im having a problem with the idle of the car and have tried a few different combinations. During the last 2 hours of driving around and trying new thing I taught myself on how injector scaling and injector latency and how they relate to each other.

But I have had this problem from the start and can't quite figure it out. My cruising trims are good and within 2-3% of being perfect but my idle is suffering. The car goes lean within about 10 seconds of being at a dead stop and the trims go to about 20% on the positive side.

Now looking back at the logs i was looking at the battery voltage and can't quite figure out how to richen up the idle without doing the same to the entire fuel table. My battery voltage during cruising is 13.85 to 14.00003. Now during idle it ranges from 13.78 to 13.95.

Is there anything I can do?

I have Precision 780 injectors
scaled to 665
latency is:
3.312
1.68
1.296
0.888
0.6
0.36
0.24

I don't want to tune the rest of the map until this is figured out. Thanks.

 

Don't worry about the voltages too much. The different voltages are just there because the injectors are going to react quicker or slower for higher or lower voltages. Since we idle and cruise at about the same voltage, there really isn't one entry for idle and one for cruise. I would adjust the entire table by the same amount or percentage change.

However, you can adjust idle more than cruise by using this table. Adjusting these latency values is really adjusting the deadtime of the injectors. This latency or deadtime adjustment will affect low pulsewidth situations (like idle) more than larger pulsewidth situation (like cruise or WOT). High RPM WOT has such large pulsewidths that adjusting these latency values has very little to negligible effects.

To find out exactly how much you will affect idle in relation to cruise, just go out and log your IPW at idle for a few minutes, then go log your IPW at your normal cruising RPM/speed/load. If your IPW is twice as big at cruise, then the latency adjustments will affect cruise half as much.

In your situation, though, I would recommend to start increasing the latency values even more. Your idle wants a lot more fuel and adjusting these latency values will do that. But, this will also affect your cruise trims. Not as much as your idle (the exact ratio will be based on IPW), but your cruise trims will still be affected. Maybe start small and add in another 100us (.1 ms in the table) to all of the values and see where your trims settle for both cruise and idle. Don't worry about the rest of your open-loop fuel map as it will be affected very little. Also, you can use the STFT to really see where you long term trims are going to end up.

Also, keep in mind for cruise, these trims will vary a lot over time. You don't always cruise at the same RPM or speed or load. Cruise varies a lot, so don't expect your trims to always be spot on. Trims are trims, they will vary. But, that being said, you should be able to dial in your idle trims a bit better.


Eric

 

I figured I would throw a little more info into the loop here.

Situations like this are going to come up more and more and in some instances there will nothing that we can do to completely dial in the injectors correctly with these tables. That's because a lot of aftermarket intakes change the airflow readings by our MAF.

So, in those cases, the only way to get trims back in check correctly (instead of messign with the injector settings) is to rescale the MAF reading, especially at idle. I know we have a few MAF tables listed in ECUFlash, but I don't think anyone ever really messed with them yet??

If we knew what these tables were, situations like what you explained (if it actually is a MAF reading issue at idle) would be corrected very easily. All you would need to do is adjust the airflow reading at the idle HZ level that is being read by the MAF. For example, the table may have a value for airflow at 50Hz and bumping that up would richen the mixture and bring down positive trims at idle (assuming idle was around 50Hz as an example).


Eric

 

Mike,
I have PTE 780cc injectors and found that setting my 11v latency to .888, 14v latency to 0.672 and my scaling to 713 got me pretty good fuel trims. With your fuel trims being where they are, I would try increasing the 14v latency and leave the scaling where it is for now.

-Paul

 

Before you get too nuts.. If it begins to happen after the car sits at idle, see if you can get the DV not to open at idle..

If your VTA on your setup with the blowthrough, put your hand over the valve and see if idle improves.. I'll bet you it does..

What ends up happening is the engine vacuum causes the DV to open, and the airflow is very low with respect to the sensor.. So if you get good trims with the valve open and venting partially all the time, cruise will run lean, if you get your cruise trims dead on, your idle will run lean when the valve opens under vacuum....

Start by covering your valve with your hand (if you can) This is the reason why I use an HKS SSQV..

 

Yes, Jack makes a good point.

I was just assuming that you are recirculating and that you have no leaks. Also, make sure that you have your breather tube connected to the intake pipe and not vented.


Eric

 

FWIW, the two MAF related tables I've looked at, have digital values in the left side. HOWEVER, i applied the formula for Airflow from mitsulogger and guess what? It translated conveniently to a frequency..

 

http://www.aktivematrix.com/forum/al....php?pic_id=18

Click on the link..

Notice the value instead of being 16-32768 or whatever it typically reads, has been translated using the same formula that converts the airflow reading in MUT, and it seems to correlate.. Not only that, but the max value seems to correlate to the max flow the MAF seems to read before it starts becoming a restriction or reading incorrectly.

 

Let me add that I don't know how valid that translation is, only that the values created go along pretty well with what the MAF output seems to indicate, and the ECU seems to show for load at those frequencies.

 

That does seem to look pretty good...thanks for the link.

If the numbers are true, then it looks like the fist value should affect the airflow scaling at idle. I would be curious to see someone who is having troubles with dialing in their trims because of aftermarket intakes alter this value to see if it affects idle trims.

If so, then dialing in aftermarket intakes would be a breeze. Fuel trims can be affected by either the injectors not being dialed in correctly or because of the MAF reading being off at certain airflow values. If someone is positive that their injectors are dialed in correctly, they can then use this table to dial in the airflow correctly for their intake setup.

That's what I would do when changing intakes, filters, pipes. Re-calibrate the airflow, especially at idle, so that everything is in check again before worrying about injector scaling and latencies.



Eric

 

Looking at the maf scaling table, it looks like the right side should be scaled different, too.

It can't be g/s because the scale doesn't go low enough. I wonder if it is g/rev or something similar, but not scaled correctly. For example, if using g/rev, values from about .3 at idle all the way to 4.0 or so should represent the capabilities of our MAF.


Eric

 

Jack,

Maybe we should open a new thread on these maf tables to continue the discussion, but I looked at them quickly to see if I can correlate the right hand side values to valuest hat I was used to seeing logging g/rev airflows from DSMLink.

If I divided the values by 30, then subtracted 5.8, the values looked a lot more like the g/rev values that I am used to. I wonder if these values actually are a mass airflow/rev?

Interesting nonetheless.


Eric

 

Start a new thread, I'll post my findings there.. It seems like my older postings dropped off anyway.. I never touched the right columns scaling, only that the left column was always INT16 and I just applied something that seemed to make sense to me and it appeared to apply.

 

First of thanks a lot Eric, Paul and Jack for helping.

I was thinking a lot about the BOV and it venting out at idle and thought that could be the culprit. I think its the wrong spring for that sized turbo anyways. I need at least a 9lb. That should keep the valve shut during cruise and idle and make life a bit easier on me.

 

Just saw this thread...I have the same issue. My car runs around 13.85V at idle and cruise. It also runs very rich at idle, and if I back the latency out, it affects global fuel...and only really helps idle a little. My cruise trims are +-5% across the RPM band. I'm on Precision 680's, scaled to 585 with .864 at 13.85V.

So I too am quite eager to find out more about what can be done with the MAF signal at idle.