Placebo
Full Member
Karma: +3/-0
 Offline
Posts: 108
|
 |
« Reply #30 on: August 11, 2016, 06:45:10 AM »
|
|
|
Don't touch the spool area or you may get a pos deviation code.
Understood. Makes good sense to me, thanks. |
|
|
|
|
Logged
|
|
|
|
Placebo
Full Member
Karma: +3/-0
Offline
Posts: 108
|
|
« Reply #31 on: August 11, 2016, 06:55:09 AM »
|
|
|
I tend to max KFLDHBN until my boost response is where I want it, and my boost pid's are inline. Then, I go back and set it slightly higher, so it does not limit charge pressure, but still provides a margin of safety.
Ill try maxing out KFLDHBN to see if this works and then tune for safety. Thanks. Are you sure you're requesting enough load? Also, in MED9 files there are multiple LDRXN's, and some other maps. Either modify them all, or figure which ones pertain to your variant coding.
Im not convinced I am requesting enough load so looking into it more but I am pretty sure my bin only has one set of maps. Dunno why but believe this is because it is an early bin and also because it is manual transmission. Ill double check again for other LDRXN to be sure. Thanks. Got a busted window regulator that I have to fix before more tuning but should be able to tweak and log sometime this weekend. |
|
|
|
|
Logged
|
|
|
|
Placebo
Full Member
Karma: +3/-0
Offline
Posts: 108
|
|
« Reply #32 on: August 11, 2016, 11:13:40 AM »
|
|
|
Okay, was able to squeeze in another flash and log. Some improvements obtained by adjusting KFLHDBN ~+30% and KLDRXN +~20% excluding tails.
Loads go up and approach but fall short of the KLDRXN values. Actual boost is below requested boost by significant amount. Somewhat surprised I did not get a code thrown.
Maybe something else is limiting my boost pressure? Sure looks like a correction factor is being applied when looking at how the actual boost scales with requested boost for both profiles. Perhaps I need to move on and start playing with PID? I guess it could also be that my tiny turbo cant put out what I request in KLDRXN and I need to rescale this to more appropriate values. I think all is okay mechanical as I have a new diverter valve installed.
Thoughts and suggestions about what to do next would be very much appreciated.
|
|
|
|
|
Logged
|
|
|
|
|
vwaudiguy
|
|
« Reply #33 on: August 11, 2016, 11:26:02 AM »
|
|
|
Log wastegate duty cycle. There are ways to get a higher sampling rate as well with VCDS. You should look into that.
|
|
|
|
|
Logged
|
"If you have a chinese turbo, that you are worried is going to blow up when you floor it, then LOL."
|
|
|
|
SB_GLI
|
|
« Reply #34 on: August 11, 2016, 11:44:17 AM »
|
|
|
There are ways to get a higher sampling rate as well with VCDS.
Can you elaborate on this? linky? please and thanks. EDIT - from ross tech site: 4.8 How can I speed up my sample rate in VCDS? The sample rate for viewing and logging data is largely determined by each controller in the car. There are a few things you can do to speed up the rate for a given controller. First, logging one group at a time, rather than two or three, will help greatly. In order to get the highest sample rate possible go to the Options screen and set Blk Int to 25 and Char Int to 0. Note, however, that this may cause unreliable communications with some controllers. Changing KP2 time to a smaller value speeds up the sample rate in Measuring Blocks on control modules that use the KWP-2000 protocols. Try setting it to 0. If any of these adjustments cause unreliable communications, change them back closer to the defaults. If in doubt, use the Restore Defaults button! In Engine controllers using KWP-2000, there is a [Turbo] button when using a HEX interface.. This can significantly speed up sampling, in some cases to over 30 samples per second. See the Measuring Blocks page for more information. In Engine controllers using UDS, there is a "Group UDS requests" checkbox that can improve sample rates. See the Advanced Measuring Values page for more information. Sorry to clutter the thread... again. |
|
|
|
« Last Edit: August 11, 2016, 11:57:05 AM by SB_GLI »
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #35 on: August 11, 2016, 11:47:55 AM »
|
|
|
PID I max.
Look at the 1.8 and 2.7 community tune threads and the tuning wiki regarding i limiter.
|
|
|
|
|
Logged
|
|
|
|
|
vwaudiguy
|
|
« Reply #36 on: August 11, 2016, 12:06:34 PM »
|
|
|
Can you elaborate on this? linky? please and thanks.
The info you posted was what I was referring to. |
|
|
|
|
Logged
|
"If you have a chinese turbo, that you are worried is going to blow up when you floor it, then LOL."
|
|
|
Placebo
Full Member
Karma: +3/-0
Offline
Posts: 108
|
|
« Reply #37 on: August 13, 2016, 11:10:23 AM »
|
|
|
Thanks all for the helpful suggestions.
I modified the KFLDIMX 800, 1000 table data by the same 20% that I had adjusted LDRXN; while capping it at a max of 95 I saw elsewhere in the stock table. Ive read KFLDIMX does not scale linear but figured this would be a good first attempt.
Updated XDF that has this map is attached along with the updated bin and log.
Improvements are seen in the log and actual boost tracks well with requested boost up to about 3500 rpm. Duty cycle after 4500 rpm is 95%. Not sure if futher adjusment of KFLDIMX will help as I interpret this to mean the turbo can no longer provide requested boost at higher rpm based upon physical limitation. Please let me know if this is wrong and further KFLDIMX adjustments are needed.
Unless someone suggests otherwise, I will taper my LDRXN so the actual boost can match requested boost. Perhaps I should be looking at a turbo efficiency map?
|
|
|
|
|
Logged
|
|
|
|
Placebo
Full Member
Karma: +3/-0
Offline
Posts: 108
|
|
« Reply #38 on: August 13, 2016, 11:17:55 AM »
|
|
|
Forgot to attach the log.
|
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #39 on: August 13, 2016, 02:01:56 PM »
|
|
|
the turbo can no longer provide requested boost at higher rpm based upon physical limitation. Please let me know if this is wrong and further KFLDIMX adjustments are needed.
Unless someone suggests otherwise, I will taper my LDRXN so the actual boost can match requested boost.
Yep. Correct on both accounts. Perhaps I should be looking at a turbo efficiency map?
I think that will just tell you what you already know: the turbo is maxed out. Next step is to make fueling is ok, then timing. |
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #40 on: August 13, 2016, 03:06:47 PM »
|
|
|
Wastegate actuator is loose or boost leak in that log, they easily hold 1.1-1.2 bar to 5000 and 0.95 bar at 6000.
|
|
|
|
|
Logged
|
|
|
|
|
vwaudiguy
|
|
« Reply #41 on: August 13, 2016, 04:18:02 PM »
|
|
|
Wastegate actuator is loose or boost leak in that log, they easily hold 1.1-1.2 bar to 5000 and 0.95 bar at 6000.
Thanks for the info. At what range of duty cycle? |
|
|
|
|
Logged
|
"If you have a chinese turbo, that you are worried is going to blow up when you floor it, then LOL."
|
|
|
|
prj
|
|
« Reply #42 on: August 13, 2016, 06:12:06 PM »
|
|
|
1.1 bar at 5000 rpm at 65% for me with slightly stiffened actuator.
0.95 bar at 6000 rpm with 95% dc.
These cars suffer (among other things) from loose wastegate flaps. That issue needs to be fixed before tuning.
|
|
|
|
|
Logged
|
|
|
|
Placebo
Full Member
Karma: +3/-0
Offline
Posts: 108
|
|
« Reply #43 on: August 13, 2016, 08:47:01 PM »
|
|
|
Thanks for the waste gate alert, will return once whatever is wrong is fixed.
|
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #44 on: August 13, 2016, 10:13:18 PM »
|
|
|
Thanks for the waste gate alert, will return once whatever is wrong is fixed.
When you fix wg you will want to back out IMX significantly to avoid overboosting.. |
|
|
|
|
Logged
|
|
|
|
|
