What is this (picture)

[mattag]

It's the little resistor thing soldered to the red and green wires. It is the only component in a thermostat I'm trying to modify to a different temperature range that changes resistance with temp. At room temp it is ~40K, it drops to ~32K at body temp (closed in my hand). It behaves as a thermistor but doesn't look like one. All the others I find are lollipop shaped. Like the little blue thing in the second pic. which doesn't show continuity or is of infinite Ω. There are no number or other identifiers on it.

While I'm at it, is this all a waste of time? This is an off-the-shelf heat only thermostat from home depot. It's intended for low voltage control of a home furnace. It's range is ~45-90 deg. f. and the resistance of the unidentified thingy is ~40 at 65 deg f. I need a range of ~120-180 deg. f. I'm assuming that if I substitute a thermistor that shows the same resistance of ~40K in the middle of my desired temp range at say 150 and has a similar reaction curve the IC will behave as it does at lower temps. Is that crazy?

Obviously, I'm out of my league when it comes to electronic theory. Any help is greatly appreciated.

 

[theo92]

I think the first one is a diode; I think it's a zener diode or small signal diode.
And the second one is a SMD circuit board with really dirty solder. I can't gues what is it..

 

[mvs sarma]

it is a n MOV I feel. Metal oxide Varisters shoe infinite impedance until they reach the breach their breakdown voltage and at that point it will suddenly fall.
It might have a value marked on its other face.

 

[mattag]

Yeah, ran out of electronic solder and had to use plumbing paste. Made a mess. It might be a diode but it shows the same resistance in both directions and the resistance changes with temp. The second picture was only to show the blue round thing on the upper edge that looks more like what I associate with a thermistor. When I heat or cool the diode looking thing the thermistat cycles. That why it's soldered to leads. I wanted to be able to isolate temp changes to be sure that is the sensor. From what I can tell, it is how the IC senses temps. If I heat or cool the PCB without the component in question nothing happens. But when I change the temp of the diode the thermistat cycles. It has to be the sensor.

 

[mattag]

mvs sarma, I've looked at it with a magnifying glass. There are no numbers. What is a n MOV? I just replied to theo92 that it is definitely the temp sensor.

Thanks.

 

[mattag]

Oh sorry metal oxide varistors. Looking that up now. Thanks!

 

[mvs sarma]

Oh sorry metal oxide varistors. Looking that up now. Thanks!

Check for NTC thermistor.(the one on the board)

The hanging one appears a diode.

 

[rezer]

I don't think it's an MOV. I have never seen an MOV in this style case before. But then, that doesn't mean it isn't. It does look like a diode. Diodes are often used in smoke detectors to detect heat, so it probably serves the same purpose here (without the fire).

 

[mattag]

Yeah the Wiki definition of MOV doesn't work. Shouldn't a diode show very low resistance in one direction and very high in the other? This one is exactly the same in both polarities and definitely changes a great deal (relative) in response to heat and cold. The the blue thing is labeled as SAS 101KD05 which as search says is a varistor. The first thing I did was to isolate that component but it has almost no resistance (it flickers between .001 and .002 ohm) and the resistance doesn't change with temp. I just held a lighter to it for 3 or 4 seconds and the meter didn't change at all.

I guess whatever the case, modding this unit won't work if the components are that exotic.

What about this **broken link removed**. It does use a standard NTC thermistor. Would substituting another thermistor of a different value change the the temp range. I tried changing to a higher resistance pot and it didn't do much. I got the max to ~100 with a 2 meg ohm pot which was the most resistant I could find.

 

[rezer]

Did you make your measurements with the device removed from the circuit? Have you tried to test it with a DMM using the diode test function? Yes, you should meas. a low resistance in one direction compared to the other.
One thing to remember, you may meas. the same resistance across the diode if it's in-circuit.

 

[mattag]

I just chopped the leads. The resistance jumped to 55K?? but it still reads the same in both directions and goes down by 8K when I hold between my fingers then goes back up when I let it cool. How could eliminating the PCB increase resistance? I don't know what a DMM is. I have what was a very expensive VOM when I bought it in 1985 and, up to this point, has served me well.

Let me cut to the chase. I need to control the temperature of a mash tun (beer brewing) via a 20 amp 240 volt relay with a 12 vdc trigger and a 3800 watt hot water heater element. I initially tried a hot water heater thermostat but they're all thermo-o-discs which is mechanical and means a bimetal strip which means huge hysteresis and an accuracy of 10 deg f if you're lucky. I then ordered 2 MK138 thermostat kits. I assembled the first one and it did exactly what the manufacturer said it would, that is it very accurately controlled temps between 40 and 90 f. I changed out the pot and fried something while 10 thumb soldering. I assembled the second but substituted a 2 meg ohm pot in place of the 100K one included. That upped the range by 10 or 15 degrees f. I ordered a grab bag of thermistors from allelectronics and was in the testing phase when I set the voltage of my borrowed DC power supply to 48 and fried that one. I had this home depot therm sitting around and couldn't find the thermistor and we are back at the present.

Will substituting a different value thermistor in the MK 138 kit therm **broken link removed** do what I want?

I know I'm asking a lot of stupid questions but rezer I'm one of those who doesn't understand binary. My trig prof had one of those clocks. What time is it, I don't know, just a bunch of pretty blinking lights to me.

Thanks for your time so far man.

 

[rezer]

I would say your glass device is most likely a thermistor.
By looking at the MK 138 device, it appear to use an op-amp in a comparator application. The thermistor will go to one of the inputs with the other input set to a specific reference controlled by the pot. You can use a different thermistor, as long as it has a negative temp. coefficient. This means the resistance of the device will decrease with a rise in temperature or increase as temperature decreases. You could use one with a positive temp coefficient but the relay would work in the reverse order.
You will need to take a look at the resistance of the thermistor you want to use compared with the one that comes with the kit. Chances are, yours has a wider resistance range but you can verify this with your VOM. If you are needing a wider temperature range, than you will have to modify the module. Judging by the simplicity of the device, it should not be overly complicated to reverse engineer it. This may only require a resistor change.
This may be a topic for a new thread.

 

[Andrew Leigh]

Hi,

not an expert but I have made a couple of thermally controlled fan circuits using a diode as the sensor. Not my design, I build by numbers. I would not at all be surprised if this is not a diode. Have built the following circuit which uses diodes as temperature sensors, working to this day.

**broken link removed**

If you want a nice circuit to control hysteresis to the nth degree have a look here. You will need to upgrade your switching relay. You dont want the hysteresis too tight else the switching cycle gets too much. Read the following article with schematics, about to build one myself.

https://www.craig.copperleife.com/tech/thermo/

Cheers
Andrew

 

[rezer]

Hi,

not an expert but I have made a couple of thermally controlled fan circuits using a diode as the sensor. Not my design, I build by numbers. I would not at all be surprised if this is not a diode.

The problem I have with it being a diode is the absense of the cathode band. Thermistors will usually come encapsulated in a glass housing such as this.

 

[killivolt]

It's the little resistor thing soldered to the red and green wires. It is the only component in a thermostat I'm trying to modify to a different temperature range that changes resistance with temp. At room temp it is ~40K, it drops to ~32K at body temp (closed in my hand). It behaves as a thermistor but doesn't look like one. All the others I find are lollipop shaped. Like the little blue thing in the second pic. which doesn't show continuity or is of infinite Ω. There are no number or other identifiers on it.

While I'm at it, is this all a waste of time? This is an off-the-shelf heat only thermostat from home depot. It's intended for low voltage control of a home furnace. It's range is ~45-90 deg. f. and the resistance of the unidentified thingy is ~40 at 65 deg f. I need a range of ~120-180 deg. f. I'm assuming that if I substitute a thermistor that shows the same resistance of ~40K in the middle of my desired temp range at say 150 and has a similar reaction curve the IC will behave as it does at lower temps. Is that crazy?

Obviously, I'm out of my league when it comes to electronic theory. Any help is greatly appreciated.

Typical for an MOV in a circuit like this and if the Zener is in line with the MOV it is a safety. The Zener shorts and the MOV opens.

 

[rezer]

Typical for an MOV in a circuit like this and if the Zener is in line with the MOV it is a safety. The Zener shorts and the MOV opens.

Not so, an MOV is designed to short out when the clamping voltage is reached. Typically, it shorts to ground. The blue component in the 2nd pic is probably an MOV.

 

[killivolt]

Not so, an MOV is designed to short out when the clamping voltage is reached. Typically, it shorts to ground. The blue component in the 2nd pic is probably an MOV.

Right.

Once the operational voltage in the Zener is reached and it saturates the MOV does short blowing the fuse opening the circuit oops.

kv

Edit: I didn't include the fuse portion.