Hi,
Recently the wine cooler is having very hard time cooling, and it fails to get to the desired temperature.
I replaced the TEC with a new one, but still no change.
Can you please recommend a common failing point that would make the HYS80-12J have hard time cooling?

A premature guess would be something about the ability to provide enough current to the TEC, but I'd love to know if there are known issues.

Sidenote: the wine cooler wakes up and works without any issues except for inability to reach the desired temperature.

 

Check to see if the airflow is clear of lint, etc. and that the fans are spinning. A perfectly functioning unit can fail due to fouling.

That said, the most common mode of failure of similar units (I don't know about your exact appliance) is capacitor C8 on the power supply board. If you can test the voltage on the TEC under load, anything less than the nominal voltage (probably 12v?) indicates imminent failure of the power supply.

There are several threads in these forums with lots of details: schematics, pictures, etc.

 

Everything looks perfect, fans are spinning in and out, filter is of course clean.

To my knowledge, the common C8 capacitor fix comes handy when the device doesn't power up, some form of an "analog" boot loop I did replace some myself over the years (I have two coolers)

Have you seen cases where a C8 failure leads to a "not enough cooling power" condition?

Schematics and etc. : I'm trying the lazy approach first, picking people brain on historical incidents that led to similar conditions.

 

Have you seen cases where a C8 failure leads to a "not enough cooling power" condition?

I have not. Like many of us, I picked mine up free because it was being tossed, presumably because it wasn't working.

But there is a period of time where the DC voltage is falling but not yet so low that the other components stop working. During that time, the TEC is getting maybe 5-6V. I'm pretty sure half voltage would mean 1/4 power and therefore nowhere near enough cooling capacity to do the job.

Again, see if you can read the TEC voltage while it's running. That should reveal a lot.

 

You can test the TEC (thermo-electric cooler) independently if you have a bench power supply or a 12 V battery charger.
Connect 12 V to the TEC for no longer that 5 seconds. You should be able to feel one side getting cold while the other side gets hot.

Also with the TEC in the wine cooler, measure the voltage going to the TEC. You should expect to see about 12 VDC.

 

Thank you. I checked my TEC voltage, and it's about 12v. I also checked it's current, and here I found 2.7A instead of the expected 6A.

Since it's a new TEC and the behavior is the same as the previous, I don't suspect the TEC itself but rather the circuit that feeds it.

Anyone got a hint where to go from here?

 

Update: I disconnected the main fans, and received another 100mA to the TEC. If it was the TEC, then the current now would have stayed the same.

It looks like an inability to provide the required current. Perhaps something about a faulty temperature control that limits it?

 

Update: I disconnected the main fans, and received another 100mA to the TEC. If it was the TEC, then the current now would have stayed the same.

It looks like an inability to provide the required current. Perhaps something about a faulty temperature control that limits it?

Not the temperature control.
The power supply is not working properly. Start by replacing C8.

 

I tested C8 and it looks OK;
47uF / ESR=0.86ohm / Vloss=1.8v.

If the power supply doesn't work properly, would you expect a voltage drop?

 

The capacitor costs like 10 cents. Replace it anyway.

 

I checked my TEC voltage, and it's about 12v. I also checked its current, and here I found 2.7A instead of the expected 6A.

Well there you go. That tells me the TEC has failed. It needs to do its job at the rated voltage, 12V, which it is apparently getting. If it's not drawing the current and power it needs, there's nothing you can do about that. It's like a 100W bulb drawing 40W.

I'm not familiar with how these things fail and wasn't aware that they can partially fail. I would have though all or nothing, but maybe that's not right.

That said, I agree with the suggestion to just replace C8 again and see what happens.

 

TECs usually fail from excessive power dissipation. The TEC consists of hundreds of tiny individual p-type and n-type elements in parallel. When the hot side gets too hot, the electrical bond can melt causing open circuit at that element. Hence the typical failure of the whole TEC is partial failure and reduced power.

 

If the controller is PWM then I would suspect one or both of those readings could be incorrect.

Thermal fatigue is also a common failure mode in TECs.

 

Hence the typical failure of the whole TEC is partial failure and reduced power.

That virtually clinches it as far as I'm concerned.

 

I believe that I have two reasons to believe that this is not a faulty TEC:
1. The same phenomena happens with the old TEC and with a new TEC. I didn't check the current consumption of the old, but the inability to cool properly is the same.

2. The thing with disconnecting the fans, which allowed the TEC to consume 100mA more. This is a possible indication that it's not the TEC that cannot consume more than a certain current, but about an inability to provide that current - either for a power supply failure, or due to something bad in the thermal loop that limits it to total of 2.8A instead of 6A.

I know that power supply issues would be on the main board itself. Would a thermal loop failure be also rooted in the main board, or are there external components that can fail as well within the cabin? (the display and buttons work perfectly)

 

This is a possible indication that it's not the TEC that cannot consume more than a certain current, but about an inability to provide that current ...

That's testable. Substitute a car or tractor battery for the 12V supply to the TEC. I know it's not the most convenient thing to lug a battery around but it removes all doubt. Of course a power supply able to supply 12VDC and at least 6A would also work. Not as many people have one of those handy, though. A beefy computer PSU, pulled from an old computer on its way to e-waste, can often do the job.

 

Calculate the power consumed by the TEC.

12 V x 2.8 A = 33.6 W
12 V x 6 A = 72 W

A TEC that ought to draw 72 W but is drawing 2.8 A @ 12 V is bad.

 

That's testable. Substitute a car or tractor battery for the 12V supply to the TEC. I know it's not the most convenient thing to lug a battery around but it removes all doubt. Of course a power supply able to supply 12VDC and at least 6A would also work. Not as many people have one of those handy, though. A beefy computer PSU, pulled from an old computer on its way to e-waste, can often do the job.

Unfortunately I don't have access for such a power supply. I'll give it a try and order a new board hoping it would fix the issue. I have high belief the issue is in the power supply part of the board.

Calculate the power consumed by the TEC.

12 V x 2.8 A = 33.6 W
12 V x 6 A = 72 W

A TEC that ought to draw 72 W but is drawing 2.8 A @ 12 V is bad.

At an ideal world (power supply that would provide infinite current), that's right. However, I believe you're ignoring other facts I provided that would point to other possible directions.

 

You have either a bad PSU or a bad TEC. Both are easy to test.
What is the voltage across the TEC while the two are connected?
If the voltage falls below 12 V the PSU is bad.
If the voltage is steady at 12 V the TEC is bad.
(This is assuming that it is a 12 V TEC.)

The board cannot limit the current. It can only limit the voltage.
It controls the TEC temperature (power) by either reducing the voltage or turning it off (PWM).
It is the TEC itself that determines the current (Ohm’s Law).