we have no hills here but if it is cold and my car is idling with the heater going the temperature drops. Totally normal. If the car is moving quickly through the air then the cooling of the engine will be greater. If the heater works then you know you do not have a totally blocked circulation of water thru the engine that can force the thermostat open. But otherwise the gauge should be rock solid on the middle mark. It is not normal to have the gauge habitually half way.
Technical 169 1.2 thermostat 'bypass'?
- Thread starter mjc506
- Start date
-
- Tags
- thermostat
Currently reading:
Technical 169 1.2 thermostat 'bypass'?
Popitinpete
Member
- Joined
- Sep 1, 2021
- Messages
- 292
- Points
- 142
Yes exactly the same as ours, it's just a thing some do, as I say, as long as the heater is warm and it doesn't overheat its as fiat intended, when I check when the quage is a third of the way up the engine and top hose are always piping hot, really hot. I've got no worries with the engine running cold even when the guage says other wise, change your stat if it really bothers you but I suggest some stick in the middle some don't, I changed the stat once about 5 years ago, didn't make a difference but swapped it just in case it was faulty but the second one was exactly the same. Just one of its quirks. As I've mentioned as long as its not using water, not overheating and the heater is hot it's nothing to worry about.
When the rusty water pipe fails at the front of the car the guage doesn't react as you would think, it still shows everything is OK but you end up low on coolant wrecking the head gasket so that tells you all you need to know about how the guage tells you what's going on
Last edited:
Hello, Jock here. You are describing absolutely classic symptoms of a failing thermostat. It's nothing to get "worried" about but I've seen this on all (if I remember correctly because we've had several Pandas over the years) of the Pandas we've had in the family. Fit a new thermostat and the running temp will go up to normal. On the other hand you can just run around with it like that, as I did for over a year with my present Panda, and nothing very spectacular will happen. Long term it will have an affect on fuel consumption and it's not all that good to run an engine "cold" - it will degrade the oil more quickly and may eventually cause more internal corrosion as it won't tend to drive off internal condensation so well.
I know you are intending to come by so we can check out your "dogbone problem" and we can have a look at how your thermostat is working at the same time. I'm just going to PM you now.
absolutely not as Fiat intended I have over 100k miles with the FIRE engine
Punto, has been in the hills of Scotland and Wales in winter -10
05 Panda been in the hills of Wales -10
2010 Panda been -2C to +35C
They all stay bang in the middle give or take the thickness of the needle with normal driving
Yes they will drop down if you put the fans on max in winter, Blower I and II are fine manual heaters
There has to be something different, don't know what, driving conditions, mechanical or operating the heats.
Punto, has been in the hills of Scotland and Wales in winter -10
05 Panda been in the hills of Wales -10
2010 Panda been -2C to +35C
They all stay bang in the middle give or take the thickness of the needle with normal driving
Yes they will drop down if you put the fans on max in winter, Blower I and II are fine manual heaters
There has to be something different, don't know what, driving conditions, mechanical or operating the heats.
Last edited:
Pepandaros
Member
I forgot to mention the cabin gets up to comfy temperature quickly and yesterday I filled up the tank, it appears the car does 53mpg with mixed driving.
Must say on the motorway I stick around 60. yes I know.. BIG Yawn ...but economical!
Must say on the motorway I stick around 60. yes I know.. BIG Yawn ...but economical!
ahhh
one thing I missed
if your car overheats and the radiator fans kick in on max. There is a drop on the gauge.
Its pretty hard to get the radiator fans to kick in on max, if the car is running correctly
one thing I missed
if your car overheats and the radiator fans kick in on max. There is a drop on the gauge.
Its pretty hard to get the radiator fans to kick in on max, if the car is running correctly
If the needle drops a little under certain driving conditions, this may be perfectly normal behaviour. But if it consistently fails to reach the halfway point on the gauge, chances are you need a new thermostat.
Whatever economy you may be getting now, if the thermostat isn't sealing properly and you replace it, it'll improve. Thermostats failing to seat fully due to a worn or perished internal seal are extremely common.
By the time I got round to replacing mine, the gauge was consistently about two needle widths below the midway point at modest motorway cruise, and economy was down by, I'd say, around 5mpg. Changing the thermostat cost less than a tenner, took less than an hour, and restored both gauge reading and economy to their previous values.
Note also the the Panda is massively overcooled for the UK winter, and cold air flowing into the engine compartment when the car is moving forward has a substantially greater cooling effect than when it's stationary. You can notice this by observing how much faster it cools down when restarting after parking for a short while in a strong wind on a freezing cold day. It's one of those areas where the Panda is showing its age; some more modern designs have much better underbonnet thermal management.
Whatever economy you may be getting now, if the thermostat isn't sealing properly and you replace it, it'll improve. Thermostats failing to seat fully due to a worn or perished internal seal are extremely common.
By the time I got round to replacing mine, the gauge was consistently about two needle widths below the midway point at modest motorway cruise, and economy was down by, I'd say, around 5mpg. Changing the thermostat cost less than a tenner, took less than an hour, and restored both gauge reading and economy to their previous values.
Note also the the Panda is massively overcooled for the UK winter, and cold air flowing into the engine compartment when the car is moving forward has a substantially greater cooling effect than when it's stationary. You can notice this by observing how much faster it cools down when restarting after parking for a short while in a strong wind on a freezing cold day. It's one of those areas where the Panda is showing its age; some more modern designs have much better underbonnet thermal management.
Last edited:
Pepandaros
Member
Don't know about the others with the same symptoms?
But I have not heard the fan kicking in yet, since my short ownership of the car I have been stuck only once in traffic due to snow on the M80.
As it was obvious we would be stuck for a while I just switched the engine off. (M80 is not the place for fault finding and testing haha).
So far we had very mild weather so my heater always stays on setting 1, after a few miles I turn it to full and then it goes to 3/4 which keeps the cabin nicely warm in city and motorway.
Pepandaros
Member
Excellent! Didn't realise it is that cheap, sounds like investing a tenner is going to save me on the long run.
Was already impressed with the mpg as it is right now with the colder months.
This is it I think? https://www.shop4parts.co.uk/?name=store&op=Product&ProdID=470 It was the one that fitted mine. Unfortunately S4p seem to be out of stock with the Original Birth branded one (which I bought) but aftermarket examples are available almost everywhere. Interestingly if you buy an aftermarket brand they usually come with an included gasket but the genuine Fiat/Chrysler part does not so not only is it more expensive but you then have to lay out for a gasket too! Probably not worth buying a single item like this from S4p anyway because under £25 and you'll be hit with shipping.
Pepandaros
Member
I had a look, as mine is a 59 reg it seems to sit exactly on the edge of 2004/2009 or 2009/2012 model years so tricky to order, I rather let the supplier decide based on my vin number.
There's only 2 possibilities. The easiest way is to look for the electric wire going to the sensor on the later model example which has the temp sensor screwed into it. Can be seen quite easily with a mirror. The older type, like the one that fit's mine is a simpler casting and has no sensor (it's round the back where the top heater hose joins the head). If I was a betting man I'd be betting it's the earlier simpler one for yours?
Me too; it's almost certainly the type without the temperature sensor; fortunately that's the cheap one.
Attachments
Pepandaros
Member
Thanks for the info, it's currently dark wet and very windy so have a look under the bonnet tomorrow.
- Joined
- Jul 1, 2012
- Messages
- 426
- Points
- 226
Just regarding the temperature range of the "middle" point, it is rather a wide range.
A couple of years ago when I drove through Spain, I found out to my peril that my Panda doesn't like doing 70mph up a fairly steep incline for several miles while in nearly 50 degree air temperature
It didn't overheat, but got up to 3/4 on gauge, fan on, and it cut engine power right back.
So the following year I had the Torque app running on my phone all the way.
This is on a MJ, so possibly slightly different.
My temperature gauge reached the middle at 83 degrees, and stayed there until 104 degrees.
3/4 on the gauge was 112.....at which point fortunately I just made it to the top, and it was all downhill for 20+ miles, and within a mile it dropped down to 79 degrees and just under the middle on the gauge going down hill.
I now plan the route so I go up that sectionat night when it's cooler......only gets up to 104.....plus I hate the stress of watching the temperature gauge rise......and having the heater on full blast while 45 degree red hot air blows in the windows is no fun.
A couple of years ago when I drove through Spain, I found out to my peril that my Panda doesn't like doing 70mph up a fairly steep incline for several miles while in nearly 50 degree air temperature
It didn't overheat, but got up to 3/4 on gauge, fan on, and it cut engine power right back.
So the following year I had the Torque app running on my phone all the way.
This is on a MJ, so possibly slightly different.
My temperature gauge reached the middle at 83 degrees, and stayed there until 104 degrees.
3/4 on the gauge was 112.....at which point fortunately I just made it to the top, and it was all downhill for 20+ miles, and within a mile it dropped down to 79 degrees and just under the middle on the gauge going down hill.
I now plan the route so I go up that sectionat night when it's cooler......only gets up to 104.....plus I hate the stress of watching the temperature gauge rise......and having the heater on full blast while 45 degree red hot air blows in the windows is no fun.
Popitinpete
Member
- Joined
- Sep 1, 2021
- Messages
- 292
- Points
- 142
Sounds fine, it's actually telling you what's going on, better than just the needle sitting in the middle not telling you what's happening temperature wise. How it should be.Just regarding the temperature range of the "middle" point, it is rather a wide range.
A couple of years ago when I drove through Spain, I found out to my peril that my Panda doesn't like doing 70mph up a fairly steep incline for several miles while in nearly 50 degree air temperature
It didn't overheat, but got up to 3/4 on gauge, fan on, and it cut engine power right back.
So the following year I had the Torque app running on my phone all the way.
This is on a MJ, so possibly slightly different.
My temperature gauge reached the middle at 83 degrees, and stayed there until 104 degrees.
3/4 on the gauge was 112.....at which point fortunately I just made it to the top, and it was all downhill for 20+ miles, and within a mile it dropped down to 79 degrees and just under the middle on the gauge going down hill.
I now plan the route so I go up that sectionat night when it's cooler......only gets up to 104.....plus I hate the stress of watching the temperature gauge rise......and having the heater on full blast while 45 degree red hot air blows in the windows is no fun.
Interesting, thanks all.
Finally got to 'play' this morning. The car had been sat since Friday, so the engine was (should have been) at Ambient. However, OAT on the dash read 9°C, 'air temp' (intake air I think? or under bonnet ambient) was 10°C, and the coolant temp read 7°C. Not massive differences, but notable perhaps. I had hoped that the Climate ECU would also read coolant temp separately too, to 'verify' the reported temp, but no such luck.
I set up MES, cabin heat turned completely off, and started the engine and let it idle. I kept a visual check on the temperature gauge, and compared to the MES reading as it passed divisions:
The needle lifts off the bottom at 50°C
1/8 is ~58°C
1/4 is ~65°C
3/8 is ~73°C
1/2 starts at ~80°C, and the needle didn't lift above this position despite the coolant temp reaching 97°C. I note Cinq999 above reports the temp gauge reaching the middle at ~83°C and going higher at ~104°C. Whether this is just variation between cars (in the dash, I assume) or the difference is due to MJ/1.2 I don't know. It would be quite easy for each needle to be fitted at slightly different angles during manufacturing (It seems the dashboard/needle only reads integer °C - I could see it 'step' as the value reported by MES incremented. I also noted that my needle would 'step' just below the divisors on the gauge, and settled towards the bottom half of the thicker 1/2 division) but gross errors would result in the needle settling away from the half way mark on the gauge.
So, some results! Time in seconds along the bottom, coolant temp in °C on the y axis
To begin with, coolant temp rises as expected (not quite logarythmically, as the engine speed and load is higher to begin with until the ECU lets the idle speed drop) But at ~750sec, there's a dip, and temperature starts increasing more slowly. This has got to be the thermostat opening, allowing cooler water from the radiator into the system (the initial dip) and also increasing the thermal mass (additional coolant, the rad, etc). But note the temperature that this happens - ~70°C. This seems a little low for an 87.5°C thermostat? I must admit in my earlier experiments I didn't pay too much attention to exactly when the stats started to open, but I will recheck, and will be surprised if it's even as low as 80°C...
At ~1800sec (97°C), the rad fan turns on at low speed, quickly pulling the coolant temp down to 93°C when it turns off (the coolant temp drops by another degree after this, just thermal inertia). I let the coolant temp rise by a couple degrees, then turn on the cabin heat, eventually turning it off ~2400sec (I was getting a bit warm haha). I hadn't appreciated how much heat that little heat exchanger could extract from the system! (Nearly 15°C in a relatively short time, and this is sat on the driveway with very little assistance from air passing through the radiator)
So yes, the cabin heat does have the capability of pulling the coolant temp down significantly, especially with low engine load - quite easily from ~90°C to ~75°C, especially with the help of air flowing through the radiator down a hill, which would register on the temperature gauge.
I am concerned that my 'stat appeared to open at 70°C. Or rather, concerned that when the stat opened, the ECU reported the coolant temp as 70°C. I would like to verify that the temperature reported by the ECU is reasonably accurate (and I've lent my thermal camera to someone!). For now, I'll recheck when the thermostat opens in the pan on the stovetop (they were all the same). If there is a ~10°C difference between reported and actual temperatures, the ECU will think the engine is 'cold' for longer (for high idle, fuel enrichment etc) and will also turn the fan on late... The sensor is only an NTC resistor, so any additional resistance between ECU and sensor would register as a lower temperature. I couldn't see any visual corrosion on the sensor connector, but will need to check resistance between the ecu and sensor terminals. Perhaps a new sensor would be worth trying, they are made to a tolerance.
Finally got to 'play' this morning. The car had been sat since Friday, so the engine was (should have been) at Ambient. However, OAT on the dash read 9°C, 'air temp' (intake air I think? or under bonnet ambient) was 10°C, and the coolant temp read 7°C. Not massive differences, but notable perhaps. I had hoped that the Climate ECU would also read coolant temp separately too, to 'verify' the reported temp, but no such luck.
I set up MES, cabin heat turned completely off, and started the engine and let it idle. I kept a visual check on the temperature gauge, and compared to the MES reading as it passed divisions:
The needle lifts off the bottom at 50°C
1/8 is ~58°C
1/4 is ~65°C
3/8 is ~73°C
1/2 starts at ~80°C, and the needle didn't lift above this position despite the coolant temp reaching 97°C. I note Cinq999 above reports the temp gauge reaching the middle at ~83°C and going higher at ~104°C. Whether this is just variation between cars (in the dash, I assume) or the difference is due to MJ/1.2 I don't know. It would be quite easy for each needle to be fitted at slightly different angles during manufacturing (It seems the dashboard/needle only reads integer °C - I could see it 'step' as the value reported by MES incremented. I also noted that my needle would 'step' just below the divisors on the gauge, and settled towards the bottom half of the thicker 1/2 division) but gross errors would result in the needle settling away from the half way mark on the gauge.
So, some results! Time in seconds along the bottom, coolant temp in °C on the y axis
To begin with, coolant temp rises as expected (not quite logarythmically, as the engine speed and load is higher to begin with until the ECU lets the idle speed drop) But at ~750sec, there's a dip, and temperature starts increasing more slowly. This has got to be the thermostat opening, allowing cooler water from the radiator into the system (the initial dip) and also increasing the thermal mass (additional coolant, the rad, etc). But note the temperature that this happens - ~70°C. This seems a little low for an 87.5°C thermostat? I must admit in my earlier experiments I didn't pay too much attention to exactly when the stats started to open, but I will recheck, and will be surprised if it's even as low as 80°C...
At ~1800sec (97°C), the rad fan turns on at low speed, quickly pulling the coolant temp down to 93°C when it turns off (the coolant temp drops by another degree after this, just thermal inertia). I let the coolant temp rise by a couple degrees, then turn on the cabin heat, eventually turning it off ~2400sec (I was getting a bit warm haha). I hadn't appreciated how much heat that little heat exchanger could extract from the system! (Nearly 15°C in a relatively short time, and this is sat on the driveway with very little assistance from air passing through the radiator)
So yes, the cabin heat does have the capability of pulling the coolant temp down significantly, especially with low engine load - quite easily from ~90°C to ~75°C, especially with the help of air flowing through the radiator down a hill, which would register on the temperature gauge.
I am concerned that my 'stat appeared to open at 70°C. Or rather, concerned that when the stat opened, the ECU reported the coolant temp as 70°C. I would like to verify that the temperature reported by the ECU is reasonably accurate (and I've lent my thermal camera to someone!). For now, I'll recheck when the thermostat opens in the pan on the stovetop (they were all the same). If there is a ~10°C difference between reported and actual temperatures, the ECU will think the engine is 'cold' for longer (for high idle, fuel enrichment etc) and will also turn the fan on late... The sensor is only an NTC resistor, so any additional resistance between ECU and sensor would register as a lower temperature. I couldn't see any visual corrosion on the sensor connector, but will need to check resistance between the ecu and sensor terminals. Perhaps a new sensor would be worth trying, they are made to a tolerance.
Popitinpete
Member
- Joined
- Sep 1, 2021
- Messages
- 292
- Points
- 142
Mmmmmmmm, seems the stat is opening at 70c, as you say it should open at the stamped on number, ie, 82c or whatever. Definitely a contender for a leaky thermostat seal? Is it a new stat?Interesting, thanks all.
Finally got to 'play' this morning. The car had been sat since Friday, so the engine was (should have been) at Ambient. However, OAT on the dash read 9°C, 'air temp' (intake air I think? or under bonnet ambient) was 10°C, and the coolant temp read 7°C. Not massive differences, but notable perhaps. I had hoped that the Climate ECU would also read coolant temp separately too, to 'verify' the reported temp, but no such luck.
I set up MES, cabin heat turned completely off, and started the engine and let it idle. I kept a visual check on the temperature gauge, and compared to the MES reading as it passed divisions:
The needle lifts off the bottom at 50°C
1/8 is ~58°C
1/4 is ~65°C
3/8 is ~73°C
1/2 starts at ~80°C, and the needle didn't lift above this position despite the coolant temp reaching 97°C. I note Cinq999 above reports the temp gauge reaching the middle at ~83°C and going higher at ~104°C. Whether this is just variation between cars (in the dash, I assume) or the difference is due to MJ/1.2 I don't know. It would be quite easy for each needle to be fitted at slightly different angles during manufacturing (It seems the dashboard/needle only reads integer °C - I could see it 'step' as the value reported by MES incremented. I also noted that my needle would 'step' just below the divisors on the gauge, and settled towards the bottom half of the thicker 1/2 division) but gross errors would result in the needle settling away from the half way mark on the gauge.
So, some results! Time in seconds along the bottom, coolant temp in °C on the y axis
To begin with, coolant temp rises as expected (not quite logarythmically, as the engine speed and load is higher to begin with until the ECU lets the idle speed drop) But at ~750sec, there's a dip, and temperature starts increasing more slowly. This has got to be the thermostat opening, allowing cooler water from the radiator into the system (the initial dip) and also increasing the thermal mass (additional coolant, the rad, etc). But note the temperature that this happens - ~70°C. This seems a little low for an 87.5°C thermostat? I must admit in my earlier experiments I didn't pay too much attention to exactly when the stats started to open, but I will recheck, and will be surprised if it's even as low as 80°C...
At ~1800sec (97°C), the rad fan turns on at low speed, quickly pulling the coolant temp down to 93°C when it turns off (the coolant temp drops by another degree after this, just thermal inertia). I let the coolant temp rise by a couple degrees, then turn on the cabin heat, eventually turning it off ~2400sec (I was getting a bit warm haha). I hadn't appreciated how much heat that little heat exchanger could extract from the system! (Nearly 15°C in a relatively short time, and this is sat on the driveway with very little assistance from air passing through the radiator)
So yes, the cabin heat does have the capability of pulling the coolant temp down significantly, especially with low engine load - quite easily from ~90°C to ~75°C, especially with the help of air flowing through the radiator down a hill, which would register on the temperature gauge.
I am concerned that my 'stat appeared to open at 70°C. Or rather, concerned that when the stat opened, the ECU reported the coolant temp as 70°C. I would like to verify that the temperature reported by the ECU is reasonably accurate (and I've lent my thermal camera to someone!). For now, I'll recheck when the thermostat opens in the pan on the stovetop (they were all the same). If there is a ~10°C difference between reported and actual temperatures, the ECU will think the engine is 'cold' for longer (for high idle, fuel enrichment etc) and will also turn the fan on late... The sensor is only an NTC resistor, so any additional resistance between ECU and sensor would register as a lower temperature. I couldn't see any visual corrosion on the sensor connector, but will need to check resistance between the ecu and sensor terminals. Perhaps a new sensor would be worth trying, they are made to a tolerance.
thanks for the graph
looks about right to me
it takes time for the thermostat to open more, fill the radiator. cool down, and enter the pump to be circulated
there will be a short delay from the stat opening while the engine will still get hotter
depends on brand of the thermostat most will start to open at 87C although I have seen one stamped 85C
you could put your hand on the top Hose and time when the thermostat opens I doubt it will be at the little glitch in the graph
looks about right to me
it takes time for the thermostat to open more, fill the radiator. cool down, and enter the pump to be circulated
there will be a short delay from the stat opening while the engine will still get hotter
depends on brand of the thermostat most will start to open at 87C although I have seen one stamped 85C
you could put your hand on the top Hose and time when the thermostat opens I doubt it will be at the little glitch in the graph
Last edited:
