JMS-Four

Software?

I'm using the spacers with Revo. No codes and passed WA state emmissions twice thus far.

Revo, I believe, eliminates codes that has to do with pre cats.

hans j

The diode won't work because of the resistance built into the diode. It should be below .5K ohm not to throw the CEL. I tried it too and found the resistance was never low enough but the voltages were correct. Just get some spark plug anti foulers (i had to stack two). That worked the best for me.

complacentsee

Well if this is the case, than there should be an easy fix.

I will try to keep the jargon at a minimum.
In forward bias mode, the case where your getting the 450mV reading, the diode is dropping the voltage by 750mV. (1.2V-0.45V=0.75V) At 750mV, based on a standard diode spec sheet for the 1n4001-4007 rec. diodes, the forward current will be about 0.1A. (this seams a bit high but lets run with it, after all the curve should be pretty close since the current is on a log scale) At this current level, the diode will at 7.5 Ohms of resistance. Lets assume this current is off by a factor of ten, the we would have 75 Ohms, still probabaly fine.

This is most likly telling us that the ECU applies a voltage inbetween readings(or some other time like startup while its heating) to determine the resistance. In this senario the diode method will fail because it will be switch into its reversed bias mode and become an "infinite resistance."

To counter act this you could add a parallel resistance (might work would have to run some math) to the diode, or add a second diode in the oposite direction (preferably one with a low turn on voltage). This should allow a secondary path such that the ECU could still check the resistance.

S4ilorJerry's

So you are saying to run 8 total diodes, one with silver to the ECM side then one with the silver to the sensor side after that? I have the same software and defoulers but still have a CEL. I was going to Radioshack tomorrow morning to do this, but I want to figure this out first.

complacentsee

It's a real bitch trying to figure out one that would work for this application. Unfortunatly since your keep flying around the .9V to .2V range, according to what I see on my car. You'll need to come up with some fancy diodes or use some other circuitry to come up with a true fix for the problem. If I have some free time at work tommorow I'll take a look/ try and design a little something. But my guess if Vast is using defoulers that it probabaly can't be done very easily. Consitterin Mike has his masters in electrical engineering.

complacentsee

This is my best guess to a solution for this without going nuts and just conditioning the signal. Remember, you only modify the secondary O2 sensors. I refuse to make any promises since I don't have any tools besides vag to see whats happening with the system.

(1) : use the diode and across the diode add a 100-200 Ohm resistance. Might work....

(2) : make a voltage divider, use 2 100k ohm resistors. Picture below. It's not my fault if your car lights on fire or some ****.

S4ilorJerry's

Hmmm, very tempting.... I would love test this but my wallet can't handle another car catastrophe...

complacentsee

I'll test it one day I'm sure. My idea shouldn't do any thing harmful. Worst case it would cause early failure of an O2 that isn't doing anything anyway.

perpetualmotion

Using a diode to do this strikes me as very odd. I haven't thought through it fully, because I can't stand the style of writing used in that tech article. But assuming that I correctly understand what it is trying to accomplish (which is a lot to assume), the approach that I would take is that of a simple, standard voltage divider, using a potentiometer. To do this, you select a potentiometer have a total resistance several times greater than the input resistance of the circuit in the ECU. You wire the potentiometer in parallel with the input to the ECU, and you wire the ECU to the sliding contact of the potentiometer. When the sliding contact is fully at one end, the full resistance of the potentiometer is in parallel with the load. The series resistance then determines the voltage drop across both the potentiometer and the ECU circuit, and since that series resistance will be somewhat less than the resistance of the ECU by itself, the voltage drop will be less. When the sliding contact is fully at the other end, the full resistance of the potentiometer is in series with the load, and if that resistance is appreciable greater than the resistance of the load (the resistance of the input circuit of the ECU), the appreciable portion of the voltage drop will occur across the potentiometer, thus lowering the voltage drop across the load. This is simple the standard way of using a variable voltage divider to do something of this sort. Were talking very low current, so a small log-slope audio grade potentiometer should work perfectly. Use your voltmeter to the read the resistance across the input circuit at the ECU, and adjust the potentiometer as needed to get the desired voltage.

As for the "resistance too high" code, I was probably wrong, but off the top of my head I thought that this was the resistance of the heating element within the sensor, not the sensor itself.

complacentsee

Yeah, I drew up a schematic of this. And posted it. Seems people feared breaking stuff. I personally would say that you want a Pot with a total resistance around 1 order of magnitude below the input resistance of the ECU. This way you still have your leverage. And will have a more linear feel as the ECU is not loading your voltage divider. But there is an argument to be made for not increasing the load on your O2 sensor as well.


https://www.audiworld.com/forums/sho...php?p=23785824