nubie pot question

[4x40ranger]

I have a project that I am think of using a pot to average between two thermistors . My question is what ohm pot should I get? and what ohm thermistors do i need? 10K or closer to .5ohm as my controller is looking for 10k

 

[4x40ranger]

I would think that i need 10k thermistors and i'm not sure about the pot. thank you.

 

[(*steve*)]

Just connect the two thermistors in series or parallel. neither will give you a true average, but without more information on what you're doing it's all I can suggest.

 

[4x40ranger]

I have two room sensors that are 10k thermistors that I would like to control one damper but i would like to control 25% / 75% input from the rooms as the rooms are not the same size or the same heat load. so i was thinking to usa a pot to dial in a reading from the two thermistors.

 

[shrtrnd]

My 2-cents worth is to be aware, if you're not, that they make linear potentiometers and also audio taper potentiometers. You don't want the log pots for this.
Like *steve* said. If you can give us a little more information about the project, we can probably give more
complete advice.

 

[(*steve*)]

If you have 4 thermistors arranged in series/parallel and you place the pairs of series connected thermistors together, the overall resistance will follow a pattern similar to that of a single thermistor placed in an environment with the average temperature (where average has an undefined meaning but both arms have equal impact).

Tho simplify this, instead of four 10k thermistors, you could use two 20k thermistors in parallel.

To extend this further, you could use 3 30k thermistors or 4 40k thermistors. (neither 30k or 40k are values you're likely to find)

A single 40k thermistor in parallel with a 13.33k thermistor would also acct in some respects like a single 10k thermistor, except that the lower value thermistor would have about three times the effect on the overall as the 40k thermistor.

While that works in theory, getting thermistors of those values would require(probably) connecting several in series/parallel.

There is also a question as to whether the averaging would do what you expect.

Assuming the averaging does what I expect, then with the thermostat set for 24C, the two locations could be 24 and 24, or 23 and 25, or 26 and 22

 

[4x40ranger]

i guess i just don't understand how the Ω on the pot will affect the output. this schematic is what i was thinking about doing.

 

[4x40ranger]

i think i see what you are saying that so long as the two rooms are averaging the setpoint then the thermostat would be satisfied. if it were set for 70F theoretically the two rooms could be 40F and 100F which would be undesirable. i would need some way to limit the sensors say if they are more than 20% different then each other to default to one or the other sensors.

 

[AnalogKid]

What is the "input"? A thermistor needs a source of current to bias it so a voltage can be produced across it. If you know the voltage, or current, or input impedance, or just about anything about the controller input, it is critical to working out what you are trying to do.

ak

 

[4x40ranger]

measured 4.88vdc i would assume that this is 5vdc across the input and com thermals. i didn't get anything trying to measure current through the meter. and i'm not sure how to measure impedance. normally the this controller is connected to one tstat that is a 10k thermistor

p.s. thank you for putting up with my inexperience. I'm an HVAC guy trying to solve a problem I'm having with a couple rooms that are supplied by one vav and one tstat.

 

[(*steve*)]

And we're just poor electronics guys. And this one wouldn't know a vav from a tstat.

 

[AnalogKid]

If in is designed to work with a thermistor, that says many things about the input circuit. If it expects a 10K, then as a first-pass guess I's use two 10K's in your parallel circuit with a 20K pot. The problem is that you need the series resistance of the pot to force the averaging, but that series resistance significantly alters the circuit's response to temperature since much of the total resistance now does not change with temperature. Not sure how to overcome that without the added complexity of an opamp circuit.

ak

 

[4x40ranger]

i'm going to give a try 13.3k in parallel with a 40k using the 13.3k in the main room and 40k in the other room. I just ordered a 30K and a 3.3k to wire in series with my existing 10k sensors.