Audi A4 B6 1.8T P2181 AUDI - Cooling System Performance - Diagnostic Help?

[CortlandCH]

Happy Friday everyone (at least where it's still Friday)!

Being a new member here, I wanted to make absolute certain I did quite a bit of searching for this issue, and while there's plenty of info and threads out there similar to this, a lot of them end inconclusively or start to trail-off and I never seem to find the answers I need, so I'm going to give this my best go. Briefly before explaining, I'll give a small bit of background. I've worked in automotive for a few years, and I currently work for an engineering company for electric propulsion systems and energy solutions. I'm an engineering technician that troubleshoots a large range of mechanical and electrical issues, but sometimes (especially with cars like Audi, it seems), it's best to refer to the pros with all their wisdom to help diagnose an issue.
I bought my 2005 Audi A4 1.8T Quattro (Automatic) with 160,000 miles back in October 2019, and it's needed quite a bit of love, given where I bought it from. I've almost finished all the repairs needed to get it into great shape, but now I have a new CEL that I just can't seem to understand: P2181 AUDI - Cooling System Performance. It's a pretty vague code, and unfortunately, it's the only code stored in the ECU. Clearing the DTC does nothing, as it comes back within a day.

So what have I done, you may ask?
When I bought the car, it had a small coolant leak, struggled to get up to temp, and with research, I found a crack in the coolant reservoir, and of course, the coolant flange. Since I bought the car near winter, it had quite a hard time through the cold winter months of Colorado, as it would take a very long time to reach operating temps. The symptoms observed looked like a faulty thermostat, so eventually I was able to replace that, along with some other related hardware (J-Plug, hoses, so on). After this, the engine had no problem getting up to operating temperature, even at 10 degrees F outside. Last month, I replaced the coolant flange while changing the timing belt and water pump, and now my leak is gone! Hooray!
But now, after doing all this, the P2181 code has appeared.

I decided to do some checks with VAGCOM, and while I'm not 100% savvy with the software yet, I'll certainly take any advice or instructions for using it if anyone has any. Using Measuring Blocks 001, 130, and 132, I watched the temperatures reported by the coolant temperature sensors located on the coolant flange and the lower radiator outlet. In 001, the engine engine temp is reported, and appears to be an average of the two values from the upper and lower coolant temperature sensors. In my case, this value tends to read around 74C-79C when warmed up on a hot day, which seems a little low for the desired 80C temperature. If I sit and idle, it'll get up to 80C or higher, but at highway speeds, it tends to sit around 74C. In 130 and 132, I have logged and charted the two coolant temperature sensors, from cold-start to operating temp while driving on the highway. For the most-part, you can see a distinct rise in the engine temp before the thermostat opens and the lower temperature sensor begins to report hotter temps, so I know the coolant is flowing, and the thermostat is opening. In short, the only time I really see the coolant temperature go above 80C is when I sit and idle for >5 min or so (and the cooling fans do turn on at the correct temps). I did notice that my buddy and I filled the coolant/water mixture a little rich, since it's about a 30/70 water/antifreeze mixture, so I'll attempt to dilute it a bit, but I'm worried I'll run even cooler with the added benefits of water!

So, does anyone know any secrets about this code?
In summary, or TLDR:
- I have a P2181 AUDI - Cooling System Performance DTC stored that won't go away.
- The temp gauge shows a proper operating temp, but data logging suggests the engine temps are slightly cooler than desired (Average 74C)
- Thermostat has been replaced
- Coolant flange has been replaced (along with green CTS)
- Coolant hoses have been replaced
- Water pump has been replaced
- Coolant has been flushed and replaced (though may be a little heavy on the G12 antifreeze side, so will need to be diluted)
- Cooling fans operate normally
- I have VCDS VAGCOM to help with diagnostics
- I have not changed the lower CTS, though it doesn't behave erratically, as suggested by data logging in VAGCOM.

Really, the only thing I have left to try right now is to replace the lower CTS, which isn't that hard, just need to get more coolant and whatnot and make a mess again.
I've just tried very hard to check the sensors with VAGCOM first instead of draining everything and replacing good parts if it's not needed.

Please, any advice is good advice (well, at least related to the topic). I'll be happy to provide anything you experts out there need for information, and I apologize if I have missed anything that's standard procedure around here.

Thanks!

[caluma4x4]

What brand of coolant temperature sensor did you fit?
Cheap or unbranded are more trouble than they're worth.

Sent from my MAR-LX1A using Tapatalk

[old guy]

75→80c is too low. Reads through this thread to get a good understanding as to what to look for. Clicky click®

[CortlandCH]

What brand of coolant temperature sensor did you fit?
Cheap or unbranded are more trouble than they're worth.

Thank you very much for your reply,
The CTS I received came with the coolant flange kit from ECS: https://www.ecstuning.com/b-assemble.../06b121132ekt/
I can try and contact ECS to find out which sensor exactly comes with that kit, but I know it's the URO brand, and believe it's this one: https://www.ecstuning.com/b-uro-part...059919501aktu/

I do recall that the fitment of that sensor was nearly impossible into the new coolant flange, but after some light sanding of the flange sensor opening, it eventually squeezed in there. Didn't know if it was the fault of the sensor or the flange, but holy crap is the flange a rough job to replace.

[CortlandCH]

75→80c is too low. Reads through this thread to get a good understanding as to what to look for. Clicky click®

Must be my lucky day to have my first thread get a response from you.
As I read your post, I decided to do another scan with VAGCOM to make sure I didn't have any other clues.
I'll go ahead and do another temperature log when I hit the highway again today to see how the increase in temperature looks for the lower CTS.
I feel like some increase is normal, but if it's substantial, I'll shake my head knowing I got a defective thermostat.

I have a new, intermittent DTC that I saw in the past with the old thermostat.

17700 - Map Controlled Engine Cooling Thermostat
(F265): Open Circuit

It's intermittent, however, and watching the thermostat duty cycle, I can at least see the value change when I step on the gas. Don't know for sure if that means anything about it working, or if that's just the computer's way of saying "This is what I'm trying to do."
I've had this code previously, when the thermostat was faulty, and I didn't have the P2181 code, so are they directly related?
There also doesn't appear to be a fuse associated with F265 in the schematics. I'll attempt to remove that plug and check the pins on the thermostat plug to check for an open circuit, and also try and clean the connector.
This P2181 code has been here long before the Map Controlled Thermostat code, but anything's worth trying to fix!

Edit: Would it be worth logging the coolant temps as they increase while idling, instead of driving? I feel like this would be the most lengthy, but "purest" way to see the temperatures change since the water pump isn't spinning rapidly in moments.

[old guy]

17700 indicates an open circuit on the thermostat heater. You can check the resistance on the heater element to see if it is element. If the element checks out you have a wiring issue.

The P2181 is pretty much a generic cooling system code. The ECM knows something is wrong because the temperatures are not reacting as expected.

When you increase the engine load you should see an increase in the thermostat heater duty cycle. The ECM uses this to reduce the block temperatures.

Yes you can monitor the cooling system at idle but it may take a while to see any changes.

[seanf86]

I had a uro parts sensor and it only read 80C even when the car was above that. Ever since I only buy genuine temp sensors. What brand thermostat? Sounds like you got some bum parts.

Sent from my LM-G710 using Tapatalk

[hightime80]

I always think when engine isn't reaching operating temp its a thermostat stuck open. Hell its better than stuck closed😂

[EuroxS4]

Get a factory thermostat issue will be resolved.

[CortlandCH]

Sorry for the wait everyone, and I appreciate everyone's input!
Based on the data I've collected, it looks like the thermostat is opening fine. I left work and hooked-up my laptop to run VCDS and take some measurements while idling to temp. Since the cooling system performance fault exists, the motor will not idle high to help increase temperatures faster, so it took quite a few minutes. It was about 24C outside when I left work, and without running the engine, the Engine Outlet Temp (upper cts) reported 24C, while the Radiator Outlet Temp reported 28C. This could make some sense, being on the front of the vehicle, with the radiator taking heat from sunlight, and the sensor being closer to the hot pavement. The Radiator Outlet Temp remained steady around 24C-26C until the Engine Outlet Temp reached around 75C, at which point the Radiator Outlet Temp began to increase. This to me shows that the thermostat restricted flow until the engine temperature reached around 75C. After about 18 minutes of idling, the Engine Outlet Temp reported 81C, while the Radiator Outlet Temp was still around 43C.
I then went for a short drive, about 5 minutes at 45mph. The Engine Outlet Temp remained around 79C-81C during the drive, but the Radiator Outlet Temp stayed within a range of 44C-60C, and no-hotter.
This, to me, points more at a faulty Radiator Outlet CTS, as I'm fairly certain the radiator isn't cooling 80C coolant down to 44C, even if the thermostat was wide-open. This is especially considering that, during the short drive, the Engine Outlet Temperature barely fluctuated at all, so the extreme cooling observed by the Radiator Outlet CTS didn't seem to have a dramatic effect on the outlet temp of the engine.
Hope this link works, it's a chart depicting this data described:
https://photos.app.goo.gl/tRkDfm4Wb4uaeAZS9

Previously, I have sat and idled on a hot day after a drive on the highway, and I was logging these temps on VCDS. I did see the Radiator Outlet Temperature eventually reach 90C along with the Engine Outlet Temperature, at which point the cooling fans came on and very quickly lowered the Radiator Outlet Temp. I'm just not certain the sensor is working as expected all the time.
My next steps would be to replace the lower CTS and see if this changes anything. I'll order OEM just to be safe, and I'll keep everyone updated!

[CortlandCH]

I had a uro parts sensor and it only read 80C even when the car was above that. Ever since I only buy genuine temp sensors. What brand thermostat? Sounds like you got some bum parts.

Get a factory thermostat issue will be resolved.

I feel like OEM quality sensors is the right way to go, since everything's so picky on these cars, so lesson learned there. I did buy an OEM thermostat, so hopefully that was a smart move:
https://www.ecstuning.com/b-genuine-...ly/06b121111k/

I always think when engine isn't reaching operating temp its a thermostat stuck open. Hell its better than stuck closed😂

That's what I was thinking too! Could be far worse. The thermostat doesn't appear to be stuck open, given the data I've collected on the coolant temperature sensor changes. The lower radiator hose stays nice and cool until the engine gets up to temp.

[CortlandCH]

75→80c is too low. Reads through this thread to get a good understanding as to what to look for. Clicky click®

I gave this another read and some more thought to my issue, after seeing that my thermostat appears to be opening fine and isn't stuck open when getting the engine up to temp.
You say here that the engine temperature is measured at two points, the upper and lower CTS. Given this, I'm curious where the value for the coolant temperature in Measuring Block 001 comes from. It looks like an average of the two sensor values, but now I'm not so sure.
I decided to clear the P2181 fault code so the fans wouldn't come on full blast after starting, and logged the warm-up cycle at idle again like before. Not surprisingly, it warmed up faster with the engine idling higher, and the cooling fans off. Still, the lower CTS value did not begin to increase until the engine outlet temp (upper cts) read around 75C-79C, so the thermostat stayed closed until the gauge read warm.

You can see the data plotted in a chart below, which shows the ambient temp outside (25C), engine outlet temp (upper CTS), radiator outlet temp (lower CTS), and the coolant temperature reported in Measuring Block 001.

https://photos.app.goo.gl/Qw7rrGgiwbVUQKtq7

I then logged more data of the same type after a drive on the interstate, then sitting idle in a parking lot. This changes things a lot for me, as previously, I had only logged temperature data when warming up or driving.

https://photos.app.goo.gl/S9oiQTUDnqJpt9qH7

What baffles me is the coolant temperature reported from Measuring Block 001. The upper CTS reports engine outlet temps of around 93C-99C, and the lower CTS reports radiator outlet temperatures of around 87C-99C. The oscillating pattern in the data can be seen due to the cooling fans turning on, bringing the radiator outlet temps down. You can see where the fans are on and off, based on the thermal cycling.
Among all this, however, is the incredibly low temperature reported in 001. For some reason, even though the two coolant temperature sensors are providing readings >85C, the coolant temperature value never reaches or goes above 85C. I don't know the math the ECU is doing, and honestly I don't know if anyone knows exactly, but if the other two sensors are providing an average temperature of 93-96C, why is the coolant temperature average 80C?

At this point, I know what I'm going to replace and test to continue fixing this issue, but I also want to make sure to properly document everything I've done and the advice given so if anyone else comes across this issue, they're not lost. Hope some insight can be provided!

[old guy]

Here is a much more in depth study on how a mapped cooling system works: Clicky click® Here is a more condensed version: Clicky click-2 The Self Study link is for a 1.6L engine but the overall mapped control is the same for the 1.8T.

As you can see from reading these articles you should be recording considerably higher temperatures than you are. Lower than expected temperatures can be caused by one of two things. Your thermostat is opening at a lower temperature than the the expected maps or your thermostat heater duty cycle is excessively high because of faulty input from the temperature sensors.

The first thing I would recommend is to check your duty cycle. If it is staying around the baseline of 3→4% then the thermostat heater isn't what is causing your overall lower than expected temperatures. And if that is the case I would then strongly suspect your thermostat is opening at too low a temperature. Thus the P2181 system performance error.

Use the Measuring Block Group 130→137 Engine Cooling for performing you diagnostic work.

[CortlandCH]

Here is a much more in depth study on how a mapped cooling system works: Clicky click® Here is a more condensed version: Clicky click-2 The Self Study link is for a 1.6L engine but the overall mapped control is the same for the 1.8T.

As you can see from reading these articles you should be recording considerably higher temperatures than you are. Lower than expected temperatures can be caused by one of two things. Your thermostat is opening at a lower temperature than the the expected maps or your thermostat heater duty cycle is excessively high because of faulty input from the temperature sensors.

The first thing I would recommend is to check your duty cycle. If it is staying around the baseline of 3→4% then the thermostat heater isn't what is causing your overall lower than expected temperatures. And if that is the case I would then strongly suspect your thermostat is opening at too low a temperature. Thus the P2181 system performance error.

Use the Measuring Block Group 130→137 Engine Cooling for performing you diagnostic work.

This is super cool information, I really appreciate you sharing it, as it helped a lot in understanding the issue with the thermostat. I'm still not entirely sure why the engine computer is reporting coolant temps significantly lower than the temps provided by the two sensors, but there's likely more going on with the temperature circuits than I can comprehend right now.
Well, I haven't had a ton of garage time lately, so I don't have too much to report on the issue, but I did do a little investigating related to the information you provided.
Back when I replaced the thermostat originally, I recall the Map Controlled Thermostat connector (male-side, on the thermostat) busting off. It had a little bit of coolant in the connector as well, and was all gross and corroded.
Since the circuit for the heating element in the thermostat appears to be directly linked to the ECU with no fuse, I feared electrical overstress in the ECU (which, I suppose is still a possibility) due to shorting from the coolant.
I decided to probe the back of the connector, as I had a hard time getting any voltages touching the pins in the front of the connector. With the vehicle key in the on position, I had a steady (aprox) 12V from the ECU. Didn't appear to be a PWM signal, just a DC voltage.
I assumed this was normal, since there was no reason to start trying to open the thermostat. You can see these results here:
Connector with Probes
Voltage from Connector
Voltage on Scope

Given the confined space and difficulty reaching the thermostat connector in the engine bay, I couldn't quite check if there was really an open circuit in the heating element. I was considering looking for another female connector online to wire for testing this, instead of removing the thermostat to test for resistance. Right now, I have a feeling that the barrels in the female end of the connector are either damaged or corroded, and aren't making a good connection with the pins on the thermostat, but I can't quite confirm this yet.

EDIT: I did end up finding the connector for the thermostat heating element, and to my surprise, it's got prime delivery on Amazon! I know it isn't OEM, but for the purpose of testing for an open circuit on the heating element before pulling the trigger to buy another thermostat kit, $14 isn't that bad.
In case anyone needs it, you can find this connector on Amazon Here, or ECS Here (Without Pigtail)

I'll keep more information coming, as I really hope this will help others with diagnosing similar issues in the future. Thanks again everyone!

[CortlandCH]

Results are in!

And it's what we all expected. The thermostat heating element is busted, nearly open circuit, reading about 5.5 Megaohms.
I couldn't find anything for the life of me that lists what the resistance is supposed to be for a new thermostat, so as soon as I get the new one, I'll post it here.
I'll be sure to follow through with this until the problem is resolved.
I believe I know the root cause of the failure, but I'll confirm it when I can take apart the faulty thermostat. Since I was having cooling system issues and had to change the thermostat in the past, it's highly likely that the damage was caused by improper bleeding of air from the cooling system. An air pocket could have formed in the thermostat at one point while the heater was live, and without coolant surrounding it, it's likely that it burned itself out. Either that or I got a defective part.

So the connector came in, nice quality looks like the OEM part. I have an image of it here if anyone wants a look.
I went ahead and pulled everything out of the way to get to the connector again. For those who want to recreate this troubleshooting procedure, here you go:

Tools Needed:
- Slotted Screwdriver (Flathead)
- Socket Wrench
- Long 1/4" Extension (Handy, but not totally needed)
- 7mm/8mm Socket (My intake hose clamps required a 7mm to undo, but yours might be 8mm)
- 5mm Hex Key or Socket (For Thermostat Bolts)
- 6mm Hex Key or Socket (For Intake Support Bracket Bolts)
- Digital Multimeter or Ohmmeter (DMM)
Optional:
- Test Pigtail (If available. You can get the one I did Here!)
- Magnetic Tray (If you have one, great for holding the screws you remove!)
- Throttle Body Gasket
- Throttle Body Cleaner
- Rags
- New Thermostat (ECS sells a great kit, which is on sale for $152 right now as of 07/16/2020. You can grab this kit and all its goodies here.)

Procedure:
This is a pretty simple DIY and didn't take me long. I highly advise waiting until the engine is cool, since you'll be working near coolant hoses, and they can be pretty hot.

1. I began by removing the intake hose going to the throttle body. There is a hose clamp down behind the driver's side headlight assembly, and one on the throttle body.
If you don't have a lot of room or a 1/4" extension, I recommend removing the headlight assembly for ease of access to the lower hose clamp.
Otherwise, it's pretty easy to access with an extension, provided that the screw is facing upward. Once you undo both hose clamps, wiggle the hose off the intercooler and throttle body, and set it aside.
You might have a little oil come out from the lower portion of the hose, this is normal, to my knowledge, so don't panic.

2. With the intake hose out of the way, it is pretty easy to get to the 4 bolts holding the throttle body to the intake manifold. Unscrew these with your 5mm hex, and set them aside.
Be careful, the throttle body has a bit of weight to it, so don't drop it when removing the final screw. Make sure to undo the connector under the throttle body, careful not to break the tab. A flathead screwdriver works great to undo these clips. I find that applying a bit of pressure on the connector, by pushing it in, while undoing the retaining clip makes the connector pop off easier. Keep the throttle body somewhere safe, they're expensive.
Now would be a great time to inspect your throttle body gasket for wear, and likely replace it if you haven't before.
You can also take this time to inspect your throttle body for wear and carbon buildup. There are a lot of great guides out there on cleaning a throttle body, and it doesn't hurt to do so. Just make sure to do a throttle body alignment (or TBA) when everything's back together!

3. There is a long bracket that helps support the intake manifold, bolted under the throttle body. Follow the bracket down to its lower end for the second bolt. You'll want to undo both of these with your 6mm hex.
If you haven't removed these before, be slow and sure that you have the hex inserted all the way, as to not strip the heads of the bolts. Mine gave with decent pressure but came out easily from there the first time I did this.
There is a little plastic slip that holds some wire to the bracket. You don't have to undo this clip and should be able to let the bracket hang with plenty of slack out of the way.

4. You're pretty much there! The thermostat should be pretty exposed now. Here is an image depicting several points mentioned so-far: Click Here!
It isn't quite exposed enough to replace in this state, so if you anticipate changing it, I highly recommend this DIY here as it is very detailed.
Go ahead and undo the connector on the thermostat. This is the connector that powers the thermostat heating element. When I first changed my thermostat, my first sign of bad news was lots of corrosion and some coolant in this plug. Hopefully, yours is clean.
If you have DMM probes capable of touching the pins in the female end of the connector, you can verify that the ECU is providing voltage to the heater by putting your key in the "on" position and checking for battery voltage at the connector.
It should be pretty close to your car battery voltage, to my knowledge.

5. If you have a test pigtail, you can set it up with some probe hooks for your DMM like I did in this image.
Now, connect this test pigtail to the thermostat heater, and test for resistance. If it's shorted, you'll get a very low resistance, close to 0 ohms. If open, you'll either see no measurement, or something very high like I did (5.5 Mohms in my case).
If you get an open circuit, make sure your connector is connected completely, as it needs to be seated properly to give the most accurate results. If there's still no readings, or resistance remains high, you're in the same boat I am.
If your thermostat reads something like 15 ohms (which I believe is around what it should be, but I'll update that number when I find out), there may be an issue with the female end of the connector, wiring, or ECU.
There isn't a fuse for this, to my knowledge, and the wiring from this connector goes directly to the ECU, so there aren't too many other components to inspect if you're confident your thermostat is operating correctly.

And that's about it! Unfortunately, the heating element isn't a serviceable item, even though it appears possible to remove. You'll need to replace your thermostat to remedy this issue.

Hope this helps!

[EuroxS4]

Glad you got it sorted!


Sent from my iPhone using Tapatalk

[CortlandCH]

As promised, here is (hopefully) my final update on this issue!
I finally got around to replacing the thermostat. So much easier the second time around!

Since I could not find information on it anywhere, I will document the nominal resistance value of the thermostat heating element here:
The map controlled thermostat heater resistance is approximately 15 ohms.

I measured the pins on the thermostat housing connector, and got around 15.1 ohms.
I would use that as a general baseline going forward for troubleshooting the 1.8T map controlled thermostat heater.

After replacing the thermostat, and properly bleeding air out of the system this time, the engine gets up to temp with no issues and regulates it perfectly.
No more CEL! Hooray!

Thanks again for everyone's help, and I hope this helps anyone in the future if they stumble upon this thread.