Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

How to repair damaged SMPS 450 W computer power supply?

Status
Not open for further replies.

tanvir390

Newbie level 4
Newbie level 4
Joined
Dec 25, 2010
Messages
5
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,359
Hi . I am Tanvir. I have one damaged 450W computer power supply. I have checked all the components like MOSFETs, Transformers ,inductors,transistor, IC (ATE2005) And other components. I have found that one power BJT( MJE 13007) has damaged, and I have replaced it. After the repair I have checked the voltage output , the output was normal as before the damage. I have used 12 V cooling fan to be drive by it and it runs smoothly. But when I used it to my pc, It damaged again. I think that that It unable to supply the required power to the pc and damaged. But when I checked again, I have found that same transistor has damaged and other all the components are good except the filter capacitor for filtering the ac supply I have not checked. I do not how to check them accurately. I have checked them with the multimeter and I found as them seems to be OK. So what is the problem ? why the power supply did not able to supply the PC even if I have repaired the damaged component? Is there any problem with the large electrolytic capacitors? or there is any other problem?

Please help me in this respect with step by step instructions. Thanks
 

Need to check/change also all components related to the power transistors. Maybe better to change both power transistors.
 

Attachments

  • ATX power supply.GIF
    ATX power supply.GIF
    58.1 KB · Views: 2,355

MJE13007 is a standard switching transistor used in SMPS
Seems there may be shorted turns in the ferrite core transformer
You should also replace the large filter capacitors which is a common source of problem [capacitor top bulged upwards]
 

Is there any problem with the large electrolytic capacitors? or there is any other problem?
Maybe better to change both power transistors.
Seems there may be shorted turns in the ferrite core transformer
It's like poking about in a fog. There may be hidden problems, that can't be detected with multimeter and without a profound understanding of the circuit's operation. Although it's possible to repair PC power supplies, it's rarely done.
 

Yes you are right, but PC power supply repair isn't exactly rocket science, either we need to have some minimum knowledge about SMPS in general. :-D

The most common failure is often caused by bad capacitors. Any capacitor that appears to be bad (swollen or leaking) should be replaced. Output filter capacitors should be replaced with special "low ESR" (Equivalent Series Resistance) capacitors. On the high-voltage power supply always replace both 200V capacitors, not just one, if the mismatch is too great, one voltage could exceed the limits of the switching transistor, which can lead to transistors short-circuit.
After we found a transistors short-circuit, so need to replace both the transistor and the damper diode (fast recovery diodes that’s across its emitter-collector). In my opinion better change both transistors, as it’s not so sure about testing in circuit. Also, try to check out the pulse shaper network (resistor-diode-capacitor combination) associated with the base circuit, and to replace the resistor that’s in series with the transistor’s base ( low-value resistor), replace small capacitors (1uF-2.2uf 50/63V).
 
I agree, that your systematic approach has a good chance to catch popular faults caused by the limited lifetime of stressed components.
 

Hi FVm thanks for the replay. I want to know what kind of hidden problems maybe happen. I have checked all ferrite core transformer shortage and other components related to those two transistors. They seems to be fine. But I thinks maybe you can help me in this respect as you talking about some things that are new to me. I know the circuit theory of SMPS. But the Computer power supply seems to be more complex to me. I can not figure out what is happening? when I used it to my PC, It run but suddenly became damaged, although I have used it drive the cooling and also a CDROM too before using it to my PC. Can you please tell me that what component should I give more importance. And why the PC power supply units are damaged after using it for one or two years.What are the main causes?I have seen that most of the supply units have damaged in the same way such as suddenly burnt out of fuse and After checking the one of the transistor has damaged. when I repaired, it began to run outside the PC well and I knew that by the indication of fan running and the output voltage rating. Thanks


I waiting for your Precious replay. (please ignore any writing mistakes in my replay)

---------- Post added at 06:19 ---------- Previous post was at 06:02 ----------

Thank you Mister_rf. I have found that your replay is most reliable at this moment. Can tell how can check those Large capacitors if they are damaged with multimeter.The capacitor are leaking and they slightly out of shaped on the head. But I have opened a used and good Supply unit and I have found their the capacitors are look like as my damaged one. Please help me in this respect. I am unable to say the controller IC is well accurately but as after replacing the damaged transistor the fan started to ruining so I thought that the IC is OK and Optocupler also. Should I check power ok by using A led that point power ok in high or not. But I have checked the output voltages, they were as the rating. I am going to change the both power transistor next day , Let see what's happen. But please do not go away from me. I need your help to find the right problem and the remedy of it. Thanks
 

mister_rf's post has in fact covered many possible hidden faults. I'm generally talking of faults, that don't stop the circuit operation, but e.g. cause excessive losses of an output transistor, which get dangerous, if the power supply is fully loaded. To indentify problems like this, you usually need:
- a variable dummy load, e.g. power resistors
- an oscilloscope to check circuit waveforms
- a safety transformer to allow live measurements

There's still a risk, that a circuit problem shows in a fatal way: Normal operation up to a certain load level and "sudden death" when exceeding it.
 

Changing only the power transistors doesn’t help always. First need to replace small elements: capacitors , resistors, diodes (if any…).
 

I have changed the two transistors and checked all the resistors and diodes. But the result was same. would I need to change the capacitors too?I thought they are ok because of their uniform shape and no leakage. I want to know that does the capacity of a capacitor change due to the long time using (without leaking or shape deformation)?And a easy way to measure the condition of the capacitors with multimeter.
 

Hi,
I am new in this Edaboard Forum . This is my first post . I do not know too much about Electronics . Because I am not a student of Electronics like you . But I am working with Electronics Practically for few years . This is why - I know a little bit about Electronics and trying to Learn more .....

Tanvir Which Country are you From ?? I guess you may be from BD . May be I can help in your problem a little-bit . As in my Practical knowledge - I think you have the problem with the IC . Because ATX Power Supply uses PWM (Pulse width Modulation) IC . When The LOAD Increase - this PWM (Pulse width Modulation) IC Trigger/Feedback to 13007 Transistor to Increase their Current supply . Again When The PC is in Stand-by Mode this PWM Trigger/Feedback to 13007 Transistor to decrese their current to save Energy (Energy Saving Mode) .

As your Power Supply is running without Load . When you give the Load This PWM is not giving Trigger/Feedback to 13007 Transistor - this is why, 13007 is not increasing Current and as a result you Power supply is shutting Down at loaded condition .

Change you IC . I hope your power supply will be OK again with the LOAD . And is Your IC number Correct ?? I am not finding any datasheet of ATE2005 . Most of the Power supply uses TL494 IC .
===========================================================================

Guys wait . Please don't go . I have a great problem with a Power Supply of my Customer . I got this Power supply three days ago for Repairing . This Power supply has burnt out the Two Primary High Voltage Transistor 13007 and their Base Fusing Resistor 10 ohms . I have Replaced them . Now all the Parts are OK . But The Power Supply is not Starting at all . I have checked the Diode at Primary Side - all are OK . I am getting 5VSB and 5V PS-ON also . But the Power supply is not running . I have checked all the parts in Output Low voltage section - here is all the parts are fine . I have replaced the High Voltage and Low Voltage all Caps and IC TL494 and LM339 also, But still the Power supply is not running - Why ?????? What's the Problem is there ??? What am I missing ???? Guys can You help me to figure out this problem ............. I am really in trouble with this power supply ...:-x:-x
 
Last edited:

The IC number is correct. And I have checked it and also saw datasheet in the net but I had not downloaded it then. You can search by ATB 2005 also. You have wrote that my psu had shut down during load due to low current into the transistors but transistors are usually damaged due to excessive current flow.And there is a also question in mind that the load cannot draw high current if the there is no supply of it, so at loaded in my opinion the psu will not be provide the power but it can not damaged also. I have two Ic s TL434 And ATB/ATE2005. As i know that TL434 is a voltage comparator. Are you from BD Strange problem?. And what's your name?
Thanks for the replay. Note The ATB2005 also controls the fan.I am waiting for your replay.some questions that need to be answered, please try to hlep me to find this answers.

1.why after my PSU gives result as new without full load but partial load as CDROM after repairing it.
2.I have seen that the psu suddenly damaged after starting the PC.Why This Is Happening?
3.The PSU can gives smooth power to a 12V motor also.
 

Oooops !! I said it before - I am not a student of Electronics . I don't know so much theory like you . I am just a practical Electronics worker . I have just share my Practical experience - that's all .

Why your PC running with Partial Load But Not running with Full Load/Suddenly damaged - May be I have given the answer in the previous post . What I described there is my total knowledge, I do not know more than that . If you want to know more theory about this - Then Specialists are here already - Just ask them . They will describe to you .

3.The PSU can gives smooth power to a 12V motor also.
I am afraid to say Because I have not done/tested it before .
 

Can you measure the output voltage? Maybe the voltage is higher than normal and so, it is pulling more current than normal and destroying it.
Are any surrounding parts damaged, like resistors or diodes?
Can you see inflated capacitors?
Do all the PCB tracks/traces look okay, that is no tracks/traces are burnt?
 

hi tan......
i think you must be check stand by 5 volt suply capacitos
that was lekege and make this type falt
 

A fake transistor from china can be the culprit. 13009 is similar transistor with high current capability. Use transistor from working supply.
 

I have measured all the output voltages. The out put voltages were same as rating written on level of the casing. I will check the stand by out capacitors and the out put again as the qayyoum said.Thanks to all for supporting me .....I will info you the further result....but I will try the next level of this problem by changing the ICS
But I am afraid that may be the AT 2005B is not available in local market. Can any one give an alternative to it....And mt PSU is " D_TECH"(company name) PSU.Thanks

---------- Post added at 09:13 ---------- Previous post was at 09:02 ----------

Hey Tamid Can you give a way to increase the simulation limit from 2k of MIKROC for PIC for free? I am new to pics and using MPLab but The header files are unkown to me how can get the ANSCI C coding tutorial for PICS in MAPLAB using Hitech ANSCI C coading ? are The Header FILes are same in ALL ANSCI C Compilers like MIKROC or The MAPLAB for the pics 10x,12x,16X?

I know that this forum section is only for Power electronic but I disparately needed the answers.Please answer the questions with the path where I can get the answers. thanks
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top