coolant temperature cluster gauge test

Hello,

For some reason my temperature gauge on my 02' GTI is reading slightly above 90 celcius (canada). Looks like its at around 92c on the cluster gauge. Vagcom measuring block readings is fluctuating between 88c and 97c..

Is there an output test to do that can determine if the car is running hot or the gauge is a little off?

Ideally I would like to be able to tell the temp cluster gauge to go to 90c with the car cold and see if it goes slightly past 90c.

If it goes past the halfway mark of 90c while the car is cold, I'm assuming that the cluster gauge is just off.

Anyone know the ideal temperature ranges they cars fluctuate at? is 88c to 97c normal? I thought 90c was what its supposed to be at..

Thanks.

The cluster software has been made so, that in certain temperature range the cluster will show 90C neitherless the actual temp.
http://imageshack.us/a/img593/2682/l6tl.jpg - It's in Estonian, but i think you can read the picture.

Also, there is output test you could activate. I guess you're familiar with so vcds, so just search the Output Test button.

yeah understood. The cluster gauge is reading 91c to 93c and vcds measuring blocks have a fluctuating reading reading of 88c and 97c when the car has been running for a period of time.

I tried running an output test. I wasn't helpful for this scenario. The output test includes the temp gauge going from 0 to max. I couldn't find any command to go to 90c or another desired temp.

30secondgti said:

is 88c to 97c normal?

Click to expand...

Sounds about right to me.

30secondgti said:

I thought 90c was what its supposed to be at..

Click to expand...

Plus or minus what? Think about the differing amounts of thermal energy your cooling system has to dissipate. There's a huge difference in the amount of heat produced at idle vs. full throttle. Then there are also big differences in air flow through the radiator (depending on the vehicle's speed and state of the fans) and of course the temperature of the air varies as well. Given all those variations, the fact that they can keep it within a 9 degrees C band is seems pretty good!

-Uwe-

30secondgti said:

I tried running an output test. I wasn't helpful for this scenario. The output test includes the temp gauge going from 0 to max. I couldn't find any command to go to 90c or another desired temp.

Click to expand...

After the temp gauge reaches MAX, wont it center it to 90C?

What blocks are you reading the gage in engine ecu or cluster because the sender is split and two leads go to cluster and two go to engine ecu.

You can cross reference them to double check both are reading the same for redundancy.

no it doesn't. It goes to max then to 0. There is a way to try and go to 90c exactly. I'm just uncertain on how to do it. I've seen someone else do it before.

golfi_vend said:

After the temp gauge reaches MAX, wont it center it to 90C?

Click to expand...

Engine measuring blocks (04) under coolant temperature through Vagcom.

Vagcoms readings are fluctuating between 88c and 97c as I've observed. Per what Uwe mentioned; at idle its lower in the spectrum, around 88c - 92c. When driving continuously with throttle engaged higher on the spectrum, 92c - 97c.

The CLUSTER GAUGE is stuck at a few points past the centre (90c). Visually looks like its at 91c to 93c (definitely past 90c). The cluster gauge doesn't move/ fluctuate like the more accurate readings through vagcom does.

Is there a command through vagcom that allows the user to tell the cluster gauge to go to 90c exactly? This way I can trouble shoot the cluster gauge. If telling the gauge to go to 90c exactly and the gauge reads past 90c then I know the gauge is off/ faulty.

Thanks.

Jack@European_Parts said:

What blocks are you reading the gage in engine ecu or cluster because the sender is split and two leads go to cluster and two go to engine ecu.

You can cross reference them to double check both are reading the same for redundancy.

Click to expand...

88c (190f) to 97c (207f) is normal? I thought the car is supposed to stay around 90c (194f) plus or minus a few points. The spectrum i'm seeing through vagcom goes all the way up to 97c at times (heavy acceleration). That seems a little high to me.

Uwe said:

Sounds about right to me.


Plus or minus what? Think about the differing amounts of thermal energy your cooling system has to dissipate. There's a huge difference in the amount of heat produced at idle vs. full throttle. Then there are also big differences in air flow through the radiator (depending on the vehicle's speed and state of the fans) and of course the temperature of the air varies as well. Given all those variations, the fact that they can keep it within a 9 degrees C band is seems pretty good!

-Uwe-

Click to expand...

30secondgti said:

The cluster gauge doesn't move/ fluctuate like the more accurate readings through vagcom does.

Click to expand...

That's normal, and it's designed to behave like that.

30secondgti said:

Is there a command through vagcom that allows the user to tell the cluster gauge to go to 90c exactly? This way I can trouble shoot the cluster gauge. If telling the gauge to go to 90c exactly and the gauge reads past 90c then I know the gauge is off/ faulty.

Click to expand...

Seriously? You're worried about 1-3 degree discrepancy in the analog gauge of a 12-13 year old cluster?

I haven't run output tests on an Mk.4 cluster in a long, long, time. Most cluster output tests sweep them fully, return them to zero, and then park then needle at some intermediate location (e.g. 100 km/h on a speedometer). You cannot specify the exact intermediate position; that's hard-coded in the cluster.

-Uwe-

Yeah I understand. I did a test today. Watched the cluster gauge rise from a cold start while watching the digital reading on vagcom rise. It seems the cluster gauge only moves from 0 at around 50c per vagcom.

Yeah, its really odd to me. I have a few friends with a mk4 gti and jetta. My car is in the overall best shape out of them. Both their cluster gauge temps read right at 90c. Mine is slightly above. It's strange. And this only started happening after I replaced the coolant, flange and t stat.

I hear yah on the output test. I already did one. All the output test seems to do is run the cluster gauge from 0 to max then back to 0, then to the actual temp of the coolant.

I've seen a guy put the temp cluster gauge exactly to 90c before on another car (while the car is off and cooled). Haven't been able to talk to him about this. Thought I would give you guys a try

I'll keep on googling potential ways to do it. There must be a way.!

Uwe said:

That's normal, and it's designed to behave like that.


Seriously? You're worried about 1-3 degree discrepancy in the analog gauge of a 12-13 year old cluster?

I haven't run output tests on an Mk.4 cluster in a long, long, time. Most cluster output tests sweep them fully, return them to zero, and then park then needle at some intermediate location (e.g. 100 km/h on a speedometer). You cannot specify the exact intermediate position; that's hard-coded in the cluster.

-Uwe-

Click to expand...

That's the mid point test of outputs you are referring to.............I think when doing a test.

Some clusters do offer an ADP I don't think this one does but you can check in VCDS if channels exist. Stream through 0x17-10-000>

Is the ECT and flange factory?

When you installed are you sure it was completely cleaned surfaces.....?
Are you using G12 and with proper mix ratio?

If the system isn't at full PSI it will run slightly cold......... due to lack of proper seal of the closed cooling system.

All of this effects boiling point.

ect and flange are genuine bosch parts.

everything looked clean. the old flange wasn't too dirty.

yes g12 and proper coolant ratio.

not sure about psi but the t stat seems to be opening and closing okay. before i put the new t stat in, i put it in a boiling pot of water. to test it and see if its opening and closing for good measure.

As per vcds measuring block readings I got the car up to 98 celcius (208 fahrenheit) on the highway yesterday. IS IT NORMAL TO REACH 98c?


Im sitting in the car right now. ran an output test in instruments. rpms seem to go from 0 to max then to 3000, coolant gauge goes from 0 to max then to 92c (roughly where it goes when the car is hot), like uwe mentioned; tach goes from 0 to max then to 100 km/h.

the car is cold right now. regarding the coolant temp gauge output test. if its meant to go from 0 to max and then to 90c (the middle), then I assume my cluster is slightly off ?? it went to 92c when i did the output test just now.

Regarding "Stream through 0x17-10-000". Does this mean select in VCDS --- > Instruments (17) ---> Adaptation (10) ---> Channel 000 ?

Thanks.

Jack@European_Parts said:

That's the mid point test of outputs you are referring to.............I think when doing a test.

Some clusters do offer an ADP I don't think this one does but you can check in VCDS if channels exist. Stream through 0x17-10-000>

Is the ECT and flange factory?

When you installed are you sure it was completely cleaned surfaces.....?
Are you using G12 and with proper mix ratio?

If the system isn't at full PSI it will run slightly cold......... due to lack of proper seal of the closed cooling system.

All of this effects boiling point.

Click to expand...

30secondgti said:

As per vcds measuring block readings I got the car up to 98 celcius (208 fahrenheit) on the highway yesterday. IS IT NORMAL TO REACH 98c?

Click to expand...

I believe that's within its "nominal operating range".

30secondgti said:

if its meant to go from 0 to max and then to 90c (the middle), then I assume my cluster is slightly off ?? it went to 92c when i did the output test just now.

Click to expand...

Yup.

-Uwe-

Uwe said:

I believe that's within its "nominal operating range".

Yup.

-Uwe-

Click to expand...

Nominal or normal?


Still seems a little off, but I think its as close as I'm going to get! (it seems when the car gets hot it goes a little past 92c, while the output test looks like its at 91c. I don't think there is a whole lot I can do accept for replacing the cluster.