Intermittent P0128 After Switching to 88-Degree Thermostat

[JimLev]

Thanks, Jim and Chris! So the heater serves the opposite purpose than I imagined.

Jim, I haven't done the cardboard trick yet. That'll be next. Any advice for how to do that with my Zionsville radiator?

Fit it in any way you can, I don't have the ZV shroud, just their radiator.
I'm using the OE shroud with a fan attached to the inside of the shroud.
Try blocking of about half of the radiator if you can.
If the KMTP starts to climb too high just turn the cabin heater up to 90.

[peter540]

Fit it in any way you can, I don't have the ZV shroud, just their radiator.
I'm using the OE shroud with a fan attached to the inside of the shroud.
Try blocking of about half of the radiator if you can.
If the KMTP starts to climb too high just turn the cabin heater up to 90.

I'll give it a try. The Zionsville shroud is, I think, going to be more difficult to do this with; it really encases the core.

[peter540]

Update. With old thermostat plugged in (88° thermostat still installed) and Zionsville e-fan unplugged, and having KTMPs consistently between 90° and 94° (never below 90°), and having driven enough miles for six of eight monitors ready, I just got the P0128 code again. WTF.

Is it reasonable to assume, based on the KTMP readings, that I can rule out the dual temp sensor and water pump (again, newish pump, no leaks) as the cause of this problem?

At this point, I'm inclined to have DUDMD permanently erase the code and move on.

Thoughts?

[JimLev]

The water pump shouldn't have anything to do with it.
There may be a possibility it's your dual temp sensor, maybe.
What you are seeing in the cluster is one of the dual sensors, the other sensor goes to the DME.
I don't know which one is used to trip throw the P0128, maybe DUDMD does.
You could check the sensor for resistance and the pins for a bad connection.
Looking into the sensor you'll see the pins are labeled.
Pins 1 and 2 should read 6.3K ohms at 68F (20C), this connects to the cluster.
Pins 3 and 4 should read 2.5K ohms at 68F (20C), this connects to the DME.
As the temp rises the resistances values drop pretty fast.

[peter540]

The water pump shouldn't have anything to do with it.
There may be a possibility it's your dual temp sensor, maybe.
What you are seeing in the cluster is one of the dual sensors, the other sensor goes to the DME.
I don't know which one is used to trip throw the P0128, maybe DUDMD does.
You could check the sensor for resistance and the pins for a bad connection.
Looking into the sensor you'll see the pins are labeled.
Pins 1 and 2 should read 6.3K ohms at 68F (20C), this connects to the cluster.
Pins 3 and 4 should read 2.5K ohms at 68F (20C), this connects to the DME.
As the temp rises the resistances values drop pretty fast.

Thanks, Jim. So I'm assuming I should check the resistance when the engine isn't hot, right?

I still doubt that any of the sensors, etc., are bad, because I never had this issue before swapping in the 88° thermo. I never touched the dual sensor, other than to unplug it and plug it back in. I also cleaned both of the connectors with contact cleaner.

The only slightly suspect sensor, as you may recall, is the one in the lower radiator hose; I installed it in the new hose with its original o-ring, and it has a teeny, tiny (emphasis on "tiny") seeping leak (not even enough to drip down onto the plastic pan below it). Does that sensor even have a role in triggering the P0128 code?

Also, given the KTMP are right where they should be, with no fluctuation at all, I think it's safe to say the 88° thermostat is fine (i.e. not sticking). Would you agree?

[BMW540san]

Peter, make absolute sure to go over the dual temp sensor wiring as far back as you can see it.
The wiring might look ok from outside but it's possible that one or more wires inside the loom could have crumbling insulation and it doesn't take much to cause issues like these.
If the outside loom looks suspect and if it's starting to crumble, I'd go ahead and remove it and inspect each wire individually.
You can purchase high temp wiring loom and replace it when you're done.
According to what you just posted, I simply don't see thermostat being an issue.

[JimLev]

Thanks, Jim. So I'm assuming I should check the resistance when the engine isn't hot, right?

Yes, stone cold, otherwise the resistance reading you get will be lower.

I still doubt that any of the sensors, etc., are bad, because I never had this issue before swapping in the 88° thermo. I never touched the dual sensor, other than to unplug it and plug it back in. I also cleaned both of the connectors with contact cleaner.

A few weeks ago my cluster reading only made it up to the mid 80's, my wires were cracked so I replaced them and then put deoxit on the pins, so far it reads correct.

The only slightly suspect sensor, as you may recall, is the one in the lower radiator hose; I installed it in the new hose with its original o-ring, and it has a teeny, tiny (emphasis on "tiny") seeping leak (not even enough to drip down onto the plastic pan below it). Does that sensor even have a role in triggering the P0128 code?

Your lower hose sensor is for the aux fan in front of the AC condensor.

Also, given the KTMP are right where they should be, with no fluctuation at all, I think it's safe to say the 88° thermostat is fine (i.e. not sticking). Would you agree?

Yes, the t-stat is probably is OK, however you were only looking at the temp in the cluster, you don't know what temp signal was going to the DME from the other 1/2 of the dual sensor.

..

[Corellian Corve]

Dude. Put the old TSTAT back in and make sure the car works. Then you can pass emissions and play around all you want if you feel it's necessary.

For all you know you're chasing your tail and the problem is unrelated. And I would suggest trying a different scanner because one thing I will agree with is that INPA codes can sometimes be different from readings I get with other scanners. I had this problem with my Transmission EGS.

When you're chasing a problem like this, get back to square one. Put the OE Thermostat back in and see if you have still have a problem. These cars have hundreds of thousands of miles on them with the OE cooling components. It's not like you're on a ticking time bomb here, but you are putting unnecessary pressure on yourself when you could have had the OE tstat in and be passed already.

[Flounderasu]

Having KTMP up, doors closed, turn the key back one click and see what the readings are. SB the same or within 2-3 degrees of what you saw while the car was running. Now, I am not sure if that is truly the temp feeding into the DME as I thought it also compensates/checks the coolant outlet temp sensor - and I was seeing coolant outlet temps around the mid-high 70's while the KTMP held steady in the 90's. Coolant outlet temp would get up to about 90, then starts dropping - fan kicks on and pushes it back down to the 70's....KTMP hovered around 93-95 during that phase.

[geargrinder]

It's funny how some think that everything BMW does must be correct. It's a simple fact that 105 C thermostat in conjunction with dumb (buffered) temp gauge doesn't leave you much time to react when shit is hitting fan.
Me and some others are not advocating 79 C M5 thermostat which is way to cool for M62TU, but more reasonable 90-95 C. Unfortunately there is only 88 C available for M62TU while 85 C and 95 C are available for M62. Even with 88 C, you're looking at 90+ C average temp and that's hot enough for to keep combustion chamber free of contaminants.
Also, there is considerably more pressure with higher heat and BMW could care less when your plastic coolant components fail in relatively short period of time.

All this stuff here.

..

All the stuff here (which I can't quote but you get it) esp the bit about the dual-sensor etc.

one thing I will agree with is that INPA codes can sometimes be different from readings I get with other scanners.

And this... I'd use proper BMW coding tools with a tricky item like this. (then again I'd always use them...)

[peter540]

Update. After my last post:

-- I unplugged the original thermo, which, per Jim's suggestion, I'd plugged in to serve as a temporary resistor (I left the 88° thermo installed)

-- I plugged the Zionsville e-fan in again; it comes on at its low speed only and operates normally

-- A couple of weeks ago, my wife used the car to drive to work (20-minute drive on surface streets), then to the airport (50-minute drive on highway), parked the car at the airport for five days, then drove the car home again (another 50-minute highway drive)

-- I then drove the car for short local trips (maybe 15-minute trips each) for a couple of days

-- No more codes appeared, and, for the first time, all the monitors reached a state of readiness

-- I took the car to emissions a week ago, and it passed

-- KTMPs have remained in the 90° - 96° range, even with the e-fan

-- Still no codes a week later, again mainly short local trips only

I'm beyond relieved and happy that this issue disappeared and that I could get the car registered. Still, I'd sure like to know how the problem resolved itself. What could have made the difference? Perhaps the longer-distance drive cycles my wife logged? Or perhaps might there have been a small amount of air in the system (I'm an accomplished burper and had burped the system correctly and thoroughly) that finally worked its way out?

I haven't had a moment to check things with my multimeter, but at this point, it clearly wouldn't show anything is wrong.

Thoughts from the gallery?

Thanks again for all of your help as I worked to sort this out!

--Peter

[peter540]

Thoughts, folks?

[Uturn540]

I put the 88C thermostat in mine two weeks ago after blowing the valley pan gasket. It was due for coolant service and I wanted to try the 88c thermostat anyway. So far, no codes(put a resister in the t-heater circuit), runs at 94-96C, even the intermittent startup rattle is gone. It is in a full readiness state at 150miles of 12 mile trips. Of course the vehicle does not need to be in a full readiness state to set a code. It may have been more of a learn function that caused your code. I have had experience with several vehicles that took a long time to adapt when a component was changed. My diesel Ford acts weird every time something gets changed unless the PCM memory gets cleared and it is forced to relearn all the parameters. Clearing the codes will not do it, you actually have to disconnect the PCM memory power. I wonder if BMW might be the same way and that disconnecting power will not clear codes or learned engine parameters. You might have to have a dedicated tool to set the PCM into a relearn mode or just have to wait a long time for it to adjust by itself.

[peter540]

I put the 88C thermostat in mine two weeks ago after blowing the valley pan gasket. It was due for coolant service and I wanted to try the 88c thermostat anyway. So far, no codes(put a resister in the t-heater circuit), runs at 94-96C, even the intermittent startup rattle is gone. It is in a full readiness state at 150miles of 12 mile trips. Of course the vehicle does not need to be in a full readiness state to set a code. It may have been more of a learn function that caused your code. I have had experience with several vehicles that took a long time to adapt when a component was changed. My diesel Ford acts weird every time something gets changed unless the PCM memory gets cleared and it is forced to relearn all the parameters. Clearing the codes will not do it, you actually have to disconnect the PCM memory power. I wonder if BMW might be the same way and that disconnecting power will not clear codes or learned engine parameters. You might have to have a dedicated tool to set the PCM into a relearn mode or just have to wait a long time for it to adjust by itself.

Interesting. Perhaps it was a "learn" function. Regardless, I'm glad to be past it, at least for now. I may ultimately add a resistor to the circuit, but first I want to be certain that the P0128 code doesn't return.

Thanks for chiming in!

--Peter

[peter540]

Aaaand...

...the P0128 code came back this morning. I'd driven a very short distance to drop our girls at school, so the engine hadn't fully warmed up yet. Code was thrown as I was accelerating into traffic. KTMP was around 86-87° (again, engine still warming up). Could that acceleration cause the code if the engine's not up to temp yet? I never had that sort of problem with the original thermostat, so I can't imagine it would be an issue with the new thermostat.

I'll pull out my multimeter and check the dual temp sensor once the engine is cold and see if that tells me anything.

BTW, the KTMP did climb, right after the code was thrown, on my short drive, ranging between 88° and 92°. The temp was still rising when I reached our house, but again, it was a very short drive. Since swapping in this 88° thermo, the KTMPs have reliably been in the 90-95° range, and I'm sure it would have gotten there this morning if I'd had a longer drive.

Sigh...

At least I got through emissions and have a year to figure out a solution to this nagging problem.

--Peter

[philly98540]

Must be some minor variation in the dme code where it verifies temps are above a minimum threshold. This could explain why some cars get the code, and some dont. Or your dual temp is reading just slightly low.

[peter540]

Must be some minor variation in the dme code where it verifies temps are above a minimum threshold. This could explain why some cars get the code, and some dont. Or your dual temp is reading just slightly low.

Yeah, I agree with both of your thoughts. I'll test the dual-temp sensor soon. I think the temps were only running lower than normal because the engine hadn't fully warmed up yet. If I change the dual-temp sensor, I think I'll do so without draining the coolant -- shouldn't be more than a cupful of leakage, I imagine. Tell me if you disagree.

If all else fails, perhaps I'll just have DUDMD permanently delete the P0128 code. Only after I'm certain that none of the components (i.e. the dual-temp sensor, thermostat) are failing.

--P

[JimLev]

Peter, if you keep the cap on the expansion tank only a few drops of coolant might leak out of the hole for the dual temp sensor.
Before you change it put a resistor on the connector that use to plug into the t-stat heater.

[peter540]

One more observation, after driving the car some longer (for me) drive cycles: ever odd once in a while, the car takes much longer to warm up after its first start of the day. For example, several days ago, with outside temperatures in the mid 50s, and after having driven the car a good 30 minutes in stop-and-go traffic, the KTMPs were only ranging between 87 degrees and 89/90 degrees. No codes popped up, at least. Compare that to this morning: with outside temps in the mid 60s, the car got up to 92 degrees within ten minutes on a short drive.

All the monitors are ready again. (Not that it matters, other than to note that the car is capable of achieving code-/error-free status.)

Weird. What could cause the anomaly several days back? Perhaps this is what caused the SES code (P0128) last week. Thoughts, guys?

Still need to put the multimeter on the dual-temp sensor. I promise to do that this week.

--P

[philly98540]

Inconsistent warm up could be due to sticky stat. ie. it does not fully close on the last cool down cycle, so when you start it up next time around its bypassing a little too much. It may go away on its own, but I'm not sure.

[peter540]

Inconsistent warm up could be due to sticky stat. ie. it does not fully close on the last cool down cycle, so when you start it up next time around its bypassing a little too much. It may go away on its own, but I'm not sure.

Hmmm. Are you saying that if the new thermostat is sticking, it may eventually self-correct to where it doesn't stick any longer?

--P

[philly98540]

Just speculation on my part if it's sticking and also if it will self correct over time.


The slower warm up can be explained by a stat not fully closing, so I mentioned it. As far as self correcting, if the stat move through a bunch of cyles, maybe it will smooth out what's preventing it from fully seating. Total speculation though...

[geargrinder]

See more on this error code here - particularly a warning that this code can be dangerous:

https://www.bimmerforums.com/forum/s...ee-code-P0128)