|
nyet
|
 |
« Reply #60 on: September 10, 2019, 09:42:36 PM »
|
|
|
No worries this can be a bit tricky.
Bottom line, the pressure axis is 'lde' or in other words, pressure difference between actual and requested.
If you are underboosting, lde is positive
If you are overboosting, lde is negative
this means that if you are overshooting, the output will follow the lowest lde line (iirc)
ldrkd also cuts out on B_lddy=0 (which happens when lde drops below 10 or close to overshoot, and then turns back on when lde comes back over 100 or lots of underboost) so it could be that that is the issue.
ldrkp will follow LDRQ0S and not KFLDRQ0 on B_lddy=0 as well, so i think we should just start with increasing LDRQ0S and seeing what happens.
I dont recall if you are logging B_lddy but if you aren't, please do. Also log lde if you aren't.
I'm also thinking a simpler alternative would be to cut IMX a bit more to prevent overshoot in the first place.
|
|
|
|
« Last Edit: September 10, 2019, 09:56:42 PM by nyet »
|
Logged
|
ME7.1 tuning guideECUx PlotME7Sum checksumTrim heatmap toolPlease do not ask me for tunes. I'm here to help people make their own. Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
|
|
|
|
nyet
|
|
« Reply #61 on: September 10, 2019, 10:01:16 PM »
|
|
|
Yea, scratch all of the above, your IMX is no good.
You really dont want it all the way up at 90 during spool, and 85 at boost crossover is too high. You need it around 80-81 or so the whole time.
|
|
|
|
|
Logged
|
ME7.1 tuning guideECUx PlotME7Sum checksumTrim heatmap toolPlease do not ask me for tunes. I'm here to help people make their own. Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
|
|
|
seeeeergiu
Jr. Member
Karma: +0/-0
 Offline
Posts: 42
|
|
« Reply #62 on: September 11, 2019, 02:14:14 AM »
|
|
|
Yea, scratch all of the above, your IMX is no good.
You really dont want it all the way up at 90 during spool, and 85 at boost crossover is too high. You need it around 80-81 or so the whole time.
I did not find B_lddy in my logger .cfg file, i dont know how to find the address and other details to add it manually in .ecu file either. Regarding KFLDIMX, it is 75.33 at 2500rpm, 69.95 at 3000rpm, 70.78 at 3500rpm, i`m a bit confuse...  Maybe i sould increase it to the values mentioned by you? Or because of my lack of knowledge i should have understood that in fact, KFLDRL should be touched (being closely related to KFLDIMX) ? |
|
|
|
|
Logged
|
|
|
|
mutzicu
Full Member
Karma: +5/-1
Offline
Posts: 66
|
|
« Reply #63 on: September 11, 2019, 02:40:06 AM »
|
|
|
I think he's saying you are pushing the turbo too much and the PID is not happy. I think he wants you to lower your boost request and adjust IMX so that WGdc floats around 80%
|
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #64 on: September 11, 2019, 02:56:55 AM »
|
|
|
I did not find B_lddy in my logger .cfg file, i dont know how to find the address and other details to add it manually in .ecu file either. Regarding KFLDIMX, it is 75.33 at 2500rpm, 69.95 at 3000rpm, 70.78 at 3500rpm, i`m a bit confuse... Maybe i sould increase it to the values mentioned by you? Or because of my lack of knowledge i should have understood that in fact, KFLDRL should be touched (being closely related to KFLDIMX) ? No, its fine, just pull IMX, its being bumped up after it is fed through DRL |
|
|
|
|
Logged
|
ME7.1 tuning guideECUx PlotME7Sum checksumTrim heatmap toolPlease do not ask me for tunes. I'm here to help people make their own. Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
|
|
|
seeeeergiu
Jr. Member
Karma: +0/-0
Offline
Posts: 42
|
|
« Reply #65 on: September 12, 2019, 01:16:40 PM »
|
|
|
No, its fine, just pull IMX, its being bumped up after it is fed through DRL
Hello, I tried a new map with KFLDIMX lowered by 10 at [2000 rpm - 4000 rpm]. Results: -issue still there -seems to oscilate a little faster than before -boost spikes are lower, doesnt go above requested so much as before Any other ideeas?  LE:as i now see a lot of oscilation in D, maybe i should also touch KFLDRQ2? |
|
|
|
« Last Edit: September 12, 2019, 02:59:05 PM by seeeeergiu »
|
Logged
|
|
|
|
|
BlackT
|
|
« Reply #66 on: September 12, 2019, 03:25:24 PM »
|
|
|
What turbo do you have?
|
|
|
|
|
Logged
|
|
|
|
seeeeergiu
Jr. Member
Karma: +0/-0
Offline
Posts: 42
|
|
« Reply #67 on: September 12, 2019, 10:37:03 PM »
|
|
|
What turbo do you have?
Turbo is stock, K03S. |
|
|
|
|
Logged
|
|
|
|
|
BlackT
|
|
« Reply #68 on: September 13, 2019, 01:04:37 AM »
|
|
|
Don't be offended, but this is a really shit tune. As begginer you can't yust put values in table. You are only gething more and more problems.
Go get to beggining... upload stock file, and match fuel for your injectors. Once you do that, step by step change values for load,and make logs. Then we can fix your problem properly and make a good tune
BDW. take a read a little about KFMIRL and KFMIOP, they look awful
|
|
|
|
|
Logged
|
|
|
|
|
sonique
|
|
« Reply #69 on: September 13, 2019, 08:09:06 AM »
|
|
|
Need correct boost pid all auq binary
factory not good after mod downpipe or add more boost
nyet solution working other s4 or other binary but not auq
best regards
|
|
|
|
|
Logged
|
|
|
|
seeeeergiu
Jr. Member
Karma: +0/-0
Offline
Posts: 42
|
|
« Reply #70 on: September 14, 2019, 02:01:43 AM »
|
|
|
Don't be offended, but this is a really shit tune. As begginer you can't yust put values in table. You are only gething more and more problems.
Go get to beggining... upload stock file, and match fuel for your injectors. Once you do that, step by step change values for load,and make logs. Then we can fix your problem properly and make a good tune
BDW. take a read a little about KFMIRL and KFMIOP, they look awful
Its ok, i take criticism in a constructive way, a slap on the wrist drives me forward.  I did my best to adapt the injectors, is something looking wrong in logs? The fact that actual lambda oscilatew too much or because is overall a little below requested? |
|
|
|
|
Logged
|
|
|
|
|
adam-
|
|
« Reply #71 on: September 14, 2019, 02:30:26 PM »
|
|
|
Fuelling is now in control, your PID is a shitstorm, as is your request.
|
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #72 on: September 16, 2019, 08:31:32 AM »
|
|
|
Honestly, i've seen worse. His req is what you'd expect if you are using HBN to limit boost rather than LDRXN
I am unsure the best way of fixing his PID issues though.
either drastically cutting Q0 and Q2 or flattening DRL in the affected area
|
|
|
|
|
Logged
|
ME7.1 tuning guideECUx PlotME7Sum checksumTrim heatmap toolPlease do not ask me for tunes. I'm here to help people make their own. Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
|
|
|
seeeeergiu
Jr. Member
Karma: +0/-0
Offline
Posts: 42
|
|
« Reply #73 on: September 18, 2019, 10:37:46 AM »
|
|
|
Honestly, i've seen worse. His req is what you'd expect if you are using HBN to limit boost rather than LDRXN
I am unsure the best way of fixing his PID issues though.
either drastically cutting Q0 and Q2 or flattening DRL in the affected area
Since the parameters involved are: P-Anteil: ( LDRQ0DY ( bzw. LDRQ0S ) - KFLDRQ2 ( bzw. 0 ) ) * lde = ldptv I-Anteil: lditv(i-1) + KFLDRQ1 ( bzw. LDRQ1ST ) * lde(i-1) = lditv D-Anteil: ( lde - lde(i-1) ) * KFLDRQ2 ( bzw. 0 ) = ldrdtv 1.I am not sure if the "0" is referring to actual 0 or KFLDRQ0. Guessing it is referring to value "0" and its switched by B_lddy if i understood the FS correctly. Can someone please confirm? 2.As I see high oscilation in P and lots of it in D, shouldnt i take small steps like modifying KFLDRQ2 first (and Q0 later)? 3.If so, i am unsure about how much i should lower KFLDRQ2. Judging by the formula vs logs vs TT225 values, maybe like a 30% reduction? Is my judgement correct? ->i`ve attached my ideea of a reduced KFLDRQ2 |
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #74 on: September 18, 2019, 10:55:55 AM »
|
|
|
Hard to say, tuning out occilations in PIDs is more art that science.
yes, b_lddy completely turns off Q2 when lde is close to zero.
Looking at your logs it is tough to see what the cause of the oscillation is, I'm hoping just reducing Q0S is enough.
maybe start with that -50% and log and compare with the last one.
|
|
|
|
|
Logged
|
ME7.1 tuning guideECUx PlotME7Sum checksumTrim heatmap toolPlease do not ask me for tunes. I'm here to help people make their own. Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
|
|
|
|
