If you want to know why the stock temp gauge is useless for measuring coolant temp read this
http://www.supramania.com/forums/showpost.php?p=655455&postcount=1
http://www.supramania.com/forums/showpost.php?p=655455&postcount=1
A/C fan! why does it turn on if my a/c is off?
- Thread starter 3tc power
- Start date
Alec;1493797 said:
In that case it's simply heat soak, don't worry too much about it. You do have a stock radiator though, so keep an eye on it...
Alec: It's there as a backup to assist the stock clutch fan, and as nosechunks said, they are set to come on at 212*F, so I would highly recommend installing an aftermarket coolant temp gauge, especially after looking at the graph that 3p just posted.
You could try installing an aftermarket gauge into the cluster and use the stock needle, like some have done with their boost gauges.
3p141592654;1493783 said:
For the "metric challenged" - here's a version in Deg F:
Non-linear behavior (inaccuracy) is built into the gauge design.
In other words, it's just an idiot light.
:biglaugh: @ "metric challenged"
Well I knew it functioned as more of an idiot light then a gauge but I didn't think it was that drastic. Thanks 3P for the graph.
Well I knew it functioned as more of an idiot light then a gauge but I didn't think it was that drastic. Thanks 3P for the graph.
Alec;1493803 said:
Unless you've changed since your pics, it's a stock replacement radiator. It may be made by CSF but it's still a stock size and quality.
nosechunks;1493875 said:
hence the countdown......
If the fan is coming on it's running HOT no matter what the stock gauge is saying..
I measured it myself. Feel free to measure your own if your don't "buy it".
It could be made linear. Its actually a very high quality movement. There is a zener diode purposely added to it to make it non-linear. Another project I will never get around to doing.
Car companies have learned that gauges that move around during normal operation generate warranty claims. Hence, the famous Ford oil pressure gauge connected to a switch, its either A-okay, or 0 pressure. Same with these temp gauges, designed to stay put until something goes really wrong.
It could be made linear. Its actually a very high quality movement. There is a zener diode purposely added to it to make it non-linear. Another project I will never get around to doing.
Car companies have learned that gauges that move around during normal operation generate warranty claims. Hence, the famous Ford oil pressure gauge connected to a switch, its either A-okay, or 0 pressure. Same with these temp gauges, designed to stay put until something goes really wrong.
3p141592654;1493920 said:
i didn't mean any offense. just that my stock gauge usually sits at around 40% at normal operation in the summer. according to 2 different aftermarket gauges i've had (greddy and autometer cobalt), both tapped at the tstat housing, usually peak at 200F and typically run ~185F. unless i am reading your graph wrong, it is well over 212F according to the stock gauge?
3p141592654;1493920 said:
Early and late miata oil pressure gauge is the say way. Early was an actual guage, late was an idiot light looking like a guage...
None taken.
To get this data took some pain. I measured deflection as a function of thermocouple resistance assuming an operating voltage of 13.8V. I used a programmable power supply to simulate the thermosensor. I made a graduated scale for the gauge so I could tell the difference between 20% and 30% since the factory scale is useless.
I then fit a curve to the thermosensor resistance vs temperature. I used the data from the TSRM, since I was too lazy to measure the real sensor.
temp Resistance
degC ohms
==== =================
50 198.5
60 152.7
115 26.4
120 19.6
I fit a nonlinear curve to the data, inverted it in excel and computed temperature vs deflection. I don't claim to know if the TSRM sensor data is accurate, I assumed it is. If your gauge or sensor deviate from what I used you could expect some variation in needle position at op temp.
Here's a plot of the sender resistance, and my fit to it.
To get this data took some pain. I measured deflection as a function of thermocouple resistance assuming an operating voltage of 13.8V. I used a programmable power supply to simulate the thermosensor. I made a graduated scale for the gauge so I could tell the difference between 20% and 30% since the factory scale is useless.
I then fit a curve to the thermosensor resistance vs temperature. I used the data from the TSRM, since I was too lazy to measure the real sensor.
temp Resistance
degC ohms
==== =================
50 198.5
60 152.7
115 26.4
120 19.6
I fit a nonlinear curve to the data, inverted it in excel and computed temperature vs deflection. I don't claim to know if the TSRM sensor data is accurate, I assumed it is. If your gauge or sensor deviate from what I used you could expect some variation in needle position at op temp.
Here's a plot of the sender resistance, and my fit to it.