I've got my prototype PCB and should be installing it soon in my car. I've made my DIP switch select setting the fan on temps in 5 degree F steps. The fan off temps are set at 5 degree F below that setting. I set the alarm (buzzer) at 10 degree above the fan on temps. These are all just my guess of what should work so testing in my car will determine if those values need to change.
I was also thinking of adding a voltage alarm in this unit looking for a sudden drop below 12.5 volts since that would result if your fan belt broke and the alternator and water pump stopped turning. If the water pump stops, then the coolant flow would stop and therefor the coolant temp would not rise to trigger an over temp alarm.
The "buzzer" seems pretty loud on bench testing. Not sure when mounted in the car with the relay cover plate on if the volume will suffice. Also needs to overcome radio background music.
I wrote the software to see a sensor failure and turn the fans on and beep 5 times. That works good but then I put the sensor into a glass of water and the sensor pulse level changes due to the water conducting some current away from the sensor. It still works but I don't like the levels changing. The sensor puts out current pulses (very small currents) related to the temp. So my software counts the pulses to know the temp.
So I either have to water proof the sensor or mount it on the outside of the coolant pipe. I'm thinking mounting outside the pipe should read the same temp as the coolant inside as long as not much air is blowing over that sensor. Just wrapping it with Teflon tape should prevent any air cooling or heating.
Now this is an interesting test. I taped (wire wrap tape, no sticky) a thermocouple on the metal pipe passenger side feeding the water pump. Air temp is 60 deg. F so fans never came on and dash temp gauge peaked at just below the 1/4 mark.
I was not expecting the temp to rise until the thermostat opened so I missed the first few minuets.
2:39 = 101 deg F
2:40 = 114 deg F
2:41 = 126 deg F
2:42 = 137 deg F
2:43 = 147 deg F
2:44 = 155 deg F
2:45 = 163 def F
Then I saw a peak around 166 deg F and it quickly dropped to 141 deg F and it kept doing that cycle. So the passenger side is following the engine temp. I assume the pump keeps water flowing in the heater core. Then when the thermostat opens the flow through the radiator cools the engine down below the thermostat temp and closes the thermostat.
I thought the thermostat was designed to open at 188F...
And that the water pump sends the coolant in a clockwise direction to and from the radiator.
I wonder what results you would get if you did the same test on the drivers side metal coolant pipe.
Here is the same run with a thermocouple taped to the outside of the otterstat pipe.
3:05 = 58.2 deg F
3:06 = 59.1 deg F
3:07 = 60.0 deg F
3:08 = 61.5 deg F
3:09 = 62.7 deg F
3:10 = 64.5 deg F
3:11 = 66.0 deg F
3:12 = 67.2 deg F
3:13 = 68.1 deg F
3:14 = 69.6 deg F
3:15 = 71.4 deg F
3:16 = 72.0 deg F
Then right at 3:16:01 the temp rapidly climbed to 167 deg F in about 55 seconds, then slowly dropped to 165, then climbed to 174, dropped to 171, climbed to 176 deg F.
So my thermostat is cutting all the flow off until it opens. Yes probably the coolant temp is a little higher than reading the outside of the pipe. It also is another cool day with air temp at 58 deg F. I assume my thermostat is stock. Normal dash gauge stays right around the 1/4 mark. Fans did not kick on today and temp gauge stayed a little below the 1/4 mark.