Page Listing
Project Description
Looking to install a v3.2 Genboard onto a naturally aspirated Audi inline 5 cylinder engine (KV).
- Genboard v3.2
- 4x20 LCD
- sequential fuel injection (5 cyls)
- ignition via distributor (initially, progess to direct later)
- WBO2
- EGT
- MAP sensor on Genboard
- 60-2 crank sensor
- existing cam sensor in distributor
Questions
WB02 calibration
Following the procedure for calibration, and the voltage checks for the pins are giving me trouble. Docs specify 3.96v from GND to Pump -ve, and 4.41 from GND to Nernst, with 450mv from -ve to nernst.
Following are the measurements I have been getting. Is there any way to get the pump -ve down to 3.96 volts? Also looking for a little more feedback on the PID settings. What sort of time constant are we modelling for, and is there any other info that might be useful for tuning the PID loop?
{{pump testing
wbo2_nernstdc_target 134 160 160 160
wbo2_pump_pid_kp 64 64 64 64
wbo2_pump_pid_ki 44 44 44 124
wbo2_pump_pid_kd 8 8 0 0
wbo2_pump_pid_ilimit 132 132 132 132
gnd - pump- V 4.06 4.03 4.03 4.05
gnd - nernst V 4.3 4.36 4.39 4.5
nernst - pump- mV -240 -330 -360 -450
}}
Answered Questions
Crank sensor
I'm looking at using a VR sensor to pick up on a 60-2 toothed wheel which will be attached to the crank pulley.
- From what I can gather, it's best to count the gaps, which leads to a wheel with a 3 degree tooth followed by a 3 degree gap, all the way around the wheel. Then the -2 is achieved by not machining two of these gaps. Is this correct? It would be nice if you could check a few times during machining the missing tooth (before full depth is reached) with a notebook soundcard (and some temporary mount for the VR sensor). If this is not possible, examining a factory 60-2 to get a feeling of it should be the minimum.
If you have a space EC36 pin, it might be a good investment to have a separate pin for VR- that is connected to GND near the LM1815 chip (not through the EC36-pin26 GND). The idea is to keep signal seen by LM1815 very clean for best lowRPM (<200 RPM , small signal) behaviour. Although EC36pin26 GND is not as noisy (by far) as the GND5 decreasing it further might be a good investment if the free EC.. pin is available and ECM unassembled. EC.. is perfect, just sharing current paths might (which is common for GND) inject some noise (that is usually below problematic limits).
At this stage it looks like I'll need a number of extra connections through the endplate, so far:
- MAP pneumatic tube
- VR sensor
- PS2 keyboard
- LCD
Looks like it may be necessary to mount a number of these on the rear endplate with long cables so I can remove the board from the case if necessary.
MAP Sensor
- in manual the sensor is mounted on the BOTTOM side of the board
- but I have since seen another image which shows the sensor mounted on the TOP side of the board
I recommend mounting MAP on the top of the board. Better for getting the tube through the frontplate. Note that MPX4250AP pins must be bent to the other direction depending on the top/bottom (so that notched pin is close to AVR in any case)
Progress
- 29 May 2005 Connected PS2 keyboard. Can now change screens etc without use of a laptop.
- 23 May 2005 Have an LCD attached, got the LCD going, and added an inline fuse to the power supply.
- 18 May 2005 Got my "Hello>" back from the board as well as the test screen. My variable power supply only allows current limiting at 35mA or 1.2A, so will have to get a fuse before powering it up again.
- 16 May 2005 Recently received my v3.2. Currently in the process of soldering on the components required to get power onto the board.