BMW M5 Forum and M6 Forums banner

1 - 19 of 19 Posts

·
Registered
Joined
·
1,163 Posts
Discussion Starter #1 (Edited)
It seemed like it would be such an easy fix!? lol

Here's the story: I recently purchased a 01 M5 with some minor front bumper damage (PO ran into a parking block). The Ambient Temp Sensor was ok and worked perfectly when I bought the car. Somewhere during the 900 mile trip home with the car, the sensor's little "cubby" worked it's way loose from the splash shield and the sensor made contact with the tire at 80mph+. As you can imagine, the sensor and electrical plug were ripped off. The temp display on the dash read "-40 F" and the A/C stopped working.

I stopped and made a road-side repair to get the A/C going again. I shorted the two leads together, wrapped it up with electrical tape, and zip-tied it up and out of the way. The temp display now read "+122 F", the A/C worked better, but "Max A/C" wouldn't work and the Auto Fan speed would never move off of a fairly low setting (4 blocks from off). Regardless, I could manually run the HVAC and have A/C, so it got the job done.

Among other restoration parts I ordered from ECS Tuning, I ordered their Temp Sensor Repair Kit: BMW E39 M5 S62 5.0L Search SiteSearch ES2635473 - 65816905133KT - Exterior Temperature Sensor Repair Kit - ES#2635473

I figured this would be a piece of cake, but……I'm still reading "+122 F" on the cluster. BUT…"Max A/C" and Auto Fan control both work with this sensor plugged in, so the symptom has changed slightly by installing the new sensor.

Here's what I've done so far:

- Inspected all wiring. No shorts, no opens, etc. If I unplug the sensor, I immediately read "-40 F". As soon as I plug the sensor back in…122.
- Read resistance of the sensor. At approximately 90F ambient, I get about 1.25k ohms and holding the sensor in front of my open garage freezer door (50-60ish F), I get 1.55k ohms.
- Activated the "secret" OBC mode and read "ATemp". It reads +50.0 C with the sensor plugged in.
- I put the end of the sensor into a cup of ice water. After a few minutes, the ATemp dropped steadily to +22.5 F and the cluster display dropped from 122 to 72. Took the sensor out of the ice water, and ATemp very quickly returned to +50.0 and the cluster is slowly working it's way back up with drive time. It was at 119 currently, but I expect it will hit 122 sometime tomorrow.

I tend to think I have a bad sensor, but I'm reluctant to blame ECS without some proof. Does anyone know the proper resistance range for this sensor?

Any other suggestions? I've searched and found conflicting info regarding "calibration" for a new sensor, cluster resets, etc but haven't really gone down those paths just yet. My thought is that the cluster will never read the right temp until ATemp reads a realistic value.

Thanks in advance!
 

·
Registered
Joined
·
10,116 Posts
There was a post within the last month on this problem with all the specs you are looking for. Use the boards advanced search and add user name 68FB as he added what you are looking for.
 

·
Registered
Joined
·
10,116 Posts
I searched for you, bored. He did not give the specs but here is how it works.
The sensor is a negative temperature coefficient thermistor which means its resistance drops as the temperature increases. It reads from -40C to +50C. Since you are reading temps too high, almost at the upper limit, the resistance in the circuit between the DME and the sensor is too low. It sounds like you have tried a new sensor with no improvement. So you have a short to ground in the signal wiring between the DME and the sensor.

What voltage do you read at pin 2 of the harness connector at the sensor with the sensor disconnected?
This is an older thread.
The sensor is a simple negative temperature coefficient thermistor which means as the temperature goes up, the resistance goes down. It sounds like the shop knows what the resistance should be at the ambient temp the sensor is seeing. (You might want to confirm that.) If they are measuring that resistance across the two pins of the disconnected connector at the cluster, with neither pin shorted to ground, then it does sound like it must be the cluster.

They could check the cluster by connecting a variable resistor across the cluster connections and adjusting it to various resistances to see if that gives good readings on the dash.

Any chance the sensor wiring got shorted to a 12V supply wire during the accident? That may have blown the cluster electronics.
Hmm..
My cluster did do some funny things directly after the accident, tire pressure gauge light came on, some other lights came on but then I turned it off and on then it was ok..
Everything else functions fine.. Can't remember if it was -40 or +122 the whole time after. Something is awry. Insurance should still cover the cluster if it was from the accident but I don't want a new cluster if that isn't my problem!
I'm going to go over there in an hour and cut the new wiring they did and see if it gives me a -40 when I hold the 2 wires separately.
If it does that then the issue is from under the wiring.. if it stays at +122 then its gotta be something above where I cut..
I am guessing you fried your cluster. You can easily check the sensor by using a volt meter on it and seeing if there is a resistance change when you dip it in hot water.

Here is the second thread.
http://www.m5board.com/vbulletin/e39-m5-e52-z8-discussion/293921-ambient-temp-sensor-122-after-replacement-need-advice.html
 

·
Registered
Joined
·
1,163 Posts
Discussion Starter #4
Thanks! I also remembered reading that post from 68FB, but couldn't manage to track it down.

I did eventually find two bits of info online last night that may point to a bad sensor. I'll stop by Radio Shack today to pick up some small resistors to confirm.

Thanks again!
 

·
Registered
Joined
·
1,163 Posts
Discussion Starter #5 (Edited)
Update…

I found some info that the "BMW Ambient Temp Sensor" (although not E39-specific) is a K164NK004.7 thermistor. The blog below has a nice chart which shows several resistance/temperature points.

Converting BMW temperature sensor resistance to degrees Celsius with an IC - jaroslavklima.com

The one point that is relatively easy to reproduce is 4700 ohm / 25 C (77 F). I stopped by Radio Shack this morning and picked up a $1.50 pack of 4.7K (4700) ohm 1/2-watt resistors and substituted one for the questionable temp sensor. Sure enough, ATemp now reads 27.5 C and the cluster display shows 81 F. It reads a little higher than I expected, but there's several reasons why that might be - added resistance due to wire length or crimp connectors, component tolerance, etc. Regardless, it shows that my brand new sensor is indeed bad as it's reading 1.5k ohms at roughly the same temperature. Using the chart, 1.5 ohms would put me right at 50 C/ 122 F, so no surprise that's what the cluster was showing.

Cliffs Notes: I received a faulty sensor from ECSTuning which coincidently gave the same temp reading as I had with no sensor. If you ever need to test your ambient temp circuit, substitute a 4.7K ohm resistor in place of the sensor and it should read ~25 C/ 77 F, give or take a few degrees. This should allow you to isolate a bad sensor vs. a wiring issue.

Thanks for the help and hope this helps someone else in the future!
 

·
Registered
Joined
·
4,488 Posts
Update…

I found some info that the "BMW Ambient Temp Sensor" (although not E39-specific) is a K164NK004.7 thermistor. The blog below has a nice chart which shows several resistance/temperature points.

Converting BMW temperature sensor resistance to degrees Celsius with an IC - jaroslavklima.com

The one point that is relatively easy to reproduce is 4700 ohm / 25 C (77 F). I stopped by Radio Shack this morning and picked up a $1.50 pack of 4.7K (4700) ohm 1/2-watt resistors and substituted one for the questionable temp sensor. Sure enough, ATemp now reads 27.5 C and the cluster display shows 81 F. It reads a little higher than I expected, but there's several reasons why that might be - added resistance due to wire length or crimp connectors, component tolerance, etc. Regardless, it shows that my brand new sensor is indeed bad as it's reading 1.5k ohms at roughly the same temperature. Using the chart, 1.5 ohms would put me right at 50 C/ 122 F, so no surprise that's what the cluster was showing.

Cliffs Notes: I received a faulty sensor from ECSTuning which coincidently gave the same temp reading as I had with no sensor. If you ever need to test your ambient temp circuit, substitute a 4.7K ohm resistor in place of the sensor and it should read ~25 C/ 77 F, give or take a few degrees. This should allow you to isolate a bad sensor vs. a wiring issue.

Thanks for the help and hope this helps someone else in the future!
Very nice bit of troubleshooting. Thanks for the info.
 

·
Registered
Joined
·
1,163 Posts
Discussion Starter #7
The saga continues…. :)

I contacted ECSTuning a few days ago and they were extremely helpful. No questions asked and no issues having a replacement sensor shipped out. They also told me I didn't have to return anything, which was nice seeing as I'd already installed all the wiring, connectors, etc. I was (and still am) very happy with their customer service and willingness to make this right.

However, the replacement sensor arrived today: 65816905133 - Outdoor Temperature Sensor / Probe - ES#2635648 …..and it reads the same resistance as the first one. Arrgh! I suspect they have a bad batch of sensors. I contacted ECSTuning again, and they are forwarding my issue to the parts dept for a resolution. Since it's the weekend, they'll get back with me on Monday so we'll see…
 

·
Registered
Joined
·
4,488 Posts
The sensor is not hard to get to. I could check the temp-resistance curve of mine to confirm its the same as the curve in the link you attached if you want.
 

·
Registered
Joined
·
1,163 Posts
Discussion Starter #9
If you wouldn't mind, that would be great! Thank you!
 

·
Registered
Joined
·
4,488 Posts
I got the following results so the temp-resistance curve in the link is correct.

0C 15.5 kohms (iced water)
6C 11.9
15C 7.5
25C 4.8
35C 3.1
45C 2.08
50C 1.75

The differences between my curve and the one in the link at lower temps will be because of the huge decreases in resistance with small changes in temp showing up the inaccuracy of my temp reading, maybe +/-1 deg.
 

·
Registered
Joined
·
10,116 Posts
Have you tested your new sensors in water to see if there is movement? Simple fix might be to install a resistor in the circuit with the sender. At least until you get the proper sensor, the car will be functioning.
 

·
Registered
Joined
·
1,163 Posts
Discussion Starter #13
I used ice water with the new sensor and it does indeed move. However, both sensors return to <1.6k ohms at normal room temp.

It turns out that my Radio Shack 4.7k ohm resistors are actually 4.58k ohm, so that explains why I'm reading a little bit high as compared to the chart. I currently have one of those resistors in place so my cluster reads 81 degrees F, which isn't too far from the actual outdoor temp this week. As long as the A/C works right, I can live with a bogus temp display for a little while, but I appreciate the suggestion!
 

·
Registered
Joined
·
1,163 Posts
Discussion Starter #14
In case anyone finds this thread in the future…

ECS Tuning let me know that the entire batch they had in stock from that supplier was faulty. The alternate brand (Febi) was on backorder, so they substituted a OE BMW sensor at no charge. It arrived last week and I got it installed and it works great. Not surprisingly, the resistance readings from the BMW sensor follow the readings mentioned above.

Again, in spite of the problems with the OEM parts, I'm very happy with ECS Tuning. Things like this happen sometimes (particularly with aftermarket parts) and what really matters to me is how the situation is handled and eventually resolved. Throughout this entire adventure I've not paid a dime more than my initial purchase price and I was never asked to send any parts back, nor cover shipping charges. I can't ask for much more than that!
 

·
Registered
Joined
·
1,515 Posts
In case anyone finds this thread in the future…

ECS Tuning let me know that the entire batch they had in stock from that supplier was faulty. The alternate brand (Febi) was on backorder, so they substituted a OE BMW sensor at no charge. It arrived last week and I got it installed and it works great. Not surprisingly, the resistance readings from the BMW sensor follow the readings mentioned above.

Again, in spite of the problems with the OEM parts, I'm very happy with ECS Tuning. Things like this happen sometimes (particularly with aftermarket parts) and what really matters to me is how the situation is handled and eventually resolved. Throughout this entire adventure I've not paid a dime more than my initial purchase price and I was never asked to send any parts back, nor cover shipping charges. I can't ask for much more than that!
Thanks! Just called and they will ship a new one on Friday. I had to pay 7.11 for shipping though. Again thanks for posting the info.
 

·
Registered
Joined
·
1,515 Posts
Recieved and replaced. Ambient Temp is working as it should. Thanks again for posting this. I would have never known as ECS Tuning is not informing customers. Even so I will continue to purchase from as they gave me no issues.
 
1 - 19 of 19 Posts
Top