|
nyet
|
 |
« Reply #45 on: April 25, 2016, 10:03:55 AM »
|
|
|
Doesn't the Motronic substitute the PS_W value directly for the contrived MAP value past a certain throttle % anyway (once WDKUGDN is exceeded, I think)?
ps_w is used pretty much everywhere for fueling... and, in fact, for calculating load. So it really is a case of just hacking real MAP output into ps_w. At least, that is my (purely theoretical understanding) of what prj is suggesting. Of course, without publicly testable code, it will remain theory. |
|
|
|
|
Logged
|
|
|
|
mark_r33
Jr. Member
Karma: +0/-0
 Offline
Posts: 30
|
|
« Reply #46 on: April 25, 2016, 10:37:17 AM »
|
|
|
Ahhh, sorry, I'm misremembering my acronyms, there are so many! pvdks_w I think I mean, the pressure in the boost pipe from the boost sensor  What I meant to mean above is; doesn't the manifold modelled pressure (PS_W / PSP_W) get replaced by the actual measured pressure in the boost pipe after the ecu goes over to the "unthrottled" mode, as in past the point at which the ecu is programmed to think the throttle has no effect? Could this be a hacky way to achieve the MAP -> PS_W we seek here, mechanically by plumbing the boost sensor into the manifold, and hacktastically by telling the ecu it is always in unthrottled mode? |
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #47 on: April 25, 2016, 10:40:12 AM »
|
|
|
Could this be a hacky way to achieve the MAP -> PS_W we seek here, mechanically by plumbing the boost sensor into the manifold, and hacktastically by telling the ecu it is always in unthrottled mode?
OH! Ok, I see what you are getting at. That seems like it should work, except there might be a lot of places where upstream throttle plate pressure is expected to be different from MAP... Not saying it isn't possible, just a bit more unsure of the repercussions vs straight replacing ps_w. |
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #48 on: May 12, 2016, 08:23:29 AM »
|
|
|
Some interesting stuff going on in here!
Doesn't the Motronic substitute the PS_W value directly for the contrived MAP value past a certain throttle % anyway (once WDKUGDN is exceeded, I think)? No. It never does this. For hints on how to do the conversion, look at 2.5NA Jetta ECU's... They have MAP and MAF and can run without the MAF from manifold sensor... I am not interested in posting this publicly. Besides, it is a new implementation for every ECU. So far I have done 1.8T, 2.7TT and R32. |
|
|
|
|
Logged
|
|
|
|
|
masterj
|
|
« Reply #49 on: October 17, 2016, 12:18:07 PM »
|
|
|
How do people overcome slower map vs maf signal response when in mafless? Any hints on which maps to tackle in ESUK module? Initially when transitioning from idle to wot I see a big lambda correction request of ~30% for a fraction of second. Anyone encountered similar situation?
|
|
|
|
|
Logged
|
|
|
|
|
vwaudiguy
|
|
« Reply #50 on: October 17, 2016, 01:44:18 PM »
|
|
|
How do people overcome slower map vs maf signal response when in mafless? Any hints on which maps to tackle in ESUK module? Initially when transitioning from idle to wot I see a big lambda correction request of ~30% for a fraction of second. Anyone encountered similar situation?
I was under the impression this was one of the downsides of no maf. I believe you need to adjust the throttle plate maps, so the ecu can predict in advance (as best as it can) what the airflow is going to be. |
|
|
|
|
Logged
|
"If you have a chinese turbo, that you are worried is going to blow up when you floor it, then LOL."
|
|
|
|
_nameless
|
|
« Reply #51 on: October 17, 2016, 04:36:04 PM »
|
|
|
log with a maf and correct air over throttle maps based off actual air flow
|
|
|
|
|
Logged
|
If you are in the market for a tune and would like the ease of downloading and flashing a dyno tested tune for a fair price check out https://instatune.sellfy.store/ |
|
|
dragon187
Full Member
Karma: +13/-15
Offline
Posts: 106
|
|
« Reply #52 on: October 29, 2016, 01:18:33 AM »
|
|
|
Ahhh, sorry, I'm misremembering my acronyms, there are so many! pvdks_w I think I mean, the pressure in the boost pipe from the boost sensor What I meant to mean above is; doesn't the manifold modelled pressure (PS_W / PSP_W) get replaced by the actual measured pressure in the boost pipe after the ecu goes over to the "unthrottled" mode, as in past the point at which the ecu is programmed to think the throttle has no effect? Could this be a hacky way to achieve the MAP -> PS_W we seek here, mechanically by plumbing the boost sensor into the manifold, and hacktastically by telling the ecu it is always in unthrottled mode? Hi Are there any news on this ps_w thing? If we can just can change ps_w to max 5120 it would be great. I have also seen a r32 working with map instead of maf. |
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #53 on: October 29, 2016, 01:26:13 AM »
|
|
|
If we can just can change ps_w to max 5120 it would be great.
Think so? Maybe somebody should discuss it here on nefmoto. Might even result in a link to it in in the s4wiki. Cray. |
|
|
|
|
Logged
|
|
|
|
|
masterj
|
|
« Reply #54 on: November 02, 2016, 02:09:54 PM »
|
|
|
log with a maf and correct air over throttle maps based off actual air flow
If you're talking about the KFMSNWDK & KFWDKMSN then they're fixed long time ago. My AFR overall is ok in whole rpm range on wot and idle. Problem now is with these lean conditions when letting of gas pedal (as in throttle plate close). I thought I could workaround them by lowering KFVAKL in ESUK module (wall wetting) on deceleration, but that did not help a bit (see attached log). |
|
|
|
« Last Edit: November 02, 2016, 02:12:06 PM by masterj »
|
Logged
|
|
|
|
|
Rick
|
|
« Reply #55 on: November 04, 2016, 10:42:37 AM »
|
|
|
You have no MAP sensor, only boost, so there is no way of accurately determining partial filling.
Rick
|
|
|
|
|
Logged
|
|
|
|
|
masterj
|
|
« Reply #56 on: November 05, 2016, 12:02:47 PM »
|
|
|
Thanks goes to fukenbroken for pointing me to b_statmd variable. So I guess what prj said about easier fix is actually true. Basically if someone wants to remove maf here's what you have to do: set CLALM error class all bits to 0. Then apply one patch to logic in CAN module (look at the attached image).
Hopefully this is finally completely open to the public without anything hidden.
|
|
|
|
« Last Edit: November 05, 2016, 12:10:33 PM by masterj »
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #57 on: November 05, 2016, 01:41:06 PM »
|
|
|
How does this solve the no MAP problem?
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #58 on: November 05, 2016, 02:05:31 PM »
|
|
|
Yup, that's exactly how to do it.
However, if you are actually doing it correct with two MAP sensors, then you do not need to do this, as the ECU is still running from the primary load signal.
So E_lm is never set, except for if when something happens to the MAP sensor (if you configured the DHFM module correctly, to diagnose the sensor instead of the MAF)...
|
|
|
|
|
Logged
|
|
|
|
|
masterj
|
|
« Reply #59 on: November 05, 2016, 02:17:52 PM »
|
|
|
Yup, that's exactly how to do it.
However, if you are actually doing it correct with two MAP sensors, then you do not need to do this, as the ECU is still running from the primary load signal.
So E_lm is never set, except for if when something happens to the MAP sensor (if you configured the DHFM module correctly, to diagnose the sensor instead of the MAF)...
I have few questions though. I tracked few things from DHFM and want someone to correct me if I'm wrong. 1. Throttle plate diagnostics @ GGDVE-GLPWDK13? Haven't logged it but looks like when b_ehfm = 1 it sets the b_dkp1e & b_dkp1np and I guess they should set error for throttle plate signal ( DDVE-DK1P)? Same goes for the GGDVE-GLPWDK23, no? 2. Situation @ BGSRM-BPIRG. When b_hfm = 0, then psmx_w is not 4, but rather map FPVMXN2. Does that look ok, or we should fix it back to 4 for mafless? |
|
|
|
|
Logged
|
|
|
|
|
