I don't have any specific miata experience, but have you checked the grounds? A couple bucks or some left over battery cables will give you a good solid ground to the engine.
I've also had luck tracking down codes by having a buddy watch the live data on a scan tool and wiggling and banging on sensors and wiring under the hood.
I have yet to see an OBD II car that really needed to have a PCM replaced, I'm sure that it does happen (components on board failing such as leaky capacitors), I've just never seen it in a couple years of dealership work (although that was almost 10 years ago). Have they tried reflashing the PCM?
Keith
SuperDork
9/28/10 6:33 p.m.
Not an option on these cars. Not in the car, anyhow.
Keith wrote:
Here's what the manual says.
P0300 (random misfire)
- CKP sensor (crank position, I think)
- CMP sensor (camshaft position)
- ignition coil
- plug wires
- MAF sensor contaminated
- "excess air suction in intake-air system" between MAF and dynamic chamber
- fuel pressure
- fuel line clogged
- fuel filter clogged
- fuel leak
- purge control solenoid valve
- PCV valve
- EGR valve
- vacuum hoses
- connectors, terminals, wiring, etc.
There are 22 steps to checking this out, the last of which is "replace PCM". That's always the last step Compression check comes before that.
P0420 is usually a result of the misfires. Sadly, that unburnt fuel likes to kill cats. Here's what the manual says.
- cat deterioration
- exhaust leak
- loose O2 sensor
- front O2 sensor malfunction
I'd dig into that P0300 cause, checking fuel pressure and a few other bits. Leave the P0420 until after, although it's possible the cat is dead by now. A Mazda dealer should be able to plug their magic tool in and check the condition of all the sensor inputs. If there's one out of range (such as the cam position), then fix that. As I mentioned, the trigger wheel has been known to come adrift.
(edited to overcome the retarded formatting of the GRM forum)
Just so we're clear, are you suggesting I take the car to the Mazda dealer, have them diagnosis something, and go from there. Because I am not nearly talented enough to diagnosis hardly anything on that list without help.
Keith
SuperDork
9/28/10 9:35 p.m.
If you're going to pay someone else to do the work for you, then yes. A Mazda dealer with the right tools and the manuals would be a good place to start. And make sure they follow all the steps before skipping to the end and saying "buy a new PCM". Because one of the steps is to hook up a tool and make sure all the sensors are giving legit readings. Stay on top of them. If you can't do the work yourself, make sure they do the work. If they're following all the steps, they'll be generating data. Ask to see it.
I gave that list partly to show how many things could cause your error codes besides a PCM. There's been no mention of a fuel pressure check in this thread, but Mazda feels it's important. It could also be as simple as an air leak in an intake tube, although I'd expect it to run rich if that were the case. I'd hate to see you drop $1300 on a PCM just to find out it was a broken wire on a cam sensor.
Page one of the Mazda Bulletin
Page two
And a list of what they checked for at the shop.
Keith
SuperDork
9/29/10 1:50 p.m.
So this is the first time you've mentioned P0012.
So, they started the car and saw oil pressure on the gauge. Then they verified the function of the oil pressure controlled valve and control unit both on and off the car and drove it 20 miles in under an hour. Quick work on the testing! I don't have my manuals handy but I'd be interested to know just how to test those parts on and off the car.
My concern here? The bulletin mentions the "PCM logic" is incorrect. Why would that suddenly change on a 9-year-old car? Personally, I'd confirm that trigger wheel on the cam first, as I've seen one of those come loose and that'll really hose a PCM's thought process. But if you tell the Mazda dealership that, they'll probably just look at you like you have a loose gear yourself.
Let me have a look at this tomorrow in the manual.
Keith wrote:
My concern here? The bulletin mentions the "PCM logic" is incorrect. Why would that suddenly change on a 9-year-old car?
I just wanted to quote that. I can't tell you how many times we've been called to tell us that the "calibration broke" when it was some part. Yea, some bit in YOUR module changed so that one cylinder is now misfiring... sure....
ok- sorry that this didn't help here, but Keith just made me laugh.
What can help- are you really getting a P0012 code? If not, then it's not cam/ECU. I agree with keith that they are just throwing ideas out, and not going through the process that Mazda is telling them to do.
I agree with Keith! You'd kind of be stupid not to, anyway, but I seriously doubt the PCM is junk.
Keith
SuperDork
9/29/10 2:49 p.m.
The last scan from the repair shop does identify a P0012 code as well as the p0300 (which I can see coming as a result) and a P0420 (which is the result of the misfires reported in P0300). I'll check that P0012 tomorrow.
Keith wrote:
So this is the first time you've mentioned P0012.
It's the first I've heard of a P0012 code. From the previous shop visits and me going to Autozone I only ever saw the P0300 and P0420 codes. So hopefully this is a good new development.
Keith
SuperDork
9/30/10 5:04 p.m.
Okay, a P0012 code. Looks like the shop followed the diagnosis procedure correctly - almost. One of the possible causes is an incorrectly timed timing belt. Before checking the function of the oil control valves, the shop should have pulled the valve cover off and confirmed the belt timing.
If the car had a recent timing belt change, I'd zero right in on that.
Aren't those timing belts due at around 60k?
Ok, so I took the car to the Mazda dealership today. The P0012 and P0300 code are now gone. They couldn't find them, at all. WTF.
So Mazda says I just need a new cat (keep in mind this one is two months old) and I told them that. They said, "Hm, weird. So should we order parts?". I told them no. Mainly cause they want $1600 to do it! Apparently the last shop didn't use an OEM piece like they said they would. But I'm not paying that kind of cash for stock. I figure $1600 should be able to get me something a little better.
But it still doesn't explain why the cat went bad in the first place. Help?
If misfires were real, it would kill a catalyst in short order.
Here is my plugged nickel of auto mechanic knowledge in this situation....
All of the codes are related. The TSB, yea it applies, but really only if it has that one code present. One of the caveats of TSB's, it has to be exactly as described to apply. With the other codes present, that TSB doesn't apply until the MIL/CEL comes back on with only that code. As said earlier, I would look at the timing belt and gears for misalignment. If the belt is off, it leads to a misfire, which can lead to a misread O2 reading post-cat.
BUT. The aftermarket cats are just above passable JUNK emissions wise. Most of them don't flow properly as stockers, in my experience. So it is VERY EASY to get the P0420 code with a new AM cat.
Brian
Keith
SuperDork
10/14/10 4:33 p.m.
I'm with alfadriver - with the P0300 misfire code, a dead cat is fairly close behind. The raw fuel from the misfires kills the cat.
Keith wrote:
I'm with alfadriver - with the P0300 misfire code, a dead cat is fairly close behind. The raw fuel from the misfires kills the cat.
(it's not the raw fuel alone, it's the burning... .)
I tried to tell the mazda people that but they wouldn't listen to me. They just said they couldn't find anything else wrong. So I guess I'm rather stuck. I can't fix a problem I can't find (Nor can I hire someone too, too expensive). So I guess I'll just have to sort the P0420 code first, and see what happens.