New CTS, temperature readings way off

[pr0xZen]

I tried posting about this on another forum, but it seems american dominated, and it seems mentioning the 2.4 (1999 A6 quattro avant) tends to scare them away, even though its almost identical to the 2.8 in most areas. So hopefully you guys can chime in on this, and pardon me for not completely rewriting the whole post.


I was having warm start issues, I had a code for a faulty CTS, and the cluster needle would stay at bottom. The code (16502 / P0118) would return instantly when cleared, so I decided to change it (059 919 501, square plug type). Viewing on the cluster, it will get fairly fast up to 90'C, and remain aorund there. Code has not showed up since.


However, I'm still having some warm start issues (cranks fine but takes ages to fire up), and fuel trim is heavily adding fuel (+15-25%) regardelss of cold or warm engine, especially on idle.


After hours and hours and hours of crawling forums, I was still not quite sure about the temperatures. So I left it sitting, with rather little temperature change, for 24 hours. I then proceeded to go through all measuring blocks (VCDS) looking for temp values, on both engine and cluster (as they both share the CTS housing, but separate circuitry), without starting the engine. A temperature control test with a thermometer, showed 7.0'C ambient and 8.0'C for the coolant in the expansion tank.


I chopped together the readings from screenshots. When I got to engine group 11, the coolant temp was jumping back and forth between -7.5'C and -6.0C, it seemed to stabilize back on -7.5'C right after the screenshot was taken (might be down to minute voltage changes?).


As you can see from the values, intake air temp and ambient temp is pretty much on the money (well, close enough to not be a malfunction). But While the actual coolant temp was 8.0C, the engine reported -7.5'C and the cluster reported 30'0'C.


Any ideas? And could this be related to the fuel trim / warm start issues?

[Doctle Odd]

Genuine CTS?

[ceilidhalfie]

Looking at the above control blocks, I would say the new G62 sensor is also faulty. Can't think where else the EMS is getting this temp from. Have you tried putting the old sensor back in just to see if you get the same readings? As other poster asked, is it genuine Audi? I am now on my 3rd replacement (last one genuine Audi) in my 2002 Allroad. The first 2 lasted only a matter of months (symptoms were temp on gauge dropping to <75% then recovering to 90%). So far genuine G62 is fine.

[pr0xZen]

I haven't pulled the new CTS yet (weather is horrible), and I don't have a picture of the stamping. But given low price (ebay), i'm starting to assume its a knock-off. The seller do push a 2-year warranty on it, so I intend to hold them to that. Just a shame outbound shipping (with tracking, as pretty much all ebay returns require) from Norway is absurdly expensive - looking at ~£35 for that small piece of headache in a small padded envelope.

After so many years - are there no aftermarket suppliers that have conjured up a replacement with a reasonable degree of reliability? Because genuine OE price is (here anyway) about £50, and VAG CTSs seems to be a very common failure. One might be lead to belive that doing a proper aftermarket part, would generate some hefty sales.

[BRAINFART ] Maybe even VAG themselves cannot manufacture them with a dependable quality/tolerance output, and an atypical large portion of production runs gets discarded through QC? [/BRAINFART ]

[ceilidhalfie]

Hi. I know the problem - I have a Mercury outboard down in France with parts new/second user supplied from USA and it costs a packet for shipping. I have had a look on eBay UK and have come up with a NEW genuine Audi CTS for you at £14.95 plus FREE shipping to UK. Have a look at 4 pin coolant temperature sensor VW Audi 059919501 New genuine VW part | eBay and see if this is what you need.
If you wish, you can get it sent me and I will forward on to you (in Norway?? - can't be that expensive to post such a small package).
Alternatively, I can first find out the cost of tracked postage, you pay me the full amount, I get it delivered to me and forward on.
Let me know if you wish to take up this offer and I will pm you my email address to discuss further.

[pr0xZen]

I very, VERY much appreciate the very friendly gesture, and thank you quite a generous bunch - but I've already ordered another one off Ebay, from Poland. I found two "auctions" - the one you linked in UK (£14.95, £21.94 incl. shipping to Norway), and one from Poland (£13.95, free int. shipping). Both high-volume (12k+ rep) sellers. Contacted the seller in Poland (£8 is £8), said the part is genuine, new Audi stock, with two year warranty, comes in official packaging etc. Also noted that, for my peace of mind - since the defective one I pulled was the same, it should be easy to spot a knock-off when examening the small details off engraving/stamping, molding etc. - and if I still felt uncertaing, I'd get a refund (sans shipping) if returned unused.

So now I just wait. Meanwhile I did a new log with warm engine first at idle, then climb and hold at 2500, 3500, 2500, then idle again. Something seems very much wrong with my fuel trims. Log is here (.CSV), feel free to have a look / playback with VCScope if anyone can have a look at it. It might be down to the CTS, but I personally I suspect fuel or vacuum (no change in idle or fuel trim behaviour when pulling dipstick).

EDIT: Omg. £13.95 for (presumably) genuine Audi, new CTS - free international shipment from Poland to Norway (NOT within EU). And I just got a shipment notice from DHL Express :O

[pr0xZen]

Got the new CTS, genuine deal. Bench (kettle) tested both outputs, both working fine and withing spec values across the range. With Vag-Com / VCDS, temps at engine is now reading within 2-3 degrees celsius of control, so that part is good. But cluster temp is still stuck at 30.0'C. I even tried both an improvised rewiring of the sensor/connector - and disconnecting the CTS all together (!) - cluster temp still reads 30.0'C off coolant temp (-2 outside, sitting 48 hours). Cluster Output test shows the guage working - sweeping full and stopping dead at 90'C (straight up).

Any ideas? I've been reading all over, but not seen anything on this as of yet.

[pr0xZen]

I'll reply to myself, for future forum reference. The cluster "bottoms out" at a reading of 30.0C . Thermistors increase resistance as it gets colder - thus any coolant temp lower than 30C, or even a wire break, will show 30.0'C on the cluster (measuring block). Can be tested with a ~110-120 ohm resistor across the G2 side terminals of the CTS plug, the guage should show ~90'C (same with Vag-com/VCDS on measuring block @ cluster).

[Flash2]

Thanks for posting that info. It's easy to see how the diag tool and lead you astray if your not careful.

Well done for coming up with your resistor test to prove the circuit is actually working correctly.

Jim.