Vader

ROC-Z,

The code ‘33' is a higher than expected intake air flow as measured by the MAF. There are several possible contributing issues, but since you wanted the technical explanation, I'll lend a little background.

The MAF measures intake air mass by the measuring the temperature difference of two sensing elements in the sensor. The MAF has two elements - one that monitors incoming air temperature, and one that monitors the temperature of the hot wire or thick film heated element. The sensor electronics package measures the current necessary to maintain the heated element temperature exactly 75°C above the incoming air temperature. The amount of heater element current required to maintain this temperature is directly proportional to the amount of heat being removed from the element by the incoming air. Regardless of the temperature, density (altitude and barometric pressure), or moisture content of the air, the current is proportional to the actual mass of the air entering the engine. Since fuel is added by mass rather than volume, reading the mass of intake air allows the ECM to proportion fuel correctly to maintain the ideal mixture. So much for how the sensor works.

The ECM uses the MAF information to add fuel as described above. However, as a monitoring system verification of correct MAF readings, there are acceptable ranges for MAF readings under various circumstances. These allowable ranges are calculated by the ECM based on several pieces of input information. The conventions are that incoming air mass should be proportional to engine RPM, and that throttle angle will affect the amount of air available to the engine. The ECM measures RPM via the HEI module trigger input, and throttle angle via the TPS. A higher TPS voltage indicates that the engine should be moving more air at any given RPM. A higher RPM at any given throttle angle should also indicate that the engine would be moving more air. These calculations are compared to sample parameters stored in the ECM's EPROM/CalPak. If the calculated intake is outside the allowable range for more than 600mS (6/10ths of a second) the ECM will generate and store a fault code. Once a MAF error code is stored, the ECM will ignore the MAF input for that engine run cycle and use backup fuel and spark tables entered into the CalPak EPROM. These tables are typically very conservative for spark advance and very rich on fuel.

Back to your specific situation, it seems that your ECM is measuring a higher flow than is allowed at the indicated RPM and TPS voltage. Since the range is affected by TPS voltage, and a lower TPS would demand a lower allowable intake air, it is imperative that the TPS be set properly. 0.33V is not only well outside the factory allowable range, it is far too low for the ECM to calculate an idle mixture. The acceptable range is 0.47-0.61 VDC, with 0.54VDC being the mean, or desired setting. Set the TPS back to where it belongs. If you still experience code ‘33' DTCs, try setting the TPS to the high end of the allowable range, or around 0.60VDC. Changing the TPS upward will force the ECM to calculate a higher allowable range of intake air mass measurements, and may solve you problems.

Not to contradict what your mechanic has told you, but he is one mechanic, and there are well over 10,000 members on this one board who have had experience setting TPS at a wide range of voltages through either error or experimentation. It appears that the factory specifications are best for our engines when stock fuel systems and EPROMs are used.

And since you own a 1986 TPI car (like I do), I would be remiss if I didn't mention the problem with that particular year. That model was the first year with the Bosch analog MAF and new 1227165 ECM, so there was no experience other than a few factory test cars upon which to base these parameters. It seems that the engineers programmed allowable parameters and variables that were too tight for some TPI engines. While there was no recall, there is a factory TSB (# 86-6-GAS-66) regarding the problem. Several issues were addressed, including false error codes for the MAF, mixture, TCC lockup, rough idle, hesitation, etc. The problem was not addressed until October of 1986 - after the end of the model year. Thus ALL 1986 TPI engines are suspect. I installed the upgraded EPROM in my ‘86 and noticed an immediate improvement, even though I wasn't experiencing any of the listed symptoms nor error codes.

There was also a TSB (#86-6-GAS-57) from August of 1986 regarding the code ‘33' specifically. It seems that an improper relay was used for MAF power on some engines, and the specified repair was to install a new P/N 0067925 relay in place of the original. As with any MAF error codes, the first things to check are the connectors and relays, so I suspect you've already addressed that problem.

As for the rewiring of the MAF burn0off circuit, I don't understand exactly what you are describing or mean by "rewire", but if the MAF burn-off function is disabled on a stock Bosch MAF, the MAF hot wire will accumulate buildup over time and cause reading errors, lean mixtures, and error codes. Unless you have installed an aftermarket thick-film sensor like th eWells SU-145, the burn-off function must be operable. Even the Wells MAF has an internal ballast resistor to simulate the burn-off load so the ECM doesn't detect burn-off problems and set an error code.

My advice would be:

1. Reset the TPS to 0.54VDC as a starting point;

2. Set the base timing at the factory specification of 6° BTDC;

3. Set the minimum air position of the throttle then re-check the TPS voltage and adjust as necessary;

4. Verify that you have the proper MAF power relay in good condition;

5. Restore the MAF burn-off circuit to the factory design;

6. Disconnect the battery for at least a five minutes to clear all stored codes and BLM data tables. This is usually just long enough to clean the battery terminals and prevent other problems;

7. Order a replacement upgrade EPROM is your engine is still stock. They're about $53.00 from the dealer.

The clearing of all accumulated BLM tables and codes will allow the ECM to build new table based on the correct input information sensors, and allow you to retrieve only any recurring DTCs instead of those from 50 start/run cycles ago.

The correction of the TPS and MAF readings and subsequent use of the MAF by the ECM should prevent the ECM from using the backup fuel mode and should help the code ‘45' (rich OČ sensor readings) since the engine control system will adjust fuel as intended.

The correct setting of the TPS should eliminate your need to push the accelerator on startup, since all you were doing was raising the TPS to an acceptable level.

As for the MAF, it is not ignored in open-loop mode, but only when there is a MAF error code or at WOT.

I hope I addressed most of your questions. If not, please reply with a more specific question or any that I may have missed.

ROC-Z

Vader-

Hey, I greatly appreciate the long explanation.

It originally had codes: 33, 36, & 45 set.


I replaced the MAF sensor with a Borg-Warner unit about a year ago. It uses the film instead of the platinum filament wires like the orig. Bosch unit. The MAF is also "descreened"....
My mechanic believes the MAF has intermitten problems. He said, he's had it to "44 gm/s"...
Right now, the car will stumble and stumble if you try to drive it without letting it warm up. The idle is very harsh before it warms up also.
Once it warms up though, it idles and drives good- except right before the Check Engine Light comes on, it starts to stumble. After that, each time you quickly hit the gas it will briefly stumble.
The machanic said it appears the car back-fired into the MAF, and this may have caused damage to it.


#1) If the MAF is reading Air Mass (until the ECM sets 33), what is the possibilty it's damaged from a previous 'back-fire'??

#2) It's perfectly possible to have Code 33 stored, but the Check Engine Light not come on till later, right??

#3) Why does the Check Engine Light not come on till later? Is it because, the ECM is getting a reading from it at first?
If so, then that means the light would not come on till the MAF code '33' parameter is met, right?
Therefore the ECM will ignore the MAF only once the light comes on, right??


In order to correct 36 (Burn-off), new wiring had to be installed, due to a short somewhere in the burn-off circuit. I had the new wiring installed & that corrected 36. I should have been more specific when I said: "Rewire the Burn-off circuit". The car had some wiring issues when I bought it, so those prob's needed to be addressed with new wiring.


#4) If the wiring for code 36 was never corrected, would the ECM ignore the MAF? Or does only code 33 & 34 make the ECM ignore the MAF for each run time?


The idle would climb after being started, to ~2000 RPMs. I had set the TPS voltage sometime back before this incident, to .54 +/- .02. Instead of playing with it any longer, I brought the car to my mechanic. When I brought the car in, he said it was actually at .74. He reset the throttle blade angle & TPS voltage (.33, I believe). This fixed the high idle & code 45 (rich).


#5) What other affects does the TPS have on engine managment operation? Does it play into idle quality while in open-loop?


The upgrade issues with the MAF power and Burn-Off relays have been addressed with those latest relays.
The only other thing that hasn't, has been the 'updated' PROM. However, I'm not using the stock PROM. I have the Hypertechź Thermomaster currently in the ECM.


#6) Does this Hypertechź unit contain the updated CalPack, you mention?

#7) Once the ECM ignores the MAF, could using the "back-up fuel parameters" cause a code 45, since those parameters induce more fuel?


Once again, thanks for the help and time Vader!

Guest

The hypertech doesnt have the problems the factory prom does. There is actually no separate calpak in it. Its a piece of junk though, IMHO, but at least you dont have to replace it because of the TSB.

Most likely your code 33 is related to the TPS. Check to make sure its working properly, with a steady voltage increase from the base setting to max voltage. .33v should be fine, at least there it will be well out of a high voltage problem area. And funny as this may seem, check your spark plug wires and make sure they arent grounding to the block or y-pipe, especially the 4 on the driver side. This can screw all sorts of things up, including the TPS. Usually it will burn the TPS out though.

ROC-Z

Thanks for the reply....I'm hoping it's the TPS too & not the MAF itself....
Just curious, since the Hypertech chip doesn't contain a Calpak, what is the computer doing when it ignores the MAF? How does it know how much fuel to add?

Guest

I have no idea, it must be in the chip. There is only the prom chip on there, and nothing else besides some black goo. I took that stuff off, and there is nothing under it. I dont know what they did.

Vader

ROC,

I wasn't aware that you had an aftermarket EPROM. I have a HyperScrap paperwieght as well, but I've never hacked it open to see what the component load actually is as Max has. I suspect the backup parameters are stored in the main EPROM as well, and the overall device is wired to provide a section of ROM to the pins that would normally connect to the backup mode PROM (just a guess).

I removed the Thermomaster after I learned of the upgraded factory caibrator, and feel that the second factory cal is better than the Thermomaster. It's certainly an improvement over the original factory programming. However, in you situation, I doubt that the Hypertech is too tight as the original was. More likely, you have a TPS problem or the descreened MAF is allowing laminar airflow to fool the sensing element.

What I would do:

Reset the TPS to 0.54VDC;

If you can, reorient the MAF to sample a different portion of th eairstream or reinstall the screen so that intake air is less laminated along the duct walls.

Check the MAF to make sure it isn't damaged by debris and that the sensing element is clean.