Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

[delturcious]

I have an older Thermaltake PSU that has a squealing noise whenever it's plugged in, even when totally disconnected from everything else. It automatically powers on the 5V and 12V rails with full voltage, but the +5VSB circuit only has 0.9V on it. I would assume this is my first place to start troubleshooting?

There is nothing obviously burnt/leaking that I can identify.

I've tried to take some decent photos of the 5VSB area. Components of notes:

• U8 at the back is marked AZ431
• C38 is 102
• C47 is 104
• C34 is a Teapo 330uF 16V (I think all the cylindrical caps are Teapo)
• Big pink resistor appears to be brown, yellow, violet, gold but I'm not totally sure.

I'd be thankful for any pointers on what to replace or if there's another area I should be looking at that could impact the 5VSB circuit.

[stj]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

the 2 output caps on the 5v standby are bad - specially the smaller one.
probably got cooked by the load resistor!

if you can put heatshrink or fiberglass sleeving over that resistor before fitting the new caps it will help!

[delturcious]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

the 2 output caps on the 5v standby are bad - specially the smaller one.
probably got cooked by the load resistor!

if you can put heatshrink or fiberglass sleeving over that resistor before fitting the new caps it will help!

Thank you for the quick response--I can definitely heatshrink that load resistor. Please forgive my ignorance: the little 102 and 104 caps are the output caps? And you think that 330uF one is OK even with as close as it is to the resistor? I have the two little ones on hand; would have to order the big one though.

[stj]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

c34 and the larger one behind it closer to the heatsink.

you need something low esr.
like panasonic FR or FK series.

[delturcious]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

c34 and the larger one behind it closer to the heatsink.

you need something low esr.
like panasonic FR or FK series.

Easy enough--thank you! I have some equivalent FR-series for both from DigiKey in my cart. Should this also take care of the auto-turn-on issue, or is there anything else I should look at before ordering?

[stj]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

no idea,
it depends on how the psu works - most run the main chips from the standby supply.
so fixing the standby comes first.

[momaka]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

Should this also take care of the auto-turn-on issue, or is there anything else I should look at before ordering?

Maybe or maybe not.

When the caps on that 5VSB circuit go bad, the 5VSB will start to oscillate with crazy voltages, and that can take components with it - both inside your PSU and in your computer.

So indeed the first step is to recap the 5VSB - that means also replace that one small 10 uF 50V capacitor on the primary side. You can see it on this picture of yours:
https://www.badcaps.net/forum/attach...7&d=1613063363

After this and replacing the output caps, make sure the 5VSB is stable at 5V +/-5% on the voltage (i.e. anywhere from 4.75 to 5.25V is acceptable for the 5VSB.) Though if I recall, on these PSUs it usually sits closer to 5.15V. With that, the PSU should no longer turn On by itself anymore. If it does, there may be more troubleshooting to do... but worry not - these units are fairly simple to work on.

On that matter, if you do encounter any further issues on the 5VSB that need troubleshooting, you can refer to the schematic here:
https://www.badcaps.net/forum/showpo...postcount=3058

That aside, I suggest that you DON'T recap just the few failed capacitors. Instead, since you have the whole PSU already out for servicing now, you might as well DO a FULL recap.

I sort of rebuilt/upgraded a HEC Orion HP585d PSU not too long ago (which is what your ThermalTake TR2-430 is based on) and posted all of the detailed component list here:
https://www.badcaps.net/forum/showpo...postcount=3051
^ At the end of that post there, you can see which rail has what caps, so you can order your caps accordingly.

Besides Panasonic FR, FK, and FM, United Chemicon KY, KZE, and KYB are also fine choices. Rubycon ZLQ and ZLH along with Nichicon PW, HE and HW will also work. Rubycon YXJ is a bit on the "not quite rated good enough in terms of specs" -side, but will work too. So that should give you plenty of choices for caps now.

Thank you for the quick response--I can definitely heatshrink that load resistor.

Heat-shrinking that resistor will *not* help it or the surrounding caps, so there's no point to doing that. The main problem is that resistor is just dissipating too much heat and will continue to cook even your new caps.

I can't tell its value from your picture - the color of the bands appear off. But the one in my PSU was 27 Ohms, which on the 5V of the 5VSB gives about 1 Watt of heat... which is why it runs so hot. Try replacing it with 33 Ohms or 47 Ohms @ 2 Watts rating, and leave it with long leads so you can position it further from the caps. Alternatively, you can just solder longer leads/wires to this resistor and place it somewhere else in the PSU where it can get better cooling and where it won't cook any nearby caps.

I didn't do that on my PSU, since I only use it for testing and I keep it disconnected from wall power most of the time. But over long-term deployment, I suggest to do something about that hot resistor. Again, heat shrinking it alone will not really do anything. Either needs to be moved or replaced with a slightly higher value.

[stj]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

if the resistor had longer wires i would mount it under the board on the track-side

[bigbeark]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

I have 2 of these PSUs and I am NOT happy with them. Both work but as I discovered this morning not that great,

I had been testing an Asus P5GDC-V with the TR2-430 that I had been testing other boards with, and could not get it to detect the hard drive. I wasted a day on this and 2 other Asus Intel boards with the same problem. None of the boards could detect a SATA drive. An add-on SATA card worked no better. IDE didn't work either.
I tried a USB Thumb drive with Linux Mint on it. No detection, and locked up the board.

Finally selected a Dynex PSU and attached it to the board. It detected the SATA drive but would not load the O/S.

However, when I attached an IDE Drive it miraculously booted the OS and everything worked.

The machine I am typing this from has a TR-430 PSU and is exhibiitng some strange behaviour:

It needs a double push on the power button to start up and when I do a shutdown the board keeps running and I have to do a hard shut down with the power button.

So both of these are going to require attention. These are 10 years old but have not been heavily used.

I am going to buy a new PSU just to be sure I'm testing motherboards and not PSUs at the same time!

[ChaosLegionnaire]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

I am going to buy a new PSU just to be sure I'm testing motherboards and not PSUs at the same time!

I'm thinking for that purpose I need a completely trustable power supply, probably a new good quality one for testing mainboards.

I had no idea how Power Supplies could become so flakey with age.

lol! u need a branded all jap cap psu. thats the only ones worth spending good money on. the superflower leadex gold series and evga g2 or g3 series are all good jap cap psus with ultra low ripple outputs so the caps on the board and hard disks will last a long time. hard disk failures are sometimes caused by noisy high ripple output on the psu 5v line.

and lastly, in the psu world, if u pay peanuts, u will get monkeys, so dont be a cheapskate when it comes to psus.

[momaka]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

if the resistor had longer wires i would mount it under the board on the track-side

That still wouldn't help it run cooler or heat the nearby circuitry less. It just has to do with the power dissipated: any 3-Watt or smaller resistor dissipating more than 1 Watts (which is the case here) will tend to run hot - at least when the fan is not running (which it isn't when the PSU is in soft-off mode and the 5VSB circuit is always running.) So in cases like that, where active cooling is not used or very limited, then it's best to keep power dissipation on resistors under 1 Watt. 0.5-0.7 Watts on a 2-3 Watt -rated resistor is just about the max you should do if you want that resistor to run at less than 50-60C.

I have 2 of these PSUs and I am NOT happy with them. Both work but as I discovered this morning not that great,

I had been testing an Asus P5GDC-V with the TR2-430 that I had been testing other boards with, and could not get it to detect the hard drive. I wasted a day on this and 2 other Asus Intel boards with the same problem. None of the boards could detect a SATA drive. An add-on SATA card worked no better. IDE didn't work either.
I tried a USB Thumb drive with Linux Mint on it. No detection, and locked up the board.

Finally selected a Dynex PSU and attached it to the board. It detected the SATA drive but would not load the O/S.

Hmmm. I don't think those test point 100% as the PSU definitely being the fault here. HDD not detecting could also be intermittent cabling issue, motherboard issue, or even HDD issue. Of course, it does depend, though. If these PSUs have bad caps, then for sure they could be making the motherboard perform flaky.

The machine I am typing this from has a TR-430 PSU and is exhibiitng some strange behaviour:

It needs a double push on the power button to start up and when I do a shutdown the board keeps running and I have to do a hard shut down with the power button.

That could be an issue with the PS-On circuit... but given the crap cap brands in this PSU, I suspect bad caps on the 5VSB circuit are likely causing this, just like the OP's issues here, where the PSU would turn on by itself.

Since the PSU is otherwise at least powering up, that means the monitoring chip (which also is responsible for the PS-ON circuit) is likely fine and probably bad caps in the 5VSB are causing it to work erratically.

I am going to buy a new PSU just to be sure I'm testing motherboards and not PSUs at the same time!

Just recap them, and then you won't have to worry about that part anymore.

lol! u need a branded all jap cap psu. thats the only ones worth spending good money on.

Good Japanese caps alone does not mean good PSU design.
I can put Japanese caps in a Powmax, but that won't make it any better.
Have an Enermax Pro 82+ EPR525AWT that has all Japanese caps, and it blew its primary caps open, thanks to both APFC and undersized primary caps. Also have a ThermalTake Smart M Series M850W that failed in this same exact way, and again the PSU had Japanese caps. Both of these PSUs are posted here on badcaps, if anyone cares to search my repair logs on them. I got lucky that neither of these blew anything else other than the primary input caps. Many APFC PSUs tend to blow their APFC MOSFETs, and sometimes a dozen of small SMDs too - not fun to repair these!

So on that note, I think design matters more than caps. Caps you can replace easily. But a marginally-designed PSU is hard to re-design. And APFC, as I found out with the Enermax Pro 82+ EPR525AWT, is not possible to remove to simplify the PSU. I haven't updated that thread, but I couldn't get the PSU to draw even half its rated power without the APFC circuit. So likely I will just have to restore the APFC again and use really high-end expensive primary caps to make sure they don't blow again.

and lastly, in the psu world, if u pay peanuts, u will get monkeys, so dont be a cheapskate when it comes to psus.

That is mostly true.
IMO, you can find great cheap PSUs that are extremely reliable - but you would need to put the work in them to recap them. These would be the name-brand non-APFC PSUs with lower power output (i.e. 400W or less) from Delta, HiPro, LiteOn, and Bestec. Even this ThermalTake TR2-430 can fall in that category.
Otherwise if you do get a new PSU with APFC (and that would be pretty much any PSU you can find today), then my suggestion is to go with Silverstone. They tend to cost a lot more per Watt rating, but that's because inside they are fully packed with high-quality over-sized parts, including the primary input cap. And a good number of them come with 10-year warranty (yes *ten* - that is not a typo!)

[bigbeark]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

As suggested by Momaka, I recapped my Thermaltake TR2-430 today.

I found 3 bulged Teapo (Green) 6.3uf 2200uhf when I removed the cover.

I removed the grilles and screws securing the 2 fans, total 8 screws.

I removed the 2 screws holding the Voltage Selector switch from the chassis so the switch hung free.

I then unplugged the fan connector from the PCB.

I removed the 4 screws that secure the PCB to the chassis,

Then I could lift the PCB board up from the chassis. The PCB board was still attached to the chassis by 2 black ground wires on the edge of the secondary side of the PCB, I had enough room to stand the PCB board upright and access the two black wires from the backside of the PCB. I unsoldered them and the board could now be removed from the chassis.

I was able to unsolder the 3 bloated caps, but I had to remove 2 other caps so I had enough wiggle room to get 2 of the bloated ones out.

I had to remove the tall 10x30mm 2200uhf 16v Teapo, and the 8x20mm 1000uf 10v Teapo from the secondary, which is very crowded.

I tested them and they were good, fortunate because I did not have replacements for those two. I reinstalled them.

I used 2 10x20mm Rubycon ZL 6.3v 2200uf and one 10x30 Sanyo FC 10v 2200uf to replace the bloated Teapos.

Space was tight and I needed the longer leads and body length of the FC to position it on the PCB. I tried another ZL first but I just could not fit it to the PCB.

I attached the recapped PSU to the Abit UL-8 Socket 939 board I thought was finished.
All the bad BIOS codes and Ram problems disappeared. It booted immediately and loaded the Sata Drive which had MX Linux on it.

Everything worked with no lagging and shutdown was immediate when shutdown was selected.

It took me a few hours but the results were worth it.

I took the lid off my other TR2-430 but that one has no bad caps. It uses what looks like UCC KYs but are not.

I put the recapped PSU in the case from which I removed that second TR2-430 but that board, an Intel DG43NB still requires a double push of the Power button to start it and a

[momaka]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

I found 3 bulged Teapo (Green) 6.3uf 2200uhf when I removed the cover.

That likely means any other Teapo caps in that PSU will probably fail soon too. IME, Teapo are very hit-n-miss: if one Teapo cap from a batch fails, the rest will probably follow suit soon too. But if none had failed, then they might do OK or just fail much later.

With that said, I no longer trust Teapo SC.

I had to remove the tall 10x30mm 2200uhf 16v Teapo, and the 8x20mm 1000uf 10v Teapo from the secondary, which is very crowded.

I tested them and they were good, fortunate because I did not have replacements for those two. I reinstalled them.

That's fine.

I did something similar when I was recapping my HEC Orion HP585d PSU (same platform as this PSU) - I didn't have any 16V, 2200 uF caps at the time, but I found out that the 12V rail has an empty 8 mm dia. spot. So I installed a United Chemicon KZE 16V 680 uF in that empty spot, which happens to be in parallel with the 16V 2200 uF Teapo SC behind it. You can see it on the right side in this picture:
https://www.badcaps.net/forum/attach...1&d=1577684399
Thus, even if my Teapo SC on the 12V rail fails now, that 16V 680 uF KZE should still technically at least be able to "hold over" the 12V rail from going haywire with ripple.

I also replaced the 16V, 330 uF Teapo SC on the 5VSB with a 6.3V, 1000 uF Chemicon KY on my unit. Since this is on the 5VSB rail, going down in voltage is OK. And the increase in capacitance is of no harm either. You can see the 6.3V 1000 uF KY cap at the top-right corner of this picture in my unit:
https://www.badcaps.net/forum/attach...1&d=1577684399

I used 2 10x20mm Rubycon ZL 6.3v 2200uf and one 10x30 Sanyo FC 10v 2200uf to replace the bloated Teapos.

The Rubycon ZL caps are great replacements.

But the "Sanyo FC"... I hope you meant Panasonic FC. Legitimate Sanyo capacitors do not have FC series, as far as I know.

I attached the recapped PSU to the Abit UL-8 Socket 939 board I thought was finished.
All the bad BIOS codes and Ram problems disappeared. It booted immediately and loaded the Sata Drive which had MX Linux on it.

Sweet!

but that board, an Intel DG43NB still requires a double push of the Power button to start it and a

That may be an issue with that motherboard.
I've actually had this issue with a MSI MS-7184. The issue turned out to be a broken VGA connector that was shorting a 5V rail pin to ground. Because that 5V rail pin was protected by a poly-fuse, the first attempt to power-on the motherboard would result in the PSU short-circuit protection kicking in. But once the fuse opened, the motherboard could start.

Since yours is doing the same thing, I would suggest you check the USB ports for shorted pins, since those have poly-fuses as well and that may be the reason this is happening. If not, the motherboard may have other issues. Probably better to post pictures in its own thread for us to see, if you'd like.

[bigbeark]

Re: Thermaltake ATX TR2-430 Turns on by itself, squeals, & has 0.9V on +5VSB

Momaka, yes I meant to say Panasonic FC! Yes, I think you are on to something with shorted USB pins on the DG34NB, as one USB port does not work at all. I will add this to my to-do list.