PDA

View Full Version : Closed Loop Operation Questions



Kevin Doe
May 5th, 2008, 01:52 AM
Ok guys, I've got my VE table honed in within 0.5%, and I'm sitting at 0.995% BEN all across the entire map. I spent a ton of time getting this right. I'm getting about 18 mpg (in a 2900 lb car) which isn't great from what I expected. I have been running in open loop speed density the entire time. I have a few questions.

1. What are my options for running closed loop in speed density? I'm thinking that I can still run closed loop fueling in speed density, correct me if I'm wrong.

2. When I return to closed loop, I konw I need to monitor trims. From my understanding the short term trim is basically a trim based upon 02 readings directly. The long term trim is learned from the trends in the short term trims. Eventually the short term trims will minimize as the long term trim steadys out. Again, correct me if I'm wrong. Now the question. Is the long term trim I'll be logging a single value, or is there a map of long term trims and I'll just be logging the trim from the particular part of the map that I'm operating in?

3. From some research it looks like the long term fuel trim is added to open loop operation, but if and only if the trim is positive (adding fuel). Correct me if I'm wrong.

4. I know I should be logging trims. Are there trims for each bank, like a short term for each bank, and a long term for each bank? What are their parameter names for an LS1? What other parameters should I pay attention to when I make the switch back to closed loop speed density.

5. Will this really be beneficial if my VE map is within 0.5% in the majority of cells?

Thanks in advance for the help and direction.

joecar
May 5th, 2008, 02:56 AM
Kevin,

1. Yes you can run CLSD (LTFT trimming). If running a COS, you could run semi open loop (STFT trimming).

2. Yes, your description is correct... when STFT's reach a threshold, they get zeroed and the LTFT's get adjusted. There is a 4x4 map of LTFT's for each bank, from scantool go Calibration->Long Term Fuel Trim Cells.

3. Positive LTFT's get added to OL fueling (richer of B3605/B3647 and B3618) only at WOT (the PCM goes to OL during WOT).

4. Yes, each bank has long and short fuel trims... look in the pid list under the PIDs tab, you'll see them... LONGFTn and SHRTFTn where n is 1 or 2.

5. Yes it is, it means your VE table closely reflects the actual cylinder airmass... this matters if you want to dial in your AFR's using the what-you-set-is-what-you-get method.

Joe

Kevin Doe
May 5th, 2008, 03:54 AM
If I understand what you're saying, when running closed loop speed density (CLSD) with the stock OS I can only run with the long term fuel trims active? The short term trims arent' active? They only update the long term trim which is active?

joecar
May 5th, 2008, 05:56 AM
In CL, both STFT's and LTFT's are active... but the way it works is when the STFT's get big enough the PCM zeros them and bumps the LTFT's...

STFT's are very transient, LTFT's are more steady.

BThere is a parameter to trim using only STFT's in the stock OS, but I haven't played with it yet; see B4206.

In the COS, there is a way of trimming using only STFT's (semi open loop).

Kevin Doe
May 5th, 2008, 06:09 AM
I'm glad I misunderstood. I was thinking taht would be pretty messed up if that was the case.

TAQuickness
May 5th, 2008, 06:34 AM
18mpg does seem a bit off for the vehicle weight. I'm at a porky 3700 with a fat mix (13.8 @ cruise) on a lumpy cam (238/242 115) and get 22 on the hiway if I keep my foot out of it. On average I'm seeing 18 city/hiway with a few spirited sprints.

You may want to have a look at your DFCO settings for improved economy. I had to bump my enablers way up so it would activate with my cam. I further tweaked the settings to come on faster and ramp out smoother.

All and all, with DFCO tweaking I picked up about 3 mpg and greatly improved the engine braking capabilities.

Kevin Doe
May 5th, 2008, 06:38 AM
Yeah, my 18 mph was a mix of driving. Some highway, some stop and go, and a day of autoX. I expected to be in the mid 20s, and maybe hit 30 on the highway (70 mph puts me at 1800 rpms).

I have the stock GTO DFCO settings. I can tell that it turns on during decel, but I may be able to tweak the settings so that its in effect more often.

Can you elaborate on the settings you changed? Maybe some screen shots. Or I'll post up my tune and you coudl modify it and post it back up? Maybe post your tune so I can check our your DFCO settings?

TAQuickness
May 5th, 2008, 06:46 AM
Send me a PM with your email address and I'll send you a copy of my current tune.

I used the RR to dial the DFCO to my liking as well as resolve the "school-zone" bucking (from your other thread). In both cases, the changes required to satisfy my SOTPD were so minor I don't think I would have ever been able to nail it w/o the RR.

Kevin Doe
May 5th, 2008, 06:48 AM
SOTPD? Sending PM now. Thanks.

TAQuickness
May 5th, 2008, 06:49 AM
Seat Of The Pants Dyno

Aint Skeered
May 8th, 2008, 11:39 PM
In CL, both STFT's and LTFT's are active... but the way it works is when the STFT's get big enough the PCM zeros them and bumps the LTFT's...

STFT's are very transient, LTFT's are more steady.

BThere is a parameter to trim using only STFT's in the stock OS, but I haven't played with it yet; see B4206.

In the COS, there is a way of trimming using only STFT's (semi open loop).

joe my b4206 is set to disable. Would this need to be on enable to alow stft ve adjustments?

joecar
May 9th, 2008, 01:21 AM
Aint Skeered,

I haven't used it, but it seems that when B4206 is enabled then the PCM uses STFT's to trim to B3605... the STFT's will be able to trim only the cells in B3605 that are stoich... cells away from stoich are not trimmable because of how the NB02's work... this similar to how semi-open loop (SOL) works in the COS's.

If you're thinking of using STFT's to correct your B0101 VE table, then I think you may get better results using LTFT's... if you have a wideband then use it instead.

Cheers,
Joe

Aint Skeered
May 9th, 2008, 04:45 AM
Joe I have a lm1 wideband but have not figured out how to install it yet to efi live. DO you need a different serial cable then what comes with the kit as this one does not hook up to my lap top.

I see lots of help on the lc1 but not much on the lm1. TO bad no one has made a tutorial for the lm1 specific.

joecar
May 9th, 2008, 05:01 AM
I don't know how the LM-1 is...

Some LM-1 people will chime in.

SSpdDmon
May 9th, 2008, 06:28 AM
Joe I have a lm1 wideband but have not figured out how to install it yet to efi live. DO you need a different serial cable then what comes with the kit as this one does not hook up to my lap top.

I see lots of help on the lc1 but not much on the lm1. TO bad no one has made a tutorial for the lm1 specific.
The LM1 should have come with a cable that has 3 wires coming out one end and what looks like a head phone connection on the other end. That's what you use to hook up to EFI Live. I don't recall which color wire is which. Only that the ground should be easy to distinguish. Then, I plug one of the other two wires in as the signal wire. To make sure it doesn't matter which one, I use the LM1 software to make both outputs the same. Personally, I like it to read 0v=11:1 and 5v=16:1. That way, the formula for the scanner is easy (voltage recorded via EXT.AD# plus 11). Anything richer than 11:1 or leaner than 16:1 is too much. This also gives you good detail to the AFR. Check the LC1 guides and other threads on here about setting up custom PIDs (or alter the preset ones at your own risk). If you decide to alter the preset one, make sure you do both formulas (LM1-1 and LM1-2) and make sure you save a copy of the original file in a safe place.

joecar
May 9th, 2008, 07:44 AM
I'll see if I can find the LM-1 user manual on the Innovate site...

Here it is: http://www.innovatemotorsports.com/support/manual/LM1_Manual.pdf

There's more app notes where that came from.