49m

I have been chasing this problem for a while. I keep this car in a heated garage and it starts fine with no codes at first, however after driving down the road with temps around freezing or below it will set a code 33 most of the time. It then runs like crap on default ecm settings, sometimes I can turn the car off and restart it and the code will dissapper and run fine, other times it will continually set code 33 and I have to power the ecm down to clear the codes. I have replaced the maf relay thinking that the resistance changing of the coil due to temp could have a affect on the voltage,did not help, the tps is calibrated fine and the iac is fine, checked voltages on a11 when motor is up to temp and running around 800mv at idle, I stuck a fluke meter on it and put in record min/max mode scan rate at 100ms and got a 3.80v max on a 10deg evening with a cold start test. after runnig a few minutes voltage dropped to around 1.2volts, pulled it inside a heated building and it settled down to around 800mv. I have done a resistance test on b12, a11,d1 ect to ecm and found only .10 drop from the maf connector to the ecm. I have also checked 993 for any voltage drop found nothing wrong. I have another camaro that is far from stock and tried swapping ecms and it did the same thing with it, the car did not like the custom chip I run in it so I removed the custom chip adaptor and let it run on the stock chip and it run fine but still did the same thing on code 33. I have a 2 wire bosch in the 86 hot rod, and the 87 has the new solid state maf. This is really giving me fits. Has anyone broken 998 circuit and tried adding some resistance to drop the voltage lower? What is the voltage divider circuit in the ecm consist of?
Any info will be appreciated.

89irocMike

I have a microtech maf in my car that did similiar to yours. The block learn is real low trying to take fuel out. It would set codes when the temp was low. I exchanged it for a new one that wasn't much better. I put a gm maf out of my brothers car and it straightened out. Block learn hovered around 128 right where it should be. I am going to replace the microtech unit with a gm.

49m

I going swap my 2 wire bosch maf that I run in my hot rod and see what happens. What do you think the circuit in the micro tech is doing to cause this. I know it has a transistor and a film strip I assume is a varaible resistor? not really sure, there may be a way to change the gain of the transistor threshold voltage to improve its response. Thanks again for your info.49m

jms

iTrader: (1)

What temp is your thermostat set to open?

I had similar problems while using a 170 degree stat. Went up to a 190 degree and no more problems.

jms

Vader

Once the MAF power relay is closed, coil resistance should have no effect on the voltage supplied through the contacts. Still, it’s always a good idea to verify the relays in the MAF circuit.

A11? Is this for a 1227165 ECM? According to the schematics I have available, pin A11 should be a ground.

800mV is very typical for an idle output reading. 3.8V is not. I suspect either the ambient sensor and/or differential sensor in the MAF is not calibrated to produce a linear signal under the varying conditions. As I understand it, the MicroTech MAF has a lifetime guaranty. It may be time to exercise that. I have only used one of those, and have not had feedback regarding cold weather operation. The Florida MAF may not have been engineered and tested under the full range of operating conditions. The formerly available Wells SU-145 (Wisconsin MAF) seems to tolerate the normal temperature variations without exhibiting this kind of problem. The vehicle was designed to operate normally down to -40º, and the MAF should be expected to do the same. As long as the signal ground is intact and the voltage is not allowed to be pulled up by the ECM input biasing, the only thing driving the voltage in the circuit is the output transistor in the MAF. Unless there is an intermittent connection in the signal ground between the MAF and ECM which is being affected by temperature change, the MAF output voltage should not float up.


A tenth of an ohm is insignificant given the few milliamps of current for any of the signals. Any voltage drop due to that should not be a problem. The only circuit with any kind of load is the 12V supply, and it experiences the greatest load during the burn-off cycle. Given that the TFT type MAFs simply use a dummy load to simulate burn-off, current is even lower.


You have all but proven that the ECM is not the problem. For the remainder of the diagnosis, I would suggest you mount the ECM in its normal place and connect directly at the MAF for signal monitoring.


Adding resistance to the circuit to address the ‘33' would be only a Band-Aid fix which would likely cause ‘34' error under other operating conditions. The root cause is not the ECM, not the power supply relays or copper, but the device generating the analog voltage.

The MAF signal conditioning and handling is indicated in the attached diagram.

49m

Thank you very much for the info.
You said the wells unit worked fine. Are you indicating that there are two micro tech suppliers one from wisconsin and another from florida? If not I will exchange the micro tech for the third time and see what happens. The old 2 wire bosch is working perfectly. Thanks again 49m.