Kompiesto
Full Member
Karma: +4/-5
 Offline
Posts: 109
|
 |
« on: November 11, 2019, 11:36:09 AM »
|
|
|
Hello,
I want install new style maf instead old hfm2 maf. All mechanical work was done. I losing my mind how to scale this. I used data from s3 binary and compared datasheets. Maybe somebody done proper voltage table before me. Thanks for help.
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #1 on: November 11, 2019, 04:25:33 PM »
|
|
|
I have done this.
Remember, the wiring is different too, you need a 5V reference, the HFM2 doesn't use one - I used one from the hall sensor IIRC, and that worked fine.
Not much to say other than that I can sell you a file ... and tell you which part numbers to use.
Probably cheaper than trying to figure this out - I know I spent a lot of time on getting the cal right, so everything works.
Linearization is not straightforward due to lack of being able to calculate inverse flow and due to the load ceiling you hit if you scale it correctly, so some fudging is needed with FGAT0 and global MAF scale.
|
|
|
|
« Last Edit: November 11, 2019, 04:27:25 PM by prj »
|
Logged
|
|
|
|
Kompiesto
Full Member
Karma: +4/-5
Offline
Posts: 109
|
|
« Reply #2 on: November 11, 2019, 11:03:11 PM »
|
|
|
I done wiring. I used 5V from baro sensor. Part number of maf : 0280218065.
I modified linearization table from S3 BAM from 1x512 to 1x256 and set zero for negative flow but is too lean. I see difference. In old maf for 1.15V have something like 15-18kg/h for idle... in new table for new maf come out 1,15V and 7,5kg/h... and is too low value for idle.
Tryied uping all or moving all values forward to get proper idle, but still when tap pedal, car is going lean... Tested with o2 correction off.
I will be very appreciated for some help because i spent two days, without satisfy effects.... Last idea is connect two mafs inline and log old maf flow to voltage value new maf.
|
|
|
|
|
Logged
|
|
|
|
error1997
Newbie
Karma: +2/-0
Offline
Posts: 14
|
|
« Reply #3 on: November 11, 2019, 11:30:41 PM »
|
|
|
I have done it numerous times . Used excel to scale the linearization table by taking every other value . Negative flow is set to 0 .
If you still have lean issues after properly scaling , the sensor may be bad .
The voltage output at idle would be different .
|
|
|
|
|
Logged
|
|
|
|
Kompiesto
Full Member
Karma: +4/-5
Offline
Posts: 109
|
|
« Reply #4 on: November 12, 2019, 01:40:07 AM »
|
|
|
I try today something like in attachment... All data check with afr meter.
I try find another sensor and check... maybe sensor is bad...
|
|
|
|
|
Logged
|
|
|
|
Kompiesto
Full Member
Karma: +4/-5
Offline
Posts: 109
|
|
« Reply #5 on: November 13, 2019, 10:53:37 PM »
|
|
|
Ok, i think i found the problem.
I got maf with installed china insert with part number 039 F. Should be 049. I wait for new maf and I'll tell if that changed anything.
I set it by upping Khfm and fgat0 but maf cut power at 42g/s. Afr looks good. I must modify idle zone on maf curve.
Edit: Two mafs was damaged... works ok now. Thanks for help.
|
|
|
|
« Last Edit: November 15, 2019, 04:43:43 AM by Kompiesto »
|
Logged
|
|
|
|
360trev
Full Member
Karma: +68/-2
Offline
Posts: 235
|
|
« Reply #6 on: November 16, 2019, 02:07:30 AM »
|
|
|
Has anyone done something similar with other later spec sensors too?
E.g. Take for example the latest Bosch O2 Sensors too?
I.e. moved up to a later LSU 4.9 ? Obviously some code patches going to be necessary but its faster reacting, improved accuracy should see less deviation in the fuel trims and much closer flowing of the target lambda than earlier 4.4s....
|
|
|
|
|
Logged
|
|
|
|
|
_nameless
|
|
« Reply #7 on: November 16, 2019, 05:07:59 AM »
|
|
|
lsu4.2* and ive done this before too.
|
|
|
|
|
Logged
|
Giving your mom a tuneup
|
|
|
|
prj
|
|
« Reply #8 on: November 16, 2019, 05:43:13 AM »
|
|
|
Has anyone done something similar with other later spec sensors too?
E.g. Take for example the latest Bosch O2 Sensors too?
I.e. moved up to a later LSU 4.9 ? Obviously some code patches going to be necessary but its faster reacting, improved accuracy should see less deviation in the fuel trims and much closer flowing of the target lambda than earlier 4.4s....
Uhmm... M3.8 is narrowband. |
|
|
|
|
Logged
|
|
|
|
|
_nameless
|
|
« Reply #9 on: November 16, 2019, 05:59:38 AM »
|
|
|
hes using that as an example.  |
|
|
|
|
Logged
|
Giving your mom a tuneup
|
|
|
|
BlackT
|
|
« Reply #10 on: March 04, 2024, 03:17:39 AM »
|
|
|
Does ecus with HFM2 use offset like ones with hfm5?
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #11 on: March 04, 2024, 03:55:42 AM »
|
|
|
Does ecus with HFM2 use offset like ones with hfm5?
No, HFM2 can not measure inverse flow. |
|
|
|
|
Logged
|
|
|
|
|
