Electronic Otterstat

[Bitsyncmaster]

I'm still waiting on some parts to setup calibration of my electronic otterstat. So I wrote software that checks the 12 volts. It will warn the user if the 12 volt (car power) drops below 12.5 volts for one second by sounding 5 beeps from the "buzzer". I do that just in case your alternator belt breaks which would quickly over heat the engine. I only do that warning once just in case your alternator goes bad you would not want that constant beeping. But it looks like the temperature gauge sender responds very quickly anyway but if you get a warning a few seconds earlier all the better to let you pull off the road sooner.

I will also warn the user if the unit detects higher temps than normal. The user will set the fan on switch point and I will use that setting plus some yet to be determined increase to warn the user. I'm guessing about an 1/8 of a gauge increase to set that warning. Will probably keep that warning sounding.

[Bitsyncmaster]

I got my high power variable resistor today and recorded the resistance that the sensor must have to show each value on the dash temperature reading. I did this with the engine running but the gauge compensates for the car voltage. You can verify the compensation by stopping the engine (alternator stops) and see the gauge still reads the same. My alternator puts out about 14.5 volts running. I also recorded the voltage over the resistor just so I could compute the current in the circuit. You can not use the voltage for calibration since that varies with the car voltage. I "eyeballed" the gauge as close as I could.

For the gauge reading at 1/8 scale the resistance on the sensor is 244 ohms at 11.60 volts.
For the gauge reading at 1/4 scale the resistance on the sensor is 154 ohms at 10.49 volts.
For the gauge reading at 3/8 scale the resistance on the sensor is 117 ohms at 9.56 volts.
For the gauge reading at 220 the resistance on the sensor is 94 ohms at 8.91 volts.
For the gauge reading at 5/8 scale the resistance on the sensor is 83 ohms at 8.38 volts.
For the gauge reading at 3/4 scale the resistance on the sensor is 73 ohms at 7.83 volts.
For the gauge reading at 7/8 scale the resistance on the sensor is 64 ohms at 7.30 volts.
For the gauge reading at 260 the resistance on the sensor is 55 ohms at 6.66 volts.

[Bitsyncmaster]

So with my test data, it looks like not much change in sensor resistance results in pretty large changes in your reading on the temperature gauge. Most of us see the temperature read around the 1/4 mark for normal running temps. Some people see running around the 220 mark for normal running temps.

I don't know if the gauge calibration or the sensor calibration is different causing those differences. It seems pretty sensitive calibration for both the gauge and sensor.

You could adjust the sensor calibration by adding a resistor in series with the sensor. But it's not a linear change so your calibration would only work in a limited range.

[DMC5180]

So with my test data, it looks like not much change in sensor resistance results in pretty large changes in your reading on the temperature gauge. Most of us see the temperature read around the 1/4 mark for normal running temps. Some people see running around the 220 mark for normal running temps.

I don't know if the gauge calibration or the sensor calibration is different causing those differences. It seems pretty sensitive calibration for both the gauge and sensor.

You could adjust the sensor calibration by adding a resistor in series with the sensor. But it's not a linear change so your calibration would only work in a limited range.

it wouldn't have anything to do with this blue device, would it?




Sent from my iPhone using Tapatalk

[Bitsyncmaster]

it wouldn't have anything to do with this blue device, would it?




Sent from my iPhone using Tapatalk

That would probably be the gauge calibration resistor. How most old automotive gauges work is there are two coils inside the gauge. One coil provides magnetism using the cars 12 volts. The other coil is powered through the sensor circuit. That is how they make the gauge accurate no matter what the voltage is.

With newer cars they use a regulated voltage (5 volts) so the gauges don't require so much power to operate.

It may be coincidence but the temperature gauge full scale resistance (54 ohms) is equal to the power coil resistance and hence would have the same current in each coil.

[Bitsyncmaster]

I had bought a gauge sender from Hervey a long time ago. But I never installed it. I will test that sender but the thread size is M14 1.25 so not sure if that is the correct sender for our D. I was looking at NAPA sender specifications for resistance and they are all over the place. The important resistance you want is at 220 deg. F which I measured on the gauge to be a sender resistance of 94.4 ohms.

[Bitsyncmaster]

I tested that Hervey sensor. Glad I never installed it. It would read very low on my dash gauge.

100 F = 1,530 ohms
120 F = 1050 ohms
140 F = 666 ohms
160 F = 447 ohms
180 F = 310 ohms
200 F = 227 ohms
212 F = 168 ohms

I would need to use another fluid other than water to test higher than 212 F.

[Horsebox]

100 F = 1,530 ohms
120 F = 1050 ohms
140 F = 666 ohms
160 F = 447 ohms
180 F = 310 ohms
200 F = 227 ohms
212 F = 168 ohms

I used Excel to plot those values on a graph and it looked like a logarithmic drop in resistance as a function of increasing temperature. It would be interesting to put both Hervey's and DMC's sensors in the same water bath and compare the difference in rate of resistance decrease.

I know the dash gauge isn't linear...when I overheated my car last year the temp gauge moved upwards very quickly as soon as the coolant temp was out of 'normal'. It only took a couple of minutes in stop-start traffic, with the otterstat failed and not turning on the fans, for it to overheat.

[Bitsyncmaster]

I used Excel to plot those values on a graph and it looked like a logarithmic drop in resistance as a function of increasing temperature. It would be interesting to put both Hervey's and DMC's sensors in the same water bath and compare the difference in rate of resistance decrease.

I know the dash gauge isn't linear...when I overheated my car last year the temp gauge moved upwards very quickly as soon as the coolant temp was out of 'normal'. It only took a couple of minutes in stop-start traffic, with the otterstat failed and not turning on the fans, for it to overheat.

I would order a new DMC sensor if it wasn't for the $20 shipping charge. I would like to test it the same way (I made a fixture that stands it in the water that keeps the connection side dry). My meter sensor for reading the temp is held right near the sensor I'm testing.

Yes the sensor responds very fast to changes.

[Farrar]

My gauge reads pretty accurately below 220°F, and then shoots up dramatically as the engine warms up. A temperature gun pointed right next to the sensor showed only 219°F when the gauge was nearly pegged. I wonder if the problem would be the sensor or the calibration resistor. I've never known about the calibration resistor until now.