P2138, won’t rev above 3000 rpms, lots of smoke

[jsilva]

I was thinking that if PCM 2H and APP B are supposed to be grounded but aren’t, then I could conceivably connect 2H to ground manually and see if it makes any difference Maybe I’ll try that out tomorrow if it’s not raining too hard.

Perhaps I damaged a ground wire when I was doing the fuel injector swap and PVC replace. So I’ll also test all of the grounds I see in that region.

 

[jsilva]

An update...

I tested the continuity between the throttle body wiring harness and the corresponding pins on the PCM and everything is fine.

So far I can be sure it’s not the APP, the APP wiring harness to the PCM, and the wiring from the PCM to the throttle body. I’m fairly sure it’s not the throttle body. The only mysterious thing is the lack of ground on PCM pin 2H.

 

[jsilva]

Today I tested the ground wire on the fuel pressure sensor and there was continuity. I found this interesting because on the wiring diagram that wire is ultimately connected to sensor ground, pin 2H, on the PCM.

 

[Ssinstaller]

What is the impedance betwen pin b at the pedal and pin 2h at the pcm?

If the pcm is plugged in you should have continuity to ground on pin B, if the pcm is unplugged there should not be continity.

There is another connector between the pedal and the pcm it's c-03 in the attached pic, you might check that connector as well.

If all else fails you could just run a jumper wire and directly connect pin b to 2h.

 

[jsilva]

Thanks Ssinstaller. It’s very likely I tested 2H when it was unplugged from the PCM. I don’t remember for sure though. I can test it again with it plugged in. If there is ground is it still worthwhile to test that other connector?

 

[Ssinstaller]

Thanks Ssinstaller. It’s very likely I tested 2H when it was unplugged from the PCM. I don’t remember for sure though. I can test it again with it plugged in. If there is ground is it still worthwhile to test that other connector?

If there is ground with it plugged in, check the impedance between the two pins. If it's high that will cause issues as well.

 

[jsilva]

Ok, I tested for ground at 2H and it’s fine. I must have done something wrong previously.

I appreciate the help Ssinstaller and I’m quite happy to do whatever testing you suggest (especially since I have zero other leads...!), but this issue happened right after I was doing stuff in the engine bay. I mean literally when I tried starting the car after putting things back together.

I started another thread about having starting issues since I assumed it wasn’t related to this issue. Could it possibly be related to this? After replacing the PCV valve (again) and putting the previous fuel injector wiring harness back on (I tested all of the pins for continuity first) now the car won’t turn over. It’ll crank, but nothing more than that. Pretty much exactly like when there is no fuel. Not sure if that could be related to this. No error codes.

 

[Ssinstaller]

Ground wire at the hpfp bracket?

 

[jsilva]

Yes the lack of turning over was that I haven’t the slightest idea what’s wrong with it—I had tested from one of the ignition coil connector pins to that ground wire, and from that ground wire to the negative battery terminal, and everything was fine.

But I disconnected the ground wire and ran another wire from it directly the the negative battery terminal and the engine was trying to turn over. It didn’t succeed but it almost did and things once again seem exactly like the P2138 problem that started this thread, just with the inability to fully turn over.

So I’m back to this issue I suppose...

 

[The_big_dill]

Yes the lack of turning over was that I haven’t the slightest idea what’s wrong with it—I had tested from one of the ignition coil connector pins to that ground wire, and from that ground wire to the negative battery terminal, and everything was fine.

Doing a continuity check tells you that there is a circuit, it doesn't tell you what load that circuit can handle. This is likely what you are experiencing with that ground.

Similar to a loose battery terminal, there may be enough contact to turn on your interior lights, but as soon as you demand a lot of current (starter motor), it just dies.

Therefore a continuity test in these scenarios can be deceiving and is why I just recommended running a ground wire directly to your battery.

 

[jsilva]

Just to quickly get you up to speed on this thread big dill, the car would smoke (which smelled a lot like fuel) and if I revved it to 3000 rpms it’d die and throw the P2138 (and flash the traction control light). It would sometimes have issues starting but would eventually catch. I think that’s where I am back to now but that it’s just not fully catching anymore.

Ssinstaller was giving suggestions on where to test based on the service manual. The only thing I turned up was that the APP doesn’t increase the voltage on either PCM 2A or 3D (I can’t remember which one). The manual says to replace the throttle body with this condition, but I did swap it out with another one and no improvement. And Ssinstaller doesn’t think that’s what the manual is meaning to say anyway. I replaced the accelerator pedal as well and there was no change.

That’s where this issue is currently.

 

[The_big_dill]

Just to quickly get you up to speed on this thread big dill, the car would smoke (which smelled a lot like fuel) and if I revved it to 3000 rpms it’d die and throw the P2138 (and flash the traction control light). It would sometimes have issues starting but would eventually catch. I think that’s where I am back to now but that it’s just not fully catching anymore.

Ssinstaller was giving suggestions on where to test based on the service manual. The only thing I turned up was that the APP doesn’t increase the voltage on either PCM 2A or 3D (I can’t remember which one). The manual says to replace the throttle body with this condition, but I did swap it out with another one and no improvement. And Ssinstaller doesn’t think that’s what the manual is meaning to say anyway. I replaced the accelerator pedal as well and there was no change.

That’s where this issue is currently.

I just read your whole thread and I admire your perseverance.

I do think its highly coincidental that your issues appeared after messing with that area. Along with the fact that the pedal position sensor is not a common issue, I say you focus under the hood.

The throttle bodies seem to be less reliable from my experience. I replaced mine 2 times diagnosing an issue and the replacements had more problems... moral of story: be suspicious of used parts.

On that note, I do not see how you would have damaged the throttle body if you handled it carefully. I still believe that you are facing wiring troubles as those are more prone to damage when working in that area.

I have personally seen pins that are oversized, misaligned, bent, not retaining within the connector cavity. All of which can cause intermittent functionality. I challenge you to criticize the pins going to your throttle body. pull back the corrugated tubing, move the wires around while doing continuity testing, etc.

Another grain of salt for you: the code being thrown could be the ECUs best guess as to whats going on. And although it is usually the best lead, it could also be deceiving. In addition, when you swap parts or are doing testing, do not forget to reset the ECU (preferably via battery). It is possible that the ECU may have failsafe functions when it detects problems, and if you didn't reset, it could be giving you false information on part swaps.

 

[The_big_dill]

(Ignore/remove, this is a double post) AKA my internet sucks.

 

[jsilva]

Thanks for your thoughts Big Dill! It has been a frustrating couple of months, but I’m determined to get the Speed6 running. When it was running I found it to be a very fun car to drive.

How do I reset the ECU?

I will examine the connectors as you have suggested.

I was considering trying to start the engine with the intercooler off so that I could get a look at the throttle body. I’m not sure what it’s supposed to be doing during a start, but the exhaust was very fuel rich and so I was assuming there wasn’t enough air. So perhaps if I slightly push open the throttle body manually (assuming it’s not open) maybe I’ll see if that helps.

I was also considering taking the two throttle bodies I have to the Mazda dealership and see if they can test them. Not sure if they will/can do that though.

 

[The_big_dill]

Thanks for your thoughts Big Dill! It has been a frustrating couple of months, but I’m determined to get the Speed6 running. When it was running I found it to be a very fun car to drive.

How do I reset the ECU?

I will examine the connectors as you have suggested.

I was considering trying to start the engine with the intercooler off so that I could get a look at the throttle body. I’m not sure what it’s supposed to be doing during a start, but the exhaust was very fuel rich and so I was assuming there wasn’t enough air. So perhaps if I slightly push open the throttle body manually (assuming it’s not open) maybe I’ll see if that helps.

I was also considering taking the two throttle bodies I have to the Mazda dealership and see if they can test them. Not sure if they will/can do that though.

Just pull the battery for some time to reset the controllers on your car.

(On the Mz3/Speed3s) You can watch the butterfly valve in the TB cycle open and closed when you put the key into the second ignition position (before starting). I believe this is just a test/recalibration the ecu does regularly or preventive procedure. After you shut down the car, you can actually hear it cycle after some time. So I don't think you need to start the car to test if the ECU is functioning the TB.

Best bet is to find someone local who has spare parts or is willing to pull theirs. But I don't think this is your issue as the codes exhibited are not common to the way these normally fail. In fact, the most common TB failure allows your car to start without a problem, it just doesn't respond to throttle input and throws a different code specifically for the TB.

One other thing to mention is that you may have introduced new problems in the process. You mentioned smoke and fuel smell multiple times. Your spark plugs may have been fouled by now. I would still pull all the plugs and see if you have any spark. I just let them sit in the ignition coils with the threads touching the valve cover and watch them while someone cranks. This is how I diagnosed my ground issue.

 

[jsilva]

Just pull the battery for some time to reset the controllers on your car.

(On the Mz3/Speed3s) You can watch the butterfly valve in the TB cycle open and closed when you put the key into the second ignition position (before starting). I believe this is just a test/recalibration the ecu does regularly or preventive procedure. After you shut down the car, you can actually hear it cycle after some time. So I don't think you need to start the car to test if the ECU is functioning the TB.

Best bet is to find someone local who has spare parts or is willing to pull theirs. But I don't think this is your issue as the codes exhibited are not common to the way these normally fail. In fact, the most common TB failure allows your car to start without a problem, it just doesn't respond to throttle input and throws a different code specifically for the TB.

One other thing to mention is that you may have introduced new problems in the process. You mentioned smoke and fuel smell multiple times. Your spark plugs may have been fouled by now. I would still pull all the plugs and see if you have any spark. I just let them sit in the ignition coils with the threads touching the valve cover and watch them while someone cranks. This is how I diagnosed my ground issue.

Thanks for the tips on the TB! It’s helpful to have some idea of what it does. I’ll give that a look.

I did pull #1 plug and it was pretty black. It wasn’t a thick coating or anything and I’m not sure if it was bad enough to not spark but it doesn’t look clean. I can clean them up a bit. I’d prefer not to change them before I solve this problem lest I foul up a new set of expensive plugs.

To explain about my testing before. I’m not sure my valve cover is metal. Seems sort of like plastic. Maybe it is metal. I had tried what you said the other day but there was no spark. So I ended up running a wire from the nearby ground wire to the spark plug. Still no spark. I just assumed I was making a mistake, and when it was trying to turn over I assumed the new ground wire to the negative battery terminal solved that issue. I can give that another go when I clean up the plugs. And I’ll test them all.

 

[The_big_dill]

Thanks for the tips on the TB! It’s helpful to have some idea of what it does. I’ll give that a look.

I did pull #1 plug and it was pretty black. It wasn’t a thick coating or anything and I’m not sure if it was bad enough to not spark but it doesn’t look clean. I can clean them up a bit. I’d prefer not to change them before I solve this problem lest I foul up a new set of expensive plugs.

To explain about my testing before. I’m not sure my valve cover is metal. Seems sort of like plastic. Maybe it is metal. I had tried what you said the other day but there was no spark. So I ended up running a wire from the nearby ground wire to the spark plug. Still no spark. I just assumed I was making a mistake, and when it was trying to turn over I assumed the new ground wire to the negative battery terminal solved that issue. I can give that another go when I clean up the plugs. And I’ll test them all.

I don't recommend cleaning plugs because your efforts are more likely to damage the plug and can give you false diagnosis. Even fouled plugs should spark, may just be weak or in the wrong location. Since you are disposing the plugs eventually, you can clean them, just don't forget to replace them.

If your plugs dont fire, your engine can't start.

If you have the MZR-DISI 2.3 with the original valve cover, then it should be made of aluminum. I am even able to use my powder coated valve cover to do this. You can touch any piece of metal that isn't connected to battery + on the car. The valve cover is just easy and within reach.

You can also command clear flood mode by pushing the gas pedal to 100% and holding it there when you are cranking, that disables the fuel injectors. I use this mode to check spark, compression test and circulate oil after an oil change.

1. Take a picture of the set up you have when checking spark.
2. Take a picture of how you attached the ground to the coil ground near the HPFP and how you attached it to the battery.

 

[jsilva]

Ok, after cleaning the spark plugs I was able to get the car fully started today and it’s back to the original issue in this thread. So that’s a positive! Thanks for encouraging me to check out the plugs Big Dill. Some notes:

1. All four spark plugs were sparking in my test. (I did it the way Big Dill suggested, threads resting against the valve cover.)

2. The throttle body did open a tiny bit when the ignition was put into the position right before starting. Additionally, about 10-15 seconds after turning the ignition off the throttle body fully opened and shut. So it seems to be working...?

3. At this point the only definite clue is that terminal 2A on the PCM registers 1.3V and doesn’t change at all in response to the accelerator pedal depressing. This is what the service manual says:



There are no open/short circuits that I tested. So I’m without clue, again

 

[The_big_dill]

Ok, after cleaning the spark plugs I was able to get the car fully started today and it’s back to the original issue in this thread. So that’s a positive! Thanks for encouraging me to check out the plugs Big Dill. Some notes:

1. All four spark plugs were sparking in my test. (I did it the way Big Dill suggested, threads resting against the valve cover.)

2. The throttle body did open a tiny bit when the ignition was put into the position right before starting. Additionally, about 10-15 seconds after turning the ignition off the throttle body fully opened and shut. So it seems to be working...?

3. At this point the only definite clue is that terminal 2A on the PCM registers 1.3V and doesn’t change at all in response to the accelerator pedal depressing. This is what the service manual says:

There are no open/short circuits that I tested. So I’m without clue, again

2. That sounds the same as mine.

3. I am assuming you are 100% certain you are checking pin 2A, reading a specific side of the connector can be confusing at times, depends on how clear the manual is.

In any case, based on the information you have provided, the problem is in the wiring at the throttle body.

Have you tried to remove the corrugated tubing/insulation to make sure no wire is damaged there?

Have a VERY close inspection of the pins on the harness side throttle body connector. Move the wires while you are looking at it. Pull/push a little bit on each one. Compare the pin sizes to one another.

 

[jsilva]

2. That sounds the same as mine.

3. I am assuming you are 100% certain you are checking pin 2A, reading a specific side of the connector can be confusing at times, depends on how clear the manual is.

In any case, based on the information you have provided, the problem is in the wiring at the throttle body.

Have you tried to remove the corrugated tubing/insulation to make sure no wire is damaged there?

Have a VERY close inspection of the pins on the harness side throttle body connector. Move the wires while you are looking at it. Pull/push a little bit on each one. Compare the pin sizes to one another.

I’ve tried to be conscious of what pin is what. 2A should be the bottom left while laying on my back... So I’m 99% certain I got it right. But I’ll check again tomorrow.

All I can say about the throttle body wiring right now is that I checked continuity of all the pins on the connector to the corresponding pins on the PCM and everything was fine.

I’ll examine the wiring harness as you suggested.