[montr]

My guess is that the crystallization on urea on top of the tank is from overfilling with DEF. There is a BMW SIB advising against overfilling, this may damage the system.

I had the tank replaced under warranty previously, dealer was doing the refilling then. I am now doing the oil change and urea refill. I only add 2.5 gallons at a time to the passive tank when the car tell me it is low.

[QUIKDZL]

Success! Many thanks to Abax, Yozh, bagman, and prior posts by Quasimodem and BB Cuda. I couldn't have done this without this help.

You'll recall from my original post, my codes indicated a faulty DEF tank temperature sensor, and my dealer wanted over $2000 to replace the tank. I'm one who hates SES lights and wants systems to work as designed, so my original thought was to attach a new RTD to the tank, bypassing the faulty RTD, and allowing the system to work almost as designed, sensing tank temperature as opposed to urea temperature, but still activating heaters if my car ever saw temperatures in the teens. It probably won't, hence my reluctance to spend $2k.

From what I learned from you guys, I was able today to lower my exhaust system from its rear mounts, remove the heat shield attachment screws, and allow all this to dangle free and low. I then reached above the mufflers and removed the nuts that sandwich the urea tank to the removable trunk floor pan. I also removed the plastic nut holding the battery cable tray at the right front of the tank. Next, I removed the trunk floor pan, exposing the top of the active DEF tank. Everything was dusty, but otherwise clean.

I separated the electrical connectors and found the problem with a multimeter. The temperature sensor (RTD) was, indeed, faulty. When I probed it, it would read a reasonable number for a split second, and then show open. Repeated several times to try to understand what was going on, and it was consistent. I then probed a new RTD and it simply read a reasonable number. So the sensor was definitely bad. Apparently, the voltage from the meter would cause the circuit to open shortly after being applied. This was what the computer was seeing, as well.

So I got creative here and fished an 18 AWG pair between the cable bundle containing the OEM RTD wires and its grommet. Using silicone-filled direct-burial-type communication butt splices, I spliced a tiny new RTD to the wires and allowed it to dangle about 2/3 down into the urea, and then buttoned everything back up.

SES light and codes are gone! System works as originally designed, and I spent only $2 as opposed to $2000. If anyone else wants to do this, the RTD was purchased from Digi-Key online (p/n BC2646-ND). The only thing I would like to do differently would be to get an RTD with 12" leads. This one only has 2", so a splice was required in the tank.

[335dlci]

Any pictures from your DIY?? this fix will help those out of emission or CPO warranty

[Whitbread]

On the VW '14 passats, their urea temp sensors go bad all the time, a 10Kohm resistor soldered in the harness is the fix.

[QUIKDZL]

Originally Posted by 335dlci

Any pictures from your DIY?? this fix will help those out of emission or CPO warranty

Sorry...no photos. Prior posts by Quasimodem and BB Cuda took care of that. Any photos that I would have taken would not have provided any more information. You can see the RTD I used by searching for the part number given on their website.

[QUIKDZL]

Originally Posted by Whitbread

On the VW '14 passats, their urea temp sensors go bad all the time, a 10Kohm resistor soldered in the harness is the fix.

I agree this will work, but you are fooling the car into believing it is always 77F in the urea tank, so no protection from freezing. If you live in a warm climate like I do, that's no problem for you but could be a nightmare for a subsequent owner. I don't know about the VW, but the Beemer compares urea tank temperature against ambient. If they are far apart, plausibility errors pop up. Soldering in an RTD is just as easy.

[Omnomigon]

Holy crap he fixed it. Somebody get this guy a promotion.

[ninja_zx11]

Originally Posted by QUIKDZL

Success! Many thanks to Abax, Yozh, bagman, and prior posts by Quasimodem and BB Cuda. I couldn't have done this without this help.

You'll recall from my original post, my codes indicated a faulty DEF tank temperature sensor, and my dealer wanted over $2000 to replace the tank. I'm one who hates SES lights and wants systems to work as designed, so my original thought was to attach a new RTD to the tank, bypassing the faulty RTD, and allowing the system to work almost as designed, sensing tank temperature as opposed to urea temperature, but still activating heaters if my car ever saw temperatures in the teens. It probably won't, hence my reluctance to spend $2k.

From what I learned from you guys, I was able today to lower my exhaust system from its rear mounts, remove the heat shield attachment screws, and allow all this to dangle free and low. I then reached above the mufflers and removed the nuts that sandwich the urea tank to the removable trunk floor pan. I also removed the plastic nut holding the battery cable tray at the right front of the tank. Next, I removed the trunk floor pan, exposing the top of the active DEF tank. Everything was dusty, but otherwise clean.

I separated the electrical connectors and found the problem with a multimeter. The temperature sensor (RTD) was, indeed, faulty. When I probed it, it would read a reasonable number for a split second, and then show open. Repeated several times to try to understand what was going on, and it was consistent. I then probed a new RTD and it simply read a reasonable number. So the sensor was definitely bad. Apparently, the voltage from the meter would cause the circuit to open shortly after being applied. This was what the computer was seeing, as well.

So I got creative here and fished an 18 AWG pair between the cable bundle containing the OEM RTD wires and its grommet. Using silicone-filled direct-burial-type communication butt splices, I spliced a tiny new RTD to the wires and allowed it to dangle about 2/3 down into the urea, and then buttoned everything back up.

SES light and codes are gone! System works as originally designed, and I spent only $2 as opposed to $2000. If anyone else wants to do this, the RTD was purchased from Digi-Key online (p/n BC2646-ND). The only thing I would like to do differently would be to get an RTD with 12" leads. This one only has 2", so a splice was required in the tank.

Good job!! but sorry i didn't get it properly.Is the thermistor and its wires touching urea solution?

[QUIKDZL]

Yes, it is submerged in the urea. The OEM RTD (it is a resistance temperature detector (10kohm at 77F) and not a thermistor or thermocouple) as I understand it, is attached to the bottom of the lowest (reference) level sensor leg, so it is submerged as well. In my case, the waterproof splices are much larger and heavier than the RTD, and they all simply hang in the urea about three inches above all the stuff at the bottom. No chance of interference with the OEM stuff.

[ninja_zx11]

Originally Posted by QUIKDZL

The OEM RTD (it is a resistance temperature detector (10kohm at 77F) and not a thermistor or thermocouple)

Please don't take it in a negative way.Just clarifying.Really admire your DIY fix!!

Little confusion. Did you use digikey part no. BC2646-ND? If yes then its a NTC (negative temperature coefficient) thermistor.But they function in a similar way.

Is there any metal surface in contact with urea except the sensor or wire insulation?Asking because urea is very corrosive and will eat away any bare metal surface.

[QUIKDZL]

Thanks for validating my post. You are right and I stand corrected. I began my search on Digikey.com for an RTD, but I guess my quest for the best form factor led me to purchase a thermistor. Thankfully, it works--at least for now in moderate temperatures.

The sensor is sealed and the splices are waterproof (silicone-filled insulation-piercing 3M), so any metal components theoretically will not get wet.

[ninja_zx11]

Originally Posted by QUIKDZL

Thanks for validating my post. You are right and I stand corrected. I began my search on Digikey.com for an RTD, but I guess my quest for the best form factor led me to purchase a thermistor. Thankfully, it works--at least for now in moderate temperatures.

The sensor is sealed and the splices are waterproof (silicone-filled insulation-piercing 3M), so any metal components theoretically will not get wet.

Thanks for the clarification!!

[QUIKDZL]

I think I just found the perfect temperature sensor to replace/bypass the one in the urea tank (and yes, it is a thermistor). I ordered four as they are $4 each. Next time I take the trunk apart (which will be soon since I just ordered a water/meth injection system), I'll replace the one I just installed with this one so I won't have a splice submerged in the urea.

https://www.adafruit.com/products/37...FZEXHwodYEoMyA

[ninja_zx11]

This one looks good as its made from stainless steel and can be used with urea solution.Its 20k ntc thermistor.

http://www.vernier.com/products/sens...nsors/tmp-bta/

[QUIKDZL]

So I took my trunk apart again yesterday to begin my meth injection install (I'll start another post about that), and I took the opportunity to install my new 10k thermistor with the longer leads...and this time I took some pictures.

First photo shows the new thermistor in place. Next is a close-up of the tiny thermistor and the active tank internals. Third is a close-up of the splice bypassing the original thermistor. Finally, the last photo shows it all buttoned up. No more Service Engine Soon light!

[dixy2k]

Anybody around Los Angeles that can do this for me?

I need to reregister the car in two weeks.

[QUIKDZL]

I can send you a thermistor if you like. It's about a two hour job.

[dixy2k]

Do you need to clear the codes afterwards?
Also, not sure if that's my problem.
The DEF tank is full and it's been like that for months.

[QUIKDZL]

My codes were 0046F9 and 004D32, active temperature sensor signal interruption and low tank temperature to ambient temperature plausibility, respectively. I cleared them with Carly. I believe that, once triggered, you will have to clear them to make them go away (at least I did).

[dixy2k]

I guess that's what I got but different codes.
A forum member translated my codes and they sound like yours. Not 100% sure cause it's all in German.

[dixy2k]

On this thread

http://www.e90post.com/forums/showth...1212376&page=2

[QUIKDZL]

Wow...you really are in a different world than I am! We don't have emissions tests, and so all that readiness stuff is news to me. It sounds like my advice to clear the codes would be bad for you regarding readiness. Still, you likely need to eliminate the temperature sensor error then work on your readiness codes.

Another solution might be to send your ECU to BPC in Raleigh for his new SCR coding mod. It seems like a legit way to avoid codes while keeping the system functional. There's a recent thread about it on here.