|
ddillenger
|
 |
« Reply #15 on: January 30, 2014, 03:50:55 PM »
|
|
|
Are we talking about SY EGFE?
|
|
|
|
|
Logged
|
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 experience!
Email/Google chat:
DDillenger84(at)gmail(dot)com
Email>PM
|
|
|
|
prj
|
|
« Reply #16 on: January 30, 2014, 04:00:25 PM »
|
|
|
Are we talking about SY EGFE?
That's a compile time constant. |
|
|
|
|
Logged
|
|
|
|
Aurélien
Full Member
Karma: +4/-0
 Offline
Posts: 60
|
|
« Reply #17 on: January 31, 2014, 09:35:03 AM »
|
|
|
I think PRJ and I had the same idea. It only needs to alter 1 ram and tweak few constant for sensor diagnosis. And relocate the sensor, of course. If you really understand how motronic works and calculate load, etc, you will figure out quickly what we are talking about.  |
|
|
|
|
Logged
|
|
|
|
|
jibberjive
|
|
« Reply #18 on: February 15, 2014, 04:07:12 AM »
|
|
|
Going to be implementing dual maf's as well, though I'm taking the pains to do it in a fashion that can be widely adopted by the general B5 populace. But sub'd to this thread for more discussion.
|
|
|
|
|
Logged
|
|
|
|
|
Snow Trooper
|
|
« Reply #19 on: February 15, 2014, 09:06:25 AM »
|
|
|
Just go look at any vw / Audi application that uses dual mafs. Look at how they are wired, look at the sensor itself also. It would be really easy to build a kit for this. I just never saw a big market for it.
|
|
|
|
|
Logged
|
cartoons?
6A 61 72 65 64 40 76 6C 6D 73 70 65 63
|
|
|
|
Snow Trooper
|
|
« Reply #20 on: February 15, 2014, 09:13:07 AM »
|
|
|
I think PRJ and I had the same idea. It only needs to alter 1 ram and tweak few constant for sensor diagnosis. And relocate the sensor, of course. If you really understand how motronic works and calculate load, etc, you will figure out quickly what we are talking about. Are you guys talking about the map literally supplying the 0-5v over the maf circuit? If ky thinking is correct we could darn near swap one for one. Load is load, signal is signal, voltage is voltage. Or maybe I am way off base. |
|
|
|
|
Logged
|
cartoons?
6A 61 72 65 64 40 76 6C 6D 73 70 65 63
|
|
|
|
prj
|
|
« Reply #21 on: February 15, 2014, 04:27:19 PM »
|
|
|
Dual MAF is extremely easy.
If you look at RS6 code, all it literally does is (voltage 1 + voltage 2)/2.
Copy paste that into your binary and voila you have dual MAF.
Just need to find a free ADC, but with the removal of the rear O2's you get two free ADC's.
Speed-density is a different story.
|
|
|
|
|
Logged
|
|
|
|
|
jibberjive
|
|
« Reply #22 on: February 15, 2014, 11:52:11 PM »
|
|
|
Just go look at any vw / Audi application that uses dual mafs. Look at how they are wired, look at the sensor itself also. So you mention TDi's and say that they utilize a resistor and simple wiring with a single maf input to the ECU, and other people mention RS6's and how they do whatever combining of signals that they do within the ECU processor, not via wiring. So apparently not all Audi/VW applications accomplish this in the same manner? Also, MAF transfer functions aren't linear, so how would a simple summing of voltages maintain accuracy (for example, say in the case that different amounts of air are flowing through each maf)? |
|
|
|
« Last Edit: February 15, 2014, 11:54:44 PM by jibberjive »
|
Logged
|
|
|
|
|
ddillenger
|
|
« Reply #23 on: February 15, 2014, 11:57:00 PM »
|
|
|
Also, there is only one KFKHFM in the RS6 file...
|
|
|
|
|
Logged
|
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 experience!
Email/Google chat:
DDillenger84(at)gmail(dot)com
Email>PM
|
|
|
|
phila_dot
|
|
« Reply #24 on: February 16, 2014, 12:54:35 AM »
|
|
|
Also, there is only one KFKHFM in the RS6 file...
There's two inputs, two sets of variables, and B1 and B2 share the tables. |
|
|
|
|
Logged
|
|
|
|
|
ddillenger
|
|
« Reply #25 on: February 16, 2014, 12:59:12 AM »
|
|
|
Seems unwise to have the same correction map for each bank.
|
|
|
|
|
Logged
|
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 experience!
Email/Google chat:
DDillenger84(at)gmail(dot)com
Email>PM
|
|
|
|
jibberjive
|
|
« Reply #26 on: February 16, 2014, 03:55:54 AM »
|
|
|
There's two inputs, two sets of variables, and B1 and B2 share the tables.
You said the RS6 has two uhfm's, one MLHFM, and two mshfm's. I assume sometime shortly after that it sums both mshfm's? Curious, as this goes against prj saying that it 'sums the voltages and divides by two'. Though it makes more sense that they would sum the mass air flow values rather than summing the voltages, especially considering the type of sensors it is using. Seems unwise to have the same correction map for each bank.
Considering that both of the MAF's have essentially symmetrical intake geometry pre-maf, maf mounting, etc, it makes sense that they would share a KFKHFM. Since KFKHFM is not like a variable 'trim' for the MAF's, and it is calibrated once at the factory, you would expect MAF's that are the same sensor and same geometry to have the same KFKHFM. |
|
|
|
|
Logged
|
|
|
|
|
phila_dot
|
|
« Reply #27 on: February 16, 2014, 11:29:19 AM »
|
|
|
I did the legwork on this about a year ago, but don't remember all of the details.
I posted the details that I do remember from the diassembly. I always second guess myself when disagreeing with prj, but I can tell you for a fact that the voltages are not averaged and that seperate variables for each bank goes through the the same linearization, offset, and correction tables.
|
|
|
|
|
Logged
|
|
|
|
|
phila_dot
|
|
« Reply #28 on: February 16, 2014, 01:58:11 PM »
|
|
|
Correction, one mshfm_w / mshfms_w.
Airflow is calculated seperately for each bank, then summed becoming mshfm_w and mshfms_w.
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #29 on: February 18, 2014, 01:46:34 PM »
|
|
|
Correction, one mshfm_w / mshfms_w.
Airflow is calculated seperately for each bank, then summed becoming mshfm_w and mshfms_w.
Yes, this is true indeed, I just looked up the code. Both go through KFKHFM etc… Not terribly important though - voltages can be just summed and averaged as a quick hack. That said, this is not difficult to port over properly, as the RS6 file is right there, and it is not hard to implement the routines from RS6. |
|
|
|
|
Logged
|
|
|
|
|
