300 Watt ATX Power Supply Fixed
Today I finally fixed a Power Supply (For Home or Office use !) Model: SS-300FS Active PFC that I had looked at months earlier but couldn’t fix.
The Seasonic 300W desktop PC power supply had worked several years until it often wouldn’t start. Or suddenly did quit while working.
So I opened it and found that the Fuse still was intact. Next photo shows how easy it is to remove the pcb from the now empty cabinet.
Next photo shows the 230VAC input (dangerous side) with Graetz diodes input filters etc.
And in the following photo the safe Output side. (Black cable is FAN supply).
I replaced a defect primary Mosfet Motorola (ST) W8NC80Z by an IRFP 448. But still the Power supply didn’t come to life. And a view from the top with on the bottom left on the Cooling block the replaced Mosfet.
I also replaced an 2200uF 6.3V electrolytic on the secondary side (because it was bulged). The next photo’s show the replacement Mosfet IRFP448. (I took the Mosfet from another board so don’t look at the enhanced soldered pins. The original Mosfet still had much longer legs).
So because I couldn’t find why it still wasn’t working although all semiconductors were okay, I had put it away until today when I took another look inside and noticed a 47uF 25V capacitor that looked dirty on top. The bad 47uF component you see in the next photo’s.
When I measured the Electrolytic it was a total short circuit. And therefore I replaced it by a 47uF 50V component (it was originally an 25V volt type, but 25V extra max voltage is always better I guess!). In the next photo a better look on the 47uF now 50V replacement elco.
The component marking on the PCB was C24. (See next photo).
In the previous photo it’s position is in the top middle above the “9937” elco. (showing it’s 47uF 50V replacement). All other semiconductors I found to be 100% fine before, so I re-checked if the 47uF 25V replacement now had the success I wanted? And it did! So the next photos show the 12V and 5V resulting measurements. In the next photo you also see how I simulated the normally in a desktop pc available On switch by connecting 2 of the power connector wires with a wire. (connecting Black with Green of the 20 pole connector).
Also the 3.3V on the power connector was 100% too but I haven’t taken a photo from it, so take my word for it
And to conclude this repair I’ll show you the solder side of this Power supply!
Another Power Supply fixed, and this time it was a 300Watt pc desktop power supply with almost zero cost!
Hope you liked it !
Albert van Bemmelen, Weert, The Netherlands.
Please give a support by clicking on the social buttons below. Your feedback on the post is welcome. Please leave it in the comments.
P.S- If you enjoyed reading this, click here to subscribe to my blog (free subscription). That way, you’ll never miss a post. You can also forward this website link to your friends and colleagues-thanks!
Click HERE to become power supply repair expert!
(171)Dislikes
(1)
20 Comments
Leave a Reply
Robert Calk
April 24, 2015 at 4:45 pm
Good job Albert. I'm glad that you found the culprit and got it fixed!
Waleed Rishmawi
April 24, 2015 at 5:59 pm
usually these power supply are too cheap to repair in my country but sometimes, it becomes necessary to repair one it the power supply does not have too many shorted component in it. thanks for sharing. nice details with the photo taken in this article.
Anthony
April 24, 2015 at 6:42 pm
I also agree with Robert, congratulations! Sometimes we need to take a break when the problem persists and look at it again with fresh eyes. This usually does the trick when we've rested and are able to think more clearly.
Humberto
April 24, 2015 at 8:36 pm
Good repair Albert van Bemmelen,, but when you test the SMPS you should connect a load in the output as a HDD or anything else, if you do not do this the SMPS could break again.
Albert
April 24, 2015 at 11:39 pm
Hi Humberto, maybe that's true but I never had any trouble testing ATX power supplies without any load. I always remembered this to be true for those Power Supplies that didn't work without any load attached...
Nowadays Power Supplies lack any on/off switch but mine still had one too.
I also rebuild a 5V 1000 (200 Amps !) Watt Power supply for 240 VAC input (was originally set to USA 110 VAC). Even that one worked without any load.
Probably the controllers inside now work fine without any minimal current Load? Any over-current or over-voltage will control the Output. Too little current is hard to measure I guess?
Cheers,
Albert.
Paris Azis
April 25, 2015 at 4:31 am
Hello Humberto,
This was very likely to happen when testing much older designs. All the modern designs, especially those including active PFC in their input, are very smart designs and thus such a risk is practically eliminated. On the other hand all of them use the so called "minimum load resistors" at their outputs (you can "see" them by a simple Ohmic measurement of their outputs when the unit is off) which totally eliminate the risk of oscillation or any instability during start-up which is the main cause of such a failure.
There is only one exception to this (considering that the minimum load resistors are intact). One or more of the output filter electrolytic capacitors having high ESR, which is very likely to provoke this cause to appear and then destroy the switching semiconductors at the primary side.
Nevertheless, as a good habbit and as regards full safety in testing, I totally agree with your advice.
Best Regards
Humberto
April 29, 2015 at 12:25 am
Thanks Paris and albert, but I've been repairing Computer's SMPS and in this kid of SMPS the manufacturers do not permit to test them without a load.
Best regards.
Humberto
April 30, 2015 at 7:41 pm
Thanks Mr. Albert for the pdf sent by means of Mr. Jestine Yong. If you have some more valuable information or maybe websites, do not hesitate in send them.
Have a good day.
Albert
May 2, 2015 at 2:19 am
Hi Humberto,
Glad that you appreciated it! I hope you find it interesting.
And you're Welcome !!
The article came from the Electronics Magazine Elektor. (Maybe you can find more on the internet about ATX power supplies in Elektor. One other article I know dealed with modifying ATX supplies internally).
I wrote some published articles in Elektor too. (See 3 time-zone DCF clock reciever with adaptation to Picee board. Elektor December 2007).
Cheers.
Albert.
Merlin Marquardt
April 24, 2015 at 8:57 pm
Very nice.
moshe
April 24, 2015 at 9:10 pm
nice article...thanks for the tips
Andy Shiekh
April 24, 2015 at 9:20 pm
What is a Graetz diode?
Albert
April 24, 2015 at 11:24 pm
Hi Andy. it's a 4 diode bridge. So not just 1 diode ! Just Google for Graetz...and you'll understand.
Cheers,
Albert.
Robert Calk
April 25, 2015 at 7:50 pm
From what I read it should be called the "Pollack" diode. I guess they didn't want to give credit to a Polish man.
Robert Calk
April 26, 2015 at 11:42 am
Oops, I meant "Pollak".
Andy Shiekh
April 24, 2015 at 9:26 pm
Ah, got it; Graetz diodes are Bridge diodes...
Vicken Mardiros
April 25, 2015 at 2:16 pm
very nice work and thanks for sharing in details...
Anwar Shiekh
June 29, 2015 at 10:16 pm
I never put higher voltage capacitors in as the ESR (Equivalent series resistance) may be higher and I read that it may have problems maintaining the oxide layer.
mahienn
February 11, 2016 at 6:28 am
hi to all
very nice as well as informative
thank u Sirs
helo sir jestine how u doin?
i've been missing your articles plz share some of your experiences
if you find time
i'm your fan no.1
thank you god bless you...
Momin Khan
April 10, 2016 at 6:19 pm
Thank you for this post. This post is very helpful to people. Who want to solve PSU problem himself......