300W ATX PSU CHIRP...CHIRP... CHIRP... SOUND ON STARTUP

[ranronqx]

I may rate myself as an intermediate in the field of electronic troubleshooting. Hopefully somebody could help me out figure this one.This is my problem

I tried to repair a 300W ATX PSU. I found it having problem since it no longer power up a PC with only minimal load in it. (i.e. Athlon XP-2500, 80GB HDD, Radeon 9600, 512 of RAM[2x 256MB] and 2 chassis fan connected to PSU). I changed the PSU and the machine works just fine.

Yep that solves my problem with my PC though but I'd like to ask some questions regarding my busted PSU. during my troubleshooting of my pc ,when I turned on my PC, the chassis fan and the psu fans just spins for a second or two then it dies. When I turn it on again, I heard a chirping sound in the psu as if it tries to start psu circuitry again. I pulled the main cord and long pressed the power button for 10 to 15 seconds but still the psu still won't start up. I am aware that this startup problem is cmmonly caused by an overload in my PC or perhaps a short circuit in the load section> But considering my PC was working fine after replacing the busted psu, I ruled out those speculations of those common causes of that psu behavior. I also checked the secondary rectifiers (schottky diodes) and the primary regulators and also it's surrounding components hoping I could find the culprit but they all read OK in my analog and digital meter. I tried googilng some info about start up problem with ATX switching PSU and found some info bout PSU power startup sequence. but didn't answer any of my questions. My question is

1. what part of the psu that caused the chirping sound of the psu and how does it contributes to the problem?
2. Considering my psu has the overload and SC protection, how will i recognized it in the circuitry? Is there some parts or IC which may guide me through that section of the psu?
3. why sometimes when I do the long pressed of the power button while the main cord is detached, I get to power up my PC and keep it running for a period of time without problem?.Is it normal or is it just another symptoms of a dying PSU.

help me please...I got at least 20 PSU with same brand with same problem at workbench, I still can't figure out the problem.

[SZASZ]

Does your ATX contain an Ic like KA7500 AZ7500...etc???

[vtech]

Depending on the brands, majority of the ATX computer supplies are prone to have capacitor failure and are not very well designed. I have repaired many in the past, but most are simply not-worth-the-time. Unfortunately, not meant to be a serviceable item with today's economy.. (depending on what part of the world you are.. ofcourse)

Chirping sound could be anything from overload, driedup electrolytic capacitor(s) , bad connection or false triggering of protection ckts. In my experience, most problems start when the built-in fan(s) begin to fail & the corresponding heat will start the ensuing damage that may be simple or major..

Some used the "all in one" type regulator/management IC which is very difficult to find in small quantities. There used to be some schamatics floationg around on the net. Other than that , it is taking the time to t-shoot & hope not to come across an oddball IC.

[ranronqx]

(SZASZ) yeah it uses the KA7500B PWM ic and the LM339 comparator, does it contributes to the problem?

(vtech)Actually I'm from the philippines and this brand of psu is common here in this country for those low end and average type of PC's, so I'm pretty much sure all the cutting cost in the design of this psu is being implemented.I don't know if this brand is known to you guys but what it says in the cover is Pentium IV ATX Switching Power Supply Model AZ300W ATX. Yeah you're right, the first time i opened the PSU, i found two bulging caps in the 5V and 12V rails respectively, and the exhaust fan barely spins, so I've changed all the capacitors in the ouput rails to make sure and replaced the dying fan with new one, but that doesn't help. Further testing indicates that the 5Vsb is present in teh purple and (PS-On)green wire at PSU in off state(i.e. power cord plug-in but not turned on). Using dummy load(3 HDD connected) will turn on the PSU and gives the right voltages in all the output rails.

False triggering of protection ckts. This is what I'm concerned of. Does it mean the psu is actually good and that it only detects false irregularities in the circuitry?And unfortunately, draws all the blame in the design of the PSU?

[vtech]

Without a schematic it is hard to say. Depending on the circuit design for protection, it may be false trigering thereby causing it to shutt down.

Another possibility of bulged caps may be due to the fact that "rails" might have momentarily lost the feedback control line, causing them to momentarily shoot up & busting the caps & possibly zap any other semiconductor in line. So any "refernce type" opamp in the line is probably suspect.

May help for you to find the IC diagrams & try to draw up a diagram. Better yet, do a search on top of the page (red letter box). Put ATX and see if any of the corresponding files may be similiar.

[Guest]

I'm already 40% done in tracing the schematic of this psu, ill post this out as soon as I've done it. I'll check those files as well if there's any similarities i could find. thanks man.

[ranronqx]

I really appreciate your responses to this post vtech. This might help others too in this community.

Ok, so this is what I found, this file PC_Power_Supply_ATX-200W.gif (http://www.eserviceinfo.com/downloadsm/12764/all%20most_ATX.html) looks very similar to my psu. I started to wonder why it's rated 200W, if that's the case, this kind of psu has a minimal allowance in power with regards to my load.

Further research in the internal diagram of the KA7500B shows that the feedback line from the 12V and 5V rail goes into pin1 of the PWM ic which in turn is the positive reference of an OpAmp (if im not mistaken) that feeds directly into the input of the comparator.

So, that momentary lost of the feedback control line, does it caused by the PWM and comparator IC?

Anyway, I'll be replacing those IC one at a time tonight. I'm taking a picture of the culprit but further explanation to this PWM and Comparator section if possible would greatly enhance my understanding to SMPS troubleshooting.

thanks a lot vtech, hope to hear again from you.

[torbjorn]

Usually, the same basic design of a PC power supply is manufactured in several versions with different rated power (i.e. 200 W, 230 W, 250 W, 300 W).
The difference between the versions is that the more powerful ones have a bigger transformer and possibly somewhat more powerful semiconductors at critical positions (switching transistors, rectifiers on the secondary side etc).

If one finds bulged or leaking electrolytic capacitors in a such power supply, those capacitors have according to my experience mostly failed due to bad quality. It is good practice to replace all electrolytics of the same brand as the failed ones, otherwise the rest of them will fail within a few months.
Check out the exact type of the failed electrolytics, try to find the manufacturer's datasheet for them and check the ESR value for each part.
When buying replacement capacitors, make sure that all of them have the same or lower ESR than the original part. The capacitance and rated voltage should be the same or slightly higher than original.

[vtech]

You are correct as far as the feedback/ sample line.

There is not much detailed documentation on KA7500 since it happense to be a korean copycat of Texas Instrument TL494. (TI's documentation on the other hand is extensive)

I also found a 400W ATX supply which is proly closer to your unit.-- I've got it in a zipped file (along w/TL494 documentation). Just uploaded the file here;

http://www.eserviceinfo.com/download.php?fileid=33668

Good luck