Boost tuning woes

[cdvma]

So messing around with my roomates WRX wagon ('02) trying to tune the boost. Nothing complex, its a stock setup we just want a couple extra psi in the midrange.

We up the boost targets...lower boost than factory.
Modify the WG solenoid duty cycle. Rasing it lowers boost. Lowering it raises boost sometimes but lowers it if we lower the WGD more.

Its totally inconsistent. Linear changes are not giving linear results (in terms of direction). For some reason we cannot for the life of us get it to go below 12 or above 14. Stock map works perfect every time so I doubt its something physical. No wavering, always right on the boost target.

Ideas?

[JRSCCivic98]

Post the WDC and Boost tables please.

[paulmezz]

I've gone through MANY variations of the tables so I have no idea which to post (plus the laptop is in the car )

I will post my most recent mucking-around-with-it map from work but the jist of it is I upped the boost to 15.5psi while keeping the curve smooth. I think I had the redline boost set to around 13ish.

I bumped the WDC map in both directions just to see what is up. I think the most aggressive change I made was going up around 10 points. I don't want to be too drastic, I was looking for any kind of linear change so I could start actually tuning it to where I wanted it. It seems that any change I did just made the boost lower. There was only one time where I managed to hit 14.0psi on a modified image. Everything else dies around 12-13.

There just seems to be NO rhyme or reason to it. I know there is but I sure can't see it.

-paul

[paulmezz]

One more hopefully not important thing;

I have the rear O2 sensor removed and a wideband stuffed in there so I can watch my A/F to make sure I'm not doing anything stupid. This IS causing a CEL to be thrown for the rear O2 missing.

From what I've been told, the ECU won't enter a limp mode for just that sensor. Is that correct or is that where all my issues are coming from?

-paul

[paulmezz]

I still don't have the maps to post, but I have not touched the boost target (still at 15.5)

We mucked with the WGDC by pushing it up in stages. As the DC went up, boost went down. I think the highest DC we hit was around 87% (11psi at this point) As a test we lowered it down to somewhere around 55% (12psi down here).

I am so stumped.

-paul

[Spiider]

Have you tried the wastegate arm tightening adjustment?

[swifty]

Paul

Without seeing what changes you have made to your maps it is impossible to tell what is going on. You have mentioned upping the boost and playing with the WDC values but your post doesn't tell us where in the maps you have done this. The original post mentions 'lower boost than factory', does his mean lower than the boost limit or lower than the targets in the desired boost table ? You say the stock map works ok, is that with the oxygen sensor removed ?

Have you logged any of this data or are you reading off the boost gauge ? Also, have you tried using the stock WDC values and logging to see how much boost you are making ? I would approach it like that to start with, then analyse the data to see where the boost target is not being met. Then increase the WDC in those areas and try another run. Keep increasing gradually until you get to the point where the boost targets are being reached.

I don't think you will need to increase the duty very much, 3 or 4% per additional psi or something like that I would imagine. If you have the WDC too high then it will overshoot the boost target and the ecu will then reduce the boost but you will feel this when driving.

Are you sure the xml for your rom is correct ? If it isn't then it could explain why there doesn't seem to be any logical explanation of the effect of the WDC changes.

I see you are monitoring the afr, do you also have something monitoring detonation ?

[paulmezz]

Without seeing what changes you have made to your maps it is impossible to tell what is going on. You have mentioned upping the boost and playing with the WDC values but your post doesn't tell us where in the maps you have done this. The original post mentions 'lower boost than factory', does his mean lower than the boost limit or lower than the targets in the desired boost table ? You say the stock map works ok, is that with the oxygen sensor removed ?

Yeah, the lack of data is my bad. I'm just frustrated with this issue and getting data out of that laptop is a total PITA. I finally have a way to extract data AND I'm not home with the woman bugging me I uploaded my csv files, .bin files and some screen shots to; http://www.dodgubishi.com/wrx/tuning .

I originally was using ecuedit, but the xml for that was very very wrong. I since switched to enginuity (0.2.7.4b). Some day I will go through and fix all the addresses because I like ecuedit's interface more but that is a side issue.

_ALL_ of my testing is with the sensor removed and the CEL on. I have also started using a default IAM of 16 for all my testing so I don't need to wait for that to increase.

I have modified the following tables:
Fuel - Low Det
Boost - WG DC (MT)
Boost - Target PSI (MT)

Except for any stock runs, here are the fuel and boost maps I have in the ECU (I leaned it out a little bit up top).

Have you logged any of this data or are you reading off the boost gauge ? Also, have you tried using the stock WDC values and logging to see how much boost you are making ? I would approach it like that to start with, then analyse the data to see where the boost target is not being met. Then increase the WDC in those areas and try another run. Keep increasing gradually until you get to the point where the boost targets are being reached.

I am logging everything via ecuexplorer. Logs are in the aformentioned data directory. Quick summary of the max numbers;

stock map = 13.8psi, 74% WGDC
higher WGDC + upped boost target = 13.5psi, 78% WGDC
highest WGDC + upped boost target = 11.9psi, 86% WGDC
LOWER WGDC + upped boost target = 11psi, 58% WGDC (did this for expermental purposes, I didn't copy the logs up)

It IS hitting the max WGDC that I set but the boost target is WAY off. I've never gotten it to go above 14psi. I did that once in a very lightly modified map with a 4th gear pull. I increased the WGDC a little more and my boost went down.

I don't think you will need to increase the duty very much, 3 or 4% per additional psi or something like that I would imagine. If you have the WDC too high then it will overshoot the boost target and the ecu will then reduce the boost but you will feel this when driving.

Are you sure the xml for your rom is correct ? If it isn't then it could explain why there doesn't seem to be any logical explanation of the effect of the WDC changes.

I see you are monitoring the afr, do you also have something monitoring detonation ?

I'm not totally sure the XML is correct. During one session of testing (about 3 flashes) I copied my maps into both the AT and MT sections just to be sure. There was no change from what I'm seeing with only modifications to the MT map.

I have no additional det monitoring besides the stock setup.

Thanks a ton. If I can get this sorted out then I can save the $ I was going to spend on a utec

-paul

[mrtMY03Aust]

Solenoid needs a clean?

[swifty]

I looked at the maps and logs and I tried similar changes to my car last night followed by some logging. When I increase the boost and WGD values I get the corresponding increase in what's logged, what the boost gauge shows me and what I can feel in the way the car accelerates.

The only conclusion I can come to is there is something outside of the maps that is affecting the boost, perhaps something mechanical, or something else is happening that results in the boost being reduced.

I also found this:

http://forum.ecutek.com/topic.asp?TOPIC_ID=106

seems like it has been seen before.

[paulmezz]

I looked at the maps and logs and I tried similar changes to my car last night followed by some logging. When I increase the boost and WGD values I get the corresponding increase in what's logged, what the boost gauge shows me and what I can feel in the way the car accelerates.

The only conclusion I can come to is there is something outside of the maps that is affecting the boost, perhaps something mechanical, or something else is happening that results in the boost being reduced.

It is good to know that my changes are correct. I'll start picking through the car to see if there is anything "odd".

I'll let you guys know if I find anything.

Thanks!

-paul

[MRF582]

Code: Select all

Time   Engine Speed (RPM)   Throttle Opening Angle (%)   Mass Air Flow (g/s)   Ignition Timing (°BTDC)   Knock Correction (°BTDC)   Fuel Injector #1 Duty (%)   Air Flow Sensor Voltage (V)   Manifold Relative Pressure (PSI)   Primary Wastegate Duty Cycle (%)   Knock Signal #1 (On/Off)   Advance Multiplier (020124) ()

2463   3817   100   134.22   13   7   60.25   3.68   13.494   79.22   0   16
2834   3838   100   133.47   14   8   59.77   3.68   12.769   86.67   0   16



Post-turbo induction leak given ambient conditions were the same and your wastegate duty cycle could be modified. Copy your 100% throttle colum into all the other columns. Keep the values at any throttle position for the given rpm the same.

[paulmezz]

I did find one leak with the IC piping. A clamp came loose then the hose popped off.

I went through and gained a few pounds of boost just by putting the clamps on nice and tight

The car is behaving as expected right now. I forgot his username on here (sorry Ed!) but he is helping me get it all tuned. I've never flashed so many versions to my ECU.

Car feels great. Amazing what a few more pounds will do and tweaking the WGDC to have the boost come on faster.

I believe he mentioned that we are going to post the rom when we are done.

-paul

[west_minist]

Very bad on you.

There seems to have restrictors in the holes too and you will also need to check the wg too.

Paul, check here http://forums.openecu.org/viewtopic.php?t=645

Thanks for your kind words.

As you can see, we have not even hit target boost yet

Soon start on a 20002 JDM WRX w/ AVCS. So you will see the diff in spool and torque from 1500 till 4000+.

Wonderful stuff guys. Just Wonderful stuff.