Sudden high temperature without coolant loss, 1.4tdi
- Thread starter Robin_Cox
- Start date
thanks all ... advice, knowledge and sympathy in equal measures hugely appreciated by me as always in the same way that A2oc gives many requests for assistance their due merits with these threads.
I'm setting off at 1am .. Perth, Stirling, Glasgow, M74, M6, Tebay services for coffee, M5 Oldbury Asda fuel refill around 6, Gloucester services for breakfast around 7, Bristol, Taunton, Exeter, A30, over Dartmoor via Tavistock, there by 10-ish hopefully. Bedtime shortly.
I'm setting off at 1am .. Perth, Stirling, Glasgow, M74, M6, Tebay services for coffee, M5 Oldbury Asda fuel refill around 6, Gloucester services for breakfast around 7, Bristol, Taunton, Exeter, A30, over Dartmoor via Tavistock, there by 10-ish hopefully. Bedtime shortly.
Well that was fun.
Two things became obvious over the 545 miles.
1) the temp sensor - dashboard feed - is goosed. It maintained the temperature differential (5-10 degrees over the ECU numbers), but then as time went by there was the occasional flicker / blank reading which happened more and more often and then as we passed Tebay, even the temp gauge needle abruptly dropped to the end stop for a few minutes before reappearing. As the numbers were plausible I decided to trust the ECU temperature readout on colour DIS. Luckily I have another on order since yesterday to be delivered here and the tools / coolant to effect the swap during my holiday.
Edit after the event : temp sensor replacement resolved this issue.
2) Despite normal pump activity at idle and with some throttle, it is clear as engine load levels increase that at some point there is a restriction to the flow to the point where the heat cannot be effectively dissipated by the radiator, so the temperature then starts to increase above ideal - I have an idea what the problem is and will come back to this in a minute. However, by cranking the cabin heater up to full (3 miles after leaving home!), adjusting fan speed and driving with my window open an inch and the same on the passenger-side rear, it was then possible to drive according to ECU temperature reading on the cruise control, aiming to keep it around 86-90 degrees give or take load climbing hills. Accordingly instead of averaging 63mph or thereabouts we averaged around 53 with the journey taking just under 10 and a half hours instead of just under 9. Consequently I annihilated every fuel record I've done for the journey - 68mpg. Dundee to Birmingham alone used 26L of fuel instead of 36L for the refill. Conversely we were lucky that it was frosty from Scotland until somewhere in the Midlands, and temperatures gradually crept up in the cabin after sunrise as we were heading down the M5 as the vent outputs were hotter than the face of the sun .. just glad this wasn't the summer holidays!
So what do I think is the problem?
My hunch is that it is related to the DEFA heater installation. Not that this should be problematic in general. However, owing to my having forgotten to give the mechanic the print out from the installation instructions showing the orientation of the pipes and plumbing in of the pump to the oil cooler output for this engine (a workshop slot came available unexpectedly), I remembered that he improvised a bit putting the installation together so that it looks different to the recommended diagram when I did find the pdf document again. As I've barely used the car since, and when I did the ambient temperatures were low and the journeys short - the issues with cooling haven't manifested themselves hitherto. Now temperatures are higher and this was a proper drive, at motorway speeds this becomes obvious.
Edit. I'm taking this back. I've now located the pdf and the installation is actually how it is meant to be now that I've had a chance to look in daylight and warm conditions.
Edit after the event. All of this entire hunch was wrong and a red herring as various posts from people with more mechanical knowledge suggest over the next page or so as this is a minor circuit within the Tdi coolant system. TL;DR : the thermostat was not opening properly as some more observant people suggested in various posts. It was not the DEFA, and nor was it my mechanic's handiwork.
My solution for now would be to try to source an appropriate length of the right grade and diameter of coolant pipe from a local motor factor if I can't get an original one from TPS, and swap this in place of the two pipes flanking the heater for now using the existing spring clips, all at the same time as replacing the temperature sensor.
Does anyone know what diameter and grade of pipe is appropriate (equating to half of part number 11 running from oil cooler to temp sensor housing I think?) - the clips needed to secure the pipe are part 29 on this diagram (23mm x 12mm), and the tubing - with a 6N0 part number - is obviously not just an A2 part. Or am I barking up completely the wrong tree?
audi.7zap.com
Edit.
reading online it appears that EPDM tubing comes in various standard sizes, one of which is 23mm with the appropriate reinforcements - I have my calipers with me so I'm going to measure the pipes to confirm. It does appear to just be a case of releasing the spring clips and plumbing the pipe in to bypass the heater from oil cooler to the sensor housing .. what could possibly go wrong ?
Two things became obvious over the 545 miles.
1) the temp sensor - dashboard feed - is goosed. It maintained the temperature differential (5-10 degrees over the ECU numbers), but then as time went by there was the occasional flicker / blank reading which happened more and more often and then as we passed Tebay, even the temp gauge needle abruptly dropped to the end stop for a few minutes before reappearing. As the numbers were plausible I decided to trust the ECU temperature readout on colour DIS. Luckily I have another on order since yesterday to be delivered here and the tools / coolant to effect the swap during my holiday.
Edit after the event : temp sensor replacement resolved this issue.
2) Despite normal pump activity at idle and with some throttle, it is clear as engine load levels increase that at some point there is a restriction to the flow to the point where the heat cannot be effectively dissipated by the radiator, so the temperature then starts to increase above ideal - I have an idea what the problem is and will come back to this in a minute. However, by cranking the cabin heater up to full (3 miles after leaving home!), adjusting fan speed and driving with my window open an inch and the same on the passenger-side rear, it was then possible to drive according to ECU temperature reading on the cruise control, aiming to keep it around 86-90 degrees give or take load climbing hills. Accordingly instead of averaging 63mph or thereabouts we averaged around 53 with the journey taking just under 10 and a half hours instead of just under 9. Consequently I annihilated every fuel record I've done for the journey - 68mpg. Dundee to Birmingham alone used 26L of fuel instead of 36L for the refill. Conversely we were lucky that it was frosty from Scotland until somewhere in the Midlands, and temperatures gradually crept up in the cabin after sunrise as we were heading down the M5 as the vent outputs were hotter than the face of the sun .. just glad this wasn't the summer holidays!
So what do I think is the problem?
My hunch is that it is related to the DEFA heater installation. Not that this should be problematic in general. However, owing to my having forgotten to give the mechanic the print out from the installation instructions showing the orientation of the pipes and plumbing in of the pump to the oil cooler output for this engine (a workshop slot came available unexpectedly), I remembered that he improvised a bit putting the installation together so that it looks different to the recommended diagram when I did find the pdf document again. As I've barely used the car since, and when I did the ambient temperatures were low and the journeys short - the issues with cooling haven't manifested themselves hitherto. Now temperatures are higher and this was a proper drive, at motorway speeds this becomes obvious.
Edit. I'm taking this back. I've now located the pdf and the installation is actually how it is meant to be now that I've had a chance to look in daylight and warm conditions.
Edit after the event. All of this entire hunch was wrong and a red herring as various posts from people with more mechanical knowledge suggest over the next page or so as this is a minor circuit within the Tdi coolant system. TL;DR : the thermostat was not opening properly as some more observant people suggested in various posts. It was not the DEFA, and nor was it my mechanic's handiwork.
My solution for now would be to try to source an appropriate length of the right grade and diameter of coolant pipe from a local motor factor if I can't get an original one from TPS, and swap this in place of the two pipes flanking the heater for now using the existing spring clips, all at the same time as replacing the temperature sensor.
Does anyone know what diameter and grade of pipe is appropriate (equating to half of part number 11 running from oil cooler to temp sensor housing I think?) - the clips needed to secure the pipe are part 29 on this diagram (23mm x 12mm), and the tubing - with a 6N0 part number - is obviously not just an A2 part. Or am I barking up completely the wrong tree?
coolant hoses and pipes; flange; reservoir Audi A2 (A2) 2003 year Audi EUROPA 121038
coolant hoses and pipes<br/>flange<br/>reservoir Audi A2 (A2) 2003 year Audi EUROPA 121038 spare parts
Edit.
reading online it appears that EPDM tubing comes in various standard sizes, one of which is 23mm with the appropriate reinforcements - I have my calipers with me so I'm going to measure the pipes to confirm. It does appear to just be a case of releasing the spring clips and plumbing the pipe in to bypass the heater from oil cooler to the sensor housing .. what could possibly go wrong ?
Last edited:
Catnip64
A2OC Donor
This may help with the size of the pipe, it's a photo of the housing that attaches to the block onto which one of the pipes (No. 11) attaches, and indicates a diameter of 25mm.
https://www.buycarparts.co.uk/meat-doria/10513878
I can't help thinking though that this isn't the cause of your problems as these pipes are part of the secondary heating circuit, rather than the primary radiator/thermostat cooling loop, which is the one that is used to cool the engine. Unless the mechanic somehow linked both circuits together!
I agree that having the heater on will help in removing heat from the engine, but even if this circuit was blocked entirely, I still think the radiator would adequately control the overall temperature.
Other than the two outputs from the temperature sensor, is the engine showing any other signs of being too hot? Is it feasible that both outputs on the sensor could be giving incorrect readings?
Seems a strange one, but hats off to you in working on it whilst you are on holiday
https://www.buycarparts.co.uk/meat-doria/10513878
I can't help thinking though that this isn't the cause of your problems as these pipes are part of the secondary heating circuit, rather than the primary radiator/thermostat cooling loop, which is the one that is used to cool the engine. Unless the mechanic somehow linked both circuits together!
I agree that having the heater on will help in removing heat from the engine, but even if this circuit was blocked entirely, I still think the radiator would adequately control the overall temperature.
Other than the two outputs from the temperature sensor, is the engine showing any other signs of being too hot? Is it feasible that both outputs on the sensor could be giving incorrect readings?
Seems a strange one, but hats off to you in working on it whilst you are on holiday
My response to this one seems to not have posted - I've probably lost it without saving properly - apologies.This may help with the size of the pipe, it's a photo of the housing that attaches to the block onto which one of the pipes (No. 11) attaches, and indicates a diameter of 25mm.
https://www.buycarparts.co.uk/meat-doria/10513878
I can't help thinking though that this isn't the cause of your problems as these pipes are part of the secondary heating circuit, rather than the primary radiator/thermostat cooling loop, which is the one that is used to cool the engine. Unless the mechanic somehow linked both circuits together!
I agree that having the heater on will help in removing heat from the engine, but even if this circuit was blocked entirely, I still think the radiator would adequately control the overall temperature.
Other than the two outputs from the temperature sensor, is the engine showing any other signs of being too hot? Is it feasible that both outputs on the sensor could be giving incorrect readings?
Seems a strange one, but hats off to you in working on it whilst you are on holiday
Thank you for the housing diagram and measurements hint .. stupidly I'd forgotten until just now that I have two of them at home already.
I'm a bit vague on the coolant circuits ; am I right in thinking that there is one "inner" circuit that runs continuously while the thermostat is closed, and another "outer" circuit that incorporates the radiator once everything is warmed up enough to warrant its participation?
The engine isn't showing signs of being too hot only because I have caught it and put the cabin heater on asap .. I'm not good enough a mechanic to risk seeing what would happen if I didn't. However, one possible clue : the engine warm up time seems suspiciously rapid - it has been warming up (ECU coolant sensor channel numbers and hot air supply) in about the same distance as my other half's AUA (faster than my Webasto-heated car) recently where normally one could drive it to the city centre (about 15 minutes) and only just creep over 60 degrees.
Yes it is conceivable that the sensor has gone completely nuts on at least one channel and worth seeing what is happening by replacing with a new (different) one.
Last edited:
audifan
A2OC Donor
Yes I think you are probably spot on there about the block heater. I know why you wanted it fitted but unless installed as per the instructions you will not see benefits only issues.
Temp sensor could have been affected by the high coolant temp or lack of flow also check the wiring for damage, would also consider replacing the thermostat as you will need to replenish coolant now anyway.
Temp sensor could have been affected by the high coolant temp or lack of flow also check the wiring for damage, would also consider replacing the thermostat as you will need to replenish coolant now anyway.
PlasticMac
Admin Team
Do you have the DEFA Installation PDF with you?
Mac.
Mac.
PlasticMac
Admin Team
Or just the part number, and I'll download it from DEFA.
Mac.
PlasticMac
Admin Team
I don't think it's a block heater. If I'm right, it's plumbed into the oil cooler coolant circuit, to heat the coolant directly, with a separate, mains powered circulation pump. DEFA refer to it as a PTC/Safestart heater.
Apologies all round if I've got it wrong ...
Mac.
Yes, it is a mains-powered heater, exactly - plumbed into the oil cooler circuit. However, it doesn't have a pump. Model DEFA 411761, in the manual it is procedure 6 in location H (see attachment below). I'm must admit I'm a bit confused about the various circuits in the cooling system.
As I've commented above in an edit, I have misled you all a bit with my statement about the 'freestyling' of the DEFA installation - now that I've looked properly with the pdf and proper daylight it does appear to be in the correct orientation, so it is plausible that this is a red herring.
Conversely, if (as a temporary measure) it is possible to bypass the heater and everything returns to normal - then it would solve the problem.
The coolant levels haven't budged since departure this morning, and the expansion lid remained properly closed.
I have the replacement sensor(s) - (bought two just in case), so I'm going to swap it tomorrow - I will have time to warm it up and see if this fixes things or we need to go to plan B and find a means to bypass the heater.
Edit after the event : to make it quite clear, it wasn't the heater, but the thermostat that was not opening sufficiently.
Attachments
Last edited:
Proghound
Admin Team
I see Graham @audifan has suggested to replace the thermostat. I'd second that. It occurred to me that the while the 'stat is opening, it could be failing by not opening far enough. So it's OK at lower loads but at higher loads it can't get enough coolant into the radiator to remove the heat. Meanwhile the (rather smaller) cabin heater circuit is always available but this still limits the load you can put on the engine without it overheating.
PlasticMac
Admin Team
I was suspicious about the pump, but as there isn't one ...
The heater itself has no significant flow restriction (looking at the pic below). As long as the hoses at each end are in the original/right places, looks very unlikely to be the problem.
So, I'd say we're back to poor or no circulation through the rad, as the heater matrix has more effect than you'd expect, indicating it's the only flow path. Since you've seen evidence that the pump works into an open circuit, you'd go with the thermostat, as per @audifan and @Proghound.
Mac.
Attachments
thank you for your suggestions and thoughts along with those of @audifan and @Proghound on this thread - it is much appreciated.
The only difference I can see on the heater between the diagram and the installation is that the tube with the S-bend is not as tall vertically in the installation as it looks in the image, but I don't see how that could impact on current or flow - there are no constrictions.
In that case I have to consider the thermostat .. ironically I replaced that here on an early trip down around 3 years ago!
Worth noting that the pipes going to the radiator were hot yesterday - I didn't check the radiator matrix temperature myself.
audifan
A2OC Donor
Sorry to go on, but have you verified the radiator fan steps up its speed when the temperature is about 95 to its second speed ( speed one being climate cooling speed ) there is a third speed when the system is around 102 ish. If these faster speeds are not being enabled then the fan controller unit on the upper rear of the radiator is not working correctly.
No - I haven't done that yet as it was only hot enough the first time I noticed the issue and then today driving down and today I didn't have time to verify what was going on en route, will add that to the list of things to look at now I know what to check and how to do so. Thank you!
I think you might be onto something here.
Having replaced the temperature sensor and gone on a local test drive, the readings - particularly the dashboard channel - normalised compared to yesterday and there are no fault codes recorded pertaining to the sensor (yesterday there was an open to ground fault on the dashboard diagnostics, unsurprisingly).
However, in sustained 2nd gear hill climb, the ecu read temperature hit 90, and then carried on climbing without impediment - once it got to 98-99 I put the cabin heater on full, whereupon it stabilised and inched back to just over 90 where it stayed until the last few hundred metres home where it went back up to 96 climbing a slow hill and then the driveway. However I was then able to take off the bonnet and confirm that the fan is running - but only at what looks and sounds like the speed necessary to support a/c cooling. As soon as the Climatronic was switched to off - with coolant temp still at an indicated 96 degrees - the fan stops. At no point in the journey did I hear the audible noise of the fan speeding up (typical of sitting in traffic jams on hot summer days!).
Thinking about possible faults - fan controller being one, but does it take a cue from the temperature sensor at the bottom of the radiator - ie if that fails do we also lose the higher speeds? Secondly, does that sensor send a temperature anywhere to a measurements page that can be read using diagnostics?
ps. I had completely forgotten until just now that I have the complete workshop manual pdf sitting on my laptop - I am now going to read the tdi cooling circuit chapter to clarify a bunch of silly questions and misconceptions that I have been asking above through lack of knowledge.
audifan
A2OC Donor
Absolutely no harm in asking questions or even for clarification. From what I remember there is no other coolant temperature sensors on the diesel apart from the dual sensor next to the tandem pump.
With a warmed up engine sitting with little airflow and reasonable ambient temperature then the second fan speed can be achieved with the car stationary and the bonnet off. This is how I check for coolant leeks and all air out of the system. And yes the increase in fan speed is clearly audible ( and visual ).
Fan controllers do sometimes fail internally - as some have experienced on here, but also can suffer from broken or corroded wires and connections.
If you replace it be aware there are 2 different types. One for climate cars and the other for ones without.
As for what actually sends the speed requests to the fan controller not 100%. Obviously the climate needs the slow speed, but whether it is completely controlled by the climate controller or if speeds two and three come from the ECU not sure. So it could be there is no increase speed signal being received unless it is the fan controller itself that controls the speed. I would remove the wiring plugs from the fan controller and carefully check the wiring to it for damage, the connectors for corrosion or sitting water and then check from the manual what tests you can perform on the fan controller to see if that shows the controller faulty or not receiving inputs.
With a warmed up engine sitting with little airflow and reasonable ambient temperature then the second fan speed can be achieved with the car stationary and the bonnet off. This is how I check for coolant leeks and all air out of the system. And yes the increase in fan speed is clearly audible ( and visual ).
Fan controllers do sometimes fail internally - as some have experienced on here, but also can suffer from broken or corroded wires and connections.
If you replace it be aware there are 2 different types. One for climate cars and the other for ones without.
As for what actually sends the speed requests to the fan controller not 100%. Obviously the climate needs the slow speed, but whether it is completely controlled by the climate controller or if speeds two and three come from the ECU not sure. So it could be there is no increase speed signal being received unless it is the fan controller itself that controls the speed. I would remove the wiring plugs from the fan controller and carefully check the wiring to it for damage, the connectors for corrosion or sitting water and then check from the manual what tests you can perform on the fan controller to see if that shows the controller faulty or not receiving inputs.
PlasticMac
Admin Team
This might not be, (probably isn't), practical, but you may be able to disconnect the fan from the FCM completely, and hard wire the fan to the fused side of the fan fuse, to give permanent running of the fan at high speed, and rule the fan speed in or out as the cause.
Would probably need to get the engine hot first, then run the fan.
Sounds a bit daft now I've written it ...
Mac.