DPF Regen Effectiveness

[killerbee]

I found out the hard way, that tuning could involve killing the effectiveness of the cycle. I have no idea what I did specifically that hurt it, but it was clear that I was going to run my tune in regen, nearly full time because soot was building up as fast as the regen could remove it. I am pretty sure it was not a dirty tune, but rather, I change SOMETHING that killed regen effectiveness. As soon as I reflashed stock, the soot melted off at 10X rate.

I started this thread to discuss one thing: Regen criteria for fast effective regenerations.

If you have some educated ideas on what parameters constitute more effective (hot) regen, then please contribute. I would really like this to be an educational thread for tuners, so that we can all benefit. I wasted at least 40 hours of tuning, and would like to help others not follow in this path.

[killerbee]

ok, sleepy crowd.

Is there an injector at the dpf site?

[duramaximizer]

NO!

There is no injector. I think 3 sensors pre and post pressure, and EGT.

My guess is timing is killing it somehow, it is messing with post injection quanity or time.

[rcr1978]

Link no worky

[killerbee]

Does anyone have a regen log they can attach here, one that clearly shows post-injection?

[killerbee]

It is unfortunate that the software has NO dpf tables for timing, fuel pressure, soot limits, EGT modulation, intake valve modulation, etc. Regen is such a critical part of effective operation, especially for those trying to stay legal. :)

Is there a chance that we may see some of this?

[GMPX]

It is unfortunate that the software has NO dpf tables for timing, fuel pressure, soot limits, EGT modulation, intake valve modulation, etc.Time for you to go and buy GM's tuning software Michael.

[Brayden]

I found out the hard way, that tuning could involve killing the effectiveness of the cycle. I have no idea what I did specifically that hurt it, but it was clear that I was going to run my tune in regen, nearly full time because soot was building up as fast as the regen could remove it. I am pretty sure it was not a dirty tune, but rather, I change SOMETHING that killed regen effectiveness. As soon as I reflashed stock, the soot melted off at 10X rate.

I started this thread to discuss one thing: Regen criteria for fast effective regenerations.

If you have some educated ideas on what parameters constitute more effective (hot) regen, then please contribute. I would really like this to be an educational thread for tuners, so that we can all benefit. I wasted at least 40 hours of tuning, and would like to help others not follow in this path.

Keep your post injections in... If you kill them it will definately reduce your time between regens..

As a test, drop back to 2 injections.. Drive 20 miles and you're catalyst will be full.. Then turn on the handy dandy scan tool. You can watch soot mass in the DPF fall rather quickly with a 100% clogged filter if you go back to 5 injections.


I've got some clean 425rwhp tunes running out there with no issues. What problems were you having in particular?

[killerbee]

I don't think my wife will approve.

[Brayden]

I don't think my wife will approve.


Of you testing on her LMM ? Are you talking to me or Ross ?


My theories on efficient combustion would point someone to use more fuel pressure instead of pulse width, increasing timing for more CP etc..

My experience with these trucks shows that by reducing the torque limiting will gain plenty of power without increasing the pulsewidth much if any.

[killerbee]

Keep your post injections in... If you kill them it will definately reduce your time between regens..

As a test, drop back to 2 injections.. Drive 20 miles and you're catalyst will be full.. Then turn on the handy dandy scan tool. You can watch soot mass in the DPF fall rather quickly with a 100% clogged filter if you go back to 5 injections.


I've got some clean 425rwhp tunes running out there with no issues. What problems were you having in particular?

I figured as much. I have a learning curve now, and figure that this thread might serve to help others not make the same mistakes, but also serve as a tech thread on DPF theory. We'd all like for regens to take as little fuel as possible.

BTW, what is the table number for the number of injection events? I must be losing it, I can't seem to find it now.

[killerbee]

Of you testing on her LMM ? Are you talking to me or Ross ? Ross


My theories on efficient combustion would point someone to use more fuel pressure instead of pulse width, increasing timing for more CP etc..

My experience with these trucks shows that by reducing the torque limiting will gain plenty of power without increasing the pulsewidth much if any.


One of my questions is about regen fuel pressure. Wondering if there are sep tables for regen process.

It also seems that an accurate oxygen algorithm (boost dependent) is an important part of the reaction equation. This and other dpf theory parts, require at least I ask some of these questions. It is not unreasonable, with the state of the truck changing every 300 miles, with all new parameters.

[Brayden]

B1021.. It's a limiter table by voltage and RPM. You can use it to your advantage to shape your number of pulses by RPM, which is not ideal, but works very well for a DPF'less truck.. Set it to 2 across the board except for low voltages.

[killerbee]

Thanks.

I had set most values to 3 (don't know why). I have used 2 with the LBZ without issue. Of course I think it only ever has 2 pulses, LOL.

So this leads to 2 questions

what does voltage actually represent. Never have gotten a straight answer to this.
What is pulse 5?... 4?... 3?... 2?

Is there 2 post pulses during regen?

Trying to log all these channels is going to pose a real challenge, on top of the other required data.

[Brayden]

LBZ only runs 2 injection events.

Of the 5 events You have:

Pilot Injection 1
Pilot Injection 2
Main Injection 1
Post Injection 1
Post Injection 2

If you drop back to 4 it should kill Post2
If you drop back to 3 it kills Pilot1 and Post 2
If you drop back to 2 it kills Pilot 1 and both Post 1 & 2
And that just leaves 1 event left and that is non piloted main injection

Running only 3 was reducing one of your post injections and hindering the systems ability to create sufficient late cycle heat required for efficient DPF operation.

[killerbee]

I was thinking that 3 eliminated both post events. Was that a bad assumption of mine?

[Brayden]

Voltage plays a very important role in an electronic injection system ;)

Injection speed is dependent on voltage. If voltage drops too low, injection accuracy is compromised. That lookup table gives the ecm a threshold to abort injections if the voltage is too low to support them.

[Brayden]

From what I've read, it cuts from both ends, knocking off one post, then one pilot, etc etc.

[Brayden]

Do an experiment over an average drive cycle with 5 events, 4 events, then 3

[killerbee]

Voltage plays a very important role in an electronic injection system ;)

Injection speed is dependent on voltage. If voltage drops too low, injection accuracy is compromised. That lookup table gives the ecm a threshold to abort injections if the voltage is too low to support them.

that helps. Thank you again.

Note that post 2, has no qty table associated with it. I assume it can't be adjusted?

Also, FWIW, pilot 2 shows 1mm3 for all cells (factory tune), yet in my logs, I have not seen any pilot 2.

[killerbee]

Another disparity I just noticed. B1021 allows max of 5. Yet there are 6 total events including the main, Pilot 1,2,3 and Post 1,2

[killerbee]

One of the elements that I am really impressed with, is the ability of the induction system to lower ambient from 14.2 to under 9 psi, see boost. The MAF is highly reduced as low load/idle conditions approach.

This is why it would be valuable to see the intake valve, but I don't see a PID for it. I know next to nothing about this, and am assuming that this pressure is only possible with a metering restriction located pre-compressor?

http://forum.efilive.com/attachment.php?attachmentid=7057&d=1264307761

[bballer182]

In BBL

IAFVCMD

intake valve motor commanded %

[killerbee]

I guess you have just managed to convert me.

[bballer182]

I guess you have just managed to convert me.

It's the wave of the future. BBL takes advantage of V8 and i would assume that V8 scan too would have all the logging capabilities that BBL has currently when it's released.

[killerbee]

Another disparity I just noticed. B1021 allows max of 5 events. Yet there are 6 total events including the main, Pilot 1,2,3 and Post 1,2

does anybody have any idea?

[duramaximizer]

unless in regen? :confused:

[Brayden]

does anybody have any idea?


You are logging 6 events or you see the ability to log 6 events...

They only use 5 to the best of my knowledge but have the ability to run 6.

[bballer182]

It's interesting to compare B1021 in the LMM's to LBZ's.

LBZ's had no limitations on injection events vs voltage from the factory.
LMM's have that table refined from the factory.

Since most electrical systems run in the 14.4-13.8 vdc range while running it looks like they limit injections to (4) after 2800 rpm and (3) after 3600 rpm.

[killerbee]

Do we know that the voltage referenced is system bus voltage?

[bballer182]

I assumed that that table is based on the control module voltage, which is why i linked that table to the VPWR PID. funny i have never logged it to see what it is.

[Brayden]

It's interesting to compare B1021 in the LMM's to LBZ's.

LBZ's had no limitations on injection events vs voltage from the factory.
LMM's have that table refined from the factory.

Since most electrical systems run in the 14.4-13.8 vdc range while running it looks like they limit injections to (4) after 2800 rpm and (3) after 3600 rpm.


I would imagine it's because they knew they were only running two shots in the operating system, there was no need to cal that table.

[bballer182]

I would imagine it's because they knew they were only running two shots in the operating system, there was no need to cal that table.

You forgot about the post-injection. and there is a pilot #2 that has quantities in it but is never active... as far as i can tell.

But you make a good point.

[Brayden]

Right

[killerbee]

I can't force a regen on the DVT page. I am at 20g of soot (low). Anyone have an idea how I can control this?

[bballer182]

i think we are stuck currently.

http://forum.efilive.com/showthread.php?t=12861

[GMPX]

Well, at the moment I don't have access to an LMM to test the latest builds on, anyone who has done this before able to?

Cheers,
Ross

[bballer182]

Well, at the moment I don't have access to an LMM to test the latest builds on, anyone who has done this before able to?

Cheers,
Ross

Just tried latest build. still no workie.

fan control gets and error, and vane position DVT when set to 100% really only sets the vanes at 40%. that's kinda weird....

cant attest to the functionality of the LMM DVT's though

[killerbee]

that's odd. The other day, when the soot code set, I was able to start the regen with DVT, after clearing the code. Now I am wondering if I actually was starting it manually, or if it was doing it on its own.

This does make it impossible to test DPF function with tuning.

[killerbee]

hmmm, any idea when we might see this again? I'll keep trying. Do I have to wait till 65g for a normal regen?

[GMPX]

It's pretty tough out here to get to an LMM for testing, the only place that regularly brings them in usually pulls the DPF etc off during the conversion. If we get lucky they will have one there, if not, it could be weeks off, maybe longer before I can't try.

Cheers,
Ross

[duramaximizer]

I don't want to, but if I have to, I could put my dpf back on and load the old 7103 OS back in and see if the DVT's work. I haven't played with them at all, I had no reason to.

[killerbee]

Update.. Still no DVT regen ability. Normal regen started at 45g, ended at 7.

I found I got 5g per minute regen when I started at 90g. 2g per minute when it starts at 45g. That tells me a little about how I would LOVE to adjust the regen cycle.

I also managed to determine that lowering boost helped decrease my regen time by increasing EGT1 about 50F. This is preliminary, subject to further analysis. There is nearly no regen below 900, but this changes exponentially to 1200F.

Idle regen is possible, and does occur, but fuel consumption is more than double what normal idle is. Fun to see 1100 egt at idle.

Hope to see more tables that will allow us "compliant" do-gooders play around some more.

[killerbee]

Hope to see more tables that will allow us "compliant" do-gooders play around some more.

shaking the cobwebs off. There are some things happening with flash updates at GM that would be important to get our heads around. There are a lot of regen issues with LMM, that appear to be based in old flash.