clif Posted July 17 Report Share Posted July 17 Good day all i am kindly asking for advice or assist on my MK7 gti, its having an overheating issue, at 1st the temperature was going over 90° then water boiling, i replaced the complete water pump with thermostat, then now it stays on 90° temperature however coolant doesn't circulate , i ran a live data With VCDS, and noticed the coolant temperature outlet on radiator reads -48° which the Fans doesn't kick in. I then replaced both sensors on the head and on the radiator outlet also tested supply on the sensors . But no circulation and I still reads -48° on the radiator outlet. And the other thing I noticed yesterday i don't get supply to coolant level sensor but it was able to tell when the level is low, so other than the ecm not sure where the supply comes from, so not sure if its the one that can causes the thermostat not to open.. Quote Link to comment Share on other sites More sharing options...
dv52 Posted July 17 Report Share Posted July 17 @clif:Hi. OK- lets start by discussing the technical aspects of your post. Because you haven't entered your country into the forum bio - I must assume that you are a loyal subject of King Chuck. If my assumption is correct, this Golf mk7 GTI is a Rest-of-World model and it has a EA888 Gen 3 2.0 TSI engine. Likely the engine will have a VW designation CHHA, or CHHB - I'm not sure which, but it doesn't matter for this reply. I suggest that you confirm my understanding with your diagnostic device to ensure that I'm not telling porkies in the rest of my words below! On these engines, the Radiator outlet coolant temperature sender is called G83 in the VW wiring diagrams. This sensor is a NTC thermistor and its wired directly to the Engine Control Unit (ECU), NTC is an acronym for Negative Temperature Coefficient- which means that as the temperature rises, the resistance between the two sensor pins decreases. The 2 x sensor pins terminate on the 91 x pin connector on the ECU at pin #29 & pin #49. As I said, these are direct connections without the 2 x wires terminating on any intervening coupling points between the sensor and the ECU. Incidentally, the reason why the temperature gauge reads the usual 90°C on this car is because the signal that drives the dashboard gauge comes from the engine‑head temperature sensor - not G83. I suggest that you grab a multi-meter and start by confirming that the replacement G83 is operating correctly. As I said, this is a NTC thermister - so remove the sensor plug and measure electrical resistance across the 2 x pins on the sensor connector for a cold and hot engine. If the sensor is OK - you should see the resistance fall as the water temp rises. Typical readings for G83 on a CHHA, or CHHB engine are as follows: Cold engine (with the car sitting over-night) 3.5–5.0 kΩ at ~10 °C 2.2–3.0 kΩ at ~20 °C 1.5–2.0 kΩ at ~30 °C Warm engine (thermostat open, radiator flowing) 300–400 Ω at ~90 °C 250–300 Ω at ~100 °C If G83 is operating correctly according to the tests above - eyeball the condition on the mating wires/pins on the loom-side of the sensor connector. Finally, also confirm the integrity of the 2 wires that terminate on the ECU (use my ECU pin numbers in the early part of this reply). Don PS: If the signal from G83 is faulty on this car - I assume that your diagnostic device has recorded an error-code. Is this so? fastandy and clif 2 Quote Link to comment Share on other sites More sharing options...
clif Posted July 19 Author Report Share Posted July 19 Hi @dv52 Thanks for the informative reply, I will save the comment to my notes... Anyway the model of the engine is CHH0, so before I did test the sensor and wiring only on the plug since well I did not have diagram to the ECU, i could get the supply from ECU around 5v, then I was getting around 2.1k~ at engine cold, however wasn't able to check at warm engine as the was no flow on the radiator outlet, as it look like the thermostat wasn't opening still with the new set.. But long story short: I found the expansion tank CAP to be faulty it had some small cracks inside, which was not able to regulate the pressure, after replacing the CAP ,the was circulation to the radiator. If I may ask are you able to share the diagram if possible cliffmotsamai@hotmail.com Quote Link to comment Share on other sites More sharing options...
fastandy Posted July 19 Report Share Posted July 19 I love the intracacies of the mk7 Ea888 gen 3 cooling system.. Some good info above - as usual - I would expect no less of Don. The minus 48C is a specified radiator outlet value. I can't remember why it appears to be an error but it doesn't affect the proper operation. You need to look at rad outlet temp - actual value. One way to check the cct is operating is to unplug the g83 and see if a full scan shows open cct on the temp 2 cct. Of course, you can also prove connectivity back to the pins. If your faults are all resolved this is just sign posting for other searchers... The faulty cap would cause potential boiling at operating temps but I'd be most surprised if it is the cause of a circulation issue. Fwiw the coolant expansion tank should be replaced as well if it is original. They fracture around the level pins and can cause coolant migration all the way back to the cluster or ecu. Which is not fun. Quote Link to comment Share on other sites More sharing options...
dv52 Posted July 20 Report Share Posted July 20 20 hours ago, fastandy said: I love the intracacies of the mk7 Ea888 gen 3 cooling system. The minus 48C is a specified radiator outlet value. I can't remember why it appears to be an error but it doesn't affect the proper operation. Andrew: In my many years of observing carbon based life units - I have come to categorize the species into two opposite extremes; those that love simplicity and those that love complexity!! Clearly you are in the latter category - I'm honored to be in such rarefied company!! As for -48°C, AFAIK this is the lowest temperature that the Ohms-to-Degrees conversion algorithm will accept. I'm not sure - but I sense from your words that you are saying that OP's problem may be purely an aberration in the Live Data reading, rather than a true fault. You may well be correct (I don't know) - but because G83 is a NTC sensor - the resistance between the 2 x pins at the ECU connector for an open-circuit in the sensor circuit would result in a reading of -48°C!! Don Quote Link to comment Share on other sites More sharing options...
dv52 Posted July 20 Report Share Posted July 20 21 hours ago, clif said: Anyway the model of the engine is CHH0, Cliff: Without any offense intended - I have never heard of a "CHH0" engine! As I indicated in my earlier reply, because you haven't filled-in your location on the forum bio, I don't know your country - but I'm not aware of VAG ever installing a CHH0 engine on a MQB platform car in any of the world markets. Even if the trailing "0" has been misinterpreted as the alpha letter "O" - I'm still mighty confused! Admittedly, there's lots of this stuff about which I'm totally ignorant and I ALWAYS delight in learning new things, so maybe the error is mine (in which case, I apologize) - but how is it that you believe that this car has a CHH0 engine? Don Quote Link to comment Share on other sites More sharing options...
fastandy Posted July 21 Report Share Posted July 21 @dv52 Hi Don - Just for interest. Ponder this selective live data capture which was taken when my car was fully operational no faults and no cooling system issues (1 year after I replaced the electromechanical thermostat and water pump). What I was saying is -48C also seems to be produced in a working vehicle in the SPECIFIED value (not actual read value) for the rad outlet (temp 2 ntc). Yet, despite the crazy specified value, it does not cause any noticeable mis-operation and although I do not know for sure - I think it is like that from new. I suppose looking at it the other way - there may be no ECU reliance on Specified to Actual temp at Rad Outlet. There is of course management of the delta temp between engine and rad outlet sensors though as that is what the ECU uses to monitor cooling system effectiveness and will report a DTC fault is that drops out of expected ranges. Quote Link to comment Share on other sites More sharing options...
dv52 Posted July 22 Report Share Posted July 22 @fastandy Andrew: Excellent question - I like the way that you think! The equivalent Live-data item on VCDS is IDE00194 - Coolant temperature at radiator output: specified value. I needed the leading IDE00194 number to search through the records in my MQB database - although I can't find a Golf mk7 GTI for UK with a CHHA/B engine - ALL instances where a car has a EA888 Gen 3 2.0 TSI engine, the value for this channel =-48.0 °C. So, I agree with your suspicion that " it is like that from new". Incidentally, I'm still not sure what engine is in OP's mk7 GTI (or if this is even a UK vehicle) As I said in a previous reply,-48.0 °C is probably the lowest temperature that the conversion algorithm recognizes. My hunch is that -48 °C is not a physically meaningful coolant temperature on any EA888 Gen‑3 engine when applied to a "specified value". It's just an algorithmic placeholder, not a thermal target that is used in a practical sense. I suspect that this means that whenever ANY "specified value" on ANY engine =-48.0 °C, the ECU is not regulating the outlet temperature at the location of the sensor. In these instances, it's likely that the ECU is only monitoring the temperature (radiator‑outlet temperature in the G83 case) - but the ECU does not actively command and control a target temperature for that sensor. It also probably means that the engine cooling system is driven either by another temperature sensor, or it's entirely thermostat‑driven - not ECU‑driven. Of course if the same -48.0 °C, value appears on the "actual value" - the explanation is as in my previous reply about what happens when NTC sensors open-circuit! Don fastandy 1 Quote Link to comment Share on other sites More sharing options...
clif Posted July 26 Author Report Share Posted July 26 On 20/07/2026 at 04:42, dv52 said: Cliff: Without any offense intended - I have never heard of a "CHH0" engine! As I indicated in my earlier reply, because you haven't filled-in your location on the forum bio, I don't know your country - but I'm not aware of VAG ever installing a CHH0 engine on a MQB platform car in any of the world markets. Even if the trailing "0" has been misinterpreted as the alpha letter "O" - I'm still mighty confused! Admittedly, there's lots of this stuff about which I'm totally ignorant and I ALWAYS delight in learning new things, so maybe the error is mine (in which case, I apologize) - but how is it that you believe that this car has a CHH0 engine? Don @dv52 im I south africa and the engine it is the one I mentioned Quote Link to comment Share on other sites More sharing options...
dv52 Posted July 27 Report Share Posted July 27 4 hours ago, clif said: @dv52 im I south africa and the engine it is the one I mentioned Cliff: Arrhhh.......... that's better - I suggest that you update your forum bio with your location to minimize future confusion. With no offense intended - I'm still convinced that VW never made a version of the EA888 engine with the engine code CH00. However, the fact that you are from South Africa clears the fog somewhat. As my very brief research into this matter suggests - South Africa has a few quirks in how engine codes are printed, recorded, and displayed in dealer systems. Apparently, in South African VIN/engine databases (eNatis, insurer lookup tables, some dealer DMS systems): CHH → sometimes stored as CHH0 CHHB → sometimes stored as CHH0 (because the “B” is dropped or misread) CHHA → sometimes stored as CHH. So, almost certainly - CH00 is NOT an official VW engine code — it’s an artifice the results from local database normalization practices that occur uniquely in your country. Bottom-line= this car has a CHHB, or CHHA engine - so the stuff in my earlier words apply Don Quote Link to comment Share on other sites More sharing options...
clif Posted August 23 Author Report Share Posted August 23 On 27/07/2026 at 02:28, dv52 said: Cliff: Arrhhh.......... that's better - I suggest that you update your forum bio with your location to minimize future confusion. With no offense intended - I'm still convinced that VW never made a version of the EA888 engine with the engine code CH00. However, the fact that you are from South Africa clears the fog somewhat. As my very brief research into this matter suggests - South Africa has a few quirks in how engine codes are printed, recorded, and displayed in dealer systems. Apparently, in South African VIN/engine databases (eNatis, insurer lookup tables, some dealer DMS systems): CHH → sometimes stored as CHH0 CHHB → sometimes stored as CHH0 (because the “B” is dropped or misread) CHHA → sometimes stored as CHH. So, almost certainly - CH00 is NOT an official VW engine code — it’s an artifice the results from local database normalization practices that occur uniquely in your country. Bottom-line= this car has a CHHB, or CHHA engine - so the stuff in my earlier words apply Don @dv52 thanks , yeah will do....... on the car license paper/disc the Engine number its CHH then six code numbers starting with 0. However just a feedback after all the diagnosting I finally got to resolve the issue, I was fooled by the filler cap for the expansion tank/coolant tank.replaced it then i was back on the road dv52 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.