|
s5fourdoor
|
 |
« on: August 09, 2011, 01:26:02 AM »
|
|
|
as anyone who has tuned krkte and tvub at the same time realizes, adjusting the factors one at a time is a massive pain. i made some simple assumptions to help the average person derive their new krkte / tvub values given current krkte/tvub and the four block 032 inputs.
feel free to check my logic, but i'm pretty sure what i've done here is correct...
i'll probably refine this in the upcoming days as i test it in action.
|
|
|
|
« Last Edit: March 03, 2012, 02:33:52 PM by nehalem »
|
Logged
|
|
|
|
|
|
DonSupreme
Newbie
Karma: +3/-1
 Offline
Posts: 22
|
|
« Reply #2 on: August 09, 2011, 01:13:17 PM »
|
|
|
Not sure if its labeled wrong or if I am reading it wrong.
should KRKTE NEW, say tvub new or what?
|
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #3 on: August 09, 2011, 02:16:32 PM »
|
|
|
ohh, yes basically. the 5 point krkte old and krkte new represent TVUB using the krkte old and krkte new. i'll fix that...
fixed in the above attachment!
|
|
|
|
« Last Edit: August 09, 2011, 02:21:20 PM by nehalem »
|
Logged
|
|
|
|
|
jibberjive
|
|
« Reply #4 on: February 20, 2012, 07:23:02 PM »
|
|
|
Sorry for bumping the old, but I think this could be a good tool to keep near the top, and I have a question about it. So, according to Julex's post here, and Tony's quote in the OP of this thread: http://nefariousmotorsports.com/forum/index.php?topic=1512.msg14392#msg14392we want the multiplier to be the same across the board (assuming we have correctly interpolated TVUB values to begin with). So, if that logic is correct, according to your sheet, we would keep the mulitplier calculation from the sheet as-is for the cell in the multiplier column and 14.08V (assuming we usually operate around 14V when we log the LTFT's), and have all of the other multipliers equal that value, correct? The way you have it now, each TVUB value has a varying multiplier, which goes against what is said elsewhere. Just making sure I have my own basics correct. |
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #5 on: February 21, 2012, 02:07:27 PM »
|
|
|
Hi - The way I use this sheet has evolved, so I'll upload the corrected version tonight perhaps. I believe you are correct, but need to reread your comment more carefully. Will advise soon...
|
|
|
|
|
Logged
|
|
|
|
Hansi
Newbie
Karma: +0/-0
Offline
Posts: 3
|
|
« Reply #6 on: March 03, 2012, 07:55:11 AM »
|
|
|
Could you upload the new version please?  |
|
|
|
|
Logged
|
|
|
|
|
s5fourdoor
|
|
« Reply #7 on: March 03, 2012, 02:40:43 PM »
|
|
|
ok i updated the file.
here's how i look at the file: this isn't the best method, but it works and i get my trims to +/- .5% with relative ease and a few retunes. anyways here goes.
use: enter your current tvub and krkte values. goto your injector calibration sheet and find the base tvub values. these should be placed in your "tvub stock" columns on the right of the sheet. what you do is get everything in order and then enter your LTFT's. you then do rough adjustments until your KRKTE starts to get close. as your krkte hones in on a correct value, start using the tvub_mult located in cell E12. adjust this to then take your new "tvub dyna" values which you then move to the used column.
the basic idea is that you are tweaking the tvub master multiplier to raise or lower the entire curve. adjust this until you lower cell P12 to within +/- .005. this is your own way of indexing how rich/lean tvub has been made relative to it's own characteristics (found on the injector sheets). generally once your KRKTE gets close, you don't need to account too much for the partial effect of KRKTE on TVUB when adjusting the tvub_mult.
i'm sorry if some of this appears unprincipled. i promise you that it does work, and also if i were to spend enough time writing it out, i'm fairly confident i could give a theoretical justification why this works. if you are raging about the lack of theory i've left here, please look up gradient descent - which is the general sequential optimization method i use for most multi-variate co-integrated problems...
also... the "used column" is for values you end up picking. i would HIGHLY recommend not taking the values themselves that are labelled new. i generally take 50% of the old value and 50% of the new value. thats just how i do things, do as you please; i've had decent enough success when taking the actual "new values", but i just like things to be slow and safe. just me... tune how you will - and believe me i am far from the best tuner - i'm cautious, don't break stuff and eventually get results.
|
|
|
|
« Last Edit: March 03, 2012, 02:45:45 PM by nehalem »
|
Logged
|
|
|
|
Hansi
Newbie
Karma: +0/-0
Offline
Posts: 3
|
|
« Reply #8 on: March 09, 2012, 09:36:06 AM »
|
|
|
Great, thank you I will try it out.. |
|
|
|
|
Logged
|
|
|
|
marcello7x
Full Member
Karma: +1/-0
Offline
Posts: 126
|
|
« Reply #9 on: August 23, 2012, 06:22:50 PM »
|
|
|
going to try this tomorrow since i bump up my fpr to 4 bar so i can handle e85
|
|
|
|
|
Logged
|
|
|
|
|
masterj
|
|
« Reply #10 on: December 19, 2012, 08:26:16 AM »
|
|
|
I'm a little bit confused with this sheet... Could you clarify what should I do next... 1. I've copied my current KRKTE to D15 2. I've copied my current TVUB to C18:D22 3. I've copied base TVUB from inujector sheet to N16:R17 4. E12 = 0% 5.  Thanks |
|
|
|
|
Logged
|
|
|
|
|
aef
|
|
« Reply #11 on: July 13, 2014, 07:04:26 AM »
|
|
|
Any news here? This excel sheet looks very complicated  |
|
|
|
|
Logged
|
|
|
|
brembo
Jr. Member
Karma: +0/-0
Offline
Posts: 48
|
|
« Reply #12 on: September 07, 2014, 03:44:18 AM »
|
|
|
I always change only the KRKTE the conversion of other injectors.
How does the TVUB?In all the files I've never seen the TVUB was made.
Lg
|
|
|
|
|
Logged
|
|
|
|
igo300
Jr. Member
Karma: +2/-1
Offline
Posts: 27
|
|
« Reply #13 on: May 04, 2015, 05:05:49 AM »
|
|
|
use: enter your current tvub and krkte values. goto your injector calibration sheet and find the base tvub values. these should be placed in your "tvub stock" columns on the right of the sheet. what you do is get everything in order and then enter your LTFT's. you then do rough adjustments until your KRKTE starts to get close. as your krkte hones in on a correct value, start using the tvub_mult located in cell E12. adjust this to then take your new "tvub dyna" values which you then move to the used column.
the basic idea is that you are tweaking the tvub master multiplier to raise or lower the entire curve. adjust this until you lower cell P12 to within +/- .005. this is your own way of indexing how rich/lean tvub has been made relative to it's own characteristics (found on the injector sheets). generally once your KRKTE gets close, you don't need to account too much for the partial effect of KRKTE on TVUB when adjusting the tvub_mult.
Please correct me if I misunderstand this. Having put in the current KRKTE and TVUB numbers both in D18:22 and N17:R17 I should adjust KRKTE only, get up some miles, recheck the trims and repeat until KRKTE is close....then got0 E12 and adjust the TVUB multiplier to get close to zero in P12 again after multiple miles. Looks interesting |
|
|
|
|
Logged
|
|
|
|
|
