Maker Pro
Maker Pro

Switch Mode Power Supply makes a clicking noise

Hello Dear Reader

I'm currently working on this Switch Mode Power Supply that is used in a Satellite Receiver.

It was working fine till the day I decided to clean the dust around my TV set. In order to make life easier, I unplugged all the devices, including the receiver. After the cleaning was done, and hooked up the satellite receiver properly, the receiver never turned on again.

However I must admit I wasn't sure if the cable went into the socket all the way (in the back of the unit), so I unplugged it and plugged it back again, doing this couple of times. And I also think the cable was plugged into the 110V so it was live, I would assume this ON and OFF, ON and OFF is what caused the damage, since there is no separate switch on the back of the unit like more expensive ones would have.

Next step I took the metal cover off, looked inside. I can't see any fried components, nor any smell; However, I could hear a hardly audible clicking noise, not sure where it's coming from (maybe the transformer), that clicking sound would continue for couple of seconds even after I unplugged the unit from the 110V Ac. The sound was actually a little louder after unplugging the unit, like a spark plug noise on a lawnmower while testing.

I started with the diodes, some I had to remove from the circuits some were fine inside the circuit. Diodes are OK. I checked the resistors those are also good. I checked the capacitors especially the 64uF 400V giant CAP, that is good too.

Checked the continuity on the Coil and the Transformer (it has 4 legs on one side 5 on the other).

I even removed the power MOSFET (IRF830) from the heatsink. I tested that with a voltmeter. Put the black lead to Source, red lead to drain, it was OL. Left the black lead on source, touched the Gate with the red lead quickly, then went back to drain with red lead, I saw some numbers on the voltmeter like 1200, 1489, 1800, then OL. So it was weird, numbers going up instead of going down.

I reversed the leads, I put the red lead on the Source, black on the Drain, voltmeter read from 420 to 580, I would assume .420 and .580 Volt (this would be the voltage that makes the MOSFET work?).

I did not do any live testing with the power supply plugged in, because I would need some advise what is that I should do next.

NOTE: The bottom side of the circuit board is flipped over for easier component trace. Under the black box is a card reader for certain paid channels. (one diode was removed while the photo was taken)

If needed I can post more pictures.

Thank you for helping me out.IMG_0026.JPG IMG_0086.JPG
 
Clicking is the sound of the over-current (or over-voltage) protection circuitry operating. Common causes are the capacitors being 'dried out' and simple replacement of the electrolytics often affects a cure.

Signs to look for are leakage (fluid - maybe dried out) from the bottom of them, bulging tops and/or discolouration around them at pcb level.

If your device is well known (high sales) then Googling 'repair kit' and its make/model often turns up the kit of parts that are, again, usual suspects.
 
Capacitors look good, I checked them, none of them is OPEN or Shorted. I don't see any leak or bulging on them.

Unfortunately the satellite receiver is not a very popular model it's the SE830 for Wolrd TV - Globecast. It's not manufactured anymore, I couldn't find any schematics on this model either, only USER's Guide manual.
 
Could be something off-board that is causing the overload. Connect the board back into position but remove the DC output ribbon cable and check that the board powers up without the 'clicking'. If this is the case then you have an off-psu-board fault.
 
Sir Alektron . . . . .


If this is put back in the state initially found in, to be operational, then, if the AC power was connected and left connected, and you then get that temporary LED flash, leave the set alone for an hour of "warm up" time..
Is the large heat sink having any heat ? if so move to all of the electrolytic capacitors and place a fingertip to their metal case tops, feeling for any with warmth.
If the heat sink is stone cold, my next step would be to pull out the smaller electrolytic that is at the front of the big heat sink . . . . . with D6 diode also being spaced out from it a bit.

Next option would be to change out the electrolytic that is right beside aTL431 I.C. that is in a TO-92 plastic transistor profile case.

Further specific locating I.D. . . . .

There is ZD1 a small 750MW zener diode , then a large electrolytic capacitor, then THE capacitor and then the
TL431 on past it.

Triple check the electrolytic polarities,right at install time, to be sure that they are not installed backwards.

73's de Edd
 
Thank you for both replies.

So here is what I'm going to do. I'll reassemble the MOSFET and put it back into the PCB, then I'm going to turn it ON without hooking up the DC cables to the other board inside the unit, and see (I mean hear) if it still makes that clicking sound. (kelly's eye suggestion)

If the clicking is still their I'll follow De Edd instructions. I just don't have an LED that would FLASH, so that part does not apply here.
 
IMG_0088.JPG I reassembled the MOSFET into the board, then plugged the Power Supply Board into the 110V. It was clicking for a while like maybe 4 minutes, then the clicking went away. After that I took some measurements at the DC outputs (see photos) there are like 10 outputs. I got a reading of 0.6 V or 600 mV.

Then I plugged the Power Supply Board into the MAIN Board with the DC connectors, only to see if the clicking comes back, but there was no clicking.

After this I unplugged the PSB from the Main Board and waited to see if the Capacitors warm up, like De Edd mentioned. I did not feel any heat, however it was on for only 30 minutes.

After unplugging the Power Supply Board from the 110V, I heard one or two clicks, and then silence.
 
Success! Unit powers up!

Thank you Mr. 73's De Edd for suggesting to replace the 25uF 25V (close to the heat sink) and the 1uF 50V capacitors.
Also thank you Kelly's eye for pointing me in the right directions to check the capacitors.

Only issue while de-soldering was the 1uF 50V capacitor, one of the pads was damaged for the capacitor's leg, when re-soldering the new one, I had to use a small wire as a solution. It looks better then in the photo.
If there is a better solution for these type of situations please let me know, I'm open to ideas.

After I replaced those two capacitors, I went back and measured the DC output (those 10), the DMM read between 27V to 30V.

Then connected the power supply board to the Main Board, and the unit powered up.

Thank you so much, again.2.jpg IMG_0094.JPG IMG_0095.JPG IMG_0097.JPG IMG_0098.JPG IMG_0090.JPG
 
Sir Alektron . . . . .

It was the 22 ufd unit that fixed it, I just additionally gave you the next suspect in line.

NOW . . . . . since that unit can now operate, and cause the heat sink to warm up, check after 3 hrs or so of run time and see if any of the E-caps metal cap tops show heating up with the fingertip test.
If so, then they are suspect for replacement ALSO.

NOW . . . . on the opposite side of the central SM power transformer, on the "COLD" side of the power supply.
DO NOTE the specs on the side of all of those 7 or so caps jackets and confirm if they are all being of a 120 degrees C temperature rating instead of the other common rating of 85 degrees C. . . .quite bucolic.

SWITCH MODE power supplies wave forms and high frequencies absolutely "hammer the E-caps" the 120 degree rating is needed for any life longevity.
The SAME is true of the two caps that you just replaced.

73's de Edd
 
The old 22uF 16V and the 1uF 50V have a marking of 105 degrees Celsius on them.

The new ones 22uF 25V and the 1uF 50V also have a 105 degrees Celsius marking on them, also a -40 degrees Celsius.

I will check the other ones you mentioned in the next couple of days. Thank you De Edd for the advise, I highly appreciate it.
 
Since we are talking about detecting components for a short by monitoring their temperature.

I've seen some people use Freeze Spray on the circuit board, to see which component melts the frozen chemical solution. I've always wondered if that is a safe thing to do?

I have a laser temperature gauge, I was thinking to use that, and this way I could get an exact measurement of the temperature in Celsius of each capacitor.
 
It's usually safe to do so. I've only ever done so for specific thermal issues, not for detecting dud capacitors - the specific fault symptoms of which are usually enough to point in the right direction.

In an ideal world we'd all have access to hi-res thermal cameras! I wish!
 
Top