Some questions/ideas with ECUFlash

[XVeloX]

First things first, i was tuning my GM 3-port solenoid a few days ago, and tried the suggested WGDC's from bryan to run approximately 21.5 psi all the way accross, just to get an idea of where i would be. He gave me these values using stock RPM values:

100
100
100
100
100
100
90
60
55.5
55.5
55.5
55.5
57.5
61.5
65.5
70.5

With these values i was running in excess of 26 psi :shock:. Would the fact that im running a 10.5 hotside vs. a 9.8 be a factor? Because spoolin is near these values above and is running 22ish? on a 9.8 hotside. My other thought would be that somehow when i bought the car long ago i got an upgraded actuator, but my MBC never acted differently than spoolin's. Could it be because im running -1/-4 and he's running +2/0? I know he said his fuel maps were different than mine too. Also of importance is that when i tuned my GM 3-port i had error correction off altogether. This is where i left off cause it was waaaay too late, and my (^#Q$(&^#%(*#&Q$ throttle body leak was giving me different results every pull (P.S. tuning with a HUGE boost leak is so damn hard lol 10.3 AFR's one run, so i tune to 10.9-11.0, and once that's set the leak goes away and im running 11.5-12.0 :shock: (*#^%#*%$)#*&%^Q($&#Q*$&). On another note i have 636cc for injector scaling, but since im using 750's i think i might turn it up too 650cc in ecuflash since my fuel trims are +12% idle and -3% cruise...

Anyways this is what im at right now:

100
100
100
100
100
97.5
79.5
46
40
44.5
42
39.5
42.5
44.5
48
52.5

Here's a log showing the boost, since Zeitronix's program records so many damn datapoints, this is the best method of "smoothing" i could find using DLL... on the lower graph the wavy green is boost, thick pink is TPS, thin pink is EGT, and the dark green in AFR. I know the second one doesnt mean much but anyways there ya go...


i know it looks like a wavy boost curve, and it is, but until the last few pulls it was perfectly flat, the boost leak plays tricks on you... so now im gonna try these:

100
100
100
100
100
97.5
79.5
46
41.5
41.5
40.5
39.5
41
41.5
46
51

The reson in think there is something up with my car is because i didnt think that 44% WGDC would give me 23 psi?!?!?! I have error correction OFF and the BDEL and BCLO are setup properly anyways. Thoughts? I'll finish the tune as best i can soon, but what could be causing these values to be so low?

Next item (take this section with a grain of salt, it's simply an educated guess and has no facts behind it YET...):

I know there has been lots of discussion about wanting to tune closed loop stuff and i was curious about trying something...

In ECUFlash there are 4 maps of interest:
Open Loop Load #1
Open Loop Load #2
Open Loop Throttle High Load
Open Loop Throttle Low Load

From the looks of the throttle maps, it looks like it throws the system into open loop if you hit a certain Voltage from your throttle. For the Load #1 and #2 it says percentage on the right, but i have a feeling that it might have something to do with hitting a certain load column throws you into open loop. So if you are at a high RPM, but you are barely on the throttle, you might still be in open loop because of the setting in this table. Not 100% sure about these, just making educated guesses. So if you were to set these to 0 you would always be in open loop. I'm not sure if it is possible, but if you were to set these to 0, and then tune your fuel and timing maps accordingly, couldn't you potentially be "tuning" idle and cruise? I realize that this takes out of consideration a ton of safety features and this has probably been hased over on evom, but i hate trying to read stuff there...

I can do some testing on these maps this week since i dont have finals till next week, but it might be an interesting idea...

And lastly...everyone's favorite 2-byte load:

I think i have tried these before and had no success, just want to see what im missing...
1) create the "00" and "01" maps in ECUFlash @ 30862 and 30866 respectively for an '03 Internal ID 94170014
2) change values in 00 and 01 to 35164 and 35165 respectivley
3) using the new Evoscan (.99) make sure that "Load MUT 2Byte Mod," "RPM MUT 2Byte Mod," and "E4 RAM 2Byte Mod" are all checked. There is also "E1 RAM 1ByteMod" but im not sure about that one.
4) log away!

So what am i doing wrong??? i had something different for step 3 that i had to put in in the 0.97 of evoscan, but it looks like it's integrated already in this version? or not and i have to use the same stuff from 0.97 which didnt work either...

[earlyapex aka jack ass]

Every car will need different WGDC %'s. Also if you have a boost leak, that will throw them off as well of course.

Also, if you are tuning boost with error correction off, your WGDC's will be closer to what the real WGDC's will need to be to run the boost you want. Then you should turn on Error Correction and make sure your BDEL tables are correct. I tune them both at the same time to save headaches. If you get the BDEL's really close, your WGDC shouldn't have to be tweaked much. This is in a controlled environment, on the street with varied loads, surfaces, weather etc, is where the BDEL and Error Correction will come more into play as that is what it is there for, maintaining target boost no matter what the influences.

If you want to be in open loop all the time to fine tune your AFR, then you can set the desired temp to hit closed loop option to something super high that the car will never hit and you will be in open loop all the time.

[XVeloX]

Every car will need different WGDC %'s. Also if you have a boost leak, that will throw them off as well of course.

Also, if you are tuning boost with error correction off, your WGDC's will be closer to what the real WGDC's will need to be to run the boost you want. Then you should turn on Error Correction and make sure your BDEL tables are correct. I tune them both at the same time to save headaches. If you get the BDEL's really close, your WGDC shouldn't have to be tweaked much. This is in a controlled environment, on the street with varied loads, surfaces, weather etc, is where the BDEL and Error Correction will come more into play as that is what it is there for, maintaining target boost no matter what the influences.

If you want to be in open loop all the time to fine tune your AFR, then you can set the desired temp to hit closed loop option to something super high that the car will never hit and you will be in open loop all the time.

ya i get that about the WGDC's...but it was a HUGE difference... with a boost leak i would assume i would need higher WGDC's to create more boost to offset the leak. but at the same time you would need to open in more because there is effectively less pressure available to open up the WG?

/confused

good point about the desired temp. Didnt think of that. How well does a car run like this if tuned properly? the only difference i see other than the "safety" pieces is the fact that idle and cruise the timing jumps all over the place...

[turbotiger]

Have you tried using mitsulogger instead of evoscan?
It's free, just use it with DLL to look at the logs. I find it easier to use to tune boost when I have one log and graph to look at. I currently have the 2 byte load mod along with the faster baud rate mod on my 05 ecu. I used mitsulogger & DLL to see what my load and boost is.

And +1 for bryan's comment, each car is going to have different settings for boost. You just need to adjust for each car.

You also need to add a few lines to the evoscan xml file in order to log 2 byte load correctly. You also need to modify the DLL definition file for it to show up in DLL correctly as well.

[XVeloX]

Have you tried using mitsulogger instead of evoscan?
It's free, just use it with DLL to look at the logs. I find it easier to use to tune boost when I have one log and graph to look at. I currently have the 2 byte load mod along with the faster baud rate mod on my 05 ecu. I used mitsulogger & DLL to see what my load and boost is.

And +1 for bryan's comment, each car is going to have different settings for boost. You just need to adjust for each car.

You also need to add a few lines to the evoscan xml file in order to log 2 byte load correctly. You also need to modify the DLL definition file for it to show up in DLL correctly as well.

ya i have mitsulogger, just not as big of a fan. i guess i could easily transfer to it, but i like evoscan and have my DLL setup perfectly for it. I do like the new mitsulogger that logs boost/EGT/ etc. but i dunno, evoscan is just easier to add items etc. because that's what i used for like 6 months before mitsulogger was better.

If you dont want to post here, can you send PM with details on the 2Byte mod ...aka what im missing, and also the faster baud rate one as well? tia if you can.

As far as boost, i realize that every car is different, but i had assumed that my WGDC's wouldnt nearly 20% different... i dunno it should have been close right? I expected +/- 5% or so, not 20's...

[spoolin]

^^^ BTW, we fixed the problem that was causing his funky WGDC maps... he had a large boost leak that was going from the GM boost sensor to the wastegate.. OUCH.

[XVeloX]

ya now im running 17.5 psi lol i knew there was a problem when my maps were so far off from anyone elses. But when i originally tuned it i had no boost leak tester and no good compressor. So now i only have a throttle body leak that holds ~2 psi w/o leak. If i crack the throttle to where the seals done leak as bad i think i was able to slow the leak and hold ~5 psi.... this is not fun.... but Oliver @ WORKS said that Pete has had some improvements and they might have the new 25psi+ seals in stock very soon to do some testing. According to Pete the stock ones are rated to something like 11 psi based on the material composition that mitsu used. So if the stock ones are holding 25+ (correct bryan?) then these should be nearly indistructable according to the overseas manufacturer of the new seals that Pete ordered. I'll let works fo into more detail, but its a new service they say they'll be offering since nobody has good throttle body seals available. Supposedly the ones off of machv are meant for DSMs only really and are rated to 7psi, and i was able to get my TB up to 16 w/o leaks. Soooooooo these new seals should be dope!!!!!!!!!

i'll keep you guys updated.