Temp controlled fan circuit

[electrookie]

Calling all guru's:

This circuit was provided by a gentleman in this forum and is being used to turn on a fan when temp is 160 degree F and turn fan back off at 130 degree F when the area is cooled down a bit. I have it working for turn on, but have a heck of a time getting it dialed in for turn off. I know it’s a comparator circuit but I don’t understand very much about it. Originally, R2 was a 3.3K resistor, which would give a narrow range to get down to for turn off. I tried with as much as a 5 meg pot, down to a 10K pot and from 0 ohms to 5 meg, I can not find that point for turn off. I apparently have something wrong. It’s hard to tell what R2 is connected to in this picture. R2 does not connect to R3 but it does connect to R4.

So, my ? to you guru’s (anyone who understands more than me…) out there, can you give me any ideas how to get this thing to work properly, or, perhaps you have a different circuit. I can read and build from a schematic, but not real good at designing anything. I know there are even IC’s designed to do this task specifically, LM56 is one I tried, 4 times I built that tiny chip’s circuit, and apparently smoked the IC 4 times… and I know how to work with those tiny guys too, but with 10 thumbs, I would rather avoid SMD’s.

Any suggestions, help, guidance, anything any of you can offer, I’ll build it again. It would be better than what I have now, which is a headache for the most part.

Thanks all.

 

[Mr RB]

Hi electrookie. That looks like my hysteretic controller circuit.

It looks ok, asuuming that R1 is your thermistor. 500k is probably much too low for the R2 hyst feedback resistor. Your thermistor might be 100k or so at the switching temp. With 5 meg ohms for R2 I'm surprised it didn't work fine.

Can you tell me the voltage across C1 at room temperature, and again at HOT temperature near where you want your switching point? (Please disconnect the pot centre wiper too when testing that if you can)

I think your thermistor might be quite different to the ones I normally use with this type of circuit as it normally works with little drama.

 

[ke5frf]

An interjectional idea here, but if simplicity is what you desire and you aren't too invested in your current project, a simple thermostatic switch with a hysterisis adjustment might be the perfect no-brain solution....and you should be able to get one fairly inexpensively even at a hardware store.

You may say KISS applies here.

 

[electrookie]

Fan temp control circuit

“Hi electrookie. That looks like my hysteretic controller circuit…”

Hi Mr RB, it is that one. In the original drawing, R2 is shown as 330K. And R1 is shown as 10K (I believe R1 is the thermistor and got some of them to try out). So all things considered, the schematic looks ok to you, then I must have something wrong. Are you saying the thermistor needs to be 100K as opposed to the 10K? For R2, I have pots and tons of resistors to play with, just need the right value. I will get back to you today with the numbers you ask for, after church, we have a function to attend, then I can get back onto this. I will certainly let you know as soon as I can. Thanks for the help.

ke5frf, any idea is appreciated. KISS is the concept I prefer and the given circuit looks to be just that, but not turning out that way yet. You cite a “thermostatic switch with a hysterisis adjustment” even available at a hardware store. That IS simple if I can find such an animal. I have never heard of this, can you give me a little better idea of what it is and I will swing by Home Depot on the way home today and see if they have such a thing. I am sure if I ask some worker there for the given item, they will have no idea what I am asking for, especially when I am not real sure either. Thanks for the idea.

 

[ke5frf]

A thermostat?

Well, as far as I know they still sell electromechanical thermostats for your central heat and air, and you can get them at Home Depot or an A/C supplier. Some are mercury switches, some bimetallic switches with mechanical contacts, some even gas bulb types. You can get them for hot water heaters and furnaces too.

There is usually a MECHANICAL hysterisis adjustment that keeps the thing from cycling too quickly, you know. I'm not sure, however, if the range you mention (delta of 30 degrees F between on/off) can be accomodated without some creative manipulation of the mechanical bits.

A brief explanation of how a thermostat works: Inside, there is some sort of switch connected to a temperature sensitive COIL or SPRING. Sometimes it is bimetallic and the coefficient of expansion or contraction of the two metals is different, causing movement of the coil which makes/breaks the contacts of the switch. Think of those party favor toys that curl up tight but when you blow on them they pop out and toot. Sometimes the movement is accomplished by a gas filled bulb that is coiled in a similar way, and when the gas is heated it causes the metal tube to expand or contract and make/break the switch. Or sometimes yet, it is a simple spring of bimetallic design.
Sometimes, the make/break isn't contact points but rather conductive mercury in a glass bulb, when tilted the level of the mercury causes a complete circuit between electrodes in the bulb.

Anyhow, such mechanical switches often have springs, counterweights, or some other "resistance" to overcome that can be adjusted to keep the switch movement from being so sensitive and thus cycle too frequently. A dampening effect if you will. That is what we mean by hysterisis, kind of a delayed reaction.

 

[electrookie]

Fan temp circuit WORKS

Hi Mr RB

I built the circuit with all new parts, again, but on a bread board for easy changes. IT WORK's, go figure. OK, what I found is that the 2 R values do interact with each other and you must tweek them to get what you want, and they are both very sensitive as far as the slightest change in one or the other can be a major difference in the thresholds. So it was all about getting them dialed in just right. Now I do have one more ? for you Mr RB, is that 100uf cap going to cause any drift or effect the thresholds I set the circuit for. I used an electrolytic cap and am wondering if the circuit is better served with a tantalum cap instead. If it is the case, then I will get the tantalum as I want this thing to stay where I set it.

I appreciate the help very much. I have other projects in the wings and when I bump my head again, and I will, I know where to ask the guru's for help. More than just the help, I appreciate being taught and learning as we go. You guys are a blessing.

Take care...