Author Topic: Graphic Equaliser - SMPS (Read 4860 times)

Electro_1 · « **on:** November 27, 2016, 07:00:21 pm »

Hi all,
Wondering if somebody could guide me as to how best to proceed.
I have been asked to take a look at a graphics equaliser by a friend of mine. When he gave it to me it would not power up at all, but upon switching the mains power to off, a standby button would light up for a few seconds, indicating that there was definitely some power getting around inside.
So I opened it up last night and did some testing. My first suspicion was a capacitor gone faulty somewhere, but visual inspection showed no obvious problems. I then tested each one in circuit with an Atlas ESR meter and again nothing showed as a problem. I decided to reconnect everything to mains and start checking voltages to see if I could isolate the fault. Lo and behold it powered up perfectly ! However after several tests I find it will power up maybe 2 times out of 3. So it has now gone from not working at all, to exhibiting a sporadic fault. I have checked to see if there are any obvious solder issues that my proding around might have temporarily corrected but again I can't find any obvious issues. I'm a bit perplexed.
I did notice, while checking voltage on the high voltage capacitor (which appears to be good) that sometimes placing my probes on the cap would cause it to power off, or on, but I can't always recreate that anomaly.
The SMPS which is part of the main board appears to consist of a varsitor, some suppression capacitors, a choke coil, and an RU RS206 bridge rectifier. Voltage all seems fine on the high voltage capacitor.
This is as far as I have checked. It also has a TOP245YN Off Line Switcher which I don't know how to test properly.
The board is quite condensed with a mixture of through hole and surface mount components so the less I have to desolder in testing the better for fear of causing extra damage.
Is there anything obvious that I might be missing here ?
Any advice is appreciated.
Many thanks !

strawberry · « **Reply #1 on:** November 28, 2016, 03:45:55 am »

What TOP245YN is shoving on control pin 1 ?

BradC · « **Reply #2 on:** November 28, 2016, 08:59:59 am »

When I hit that point I break out the hair dryer and freezer spray to see if either makes it exhibit the fault. If that shows any promise then you can start breaking the board down by section to narrow down the component.

Electro_1 · « **Reply #3 on:** November 28, 2016, 02:59:55 pm »

Quote from: strawberry on November 28, 2016, 03:45:55 am

What TOP245YN is shoving on control pin 1 ?

I'm reading 326.5V DC on the control Pin when it's working and around 333V DC when it's exhibiting the fault.

Quote from: BradC on November 28, 2016, 08:59:59 am

When I hit that point I break out the hair dryer and freezer spray to see if either makes it exhibit the fault. If that shows any promise then you can start breaking the board down by section to narrow down the component.

I've never tried this method before. I have some dust off that I can invert and use the refrigerant from. I suppose a good place to start would be the capacitors around the SMPS ? Any pitfalls to this method to be aware of ?
Thank you both for replying

strawberry · « **Reply #4 on:** November 28, 2016, 04:51:43 pm »

Quote from: Electro_1 on November 28, 2016, 02:59:55 pm

Quote from: strawberry on November 28, 2016, 03:45:55 am
What TOP245YN is shoving on control pin 1 ?
I'm reading 326.5V DC on the control Pin when it's working and around 333V DC when it's exhibiting the fault.
Quote from: BradC on November 28, 2016, 08:59:59 am
When I hit that point I break out the hair dryer and freezer spray to see if either makes it exhibit the fault. If that shows any promise then you can start breaking the board down by section to narrow down the component.

I've never tried this method before. I have some dust off that I can invert and use the refrigerant from. I suppose a good place to start would be the capacitors around the SMPS ? Any pitfalls to this method to be aware of ?
Thank you both for replying

safty first, better do not spray something on live parts operating at 650Vpk!! especially at your control pin

Electro_1 · « **Reply #5 on:** November 28, 2016, 05:20:05 pm »

Agreed. I thought about it afterwards and decided against it. Too much chance of frying the circuit and myself.
I have a Fluke 179 RMS meter with a temperature probe. I have never used the temp probe before, but based on the freeze method I was wondering if it might be able to indicate to me an over heating component.
Anyone ever tried this ?
In regard to the the 245YN, that reading indicates to me that it's working correctly. Am I right in this ?
Still scratching my head here.

BradC · « **Reply #6 on:** November 28, 2016, 11:49:09 pm »

if you don't want to hit it with freezer while its live, just in-plug it, make sure the caps are discharged and hit it all over with the freezer. Wait for the frost to evaporate and plug it in. The aim is simply to find out if it starts up better when cold, or doesn't start at all. Same with the hair dryer. Once you find a reliable characteristic you can then use that to narrow down the area in question. You don't have to hit it while it's live (although I generally do).

As long as you are not making sparks, butane makes a good freezer only. I always have a can of that around for the gas powered iron and its more environmentally friendly that pumping r134a into the air. It's going to be interesting to see what happens to that now the worldwide phaseout on refrigerants has been extended to HFC.

strawberry · « **Reply #7 on:** November 29, 2016, 12:55:51 am »

Do not test flyback unpredictable high voltages, othervise you Fluke will be gone.
pin #1 is left and it has internal voltage source about 6..7V, when in normal operation it stays still at 7V, if fault mode(any short or lost feedback) then pulsing 1cycle or two by pin #1 electrolytic 22..47uF. Posible causes: Short circuits in clamp circuit. Line sense resistors broken. Broken bias diode (some glass kinda) or dry cap. Tl431 reference is broken. ..

Electron_1 · « **Reply #8 on:** November 29, 2016, 11:12:41 pm »

Still scratching my head on this one.
Haven't had much time to test today but all in circuit tests on diodes etc. show as expected.
Time to start removing components.
I presume it's best to begin with capacitors ?
Also I don't know how to properly test the TOP245YN. Perhaps somebody might be able to guide me ?
The idea of freezing or heating to see which enviroment is best for it isn't really feasible, as the fault is so sporadic it can start 10 times out of 10 and then not start the next 5, and vise versa.
I've included a few phone photos below.
Thanks for all your help.

strawberry · « **Reply #9 on:** November 30, 2016, 02:26:59 am »

Thors hammer will you help this world from cra..

Electron_1 · « **Reply #10 on:** November 30, 2016, 02:25:35 pm »

Thor's hammer didn't help so I measured component temperatures instead.
Nothing outside the ordinary.
The search continues.

strawberry · « **Reply #11 on:** November 30, 2016, 06:51:57 pm »

http://www.farnell.com/datasheets/26251.pdf page #6
Check that C5 got cavity. that capacitor should read 5.8Vdc and 9..20Vdc that neighbor capacitor. page #20

Worst and dangerous designed powersupply. Look up something insuficient that prevent normal operation. Check voltages at point when it wont startup, then apply heat or vibration ... to make it again work

Electron_1 · « **Reply #12 on:** November 30, 2016, 07:28:55 pm »

Thanks strawberry,
Will check now and post back.

Electron_1 · « **Reply #13 on:** November 30, 2016, 10:12:35 pm »

Wow Murphys Law !
Took nearly 2 hours to coax her back on.
No amount of heat, cold or vibration seems to make a difference. She just switched back on while I was taking a measurement across C4.
When functioning,
C5= 5.776v
C4= 18.12v

When not functioning,
C5= 5.25v
C4= .566v/.462v

Nothing over .5v going anywhere else on the system when not functioning.

strawberry · « **Reply #14 on:** December 01, 2016, 01:25:43 am »

TOP245 is in fault mode as in page #6
0.566V seems like diode drop, some short circuit somewhere... or break. some faulty capacitors, SMD resistors, everything else
What that diode do near C4?

Electron_1 · « **Reply #15 on:** December 01, 2016, 08:02:16 pm »

Diode D1 is taking voltage from one pin of the transformer and feeding into C4 which is in turn feeding the cathode of the 4N35 Opto.
It's reading 18.26v when working and 6v when in standby. Still waiting for it to fail so I can take a reading.
Am unable to identify what type of diode it is without removing it from circuit.

Electron_1 · « **Reply #16 on:** December 06, 2016, 11:18:08 pm »

So I took a break from this for a few days to finish some other projects and went back to it this evening.
It's been on for 5 hours and only now just failed.
While functioning I've shaken it, vibrated it, heated it, and talked sternly to it to try and get it to fail, but nothing seems to make a difference.

It only fail when it feels like it.
I bought a new can of freeze spray and as soon as it failed I went to work. Nothing !
So just wondering if any of you kind, nice, decent people might have any other tricks up their sleeve to help me isolate the expel the gremlin from this machine ?
Thanks.

strawberry · « **Reply #17 on:** December 06, 2016, 11:56:43 pm »

Get similar powersupply and begin swaping parts. Then you will know what is wrong with that puppy. Or.. that part about bad enginering and Thor's hammer

Electron_1 · « **Reply #18 on:** December 07, 2016, 12:10:05 am »

Similar power supply is not an option.
I'll have to start collecting parts and start swapping.
So be it.
Thanks.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Graphic Equaliser - SMPS (Read 4860 times)

Share me