hey matt,
or anyone else that knows.
on the late model nissan ecu's that natively use the 0-5v type afm, does the afm signal output feed direct into the chip, or does it have some filtering circuitry in there some where.
i would think not, or else the accuracy and sensitivity of the air flow meter would be effected?
only reason i wish to confirm, after fitting the z32 afm to earlier hardware (vl ecu) and reviewing the logs, the afm voltage is like pulsing a little or maybe it just so sensitive it reporting the eradic airflow, if thats the problem, maybe its the boost controller pulsing the wastegate and effecting the inlet airflow and the meter is just so sensitive.
i will put a pic up to show what i mean.
do the late model nissan ecus...
Moderator: Matt
-
- Posts: 62
- Joined: Fri Jul 20, 2007 4:43 pm
do the late model nissan ecus...
- Attachments
-
- afm_sensitive.jpg
- (99.69 KiB) Downloaded 3157 times
-
- Posts: 62
- Joined: Fri Jul 20, 2007 4:43 pm
ok, fairly sure i worked out the problem.
it looks like it the excessive timing change between the two timing columns its accessing, either i will have to spread the two columns further apart by increasing the load tp numbers, or change timing so there is only a 2 degree maximum timing change between those columns. (if the engine doesn't mind)
i think the timing was effecting the engine power (explains the light flat spots i was feeling also) and in turn the inlet airflow was going up and down in turn.
i really wish i could expand the maps in such fashion that i didn't have to lose any resolution down low.
ohh well, think i might have to drop a few of the lower timing tp columns because i really need the resolution up top, since i need to tune it on the edge to get max power from the engine setup.
it looks like it the excessive timing change between the two timing columns its accessing, either i will have to spread the two columns further apart by increasing the load tp numbers, or change timing so there is only a 2 degree maximum timing change between those columns. (if the engine doesn't mind)
i think the timing was effecting the engine power (explains the light flat spots i was feeling also) and in turn the inlet airflow was going up and down in turn.
i really wish i could expand the maps in such fashion that i didn't have to lose any resolution down low.
ohh well, think i might have to drop a few of the lower timing tp columns because i really need the resolution up top, since i need to tune it on the edge to get max power from the engine setup.
sorry i missed this post
the AFM voltage 'normally' goes via an opamp dropping the voltage straight into a 9 bit A/D converter
when you bypass the signals on the opamp board, you are feeding the A/D directly which is what later ECUs did anyway
so AFM shouldn't make any difference. Yeah let us know how you go with that. you cant drop some of the resolution coming onto boost at all? i guess its linear interpolation between all points as it jumps up
the AFM voltage 'normally' goes via an opamp dropping the voltage straight into a 9 bit A/D converter
when you bypass the signals on the opamp board, you are feeding the A/D directly which is what later ECUs did anyway
so AFM shouldn't make any difference. Yeah let us know how you go with that. you cant drop some of the resolution coming onto boost at all? i guess its linear interpolation between all points as it jumps up
-
- Posts: 62
- Joined: Fri Jul 20, 2007 4:43 pm
Hey Matt,
Sorry to bump an old thread, but i'll try to keep similar topics in the same thread.
Back to what I asked originally in this thread some time ago, have you since found out if the later model nissan ecu feed the AFM 0-5v signal to the a/d convertor directly or is it actually protected from overvoltage?
Since you are now stating a zener diode should be used even with the Z32 AFM?
Regards
Mick
Sorry to bump an old thread, but i'll try to keep similar topics in the same thread.
Back to what I asked originally in this thread some time ago, have you since found out if the later model nissan ecu feed the AFM 0-5v signal to the a/d convertor directly or is it actually protected from overvoltage?
Since you are now stating a zener diode should be used even with the Z32 AFM?
Regards
Mick
As promised. I've found some time to map this out.
Also below is my meter vs reported out of the interface on an RB30 ECU. Seems a little out
That chip where the signal is going to on the Z32 diagram may be another opamp? Its an 8 pin chip
The R32 ecu doesnt have this though, appears to go into a seven leg resistor pack looking component
Meter vs Consult display
0.379 0.388
0.449 0.458
0.502 0.508
0.711 0.728
1.020 1.038
1.524 1.548
2.048 2.068
3.215 3.288
4.11 4.108
4.92 4.898
For comparison purposes, measuring the AFR output via consult on the Z32 ECU is almost identical to multimeter readings on the wire. It may have a more accurate A/D converter perhaps?
I would like to get hold of another late model Z31 ECU again sometime to see where those resistors that get changed go to on this opamp board
Also below is my meter vs reported out of the interface on an RB30 ECU. Seems a little out
That chip where the signal is going to on the Z32 diagram may be another opamp? Its an 8 pin chip
The R32 ecu doesnt have this though, appears to go into a seven leg resistor pack looking component
Meter vs Consult display
0.379 0.388
0.449 0.458
0.502 0.508
0.711 0.728
1.020 1.038
1.524 1.548
2.048 2.068
3.215 3.288
4.11 4.108
4.92 4.898
For comparison purposes, measuring the AFR output via consult on the Z32 ECU is almost identical to multimeter readings on the wire. It may have a more accurate A/D converter perhaps?
I would like to get hold of another late model Z31 ECU again sometime to see where those resistors that get changed go to on this opamp board
- Attachments
-
- z32_afm_to_AD_schematic.JPG
- Z32 ECU schematic
- (21.11 KiB) Downloaded 3044 times
Last edited by Matt on Mon Jul 07, 2008 11:08 am, edited 1 time in total.
measuring the ECU back in its original configuration
0.443 0.420
1.033 1.020
2.234 2.280
2.945 2.940
this is the max i can reach with my 5 volt input on the bench before it gets dropped
there seems to almost be about the same amount of variance with the original circuit in place using the opamp
the modified circuit is missing a filter cap and the electrolytic cap after the modification
the original prototype VLT ecu had a space for a tant cap after the opamp daughterboard, but this doesnt exist on production models. you could put a tant cap on the same lines as the diode for filtering. it may smooth out your afm signal a little
there is definately a voltage discrepency of sorts about 0.02 volts - 0.1 volts when comparing the A/D input to the ECU determiend A/D output
One tuner has indicated to me that they have had to adjust the low end of the VQ map a little to accodate for it running leaner than it should to get the idle right
0.443 0.420
1.033 1.020
2.234 2.280
2.945 2.940
this is the max i can reach with my 5 volt input on the bench before it gets dropped
there seems to almost be about the same amount of variance with the original circuit in place using the opamp
the modified circuit is missing a filter cap and the electrolytic cap after the modification
the original prototype VLT ecu had a space for a tant cap after the opamp daughterboard, but this doesnt exist on production models. you could put a tant cap on the same lines as the diode for filtering. it may smooth out your afm signal a little
there is definately a voltage discrepency of sorts about 0.02 volts - 0.1 volts when comparing the A/D input to the ECU determiend A/D output
One tuner has indicated to me that they have had to adjust the low end of the VQ map a little to accodate for it running leaner than it should to get the idle right
- Attachments
-
- r31_afm_input.JPG
- R31/Z31 VG30/RB30 AFM circuit
- (27.35 KiB) Downloaded 3037 times
Its installed now. I'll post progress in the main thread
http://www.ibconceptions.com/nistune/po ... eply&t=606
http://www.ibconceptions.com/nistune/po ... eply&t=606