2004 Mazda 3



August, 24, 2009 AT 4:47 PM

Electrical problem
2004 Mazda 3 4 cyl Two Wheel Drive Manual 58000 miles

Until my engine is warm my car operates perfectly. Once the car is warm and goes from open-loop mode to closed-loop mode the performance becomes inconsistent (hesitation and varying power output). I believe this is due to a sensor and/or wiring problem. I have read that many people have issues with the MAF sensor wiring but I have been unable to find any broken wires and I have chased down the wiring with a multimeter (also cleaned the MAF/IAT and did fuel injection system service). One day I cleaned the PCM wire harnesses and the car drove perfectly for almost two days but since then no amount of cleaning makes a difference.

Just recently, I have been getting the P0128 code which suggests the thermostat or ECT sensor is bad. There is a TSB out for this problem but I recently had my PCM updated and the dealer said this update would have solved the problem if the PCM was the cause. The temperature gauge in my car never moves a millimeter once it is warm so I do not think the thermostat is the problem. I have read that the P0128 code is based on a calculation that includes input from the MAF and IAT sensors which seem to be the more likely suspects. Does anyone have any advise before I start replacing parts?

I have read that this P0128 code does not let the Mazda3 get out of open-loop (start-up) mode. I am not sure if this statement was accurate but the author was confident that this trouble code would impact engine performance. My car has not behaved any differently since the CEL first came on about six weeks ago due to the P0128 code. My problems are intermittent, only happen when accelerating, go away for hours (sometimes day), but always come back. When my car is experiencing the problems the gas mileage is poor (21 vs 28 mpg) and there is excessive carbon build-up on the muffler tip. I am not sure if my problems are related to the P0128 code but I sure hope so.


23 Answers


Service Writer

August, 25, 2009 AT 5:00 AM

Treat the symptoms of the hesitation and the code separately. They may or may not be r4elated to each other.

Regarding the hesitation when warm: Check engine vacuum while cold and then see if it falls off as the engine has run for a while. There are two nipples on the intake manifold near the frame rail. You may have a compression problem. Check compression when hot warmed up. Look at the plugs while they are out and see if they are loaded with carbon.

The short block on this vehicle is warranted by Mazda for 8 years and 100,000 miles. Cats have been known to fail from time to time and can give this symptom, but generally a code po420 is thrown for this.

On the code po128: I would have thought recalibration would have been right, except you did that. The temp gauge may be incorrect and giving mis-leading info. Try to establish the temp through another source, such as a scanner or a heat temp gun pointed at the t-stat housing if possible. Check with the dealer on a thermostat and thermostat housing that has been updated.

Do not Run regular fuel rather than the required premium fuel to save money or potential carboning may cause this. ( I know you did some cleaning, but it is not always effective)



August, 25, 2009 AT 5:15 PM

Thank you for your response.

After letting my car sit for a couple hours with a fan on it I performed the vacuum test as described. The initial vacuum reading was 21"Hg. The vacuum pressure did not change once the car was warm but there was a significant drop in vacuum pressure once the A/C was turned on (from 21 to 17.5"Hg). Turn off the A/C and the vacuum pressure goes from 17.5 to 20. A few seconds later the cooling fan will cut off and the pressure will go from 20 to 21"Hg. Hopefully this is unusual and indicates a problem?

The dealer has performed a compression test and all cylinders are fine. They have also checked the fuel pressure twice and it passed both tests with flying colors. I recently checked the spark plugs and there is no unusual carbon buildup.

You are the expert, but I find it hard to believe that the temperature gauge is faulty. If the car is running hot and/or the thermostat is failing there are symptoms other than a flucuating temperature gauge and I have experienced none of them. The temperature gauge behavior is like clockwork. I have absolutely no reason to believe it is faulty (and I have been staring at it for the last few weeks).

My car does not take premium fuel. The owner's manual states 87 octance and that is the only fuel I have filled with. I am also careful about where I get my fuel and 99% of the time it is from BP. The fuel injection cleaning service was performed by the dealer, was comprehensive, but not necessarily effective as you mentioned. Having good fuel pressure does guarantee proper fuel delivery. One Mazda3 owner I read about solved his hesitation problem by removing the fuel injectors and backwashing them. The intake manifold was cleaned relatively thoroughly in fuel system service but I am quite sure there is excessive carbon buildup in the exhaust/emmission systems. I recently purchased this vehicle and I think the problem has been going on for quite some time. When the performance is inconsistent the gas mileage goes down and the carbon buildup on the muffler tip is far more rapid (like the car is operating in "safe mode").



August, 25, 2009 AT 6:20 PM

Do you have access to a scanner that will allow you to monitor the ECT and IAT readings? This will be necessary to follow the troubleshooting chart. As far as the vacuum readings, those all sound normal, vacuum will lessen with engine load (a/c), but your numbers sound pretty good. This may just be a bad ECT, working, but out of range a little which would effect th emixture and time to closed loop, creating the poor mileage and soot you spoke of


Service Writer

August, 25, 2009 AT 8:20 PM

No, I'm not an expert. I know more than some and less than others. Maybe less than you. Maybe not.
Just trying to help you diagnose and unturn stones.

Having good fuel pressure does guarantee proper fuel delivery is not true. Good fuel pressure does not gaurentee fuel volume.

I agree with Jack's update.



August, 26, 2009 AT 5:05 PM

Service Writer, I did not mean to offend you. I am quite certain you know more than me based solely on your response. I did not think to check the vacuum pressure and that was a great suggestion. If injectors are clogged fuel is not being properly " delivered" to the cylinders (the common industry term is " volume" but both descriptions are accurate - " delivery" is flow - flow is quantified volumetrically - flow requires pressure - pressure does not guarantee flow).

I also agree with Jack's update. I have thought all along that a sensor was going bad but still operating within a range considered acceptable by the PCM. When I started messing around with the wiring and saw temporary improvements I began to think it was a sensor wiring problem. I guess is still could be but the fingers are now pointing at the ECT sensor. I will check the ECT sensor for continuity and open/short circuit and post the findings.

Thanks for all the help!


Service Writer

August, 27, 2009 AT 6:12 AM

LOL. It doesn't matter who knows more. I have often seen a group of very qualified techs that have a different view of a problem. Testing can narrow it down but it can still come to theory and an educated guess.

Very well may be a wiring issue. Troubleshooting isn't always black and white. There are conditions that should set codes and don't and conditions that you would not suspect being related to a particular code existing.

I have seen this scenario once before and after replacing the ect and t-stat, it ended up being the reprogramming. I do believe there was an updated stat and housing that was related to this, but again I would check with the dealer if the ect doesn't pan out.



August, 27, 2009 AT 8:41 PM

I do not have time to remove the ECT until this weekend but I just disconnect the plug, cleaned the terminals, and tested for an open circuit using the instructions below. Continuity between ECT sensor A and PCM terminal 2J was good. There was no continutiy between ECT sensor B and PCM terminal 2AX. Also, the check engine light went off after these efforts and I did not disconnect the battery. I thought the ECU waited for 3 consecutive good drive cycles before turning off the MIL.

Does the MIL going out immediately after messing with the ECT suggest anything definitive?

The manual recommends to "repair or replace the wire harness". Not exactly detailed instructions. Any suggestions?



August, 31, 2009 AT 4:58 PM

I didn't have time this weekend to pull the ECT and check for continuity and after the failed open circuit test I am not sure what to think.

The MIL came back on today, the performance consistency went from acceptable back to horrible, and I'm sure it is the same code. I have performed one test that indicates a wire harness problem. If there are any Mazda3 owners out there please perform the ECT open circuit test included in my last post and respond.

My research suggests a bad thermostat or improper PCM configuration. The TSB for the P0128 code describes the PCM fix process and I am not sure if the PCM update performed by my dealer would have specifically addressed this problem. However, the latest release of this TSB (01-041/07) suggests this problem is related to cold weather so it might be completely unrelated. Can anyone confirm that updating my PCM with the latest software would have accomplished the steps outlined in this TSB?

I am leaning towards replacing the thermostat. I have read that a stuck open thermostat causes a car to run rich resulting in poor gas mileage which I have been experiencing. I have never had a thermostat fail open on me. I would not think this would cause overheating which is probably why I have not noticed anything unusual from my temperature gauge. Has anyone had a thermostat fail open before? Are there any symptoms other than running rich (poor gas mileage) and MIL illumination?


Service Writer

August, 31, 2009 AT 7:05 PM

A stuck open thermostat does happen. It causes the engine to take much longer to warm up as there is more volume to heat up. The longer it takes, the more fuel is being used. The Rich condition may cause a performance issue that very well be related. There may be an upgraded t-stat and housing possiblility on this one. A t-stat can stick both ways. A gauge may not read accuratley either.

The TSB does indicate cold temp conditions and since it has been re-programmed, it is not relevant anymore.



September, 8, 2009 AT 5:01 PM

Well, I replaced the thermostat and went through a tank a fuel. The check engine light has not come back on but replacing the thermostat did not solve all of my problems. The hesitation/stumbling rarely happens now with the A/C off and only occassionaly with the A/C on. The weather has gotten a lot cooler lately and I think that has something to do with it. I hope we get some more 95-degree days before fall completely sets in.

The performance (acceleration) did not improve. Before the thermostat change my car would alternate between good, okay, and poor acceleration. Now, the behavior is more consistent but, unfortunately, it is between okay and poor. I know this engine has a lot more to offer - especially considering the bad fuel economy. I got only 20 mpg on the last tank which is the worst since I started keeping track. When the engine performance is good I get around 28 mpg. Something is still restricting the engine power output and at the same time hurting the fuel efficiency.

I think I have a lemon on my hands. I just got out of a Prelude with 200HP that got better gas mileage and I wore out the carpet underneath that gas pedal. I still think some sensor is causing the computer to screw up my A/F ratio. If I can't solve this problem and 21 mpg is a permanent condition I would at least like to enjoy the power I feel for the first couple minutes after start-up. That is the only time you would ever know this engine has 160HP. Is there a way to trick the computer into operating in start-up mode permanently?

I'm getting frustrated.

Please login or register to post a reply.

Crankshaft Position Sensor Replace Chevrolet
Fuel Pump Replacement
Oxygen Sensor Replacement Honda Civic 1996-20
Catalytic Converter Replacement Volkswagen
Fuel Pump Replacement Done Right