Alfa Romeo Forum banner

1 - 20 of 21 Posts

·
Registered
Joined
·
92 Posts
Discussion Starter #1
Just had my MiTo serviced (2011 1.4 Multiair 135bhp TCT) including a coolant flush.

Since then I noticed that the engine would come up to operating temperature (90°C), then dip slightly below for a short while, then go up to 90°C and stay there.

Can someone with a petrol MiTo observe their temperature gauge on their next drive and report back whether theirs does the same? I never really paid such close attention to mine so I don't know whether this is something it always did (and I'm only noticing this now because I'm scrutinising the temp gauge after the coolant flush) or if it's something new arising from the service.

Here's a timeline to help (you can follow along by resetting trip A or B and putting the trip display in elapsed time mode) -
0 minutes: car started. Outside air temp 20°C
After 5 minutes of gentle driving: temp gauge reaches 90°C for the first time
After 6 minutes: temp gauge starts to dip. Goes down to 84 or 86°C
Around 7 minutes: sometimes dips a second or third time
After 8 minutes: stabilizes at 90°C for the rest of the drive.
 

·
Registered
Joined
·
33 Posts
When you start your engine from cold, thermostat is closed.
Short cooling circuit, without main radiator.
When operating temperature is reached, thermostat is opening and cold coolant from radiator mix with the engine coolant and so you see a little drop in temperature before it warms again.

All is OK

Sorry for my bad English.
 

·
Registered
Joined
·
92 Posts
Discussion Starter #3
When you start your engine from cold, thermostat is closed.
Short cooling circuit, without main radiator.
When operating temperature is reached, thermostat is opening and cold coolant from radiator mix with the engine coolant and so you see a little drop in temperature before it warms again.

All is OK

Sorry for my bad English.
Your English is fine 🙂. And thanks for the reply 🙂.

I know how the cooling system works, your explanation makes perfect sense - I've just never noticed the dip before now. I want to make sure it's not due to air trapped in the system or because the mechanic used generic red coolant this time (instead of the brand that Alfa Romeo recommends in the manual that I'd always used in the car until now).

Does your MiTo have that temperature dip too?
 

·
Registered
Joined
·
33 Posts
Got this temperature dip too on my 159 Tbi, and it was similar on my previous 159 (1.9 & 2.4 jtd)
You can see it clearly with MultiEcuScan during the engine warm up 😉
 

·
Registered
Joined
·
9,148 Posts
It's more obvious on the 156 where the temp gauge was pretty much "live" .. on a lot of cars now the gauge is more of an "average" temp and the fluctuations are smoothed out ..
 

·
Registered
Joined
·
92 Posts
Discussion Starter #6
It's more obvious on the 156 where the temp gauge was pretty much "live" .. on a lot of cars now the gauge is more of an "average" temp and the fluctuations are smoothed out ..
On the MiTo I believe it is live, not average, because the gauge is graduated in degrees C with the centerline at 90°C. I confirmed over OBD that the needle in the MiTo moves in 2°C increments. That is, during warmup if the needle is at 60°C it will only move once temp rises to 62°C.

I assume the average ones would be those graduated without units, with one end saying simply "C" and the other end saying "H".

Thanks all for the replies, it's reassuring to see most people have it.
 

·
Registered
Joined
·
4,236 Posts
Technically it is hysteresis in the thermostat. As the thermostat is a mechanical device, it responds to change with a bit of a lag, it doesn’t smoothly open and regulate temperature precisely. If the opening temperature is (say) 85 C, it will start to open at about 85, but the temperature of the water is still climbing, so it opens rapidly and overshoots. If you had a really precise temperature gauge, you would see the over/undershoot gradually declining as an approximate sine wave. My 156 does this, the GT gauge smooths it out (90C in all circumstances), the 159 does it.
 

·
Registered
Joined
·
92 Posts
Discussion Starter #8 (Edited)
Technically it is hysteresis in the thermostat. As the thermostat is a mechanical device, it responds to change with a bit of a lag, it doesn’t smoothly open and regulate temperature precisely. If the opening temperature is (say) 85 C, it will start to open at about 85, but the temperature of the water is still climbing, so it opens rapidly and overshoots. If you had a really precise temperature gauge, you would see the over/undershoot gradually declining as an approximate sine wave. My 156 does this, the GT gauge smooths it out (90C in all circumstances), the 159 does it.
I assumed that the first time the thermostat opened it would stay open too long and thus over-cool the engine. In fact the engine typically undershoots a second time (but by a smaller amount), and finally settles at the correct temperature about 7 minutes after being started. I'm trying to think of what could have caused the system to become unstable like this (assuming that it wasn't before). I'm leaning towards the different brand of coolant having a different heat capacity?

Today morning was hot (25°C) and I had the A/C running. The gauge didn't dip at all after reaching operating temp. Might be because the radiator was being warmed up by the A/C condenser in front of it - or maybe the added load from the A/C compressor generated enough waste heat to keep the temp from dipping (assuming 33% efficiency since this is a petrol, and A/C compressor power consumption of 3 bhp/2.2 kW at the pulley, it would cause the engine to be adding about 2.2kW of additional heat into the engine coolant*. Possibly a combination of the two.

* when we dyno'd a very well-instrumented engine at engineering school, we measured that of the energy input the petrol, 33% made useful work at the flywheel, 33% ended up in the coolant, and 33% was lost in the exhaust. So for 3 bhp to come out of the pulley to turn the A/C, 3 bhp of heat would end up in the coolant, 3 bhp in the exhaust, and petrol consumption would increase by 9 bhp worth (in terms of petrol's raw energy content of 32 MJ/liter. Works out to an extra 0.75 liters per hour of petrol).
 

·
Registered
Joined
·
4,236 Posts
Could be that your new coolant has altered the lubrication properties of the thermostat plunger a bit. If it is slightly sticky, then you’re going to get a bigger overshoot effect - it will open ever so slightly late, and be attempting to fully open by the time it gets feedback about the cold water coming in. I can’t imagine that the heat capacity of the coolant has changed in any material way.
 

·
Registered
Joined
·
92 Posts
Discussion Starter #10
Could be that your new coolant has altered the lubrication properties of the thermostat plunger a bit. If it is slightly sticky, then you’re going to get a bigger overshoot effect - it will open ever so slightly late, and be attempting to fully open by the time it gets feedback about the cold water coming in. I can’t imagine that the heat capacity of the coolant has changed in any material way.
That must be it. "Sticky thermostat" occurred to me but I didn't think the lubricity of the coolant would affect it. Any ideas what can be done short of flushing the coolant? Should I just wait until the annual service next year and then flush it?
 

·
Registered
Joined
·
4,236 Posts
Possibly the answer is a new thermostat if it is really bothering you .... but I would ignore it!
 

·
Registered
Joined
·
92 Posts
Discussion Starter #12
It's more obvious on the 156 where the temp gauge was pretty much "live" .. on a lot of cars now the gauge is more of an "average" temp and the fluctuations are smoothed out ..
Well, today I've confirmed that it's exactly as you said (and had to eat my words from earlier). I had the raw coolant temp data read over OBD and compared it to the physical temperature gauge in the dashboard as the engine warmed up. My findings were eye-opening.

Up to 50°C both gauges matched
When real temp was between 50°C and 70°C, for every 1°C rise in real coolant temp the physical gauge rose 2°C.
Therefore, at 70°C actual, the physical gauge started reading 90°C
And here's the big one: for real coolant temp between 70°C and 97°C the physical gauge continued to read 90°C!

This was done with the engine idling while parked. At 97°C the radiator fan turned on and within 10 seconds the real temp was back down to 89°C at which point the fan turned off. Of course the physical gauge read 90°C the whole time...

During driving, the real temp was between 73°C and 84°C, varying a lot based on road speed and load. During the initial dip that started this post, the engine had just hit 70°C actual (90°C on the gauge), then dipped briefly to 69°C actual (88°C on the gauge) and quickly recovered.

Based on how much temp is actually varying behind the scenes I conclude that the dip I noticed is very minor in the grand scheme of coolant temp variation and hence nothing to worry about.
 

·
Premium Member
Joined
·
1,126 Posts
If the real temperature reading fluctuates between 73°C and 84°C when driving in a warm climate like Malta, I would say your thermostat can't close completely and is stuck a bit open.
 

·
Registered
Joined
·
92 Posts
Discussion Starter #14
If the real temperature reading fluctuates between 73°C and 84°C when driving in a warm climate like Malta, I would say your thermostat can't close completely and is stuck a bit open.
Here's the graph I recorded from the OBD telemetry on the drive home (time axis goes from zero to 16 minutes 40 seconds). Outisde air temp was 20°C. I started with a completely cold engine, drove at a range of speeds, and at the end I parked the car and let it idle until the radiator fan turned on.
 

Attachments

·
Premium Member
Joined
·
1,126 Posts
Looking at the graph, I would still say your thermostat is stuck a bit open.
You can verify this by starting the engine from cold and then feel on the top hose of the radiator.
It should suddenly get hot when the thermostat opens
If it gradually warms up your thermostat is stuck a bit open and a replacement will make a difference
 

·
Registered
Joined
·
92 Posts
Discussion Starter #16
Looking at the graph, I would still say your thermostat is stuck a bit open.
You can verify this by starting the engine from cold and then feel on the top hose of the radiator.
It should suddenly get hot when the thermostat opens
If it gradually warms up your thermostat is stuck a bit open and a replacement will make a difference
That's a good idea, I will do that. By top hose I take it you mean the thick one connected to the thermostat housing that passes under the coolant tank?

What benefits would I get from a more stable coolant temperature if I did replace the thermostat?

Also, I thought the thermostat was meant to leak a little (I assume if there was zero flow with the thermostat closed, the hot water from the engine would never actually reach it initially and cause it to open?)
 

·
Registered
Joined
·
92 Posts
Discussion Starter #17
So, I've done as you suggested. I identified 3 pipes coming out of the thermostat housing - a thick pipe that feeds the top of the radiator (that seems to be controlled by thermostat), a medium pipe to the heater core, and a thin pipe that connects to a T-piece that connects to the back of the expansion tank and the turbo. The pipes to the expansion tank and heater core don't seem to be shut off by thermostat.

So, immediately upon startup I started feeling the radiator hose and the small expansion tank hose. Within 2 minutes the expansion tank hose was warm but the thick radiator hose was still cold. This would imply that up to around 40°C the thermostat is still fully closed.

For what it's worth, when I first got this car 7 years ago I noticed that the heat coming from the heater depended on how hard I was pushing it (hotter when the engine was working hard). At the time it struck me as odd because my previous car (a Ford Fusion 1.4 80bhp) had remarkably constant heater output.

I'd always assumed it might be due to the different location of the tap of the heater core, or because the Alfa makes 1.5x as much power as the ford so the difference between full throttle and idle power is more pronounced. But now I'm thinking it might just be that this thermostat does a lousy job of keeping the temperature constant.
 

·
Premium Member
Joined
·
1,126 Posts
If the thermostat actually is leaking and you replace it, the engine will get to the working temperature quicker and the water temperature will be more stable.
If you have a leaking thermostat in a climate where it gets below 0°C, the engine will never get to the correct working temperature :)

The top hose is the one coming from the thermostat housing to the top of the radiator (My daughter has currently borrowed my Mito so I can't take a picture to indicate the hose).

The warm side of the heater block is coming from the thermostat housing on the engine side of the thermostat so there will be a cooling water flow through the heater block and the thermostat will be warmed up as the cooling water temperature increases.
 

·
Registered
Joined
·
92 Posts
Discussion Starter #19
If the thermostat actually is leaking and you replace it, the engine will get to the working temperature quicker and the water temperature will be more stable.
If you have a leaking thermostat in a climate where it gets below 0°C, the engine will never get to the correct working temperature :)

The top hose is the one coming from the thermostat housing to the top of the radiator (My daughter has currently borrowed my Mito so I can't take a picture to indicate the hose).

The warm side of the heater block is coming from the thermostat housing on the engine side of the thermostat so there will be a cooling water flow through the heater block and the thermostat will be warmed up as the cooling water temperature increases.
Hi Henrik, I've actually done the test you suggested - see my reply in post #17 immediately above yours ;)
 

·
Registered
Joined
·
2,329 Posts
Temperature gauges staying at or close to 90 degrees for anything between 70 and 100 is common and even normal. Modern cars are made this way to stop people worrying about running temperatures. Also note that with a 5psi radiator cap and 33% ethylene glycol mix the coolant will not boil until it reaches 110 degrees.

I have used live data on many cars and this is correct for all modern ones.

The Alfa 156 gauge is live so will see temps swing more. The Alfa 147, whilst built on the 156 floorpan and using same engines has later electronics and is damped, so gauge rarely moves from 90 deg C.
 
1 - 20 of 21 Posts
Top