First of all, I'm a neophyte here. I was a main-frame programmer until I retired, and am getting my feet wet in other pursuits. I want to repurpose an air conditioner and my first hurdle is to tell when the evaporator coil is freezing. It's done with a NTC thermistor, I think, but what method do I use to signal my SST control contacts to open, stopping the unit temporarily, until the coil has thawed and I can begin to operate again? If I have used the wrong terminology, please be patient. Thanks.

 

The NTC (Negative Temperature Coefficient) is merely the sensor and as the name implies a resistor whose resistance drops (goes negative) as the temperature increases. The actual component as used in a temperature sensing circuit could be a part of a Wheatstone Bridge or used in a Comparator circuit to control whatever you wish to control. A Google of "ntc thermistor circuit diagram" will bring up a dozen circuits.

This is a good read on Comparator Circuits and also this forum has a good read on the same. I would use a simple comparator circuit with Hysteresis so you avoid chatter when the condenser coils are hovering around 32 degrees F.

Ron

 

You can use either a Ntc or Ptc, as long as you use a Comparator chip with Hysteresis to set the Minimum and Maximum trip temperatures it will work. There are several dedicated comparator ics, or the standard op amps like Lm358, Lm324.

 

How about a thermostat from ebay?

http://www.ebay.com/itm/DC-12V-50-1...413547?hash=item1c6f7af5ab:g:9t8AAOSwBLlU~qJy

 

... unsolicited comment ... if the evaporator coil frequently freezes to the point of stopping refrigerant flow, there may be H2O contamination, suggesting a leak someplace, or that possibly a new filter-drier is needed.
.... So one permanent fix would be to have the Freon suctioned out, as per rules and regulations, and then perform a vacuum test for several hours.

 

there may be H2O contamination,

I don't agree that the first thing to do is the very expensive Freon System repair because water in the refrigerant is about the 99th most likely cause of frosting. The most likely cause is that this is a heat pump and it gets cold outside. The second most likely cause is reduced air flow, often caused by expensive air filters that the fan wasn't designed for or a dirty evaporator...anything that reduces air flow. Third most likely is a Freon leak. Not enough refrigerant in the system.

If the home-brewer wants to pursue this line of thought, he will reply.

Most commercial units use a Klixon to sense the temperature because the range of temperatures that should occur are far enough away from the temperature of a failure that a crude bi-metal is good enough.

 

Thanks Ron, so how far am I off, here? I went to the first suggested read and my path is becoming a little clearer. I'm going to attack that next read after I get my "electronics for dummies" book and become a little more conversant with all (at least a few) of the terms.

You gave me a place to start, and I am grateful. I'm sure I'll be here a lot more with a lot more questions.

How about a thermostat from ebay?

http://www.ebay.com/itm/DC-12V-50-1...413547?hash=item1c6f7af5ab:g:9t8AAOSwBLlU~qJy

I don't agree that the first thing to do is the very expensive Freon System repair because water in the refrigerant is about the 99th most likely cause of frosting. The most likely cause is that this is a heat pump and it gets cold outside. The second most likely cause is reduced air flow, often caused by expensive air filters that the fan wasn't designed for or a dirty evaporator...anything that reduces air flow. Third most likely is a Freon leak. Not enough refrigerant in the system.

If the home-brewer wants to pursue this line of thought, he will reply.

Most commercial units use a Klixon to sense the temperature because the range of temperatures that should occur are far enough away from the temperature of a failure that a crude bi-metal is good enough.

 

I don't have a problem, as such, with the evaporator coil freezing. As my user name implies, I brew beer, and am trying to repurpose an air conditioner to cool the fermentation chamber to as cool as 40 degrees f. The hurdle in doing that is that the unit has to run longer than it's designed to run, so some Rube Goldberg style modifications are necessary. The setup I have works, to a degree, but not as well as I would like. I "jumped" the thermostat to force the unit to run constantly, then inserted an on/off timer switch to have it run for a length of time, then stop for a length of time. It doesn't reach the desired temp, but gets fairly close. What I'm trying to do is to sense when the coil is starting to freeze, which can vary depending on the ambient temperature, humidity, amount of product in the cooling room, etc.. If I can implement the thermistor I can eliminate the guesswork inherent in the on/off technique. I think the sensor will allow me to achieve the goal, but am a neophyte in the realm of electronics, and that's why I'm here.

 

Then if your electronic skills are poor, i suggest buying a ready-made one from ebay like the one in post#4..

Similar searches....

http://m.ebay.co.uk/itm/272468443881

 

I don't have a problem, as such, with the evaporator coil freezing. As my user name implies, I brew beer, and am trying to repurpose an air conditioner to cool the fermentation chamber to as cool as 40 degrees f. The hurdle in doing that is that the unit has to run longer than it's designed to run, so some Rube Goldberg style modifications are necessary. The setup I have works, to a degree, but not as well as I would like. I "jumped" the thermostat to force the unit to run constantly, then inserted an on/off timer switch to have it run for a length of time, then stop for a length of time. It doesn't reach the desired temp, but gets fairly close. What I'm trying to do is to sense when the coil is starting to freeze, which can vary depending on the ambient temperature, humidity, amount of product in the cooling room, etc.. If I can implement the thermistor I can eliminate the guesswork inherent in the on/off technique. I think the sensor will allow me to achieve the goal, but am a neophyte in the realm of electronics, and that's why I'm here.

This should work for you,

and here's how it works:
V1 is a 12 volt DC source used to drive U2, (a voltage comparator) K1 (a relay), and U1, which is a 5 volt regulator used as an excitation source for the thermistor and a reference source for U2. RT1 is the thermistor, and it exhibits a resistance of 10000 ohms when it's at 25°C and some 33000 ohms when it's at 0°C:

Since the resistances of the resistors comprising voltage divider R2,R3 are equal, the voltage at their junction will be half of U1's output voltage, and that 2.5 volts will be presented to the non-inverting input of U2 (U2+) as a reference.

U2 functions, as do most comparators, by having its output go open-collector and, essentially, float, when its + input is more positive than its - input. When its - input is more positive than its + input, however, its output will be shorted to ground through one sort of transistor or another. In the case of the LT1716 it's a bipolar NPN.

OK, so let's say that U2+ is sitting at 2.5 volts, that RT1 is at room temp (25C), and that R1 is set to about 32k.

Since RT1's resistance at 25C will be 10k and R1 is set to 32k, then U2- will be sitting at V(U2-) = U1out X RT1 / R1 + Rt1 = 5V X 10000/32000 + 10000 = 5V X 10k / 10k + 32k = 1.19V.

Since 1.19 volts 1s less positive than the 2.5 volts on U2+, U2 out will be floating and there won't be enough base current through Q1 to turn it ON and energize K1.

As RT1 cools down, however, its resistance will increase and, as it does, the voltage on U2- will increase until it goes more positive than the 2.5 volts on U2+. When that happens, U2 out will go low, which will turn Q1 ON and energize K1.

R4 provides hysteresis, and determines what temperature RT1 must rise to before K1 is de-energized.

 

Aarghhh!!! I just lost a bunch of text, (mea culpa, I suspect, since I haven't yet figured out the vagaries of this site) so I'll post what hasn't

EM: Ia lot of cases on this site, you can bring text back even if the page disappears. You just have to navigate back to where you replied and the text will appear greyed. The Toshiba A665 series laptops have issues that don't seem to be acknowledged by Toshiba but wildly reported. The right shift works correctly. The left does not. So capital R's and capital W's made with the left shift are sometimes troublesome.

 

EM: Ia lot of cases on this site, you can bring text back even if the page disappears. You just have to navigate back to where you replied and the text will appear greyed. The Toshiba A665 series laptops have issues that don't seem to be acknowledged by Toshiba but wildly reported. The right shift works correctly. The left does not. So capital R's and capital W's made with the left shift are sometimes troublesome.

 

EM: Ia lot of cases on this site, you can bring text back even if the page disappears. You just have to navigate back to where you replied and the text will appear greyed. The Toshiba A665 series laptops have issues that don't seem to be acknowledged by Toshiba but wildly reported. The right shift works correctly. The left does not. So capital R's and capital W's made with the left shift are sometimes troublesome.

Thanks for the heads-up.

I tried the back-navigation and got back to the greyed text and clicked on it. My earlier text came back, but what I had written after that , didn't.

All in all, it's not a problem, just an inconvenience, since I can do what I want offline and then post the result as a chunk of text

 

To the OP. You may be able to change the expansion valve/orifice to change the temperature.
R12, for instance was use din both AC systems and refrigerators. You could also try adding a small amount of heat to the condenser. One thing you might be able to use is a defrost heater used on refrigerators.

I do have an EPA small appliance/automobile license. I put my car AC in the car from a box before the regulations were in place. This was about 14 hours of labor. It blew 2 hoses every 7 hoses like clockwork. I had the car 17 years. The final straw was a worn out clutch. I replaced it, with a bad used compressor, so it was now time to get rid of the car. I had an electronic leak detector I could borrow from work. I was given that one, but the last time I used it, it had battery issues. The batteries must have the recessed button. The company that made it, won't talk to consumers posessing the instrument. It's great for natural gas too.

The other time, it became useful, was when I was not licensed and an R-12 vehicle was diagnosed as needing an evaporator. The local shop, said it would just cost too much for him to put one in which is definitely true. So, he recovered what was left, got a new evaporator. I did the install and did the leak test and he charged it.

There's just too much stuff to invest in: Recovery system, gauges, scale?, temperature, torch, as, nittogen, license, UV lamp etc. I'd love to get the other license. It's not that I can't use the tools. The R-12 system with O-ring fittings and a sight glass without recovery was super simple to charge. Silver soldering is cake. Having UHV (Ultra High Vacuum) experience made it even easier. No one uses and RGA to find leaks in a fridge, AC system or a car. Fingerprints, water and contaminants are your enemy and you can't mix oils.

 

KISS! R12 has been gone since the mid 1990's and R22 was phased out about 10 years ago. If you're going to build your own, I recommend R134A because it's still available and is way easier to work with than 410A.

 

Agreed. That was then and this is now. Consider it an aside.

 

I think your wasting your time. It's maxed out now. You want a machine that can do twice what you want. And do it automatically.
One wants a controllable, reliable beer environment. Safe beer zone.
Except no substitutes.

 

What I'm trying to do is to sense when the coil is starting to freeze,

I think you're going at it the right way. Even if you used a machine that was designed to get to 40F, it would still frost because the evap must be below the target temperature in order to get the rest of the place to the target temperature. Just go with one of the thermostats already posted.

Three bucks in post #4? No reason to even dust off your soldering iron.