I am testing mrfred's direct boost control method on my car and I noticed this inconsistent behavior in boost correction. I have tested this three times and it keeps on happening. I noticed it yesterday in one of my logs, but I ignored it thinking it was a one time affair.

I went out logging tonight and it happened again. In some WOT runs the boost error correction works like it should and in other runs it over corrects and pulls down the boost. I am using the boost error correction that mrfred posted.

I did 4 logs. The first two were fine and boost and boost error was on target. Here they are. I hit almost the exact boost target that I set up in my boost tables and the correction was right where it should be.



And here are the second two logs. I made them about 10 minutes after the first two. Notice how the boost error goes nuts and starts pulling my boost down. Why? I have no idea.



At first I thought that something mechanical went wrong with the car. So I headed home and checked out the car and all was fine.

An hour or so later, I went logging again. The SAME thing happened. In the first log I hit the boost target and in the second log the the boost error went nuts and started pulling boost. I did not even stop the car in between runs like I did the first time.

I am attaching the logs.

 

Can you post your BWGDC tables?

 

Here are all the tables that I have modified for direct boost control

 

Thanks. What's happening is that during spool up, the boost is way overshooting the target boost level, and the ECU is pulling back a large amount of WGDC. It pulls back so much that the boost drops more than 1.4 psi below the target, and the ECU no longer tries to correct the boost. This is clearly what's happening even in your 3rd gear "hiboost" log. The same thing happens in your 4th gear high boost log, but since more boost is generated in higher gears, the boost does not drop as much when the ECU pulls back WGDC, and the ECU eventually makes up for the sub-optimum BWGDC settings.

My recommendation would be to first try changing the BWGDC value at 3500 rpm to something like 60%. Maybe lower the value at 3000 rpm to 60% also. Based on what I see in your 4th gear log I recommend trying the BWGDC table in the attached pic. Basically, I estimated that table by looking at your actual WGDC curve during your 4th gear run. It will probably need some fine-tuning but should be fairly close.

3rd gear hiboost log:


4th gear hiboost log (note that I incorrectly point to WGDCC to the boost error):


recommended BWGDC table:

 

That boost error correction table is way too much. It should be lower, like this..

-4.0---0
-3.5---3.0
-3.0---2.5
-2.5---2.0
-2.0---1.5
-1.5---1.0
-1.0---0.5
-0.5---0
0
0.5---0
1.0---(-0.5)
1.5---(-1.0)
2.0---(-1.5)
2.5---(-2.0)
3.0---(-2.5)
3.5---(-2.0)
4.0---(-2.0)

This is how i would do it. You shouldnt even try to correct .3-.8 psi off. Pullin a point of WGDC will drop boost more then 0.5 psi. Also sometime the correction lags like touringbubble said. etc.

Also having a lower number in the most lower cell will help with spikes. If you have a big number there and it spikes the correction will most likely always pull to much, so keep this number low.

 

mrfred,

I tried lower WGDC tables and got the same results. In one log the boost is very close and in the second log it pulls boost like crazy.

Here is the close one:



And here is the one where boost gets pulled like crazy. The logs were done 5 minutes apart. Both are in 4th gear.



And here is the WGDC table, boost table and BEC table.



The only table that is the same between the two maps is the TBEC table. Maybe, I need to change that like Evo_Kid suggested.

I attached the logs so you can view them using Excel since you are accoustmed to it.

 

Could you explain this one? Do you mean the number that is -18.0 in the my current table?

 

nj, have you tried a 3rd and 4th gear pull with the TBEC all set at 0 to see what your BWGDC settings are giving you in the raw? Make sure you are hitting your target boost purely off BWGDC, then fine tune the TBEC. You might find that TBEC is bringing you to your target boost, and not BWGDC. Which would make it inconsistant like you are seeing.

Also, I've found that you need to tune in less TBEC with this control method than with the stock load method.

 

I have not zeroed out the TBEC. When I tuned boost using the load method, I stuck with the stock TBEC and had no problems like this. The boost was very consistent.

I will do that with this method with safe WGDC numbers and then go from there.

 

Thanks. It looks like you tried rescaling the rpm axis of the BWGDC table. Unfortunately, that same rpm axis is used for the Baseline Boost table. What you inadvertently did was move over your Baseline Boost values by 500 rpm. Because of this, the boost went way way way past the target boost during spool, and the ECU pulled back a huge amount of WGDC.

Please try my suggested BWGDC table. I think it will work very well.

Evo_Kid has a point, but the correction table you're using will only overreact when the boost shoots way past the target. IMHO, best to fix the BWGDC table first.

 

Yes. you should try out my table i posted above

Wait.. are you using the GM solenoid?? If your using the stock solenoid, I would maybe double or triple the numbers on the right side of my BEC table i posted above, but still they would be way lower then what your running.

 

Yeah, defiantly tune like this. Zero TBEC, use WGDC to get you to your desired boost level, then re induce the TBEC.

The only thing I do different is a run a flat WGDC ie 65s from 3000-7000. I just make sure that the WGDC I use, gets me to my desired peak boost. I then let the correction take away the taper.

 

+1. I was going to suggest the same, but was hoping to skip over it with a good guess at a BWGDC table.

 

Yes, I am using the stock BCS with a #65 pill. What should the numbers be in the BEC table with the stock BCS?

 

Let me ask this question:

If I tune with the TBEC zeroed out:

1. Should I start with a conservative WGDC. Something like this:

100---1000 rpm
100
65.5
54.5
53.5
53.5
53
52.5
52.5
50
54
54
61
62.5
78.5
81.5---7500 rpm

2. If I hit my target boost and re-introduce the TBEC table that mrfred created, then why wouldn't the good boost/bad boost phenomena happen again?

Thanks guys