Thermo electric pcb power supply

[keeney123]

Re: Thermo electric pcb power supply

Uininer how cool is the unit getting to be?

[uininer]

Re: Thermo electric pcb power supply

Hi all, I just joined the site and am amazed at the level of detail you have provided for these situations. I too have a Haier cooler with two compartments (4 sets of fans with peltier devices) that quit working. This happened a couple years ago on one side so I bought a new board and replaced it...fixed the issue. This time both sides quit working so I disconnected my peltier devices after reading this post and had the same result. The fans fired up and, while still running, I plugged in the peltier devices and it all works again. I'll be replacing all of the solder on my board and heading to the store to find replacement caps over the next few days. I may have questions but I really appreciate the multiple detailed posts on this site. It allowed me to troubleshoot my system instead of buying more crappy boards.

Update...I started by replacing the larger caps since that's what the guys at the electronic store suggested, so I would know which ones failed, but the unit still didn't work. I then continued to change out all of the small capacitors and...viola...it's been working great since. It cost me $12.50 to change all (14) capacitors on each board. I greatly appreciate the feedback and direction you have been able to give me.

[keeney123]

Re: Thermo electric pcb power supply

The Peltier refrigerators are just coolers they do not actually freeze or claim to freeze. The lowest possibility on the refrig. I had was 37 degrees F when ambient is 70 F or 2.7 C at ambient 21.1. With the Chinese Peltiers it would only get down to 64 F or 17.78 C at ambient 74 F or 23.3 C This was running a total of about 4.4 to 4.6 amps at 12 volts DC. Their claim was it would run at 8.8 amps at 12 volts. and they claimed that the Peltiers could take 13 amps of current at 15.4 volts. Of course they never got hot. A person I was helping out with a wine cooler got some Peltiers made in America and they pulled about 12.4 to 12.8 amps and they put out heat which is exactly what one wants because it is virtually a heat pump. Taking the warm air from inside pumping it to the outside. If there is not enough energy doing this then the 1 to 2 cubic space will not cool down enough before the outside air heats the compartment up again. Those little Peltiers have to work awfully hard to cool such a space down. According to the specification and graph of the Peltiers the cooler the inside space is the less current the Peltiers draw. When both surfaces of the Peltiers are about the same it draws maximum current. They are self regulating according to the voltage that is applied. The temp sensor inside the refrig or cooler starts shutting down the voltage when the desired temperature is achieved. In this way one will not get cooler than what the rheostat is set to so as you get closer to the desired temp the unit will shut down and turn on in steps until the temperature is correct. When one first turns on the refrigerator or cool it goes full on trying to obtain the correct reading and it runs that way until it gets the unit cool to where the thermal couple kicks in. The last thing I heard from the person with the wine cooler was he only got his down to about 64 F and the Peltiers that were hot became warm. This could of been caused by the thermal couple or the rheostat or the Peltiers could of burned out. If the Peltiers burned out then they would be defective and not meet specs. because the supplied voltage creates the current through them. He had said he used thermal grease on both sides of the Peltier so the device should not of burned out unless the thermal grease was no good and dried out. I never heard what his result was. I think he just gave up as the cost of the device at $21 a piece kind of prohibit replacing them. The Chinese version cost about $3 a piece but they only produce 1/2 the current. All the above currents are for a total combination two Peltier devices "not" one.

[Behemot]

Re: Thermo electric pcb power supply

*Refrigerator, I must have been sleeping

Peltiers are good for cooling if you are not pushing them. Than there are huge ohmic losses and the heat flow is getting worse than the cooling effect. For reasonable power, why not.

[keeney123]

Re: Thermo electric pcb power supply

They are way less efficient than the compressor refrigerator perhaps as much as 50% less efficient. Many of the Peltier device that are sold will not even perform to specification. The real use of Peltier devices are as electrical generators in places where their is no electricity. How they are used is when a fired is placed on one side of the device then the other side evaporates the heat to the outside world. In this way a electrical voltage we be supplied to the terminals in which it can be used for a small electrical power. Kind of the reverse of how they are used in a refrigerator. I worked with Peltier devices in the mid 1980,s. They were used to control the temperature of a cuvette that was filled with triple distilled water that had a certain size sample in the water. The machine that did this was called a sub-micron particle analyzer. It work well to control such a small amount of mass. The water in the cuvettee actually went from 0 C to 90 C using a +/-5 volt supply. The temperature was controlled to with-in 0.1 C by this machine.

[Behemot]

Re: Thermo electric pcb power supply

That was only a wine fridge so starting at 10 °C. I think it can be good for that. OFC usuall fefridgeratr never, that shoudl have 4 °C in it.

But I've seen than in western Europe and USA you can get absorption fridges and quite big ones? With modern insulation materials and enhanced cycle I think it is quite capable with reasonable consumption. And it is almost immortal while 100% silent.

[kaboom]

Re: Thermo electric pcb power supply

They're garbage. Someone's capitalizing on what Bell Labs released into public domain many years ago.

Yes, TECs have a place, like cooling laser modules for F/O transcievers.

Not a fridge or cooler, where even the smallest recip or "tiny rotary" compressor will have superior EER, and reliability.

And once that polyiso insulation absorbs any moisture, the TEC can no longer keep up.

[Behemot]

Re: Thermo electric pcb power supply

Actually, I seem to got 2compartment fridge with similar boards which did not cool properly too. Maybe not exactly same but very similar. Are they making them for the whole planet? This was german fridge

Anyway almost all the small caps were bad and some of the input 200V too while all were disbalanced, so just replaced them all and it works fine since. There will be an article about that later on HWI.

[uininer]

Re: Thermo electric pcb power supply

Hi all, I just joined the site and am amazed at the level of detail you have provided for these situations. I too have a Haier cooler with two compartments (4 sets of fans with peltier devices) that quit working. This happened a couple years ago on one side so I bought a new board and replaced it...fixed the issue. This time both sides quit working so I disconnected my peltier devices after reading this post and had the same result. The fans fired up and, while still running, I plugged in the peltier devices and it all works again. I'll be replacing all of the solder on my board and heading to the store to find replacement caps over the next few days. I may have questions but I really appreciate the multiple detailed posts on this site. It allowed me to troubleshoot my system instead of buying more crappy boards.

[keeney123]

Re: Thermo electric pcb power supply

I want to make one correction to previous statement. The manufacturer still has this listed at 6.5 amps at this site. https://cdn.badcaps-static.com/pdfs/...3dec3845f9.pdf

Also, to BudM the seller whats me to hook up a battery to the device, but we have been down that road already. I have confirmed that at 2.6 amps this device will burn up.

[keeney123]

Re: Thermo electric pcb power supply

I have not heard back from the manufacturer but I do not need to. I did a little experiment on an extra Peltier I had ordered. I connected an amp-meter to it and applied the 12 volts. I did not put any heat sinks to it. The max current was 2.6 amps and would had burned up at that amperage if I had left the supply on, so the manufacturer has lied about it's current capability of 6.5 Amp Max with a Max voltage of 15.4 volts. I just returned to the web site of the manufacturer and they have taken down those specs. and have but something up quite different with much lower specs. The place I bought them from still has the same specs that are not correct. This TEC1-12706 was originally in this Black & Decker BNA17 cooler when I received it not working. If this was manufactured this way I can not see how any of them worked?

[keeney123]

Re: Thermo electric pcb power supply

I have gone through 6 Peltier devices. The first one I burn up by accident by unplugging the wrong device after I check for the hot side. The next three I had put in in the same direction as the original ones only to find out that this allow pressure down on the wires which created a stress that cracked the Peltiers. Some with a crack so small it took a 5X mag. glass to see. This is when I started measuring the current because I had the twelve volts but know current.The twelve volts fans were turning. After I arranged them 90 degree the surface was completely flat even under the wires. There is a little recess in the fixture for the wires, but in the original direction even thou the wires are in the recess the mating plate dipped down and this allowed pressure down on the wires enough to cause a sheer that crack the Peltier. I did have to find out about the power supply because I have found that sometimes what appears to be a simple understanding has many more complexities than one can readily see. I will send a note off to the manufacturer and will let you know the reply. Thank you for your help.

[budm]

Re: Thermo electric pcb power supply

Wells that verify what we know all along that the current draw from the power supply depends on the load.
So how many Peltier did you buy? Did you contact the seller and ask the questions?

[keeney123]

Re: Thermo electric pcb power supply

I hook-up a 450 watt ATX 12 volt supply and the current through one Peltier ran down to 2.0 amps at 11.89 volts. The current started out at about 2.3 amp but drifted down to 2.0 amps. This confuses me. It would seem the original Power Supply is OK. Then that questions me about the Peltier devices I bought? I still will run these Peltiers along with the inside and outside fans with the ATX supply for 24 hours and see if anything is different, but I do not see how it could be. Any suggestions?

[keeney123]

Re: Thermo electric pcb power supply

Will be getting old ATX supply in next couple of days,will let you know how it works. In the mid 80's I worked with analog engineer that designed a control circuit for a Peltier device. He only used about three op amps to control a cuvette chamber from 0 degrees C to 90 degrees C with an accuracy of 0.1 of a degree. I believe we were using a linear 5 volt supply to do this with. Thanks BudM for the information. I would think that the more resistance the Peltier offers the harder the device would want to be driven which is contrary to the way the power supply. I understand the ohms law of decreasing resistance increases current, but this Peltier seems to operate different. The less resistance the less power used and the farther apart the hot and cold temp. When the resistance increase the wattage increase and the hot and cold side are close together in temperature.

[budm]

Re: Thermo electric pcb power supply

Or to 12V batteries as I suggested earlier, or put on 50W load on the power supply.
'Something is going on where the supply senses a increase in resistance and adjust the current down" The current will drop as the resistance increase, that is normal.
This is an interesting problem to solve.
Some good info here:
https://tetech.com/faqs/#11
https://tetech.com/product-category/power-supplies/

[Behemot]

Re: Thermo electric pcb power supply

Just hook it up to ATX power supply and see…

[keeney123]

Re: Thermo electric pcb power supply

There is one more thing I remember and may be key in solving this problem. When I disconnected one Peltier device from the power supply the current in the other device drop to 1.7 amps from 2.5 amp. So, this would not support my theory that the power supply is topping out at 2.5 amps. Something is going on where the supply senses a increase in resistance and adjust the current down. As if it is looking for total resistance. I still believe this to be a design problem.

[Behemot]

Re: Thermo electric pcb power supply

Yes I applied heat sink compound to both sides hot and cold. The hot side gets a little warm but can be touch with hand no problem and the cool side is blowing cold air on both Peltiers.

I don't understand. I though the peltiers are there to cool something so if anything, cool air should be blown away from them.

Also there is this fact most peltiers are for voltage higher than 12 V so at 12 V they don't give full power. Is there any negative voltage on the PSU and if so, how powerfull?

[budm]

Re: Thermo electric pcb power supply

"The thermo-couple I believe may be a K type from the coloring? I disconnect it from the power supply measure the resistance across the wires and the resistance measures 10K to 14K " Thermocouple generates voltage, thermistor varies its resistance (NTC or PTC), so do you know if it is thermistor or Thermocouple? What does it look like.
Constant current source will vary its output voltage to maintain the current as the load changes its resistance.