Thermaltake TR2-500 Primary cap bad?

[Pentium4]

Re: Thermaltake TR2-500 Primary cap bad?

This psu doesn't have active pfc. As far as I know passive PFC doesn't boost voltage neither stresses the cap.

Ohhh yeah, you're right. Good catch.

[tom66]

Re: Thermaltake TR2-500 Primary cap bad?

The cap will still be under stress from the AC input ripple. If anything, active PFC is *kinder* as the current is typically sinusoidal, and the actual RMS current is lower than a non-PFC supply. Though in active-PFC designs, the cap is usually smaller, so its ripple current rating is lesser...

[ben7]

Re: Thermaltake TR2-500 Primary cap bad?

Power up the PSU with good caps and see if the voltage across them is [roughly] the same. If they are not, then that is what caused the cap to bulge.

You live in greece, so I beileve your power is 220v. That gets rectified and put across the two capacitors in series. If the voltage across them is uneven, one could possibly blow up (worst case would be about 300v across a 200v cap)

[dmill89]

Re: Thermaltake TR2-500 Primary cap bad?

This psu doesn't have active pfc. As far as I know passive PFC doesn't boost voltage neither stresses the cap.

It doesn't boost voltage but the PFC coil can get rather hot and since this PSU (like most CWTs) has the coil right next to the primary caps this adds additional stress. Of course the primary heatsink on the other side of the caps gets hot when the PSU is loaded and the airflow is somewhat restricted in this area which is somewhat narrow with relatively large objects on both sides that block airflow, the PFC coil being a solid object isn't exactly good for airflow.

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

I took out the Capxon primary caps and took more photos of that power supply trying to figure out why the primary capacitor in that specific spot keeps failing.

There is some burnt glue on that bleeding resistor if I am correct. I don't remember those resistors to get hot.

I also consider the possibility of the MOV getting hot, but is that possible?

My last suspect is PPFC placed near the primaries. Do they get hot usually? Is it normal or a sign of failure?

Meanwhile I had a look at the other parts of that power supply.

The bridge rectifier is a 10A KBU10J.

The PWM IC is UC3845B.

The power supply uses C5353 NPN transistor on 5vsb and 2x W14NK50Z power mosfets rated for 14A continuous at 25°C and 7.6A continuous at 100°C.

Now let's move to the secondary section:

-5vsb: 1x 1N582 3A, 2x 1000uF 10V Teapo SC 8x20mm, ripple 1080, esr 0.065

-3.3V: 1x STPS3045 30A,
1x 3300uF 10V Teapo SEK general purpose 10x25mm, ripple 950 and 1x2200uF 6.3V Teapo SC 10x20mm, ripple 1220, esr 0.051

-5V: 2x STPS3045 30A(60A total),
1x 3300uF 10V Teapo SEK general purpose 10x25mm, ripple 950 and 1x2200uF 6.3V Teapo SC 10x20mm, ripple 1220, esr 0.051

-12V: 1x STPS3045 30A, 2x 2200uF 16V Teapo SC 10x30mm, ripple 1780, esr 0.032

- -12V: 1x 470uF 16V Teapo SEK general purpose 8mm

Teapo SEK datasheet
Teapo SC datasheet

The fact that they used GP caps explains the relatively high ripple those Thermaltake psus show when high loaded.

[ben7]

Re: Thermaltake TR2-500 Primary cap bad?

I took out the Capxon primary caps and took more photos of that power supply trying to figure out why the primary capacitor in that specific spot keeps failing.

There is some burnt glue on that bleeding resistor if I am correct. I don't remember those resistors to get hot.

I also consider the possibility of the MOV getting hot, but is that possible?

My last suspect is PPFC placed near the primaries. Do they get hot usually? Is it normal or a sign of failure?

Meanwhile I had a look at the other parts of that power supply.

The bridge rectifier is a 10A KBU10J.

The PWM IC is UC3845B.

The power supply uses C5353 NPN transistor on 5vsb and 2x W14NK50Z power mosfets rated for 14A continuous at 25°C and 7.6A continuous at 100°C.

Now let's move to the secondary section:

-5vsb: 1x 1N582 3A, 2x 1000uF 10V Teapo SC 8x20mm, ripple 1080, esr 0.065

-3.3V: 1x STPS3045 30A,
1x 3300uF 10V Teapo SEK general purpose 10x25mm, ripple 950 and 1x2200uF 6.3V Teapo SC 10x20mm, ripple 1220, esr 0.051

-5V: 2x STPS3045 30A(60A total),
1x 3300uF 10V Teapo SEK general purpose 10x25mm, ripple 950 and 1x2200uF 6.3V Teapo SC 10x20mm, ripple 1220, esr 0.051

-12V: 1x STPS3045 30A, 2x 2200uF 16V Teapo SC 10x30mm, ripple 1780, esr 0.032

- -12V: 1x 470uF 16V Teapo SEK general purpose 8mm

Teapo SEK datasheet
Teapo SC datasheet

The fact that they used GP caps explains the relatively high ripple those Thermaltake psus show when high loaded.

Put two good caps in there for a moment, and power it on. Take note of the voltage across each primary cap. Occasionally, if there is a problem with the PSU, one cap will have a higher voltage across it that the other, sometimes enough (greater than 200v) to cause it to fail.

[kc8adu]

Re: Thermaltake TR2-500 Primary cap bad?

open bleeder resistor.replace both.
i see this often in the plate supply in large linears.open bleeder means blown cap.BANG!!!

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

By the way, I don't know if anyone else noticed the 5vsb critical cap is a Teapo SK 47uF 35V 85°C. Yes, it's true. They chose to put an 85°C capacitor on 2 transistor 5vsb circuit.

One would expect that CWT, responsible for the Antec psus killing motherboards back in the 00s, would take a lesson and implement an IC (for example DM311) based 5vsb circuit or at least choose a cap less prone to failure.

We are talking about the same psu maker that uses Capxon caps on 5vsb output filter on the Corsair CX power supplies.

Shame on you CWT!!!

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

Put two good caps in there for a moment, and power it on. Take note of the voltage across each primary cap. Occasionally, if there is a problem with the PSU, one cap will have a higher voltage across it that the other, sometimes enough (greater than 200v) to cause it to fail.

open bleeder resistor.replace both.
i see this often in the plate supply in large linears.open bleeder means blown cap.BANG!!!

Bingo!

I followed your advice and checked the bleeding resistors.

They should be 100kohm according to the color code and that's what I got from one of them. The other resistor, the one connected in parallel to the capacitor that failed, is open.

I put 220uF 200V primary caps and plugged it on AC. The cap with the open resistor gets 235V across its legs and the other only 85v. The total rectified voltage is 320V.

Thank you very much for the insight! I am going to replace both of them. How many kohm I should put? Should I stick to the original value?

[c_hegge]

Re: Thermaltake TR2-500 Primary cap bad?

^
https://www.badcaps.net/forum/showth...eeder+resistor

See post #6. Anywhere between 100 and 200K should be fine, but both should have the same value.

[ben7]

Re: Thermaltake TR2-500 Primary cap bad?

I would just use the same value (100K). BUT, use a higher wattage one too - hopefully it won't burnout anymore!

-Ben

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

^
https://www.badcaps.net/forum/showth...eeder+resistor

See post #6. Anywhere between 100 and 200K should be fine, but both should have the same value.

Thank you!

I would just use the same value (100K). BUT, use a higher wattage one too - hopefully it won't burnout anymore!

-Ben

Higher wattage wont fit there.

Btw I removed the resistors and plugged it in again. Every cap now has 161-165V across its legs.

[tom66]

Re: Thermaltake TR2-500 Primary cap bad?

(160^2)/100000 = 0.256W

That is 1/4W. Using a 1/4W resistor should be fine up to 70°C or so. I'd prefer a 1/2W or 1W for long term reliability though.

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

If I use 200kohm resistors wouldn't that improve efficiency and help the power supply generate less heat?

[ben7]

Re: Thermaltake TR2-500 Primary cap bad?

If I use 200kohm resistors wouldn't that improve efficiency and help the power supply generate less heat?

Yes it would. But I think the resistors even out the voltage better if they are at a lower value. (more current can flow - but also more heat is dissipated)

Is it possible to put larger 100K resistors on the underside of the board?

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

I have seen 65kohm, 100kohm, 150kohm, 200kohm and 250kohm resistors on psus. I hope they will be ok. I try not to put parts on the underside for safety reasons.

[ben7]

Re: Thermaltake TR2-500 Primary cap bad?

I have seen 65kohm, 100kohm, 150kohm, 200kohm and 250kohm resistors on psus. I hope they will be ok. I try not to put parts on the underside for safety reasons.

Safety reasons? There usually is plastic under the circuit board to isolate the bottom connections. If it does short to ground, then the fuse should blow. If there is no ground in the wall socket, then it's not your fault.

[goodpsusearch]

Re: Thermaltake TR2-500 Primary cap bad?

I placed 180kohm resistors rated for 1/4watt.

The voltage of each cap is 158V. That gives 0.139 watt, so the new resistors hopefully wont burn up.

[momaka]

Re: Thermaltake TR2-500 Primary cap bad?

I *think* I found your problem...
See the resistors and diodes next to that PFC connector (one of the resistors is labeled R88) in this picture?
...
Well, does that circuit look familiar?
Hint: *Jou-Jye primary boosting circuit*

I think you have the same problem here. I don't see the boosting caps, though. So maybe it isn't the same circuit. Any chance you can draw it out?

[c_hegge]

Re: Thermaltake TR2-500 Primary cap bad?

^
I think he removed the boost caps, which disables the boost circuit. I often do this to CWT PSUs. It's not needed and gets too hot. I doubt if the boost circuit would have caused the bleed resistors to fail, though.