Tagan TG480-U01 480W ATX PSU

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

I think it also could be due to overvoltage.

Or that cap was just lying around too long...

I assume you mean the primary capacitor FSP psu.

As user stj has suggested, I check the mains rectifier and it seems ok.

The reason could be overvoltage, but also had four adapters plugged and at the same time and no one broke .
It's curious, when I opened the source , a small insect came out. This psu is working out of a computer case, it gets so dirty
Or just died by age, it is a bad, bad brand, OST (Taiwan Ostor Corporation) and a it's only 85ºC, not a japanese 105ºC.

I had asked for a replacement for this capacitor , but now I understand that your mindset is I must find the reason why this capacitor had bloated.

I had seen in this forum that usually secondary caps get bloated, but not the primary ones. And that's strange...

Clearly I must replace both the primary condenser, because it's bloated, and secondary ones because at some point will too.
Looking online stores, I had dropped RS and Farnell for the shipping.
The shop where I had intended to order the primary condenser, a 270uf 400v 105ºC Panasonic, have minimum order of 5, 10 or 20 units in small capacitors.

Ie the cost of the capacitors is small, but buying them is expensive for people who are not professional and just wants to fix one or two sources.
You may need 7-10 capacitors for a repair and you end up buying 50-70 caps except that choose to RS, Farnell or other stores selling single units.
In local stores, you can forget to choose Japanese brands and a range/serie specified.

I also can buy them to Behemot (he specializes in repairing caps for psus and motherboards)

I would like to buy Panasonic capacitors such as those recommended by stj in other threads.

The advantage I have is that in my case I want to fix this FSP psu (few caps) and Tagan (many caps).

As the psu works, I understand that the MOSFET (FQPF13N50C) and diodes are OK. For curiosity, I will check them.

Offtopic:this psu is capable of 400w, only works between 35w and 120w. (underclocking and no video card)
Now I have to replace capacitors, if anyone knew how to increase efficiency in this range although the power source is reduced....

[stj]

Re: Tagan TG480-U01 480W ATX PSU

rubycon make nice high voltage caps.

the thing that i noticed on your bad cap is the plastic sleeve had shrunk - that's a sign of extreme heat.

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

rubycon make nice high voltage caps.

the thing that i noticed on your bad cap is the plastic sleeve had shrunk - that's a sign of extreme heat.

Yes, I've read that Rubycon manufactures excellent high voltage caps, and that I must to avoid those sold on eBay.

I understand that the primary cap has signs of extreme heat, and other component has been the cause of that problem.

So you told me to check the bridge rectifier, and after desoldered it I think it is OK , but you suspect that if I change the capacitor the problem will appear again, no?


I can upload all the photos you and other future people will need.


Thank you for your worry about this subject.

[stj]

Re: Tagan TG480-U01 480W ATX PSU

to test a rectifier,

put one meter probe on the +,
and the other probe on the ac pins one after the other

you should get about .500 on diode test in one direction, and reversing the probes - you should get nothing.

then do the same with the - pin and the ac pins.
the rectifier is actually 4 diodes.

http://www.electronics-tutorials.ws/diode/diode_6.html

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

Well, I'm stupid
I was using in the multimeter the continuity scale(200R) but not the diode scale (2KR).

I desoldered the rectifier.

If the pins are like this - ac1 ac2 + , and the multimeter cables are Red(R) and Black(B), the measures are:
-(R) ac1(B) 0.55
-(R) ac2(B) 0.55
-(R) +(B) 1.241

+(B) ac2(R) 0.544
+(B) ac1(R) 0.544
+(R) -(B) infinite

ac1(R) ac2(B) infinite
ac1(B) ac2(R) infinite

Other combinations, also infinite.

Then, Is it ok??

I will wait your answer to solder it again into the board.

Thanks

Then, the bridge is Ok, no??

[stj]

Re: Tagan TG480-U01 480W ATX PSU

looks o.k. yes.

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

Well, then I will replace the semi-broken primary cap and replace the secondary for security.

The primary cap is 270uf 400v.

Should I replace it with same capacitance or choose a 330uf 400v like similar FSP psu that I have seen? I read there's some tolerance to capacity changes but also read that other components will suffer because more current.

The primary cap is a OST serie SPS 270uf 400VDC 85ºC 30x35mm(DxH) Ripple(120Hz 85ºC)=1.70
http://www.ost.com.tw/products_ec_list.asp


I would like to replace it with this one:
Panasonic serie TS-ED EETED2G271DA 270uf 400VDC 105ºV 30x35mm(DxH) Ripple(120Hz 105ºC)=1.56 ESR(120Hz 20ºC)=0.491
https://cdn.badcaps-static.com/pdfs/...b77415dbfa.pdf


Maybe I'm obsessed with the choice of primary capacitor when it might be ok any 105ºC Japanese cap that meets similar ripple....

[stj]

Re: Tagan TG480-U01 480W ATX PSU

if you can fit 330uf or even 470uf then it's better.
i was thinking 270 sounded too small actually!
all my psu's have 680uf in them!

can you fit a 30mm wide cap in it's place?
because panasonic UQ series has a 470uf cap in 30 x 40mm

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

Yes, I can fit a 30mm wide cap (the same size as actual OST cap).

I had read in a web, maybe here, that if you increase uf also increase mA in other components because more energy every Hz, so they could suffer. It's only a caution.

Of course I can buy a bigger cap, but my doubt if then I also must replace diodes, or MOSFET, or .....


There's a proverb in my country that literally says: "walk or not walk, big horse" (the bigger the better)

I study your suggestion: Panasonic UQ series 470uf 400v
The model could be EETUQ2G471DA.
http://es.farnell.com/panasonic-elec...nap/dp/1198681

But it´s a 85ºC cap. I thought it would be 105ºC.

The attached photo is a FSP 500w (mine is 400w). It has a 330uf cap. Maybe I choose 330uf because it's easy to find than 270uf, and if possible 105ºC Japanese.

Also, as Behemot recommend to me, I must check/replace the diodes in stand-by circuitry because manufacturers often use GP 1N4007 instead of fast diodes.

Stand-by circuitry??
Luckily I've only seen 2 or 3 diodes (near Mosfets and near output cables)

I forgot to say that when I desoldered the bridge, I saw bad solder pins primary cap, so I resolder both pins (bad welt increase temperature)

Thanks



.

[Behemot]

Re: Tagan TG480-U01 480W ATX PSU

Behemot, your eshop is HV World, no??

HW World and it's noct really an e-shop…hopefully soon

What kind of PFC is that crap? Another wonder like that 2x cap+2x diode+2x resistor Chieftec (Enhance) was using which just burned power and the caps always went bad?

In that Fortron, the two diodes are between the input cap and the +5 V SB transformer. D41 and D42 IIRC. There is a thread about extensive repairs of these units you can read there what goes often wrong.

As for the input cap, I suggest using 420V. FSP often used 420uF/420V caps in Blue Storms II (retail version of that) and I never seen any failed even though they've been Teapos or CrapXons.

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

What kind of PFC is that crap? Another wonder like that 2x cap+2x diode+2x resistor Chieftec (Enhance) was using which just burned power and the caps always went bad?

I guess so

In that Fortron, the two diodes are between the input cap and the +5 V SB transformer. D41 and D42 IIRC. There is a thread about extensive repairs of these units you can read there what goes often wrong.

Ok, when I open the case, I will search and check them (better still with the references you have given me, D41 and D42 IIRC)
Through the fan grill I can see one diode, as in this photo.

I find this thread where you ask about a 500w FSP from a guy who changed the fan for 1300RPM Thermatake.

"Fortron Blue Storm II FSP500-60GLN"
https://www.badcaps.net/forum/showth...+repair&page=2

My FSP 400w is a GLY design, not GLN.

As for the input cap, I suggest using 420V. FSP often used 420uF/420V caps in Blue Storms II (retail version of that) and I never seen any failed even though they've been Teapos or CrapXons.

More voltage (420v), ok, no problem.

More capacitance (420uf instead of 270uf)?? What would happen to nearby components to the condenser (because more current)?? Less ripple?

I was thinking of buying a pico-psu because my desktop only consumes between 45-125W (84w tpd processor underclock to 48w, no graphic card, 3 HDD....) so I don't want more power, only efficiency (but this is the GLY design and this will be forever)

Also there is size limitation, only possible max 30x35mm (WxH), not 40mm Height by 2 milimeters.


I had read about FSP designs, using thin heatsinks, and seems to depends a lot of airflow.
I thought temperature was not critical in this psu because airflow is colder than, for example, Tagan (Topower).
Sometimes, before the plastic of the capacitor is partially burn, I covered with the tablet a portion of the fan grill. But not the day which happens the burnt.
It was very dirty, so maybe that was the reason.
It is now with a grid that protects the main entrance of air.


People recommend I also replace the side capacitors. I will upload photos of them. But they will be the same as other FSP boards...


Thanks



.

[stj]

Re: Tagan TG480-U01 480W ATX PSU

dousnt matter much about a heatsink being thin so much as the total surface area.

it's the surface that gets rid of the heat.

[Behemot]

Re: Tagan TG480-U01 480W ATX PSU

With higher capacitance you gwt higher inrush current and the power factor shifts a bit, on the other hand you get longer hold-up time, maybe slightly lower ripple.

[c_hegge]

Re: Tagan TG480-U01 480W ATX PSU

doesnt matter much about a heatsink being thin so much as the total surface area.

it's the surface that gets rid of the heat.

That's only true down to a certain point. Once you get really thin (like the heatsinks on some cheap no-name PSUs which I can bend and even snap with my bare hands), then they don't conduct heat up to the fins as effectively, and the bottom half of the heatsink where the parts are screwed onto gets hotter than the fins.

[Behemot]

Re: Tagan TG480-U01 480W ATX PSU

It only makes temperature higher on the silicon, on the other hand it transfers more energy in the end. I still remember the formula for heat transfer, the energy is directly proportional to delta t.

[stj]

Re: Tagan TG480-U01 480W ATX PSU

well if we are talking thermals then why not replace the ally with copper like some AC-BEL stuff uses?

[ep9es]

Re: Tagan TG480-U01 480W ATX PSU

I think it is clear that aluminum is used for price / dissipation.
And you can maximize their effectiveness using very many fins (or sheets).
Copper is much more expensive and heavier (just look how heavy bunch of external cables)

The funny thing is how simple sinks without fins such as FSP 400w.
It is also true is that the fins take up space and leaves less for components.
Add more ventilation and there is no problem. (yes, more noise)

I do not know, but I imagine that passive psus have all/full aluminum heatsinks. A very efficient design and plenty of sinks.
Or maybe they are made of copper and just I say a stupid thing.

[Stefan Payne]

Re: Tagan TG480-U01 480W ATX PSU

well if we are talking thermals then why not replace the ally with copper like some AC-BEL stuff uses?

didn't they just use a thin copper plate between the devices and the heatsink??

[stj]

Re: Tagan TG480-U01 480W ATX PSU

not in the old ones.
i have a 200w from a G3 mac that uses all-copper sinks.

i use it for testing stuff, the build is real good with all-japan caps etc.
just had to re-wire the connector to ATX it.

[Behemot]

Re: Tagan TG480-U01 480W ATX PSU

The funny thing is how simple sinks without fins such as FSP 400w.
It is also true is that the fins take up space and leaves less for components.
Add more ventilation and there is no problem. (yes, more noise)

I do not know, but I imagine that passive psus have all/full aluminum heatsinks. A very efficient design and plenty of sinks.
Or maybe they are made of copper and just I say a stupid thing.

You may note that FSPs design usually transfers heat to casing, the secondary heatsink is screwed to it.

With fanless, it is more complicated. The old Fortron Zen for exaple had huge aluminium sinks and usually died few years later with bad capacitors and completelly darkened due to overheating (even though it had only 300-400 W).

Nowadays modern power supplies with active rectification often use small copper heatsinks, you can check few of my reviews. Passive units often try to transfer as much heat to the case as possible - see the Enermax Digifanless or the Seasonic Platinum Fanless (that is in czech on Diit only). They are also intended only for cases with PSU mounted on the bottom, so the hot air can ascend freely.