aseer

Has anyone compared the temperature of the coolant at the ECU temp sensor (logged I assume) and the coolant at the input to the radiator (close to the thermostat and its thermal sensor - with the coolant valve fully open, engine running)?

cjernigan

iTrader: (13)

I haven't seen anything about it. It might exist though. There shouldn't be much difference between the two though considering the coolant in the radiator inlet and CLT sensor are both in or coming from the head. I would suspect the CLT sensor to read higher because it is in the back of the head where coolant flow is lesser and normally hotter.
Atleast that's my WAG.

aseer

Thanks, I took a modified radiator cap and stuck a thermocouple through it into the coolant flow and saw about a 20F difference (20F cooler at the radiator compared to what my MSPNP was reading, once the valve opened and long enough to reach steady state - the thermocouple was calibrated).
I did not have a feel for what to expect however.

Braineack

iTrader: (62)

why do you wanna know? id wanna know how hot the coolant just leaving is, not how cool the fluid entering is...

aseer

I wanted to get a feel for how different my fan control temperature would be if I used the MSPNP control compared to using the thermal sensor at the thermistor near the radiator (which is how I think the stock cooling fan is controlled).

{I have a 91, not sure if that makes a difference}

Braineack

iTrader: (62)

itll probably be close enough. thermoswitch sits behind the thermostat, and once open that's where most the flow goes, so im sure it's within 10° of #4.

aseer

I redid the measurements, the differential is closer to 11 degrees as suggested by Braineack; 11 degrees hotter at the head compared to the stock fan switch point (though my MSPNP seems to read CLT in 4 degree increments so the differential error is on that order - can I reduce the granularity through Megatune or some other mechanism?).

JustinHoMi

Well, it's an 8 bit variable, so the cpu is going to see it as a value between 0 and 255 (so only 256 possible values across the temperature range). Take a look at thermfactor.inc and you'll see the mapping. It looks like the range is from -96F to 408F. If you narrowed the range down to something that would work in your climate, you could get more resolution.

You'll need to reload the firmware with the new table if you make changes.

Laur3ns

My MSPNP shows 83C and my VDO gauge shows 83C when the engine is warm. The VDO sensor is just before the radiator inlet.

They only differ during warmup when the stat hasnt opened yet.