Page 4 of 4

Re: TC Running Temp

PostPosted: Mon May 22, 2017 8:58 am
by Elanman99
Chancer wrote:You are right Andy, the advantage is that (at S.T.& P.) water will boil at exactly 100?c, infra red temp guns (I have a couple) are great as comparators but the indicated value will change according to the surface finish of the item, its not the best way to measure coolant temp.

I have a problem on my Skoda, the guage which was always rock steady at 90?c now sits at 80?c and sometimes goes up a bit, used a 330 ohm resistor and got the half scale deflection so guage OK, IR temp gun inconclusive, its either the sender (been replaced once for fast idle problem) or the stat sticking open a bit, tested stat in pan of boiling water, it opens fully but inconclusive regarding 80 or 90? so I will soon be putting the sender in a bioling kettle and will know for sure what to replace.


I presume your Skoda is a VW based car?

The 'rock steady at 90C' is actually only vaguely related to coolant temperature. Well not vague as such but the value displayed on the gauge is not direct from the sensor but is modified by the software in order to present a comforting feeling to the driver. The VW gauge is an improvement on just having an over-temperature warning light on the dashboard as it will indicate both high and low temperature but it is a very non linear measuring system. VW (and other motor manufacturers I imagine) arrange the gauge reading so that as long as the coolant is within a temperature band that they consider acceptable for running conditions, will show the 90 degree mid scale value. The driver does not really need to know if the engine is at 84.2 or 91.3 degrees, the driver just needs to know there are no problems.

So, if your gauge is not showing 90 I suspect the engine is not running at the temperature its supposed to.

Ian

Re: TC Running Temp

PostPosted: Mon May 22, 2017 9:42 am
by Chancer
It isn't I'm fairly sure, another job on the to do list.

I have had the cluster on the test bench of the UK's biggest independant repairer to replace the stepper motor for the fuel guage, that is how I know that 330 ohms is needed to calibrate the half scale deflection.

The coolant temp sensor has 2 thermistors, one for the ECU the other drives the temp guage without any computer correction, I replaced the original sensor as I had a fast idling problem (a known common fault) but the temp guage was rock steady, I used a cheap pattern part and that gave the right signals to the ECU but the guage read low like now, I then found an OE one in a breakers yard but think that now has gone outside of tolerance on the temp thermistor, I have a spare one in the boot but cannot recall if its yet another I picked up or the not so good pattern part.

Re: TC Running Temp

PostPosted: Mon May 22, 2017 10:23 am
by Elanman99
My observations on the software modified temperature gauge readings obviously does not apply to older or simpler vehicles that have direct sensor to gauge wiring.

It would not surprise me to know that even with very simple sensor and gauge arrangements, the meter electromechanical design had deliberate non-linear characteristics to 'flatten' the needle deflection around mid scale.

Ian

Re: TC Running Temp

PostPosted: Mon May 22, 2017 10:49 am
by Chancer
Yes, its a very old vehicle, if you look carefully at the other temp graduations, they are all deliberately in very small font except for the 90? mid scale and they are very non linear and as Ian says most of the range of the needle is expanded around the normal running temp.

Just swapped out the sensor, I've spent more time writing about it :D I have probably refitted the original failed one though which I should have thrown out were I to admit to myself that I have no memory.