01 Mazda Millenia Remaining Bugs

I’m sure you’re right about this video not being the right engine. The article/diagrams cover the 2.3 and the 2.5, but really, the take away was that the ECU controls the relays, rather than a sensor controlling the fans directly (on the Germans seem to do this with any regularity).
Irrelevant, now other than for academic exercise.
Good shot on that photo!
And congrats on:
noticing the overheat;
getting it working again.
Now, go drive the shyte out of it and report back!
:slight_smile:
:horns: :dms: :plasmacam_smiley:

1 Like

Well, I have started that process, and here is what I know. At first, there is residual information built up in the ECU memory, and it takes a couple 2-3 resets to get a true resest. And now I know why, as I located a very informative book about this car called the SERVICE HIGHLIGHTS for the car. I paid one Great Britain Pound to have a .pdf copy. This book is the most informative book I have seen when it comes to understanding the drive cycles required tor state inspection. It completely details what the ECU does to test EGR, catalyst, evap, etc. So the multi-reset actually makes sense as soon as you understand that the ECU looks to see several drive cycles of information in succession to throw flags, both good and bad.

So far, first startup resulted in Bank 1 Sensor 2 heater CEL illumination and code. Second start resulted in Bank 1 Sensor 2 fault code, but no CEL code. Improvement? I dunno yet, I took a nap, and will go out now to find out how attempt 3 goes. I think this is a product of improper cooling. I’ll report back soon, gentlemen.

Oh, but I like to think that the point illustrated by our banter could be interpreted to mean that the student is becoming the master. Or if we were in the famous tv series “Kung Fu”, I would have snatched the pebble from the hand!.. lol…

At the very least, I have learned a bit about this car, and am now on a good path for a strong finish. Thanks to all!!

1 Like

So I took the Milly for a spin and wound up on the rack at a little-known maker community called DMS.

There I elected to take temperature measurements both before and after the catalytic converters on both sides, L & R. I kept the car idling while I did this. The idea is that when a catalyst is properly functioning, it will be around 200 F prior to the catalyst, and significantly hotter after the catalyst, maybe 400 or even 590.

Well, I got measurements of 300 and 400 before the catalyst. This was very disconcerting. Also, I am concerned as to why my ECU is not flagging catalyst codes.

But there is one thing I have not told you which is important. And that is that for this car, due to the density of componentry under the hood, Mazda has used heat shields around the headers. The shields were in place, and so it was difficult for me to take my measurements. I elected to take the infared temperature measurements at the only location not covered by heat shield, the O2 sensors. But immediately, you see the problem. These are heated O2 sensors, where the tips have heaters on them hitting 500-600 degrees F, So I think that this has affected my measurement. I now plan to retake these measurements after I get the heat shields off. I did not want to mess with them while hot.

Also, the lift was instrumental in helping me locate my last oil leak. It has gotten pretty significant, and that also helped. The last remaining leak is coming from the cams on timing belt side in the rear near firewall. I will have to remove timing belt to fix. I plan to work on this tonight at home. And I’ll work to get better temp measurements on my catalysts as well.

I had to clear my codes a few times, as I explained previously why. But once I did this, and the older residual overheating data was out of the picture, I have seen two codes: one has to do with evap and I have only seen this one as a fault (less severe than CEL code, which is confirmed code) and have not seen it since. The other this the prominent and frequently occurring P0141 Bank 1 Sensor 2 heater circuit malfunction.

There is one more thing I will tell you about the NTK sensor I have placed in slot 1,2. It has a heater resistance of 3.9 ohms per my measurement. Spec is 6 ohms. This could be the issue. I will also measure the other sensors attached and see what their coil resistances are. I think I had measured another sensor with low coll resistance. But if all the other attached devices have 6 ohm coils, then that would be noteworthy. It will take about a week for me to get a 6 ohm O2 sensor of appropriate style to match the others here in Dallas. America’s stockpile of these is in California, and shipping may be sluggish due to Santa going USPS.

Morer results soon, but I will say that fixing the overheating problem has lifted the veil of confusing and misguided codes and readings. Now, I have much more confidence that I am getting reasonable responses from the ECU. Another way of looking at this is that now I may have opened Pandora’s box, and let the fault code beasts loose to plague me with a host of little fixes I need to do…

I will also say this, as it is key:

We have all noticed from this chain that there are differences from one 2.5 to the next, and some cars have different sensing systems than others, even within the same model and engine space. This was also backed up by the Mazda Service Highlights and the Oreilly’s helpful and knowledgeable manager. The manager pointed out that the style of sensors and systems on my particular car could be quite varied, depending upon the dates it went through manufacturing in Japan. The service highlights enumerate the various styles of EGR, evap, and purge systems employed on these cars. Whats more is that the codes have different meanings depending upon they styles used for these systems on the car in question. So what may be an O2 sensor heater malfunction code on most 2.5 Millenias, may actually be an O2 sensor circuit malfunction applicable to the other sensor on that bank.

So without fully isolating the exact manufacturing times and dates and control systems equipment supplied on this particular milly, about all we can say is that we know bank 1 has some sort of O2 sensor issue. And that issue could very well be catalyst caused. I will need to read much deeper into the service highlights in order to identify the exact type of each system I have, and exactly what the ECU does to determine pass/fail for each system.

But like I said, a great starting point will be to fix my oil leak, remove the heat shields, tighten in the bank 1 sensors, and carefully measure temperatures on both sides of each cat. I’ll chime in once this is done…

I had one last thought worth mentioning. The leak drops droplets of oil at a location above bank 1 sensor 2. It would not surprise me at all of some of these drops it the sensor during operation. What would be the effect of much cooler oil hitting the sensor during operation? Remember that this sensor can only properly sense at some 600 F. Would relatively cool oil cool it off enough to create a problem? Just a question…

O2 Sensor heater coil resistance measurements:

Bank1
Sensor 1 - 4 ohms
Sensor 2 - 3.9 ohms
Bank2
Sensor 1- 6.8 ohms
Sensor 2- 6.9 ohms

Now there are two more interesting things I observed while taking these measurements:

  1. Bank 1 sensors have female connector on the wiring headed for the ECU, while the bank 2 sensors have this connector on the wiring headted toward the sensors. Since this connector is the more substantial of the mating connector pairs, the other side is just a small plug that fits within these female sockets. It appears to me the less significant connector on bank 1 sensors may have less resistance through it than that of the female connectors on Bank 2 sensors. Maybe this adds some 3 ohms of resistance due to resistances in the connector itself required to make these connectors robust. Just a thought. Either way, I am not particularly concerned about these resistances any longer. My logic is that if it were a problem to be at ~4 ohms at the sensor, then why would the ECU consistently flag the same sensor. Seems it would flag both it this were the root of the concern.

  2. The other thing I noticed has to do with the oil leak. It just so happens that Bank 1 Sensor 2 has its wire just right there were the oil leak is happening. Now this is a significant leak, as I would estimate 1-3 drops per second. And a fraction of this is hitting this wire while driving. Then, it would make sense that the oil would flow down the wire to the butt of the O2 sensor. In addition, it seems I may recall reading something about taking care to keep the wriing inputs to these sensors clear of debris, grease and oil since the operation of the sensor depends on air to seep through the small openings where the wires enter. The sensor compares oxygen content of this ambient air to the oxygen content of exhaust at the sense tip. Hmmmm… I can see that there may be a problem here.

Lets clean up this oil leak and wipe the oil off that wire before doing anything else. Then I think it may be worth clearing out the codes again and taking another run at things…

OK, so I would like to get more expert opinion on the following items, as I am still wrestling with the Milly:

I did get the oil leak stopped, but so far I am still getting codes from the O2 sensor P0141, presuming that my interpretation of the code is correct. I have checked, the coil resistance is present (low, but present…see last post). I have verified that 12 Volts exists on those pins coming from ECU. All good. And I am becoming less concerned about the catalytic converters. After removing the covers for the headers, I idled the car for a while then took some infared temp measurements. I got ~300 F before and ~400 F after each catalyst. I think that this means all good, but will test more to be very, very sure. So then, since I believe these cats are gonna be OK, now what? I’m sort of at a loss as to what my next steps should be to sort out this P0141 code. Suggestions?

Tho oil leak was the second cam on the right head, as it is supposed to be capped. Now I have capped it once, but I did not put the timing belt cover on all the way, and evidently the cover got out, and started leaking again. So I found another cover from one of my ebay kits and shoved it in the hole. Then I put a nut there to keep it pushed back (oil pressure must be shooting it out of position), then held the nut in place with the timing belt cover, this time put on with all bolts so as to hold that nut there to hold the cover in place. I know, Jimmy rig, but that’s my name anyhow…

I have noticed a squeaky cyclical sound in that area. It seems like it may be a more recent development. I hope it is not a cam journal. I think it may be a timing belt pully, in which case I need to sort it out. I sprayed WD40 behind all the serpentine belt pullies, trying to find the one that changed sound when I did so, but did not find anything. It really sounds like its coming from behind TB cover. Come to think, this is in the same area as the camshaft end cover that shot out. I wonder could it be contributing to this? More soon on this front.

And finally, after installing the new temperature sensor, the car is much, much better. However, a few weeks back I broke this temp sensor connector, so I have to just push it down onto the sensor, but nothing latches. Since I got the new one, twice it has completely disconnected, which drastically changes the car’s behavior. Anybody got ideas where I can get just that connector for soldering on a new one? Or maybe someone has a good idea for how to take old connectors that are broken and hold them onto where they go with enough force to keep solid connection. You know, perhaps this is the problem, that I have a few loose connections yielding funny data to ECU, so ECU behaves in unusual fashion. Dunno. Suggestions to hold fractured connectors together? Rubber bands? Clothes pins? Chip clips?

Easy stuff first:

zip ties

usually the dealer, or a salvage yard (if you find better sources, I would LOVE to know; it’s pretty rare to be able to find these in my experience). It shouldn’t need to be soldered; the pins push into the connector housing. Like so

not clear on this whole situation. The timing belt cover should never be holding anything in place which prevents oil from escaping. Whatever is going on here, this all doesn’t sound right, and might be the squeak as well. Hard to say, but if it appeared after “the rigging”, that seems the most likely candidate.

Basic component troubleshooting, if possible: move them, and see if the fault follows the device or stays with the circuit.

In case it helps, and you haven’t already seen it (i’m guessing you HAVE, and it won’t really be helpful):
https://www.obd-codes.com/p0141

1 Like

Try Mouser?

http://www.mouser.com/Connectors/Automotive-Connectors/_/N-1ehb5

2 Likes

Could be.
I’ve never found one I’ve needed from there, but I don’t drive Fordzdas…
:wink:

1 Like

Let me explain the oil leak better since it is becoming a pain. Imagine dual overhead cam, right head (rear). One cam has timing belt pulley, the other does not. The one with pulley has seal around it to prevent oil from escaping. The other cam just ends in a journal, and has a gear it uses to keep in sync with its neighbor. Well, the cam without pulley has a hole in the external head housing along its axis right where it would poke through the housing if it were longer. This hole is supposed to be plugged with this drain plug sorta thing. Evidently, I have lost the drain plug sorta thing, but have a piece from a rebuild kit that fits in there snug, and I believe it was supposed to go there. But the oil pressure pushes it out, as it did the original plug. I think I got a good oil pump. But the plug is still there on the other head, so not concerned. I just need to keep the piece in the hole, as it sealed for a minute, until it got pushed out again. The timing belt cover edge happens to pass over the hole and can be used as a structure to push against the plug piece from kit and hold it in place. Ridiculous enough yet? I need a better way to seal that hole. But I am not above using the cover Jimmy rig method if I can find something of appropriate thickness to keep tension on plug thing so it can’t be pushed out by pressure inside head. I know! I could cut out a few rings of that thick gasket paper I have, stuff those in there. Let the cover hold that in there. If I cut them carefully, I would then get added oil stoppage power…

Gentlemen, thank you for the input about the p0141. This code seems to follow the socket, not the sensor, as I have placed three sensors in this spot, 2 of which were new, and all of them return this code. All of these sensors had 4 ohms or less coil resistance. Also remember that this code can also mean bank 1, sensor 1 circuit malfunction. I have never changed this 1,1 sensor or done a thing to it, and this symptom of the code has not changed. And get this one, it won’t occur very frequently as long as I minimize the idling of the car. What in the world does that mean? I need to think about how to measure wiring resistance, and search for any wiring shorts. Also, maybe its worth a look to make sure ECU well connected.

I do know I have corrosion in the battery cable portions of my wiring harnesses. This could be a factor in weird electrical behavior. I am sort of leaning toward doing some grounding reviews, as being an electrical engineer I have learned that you can run yourself ragged chasing ghost problems caused by bad grounding. I just walked out to look at my grounding wires from the battery. There are two, badly corroded wires going to 1. chassis and 2. transmission (has grown some recently fresh green corrosion where it contacts transmission. What say I replace these wires next? I have some cable for this.

I also am struggling with replace/not cat decision. I spent the better part of today watching cat videos. My cats look great at idle. But I realized when watching that I need to wartch how they respond when engine cranked up, and even better under load. I don’t want to drop $400 unless I have to. BTW, online consensus is you cannot wash a cat. And I think that they lose their catalyst metals as they smolder at 1000+ degrees, shaking down the road. There is no fix for this other than $400 once enough catalyst is gone. I believe that temperature tests are deceptive, so I want to steer clear. Some folks say 200 degree swing is expected, others say as low as 30. I say depends where you measure and how your system is setup. Pressure testing can identify clogs. And if you want to know how effective they are, you need to watch rich/lean O2 data, and keep in mind the architecture of your exhaust system, in that my sort of setup like Milly will illuminate each cat performance, while third sensor combined way down the pipe merges both cat results with additional latency.

Mouser is a great idea. Now what do you call that connector that fits into those coolant sensors? Lets dig that out of Agile for the design we are working on… oops cant. Now what? I know Town North Mazda will be no help. One of their parts guys quoted my $2k for one of the wiring harnesses, $300 for another, $600 for another. He said you can only get those connectors from Maxda by ordering the harness. Gee thanks. Oh, but they would be happy for me to test drive a new Mazda… lol…

.

1 Like

Bank1, sensor 1 heater circuit malfunction SHOULD ring up a P0135
https://www.dmv.de.gov/services/Vehicle_Services/dtc_list.pdf

That plug should be held in place by friction with the hole it’s plugging. The pressure on it should be negative to zero, oil or otherwise. Anything to the contrary, and something is amiss. Could be PCV system if there’s that kind of pressure in the crankcase, as the entire crankcase should be under vacuum whilst operating properly (that’s the point of the positive crankcase ventilation system).
Oil pressure which is too high is no more tolerable than too low, though the failure is slightly different. I imagine this oil pump was furnished with the shortblock?

Ground fixing is never going to hurt.

1 Like

No, the oil pump is a top dollar Chinese ebay special. I went all out. And was careful to never measure the pressure. You see, I learned from Ostriches that if I stick my head in the sand, the danger goes away.

OK, I’ll go measure oil pressure and the pressure at that back head. And maybe I’m misunderstanding where leak is coming from. The service manual calls for liquid gasket at this area when installing the valve cover. I managed to seal the front one, which would be easy to fix, but not the back one, which requires removal of air intake to fix (per murphy’s axiom of the makerverse).

Now, the O2 heater 1,2 failure is starting to make sense. This sensor is the one furthest in the back, and last to heat up. Perhaps I need to reduce resistance of power delivery heavy wiring and try reducing load or something. That sensor warms up in time, and works just fine. But as you can see from service highlights, the ECU will insist that it be ready at a certain point. So now it makes sense why it flagged this over and over when I warmed up via idle on cold day. Check out Service Highlights:

And then the possible break thru occurs…

I’m a bit frustrated tonight, as the problem is not so easy. Changing O2 sensors does not help, there was nothing wrong with any of them. So I have got some problem, and I’m not going to get much of a pointer to it. But the car drives so smooth. As I was driving, I ran the codes, and of course P0141. I cleared it. Then I pulled over to cycle the car off then on. When I did so, I noticed that my pandora was attached to my headphones, and so I switched it to the dash mount receiver. I had exited all the way out of torque, and when I restarted torque while driving down the road, thats when it happened…

Pop!

A glitch, just like what was mentioned earlier. Nothing was damaged but I looked at my freshly reset dashboard and lo and behold, the check engine light was on. The glitch is causing the code.

I gotta get soime rest, and I can’t do anything about it until tomorrow when auto store opens so I can get some better wire to get good grounding and what not. However I am about 80-% sure the glitch is the source of this p01440 issue. More soon…

For clarity, before I sleep on this one, the POP was a popping of speakers as my new radio was on. It signifies a power glitch at the reciever (which is right above the ECU, which is powering 4 sensor heaters for sure at that moment). T’he ECU is also powering that port with ELM327 plugged in. And evidently it connected with torque and something switched pretty hard making the POP. Plenty of suspects including ELM327 bluetooth, the bus switches (if they exist). etc. There are also a lot more electronics installed in that cabin that what I imagine Mazda expected. I have a 2.1 amp charger, a camera on and recording, a TomTom on with display.

I gotta crash. I bet we lick this one tomorrow…

OK, truth be told, I have been secretly getting some expert help… see below…

2 Likes

So today is about fixing my electrical issues. I think that I have overloaded my circuits inside the cab of this car, and now the ECU faces occasional large glitch, which is generating the sensor heater code. So here is my start to this fix, including new heavy duty wiring for the corroded power cables.

@AndrewLeCody do you have any suggestions about delivering solid power rail to my componentry inside the cap of the Millenia? Obviously, I intend to improve my power and grounding to/from battery (eliminating the corroded cables). But I also think that I need to bring heavier power and ground wiring into the cab. And I think that I should consider some decoupling as well. Suggestions about good ways to do this?

I’m not sure, maybe a large capacitor would help.

Here is the complete listing of contributors to my electrical problem:

  1. Corroded cables and connections under the hood. I can see evidence of a battery that was left under the hood for a long time, spewing battery acid all over the place, particularly affecting cables near the battery, and unfortunately the transmission below. I cleaned the transmission carefully ,and now I will replace all of that cabling tonight. I will have new cabling securing high current power and ground to its destinations, and have selected even heavier gauge wiring that was original as a good measure.

2 After-market stereo installation draws 10 amps due to its power delivery capabilities and bluetooth radios, neither of which exist in the previous system. In fact, as I understand things, the old unit was not amplified at the receiver, but instead was amplified at the speakers. Some have argued that usually amplification occurs at the receiver. And while I agree that this is usually the case, I know for a fact that the Bose premium system that was an original equipment option has individually powered speakers, and while I don’t see any designation in english written anywhere on my original equipment, I have seen some weird components in locations that lead me to believe that maybe my system has individual amplification as well.

  1. TomTom screen will pull a decent draw.

  2. Camera screen will pull a decent draw as well.

  3. 2.1 Amp USB splitter can pull a fair draw.

  4. Cellphones, android, and navigation were not all the rage in 2001. So the power delivery inside the cab was expected to be minimal, I feel certain, when this car was manufactured.

I think I may approach things like this: I will replace the corroded cables under the hood. Then, get a scope shot of the glitch on the power supply in the cab. From there, I will infer the frequency of glitch (approximately), and use electronic simulation tools to figure out what capacitance is needed to build an analog filter to squelch it out of existence. Perhaps I should video this process, as it is something I would think may be useful for makers who add electronic coolness far beyond what was originally expected. There is a way to deliver your electrons without negative consequences of Jimmy-rigging. We will simply makerneer a robust supply rail…lol…

That list should be numbered 1-6, and although I have attempted 3 times to edit it, HTML and the underlying software are insisting on numbering the list as though it were 2 lists. Please forgive my butchery of sequentially numbered lists using Arabic numerals.