Connect with us

Repairing a Goodmans PVR

Discussion in 'Troubleshooting and Repair' started by nobrob, Jan 22, 2012.

Scroll to continue with content
  1. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    Hello,

    My Goodmans GHD1621F2 Freeview recorder stopped working at one point. I disassembled it to discover that a component smells burnt and the PCB around it is brown/black.

    The component is labelled:

    And looks very similar to this: http://www.dalbani.co.uk/catalogue/product_details.php?id=29200

    That shops sells the item for £3 plus £5 postage (!) and it's the only one I've found so far. Any ideas if I can replace this component with a cheaper, more readily-available alternative?

    Many thanks
     
  2. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    You could try to look for the following prefixes (instead of DM) which I believe may be identical (or at least similar, maybe different packages, I haven't checked).
    KA1H / KA1L / KA1M
    KA5H / KA5L / KA5M
    FSDH / FSDL / FSDM

    Is there no output from the PSU, or are only going by the smell?
     
  3. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    I remember getting a reply from you last time and being blown away by your knowledge. Once again, you somehow know I'm talking about the PSU board of this unit :D

    I found an unused DVD player and extracted the DL0165R, unfortunately this didn't do the trick. Could this component be too different?

    I am solely going by the smell/discolouration, as there was also a resistor next to this component which melted. The resistor measured at 56k which is what the label says it should be. I replaced that too to no avail - was a long shot!

    The PSU is not completely dead, the machine will power up and attempt to read data from the hard drive. The hard drive is a standard IDE PATA and it doesn't spin up. If I supply the hard drive with an external supply, the disk will spin up and the machine will boot successfully. This is why I assume it's the PSU, what with the seemingly burnt components on the board.

    The PSU has a cable with a molex connector which hooks up to the hard drive, which reads +11.5 (yellow) and +5.2 (red). The external power supply which works measures over +12 (yellow), could this be the problem?

    If it is, I don't really know where to look, component-wise!

    Once again, thanks :)
     
  4. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    Of course, it was after all mentioned to be a SMPS Power Switch IC in your link. :D Then it's just a matter of looking through some Google results.
    The Fairchild datasheet says there are some maximum power differences between the 1, 2, & 3 versions, 1 being the weakest and 3 the strongest.
    But there's no point in replacing that IC as long as the PSU works at all (which it does). If the IC fails it either blows up or there's no output at all.
    The voltages seems ok, but were they measured under load (boot)? Are there other outputs from the PSU in addition to the 12V & 5V?
    Provide a picture of the PSU if you can. Maybe we can spot something, or just get an idea. It could always be bad capacitors.
     
  5. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    Images of the PSU attached.

    Although the area around the DM0265R is blackened, some of that is due to my soldering on the solder side.

    Currently the DL0165R is on there, but I'll swap that back as you mentioned the lesser rating (and the fact that it's not fixed!).

    I just measured the voltages under load and they're more like the ones from the external power supply (~+12.25 and +5.00).

    It's clear that the hard drive attempts to spin up, as it clicks a couple of times. It doesn't do this when I disconnect the hard drive's power supply and power the unit up.

    There are three connectors on this board, one of them goes to the mains plug, another plugs into the hard drive and the last one plugs into another PCB (pinout +12V, GND, GND, +5V, T5V ...whatever T5V means).

    board_silkscreen and board_solderside give you an overview
    capacitors_maybeleaking show you what I think is glue, to hold the caps in place
    capacitors_notbulging should give you a profile of most caps
     

    Attached Files:

    Last edited: Jan 22, 2012
  6. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    Phenolic boards blacken very easily. It doesn't mean anything unless it's charred.
    T5V seems to be just a separately filtered branch off the +5V. What brands and series are the output capacitors?
    I'm still leaning towards high ESR's - leading to ripple & instability. The small capacitor(s) on the primary side are always suspects btw..
     
  7. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    I had to Google "psu primary side"!

    Capacitor list:

    Left (primary?)
    Chang CD263 47 microfarads 400V
    Chang CD288 47 microfarads 25V <-- suspect little one?

    Right (secondary)
    Chang CD288 2200 microfarads 10V
    Chang CD288 1000 microfarads 10V x2
    Secon CD288H 1000 microfarads 16V x2
    Secon CD288 470 microfarads 10V
     
  8. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    Yep, that's correct.
    Hmm, there's very little mention of Chang & Secon cap's on the web, but they certainly sound like Chinese brands.
    Lacking an oscilloscope, try to set your meter to a low AC V range and measure on the DC outputs (under load).
     
  9. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    Thanks again for your help!

    I'll get on that as soon as I get home this evening from work.

    What is the expected result/tolerance level?
     
  10. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    It's hard to say, multimeters are only made to measure up to a few kHz, and PSU's operate between 30-100 kHz, so results will be very inaccurate/unreliable.
    Nevertheless, measuring anything more than a few (2-20) mV could be an indication that there's too much ripple (due to too high ESR).
    Sometimes capacitors get a high ESR without an accompanying bulge. (And sometimes cheap DMM's can't measure AC on DC.)
     
  11. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    My £7 multimeter from Maplin reports that AC voltage is double the DC voltage (10V and 24V).

    I imagine this could mean one of several things: (a) there is a full fledged AC current or (b) my multimeter is a bit mental (or just cheap). I may also be entirely wrong :)

    I've been meaning to buy an oscilloscope but their cost is quite preventive. I came across the DSO Nano v2 today (http://www.seeedstudio.com/depot/dso-nano-v2-p-681.html) though which seems both affordable and effective for my beginner state. Your thoughts please?
     
  12. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    The DSO Nano starts to "alias" (?) at around 50 Hz allegedly and the range for voltage starts at 10mV. So it doesn't appear to be suitable for even its first application!
     
  13. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,404
    2,777
    Jan 21, 2010
    I have a DSO Nano. It doesn't start to alias around 50Hz, more like 50kHz :)

    Having said that, it still puts switchmode power supplies in the range of frequencies that are going to cause trouble.

    They're OK for audio frequencies and are a lot easier to carry in your pocket than a regular scope.
     
  14. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    Well spotted - 50KHz is what I meant, sorry.

    Fingers crossed for an eBay bargain on on a big beast then!

    I know it's an investment and all that, but I'll probably be spending more on an oscilloscope than on a replacement Freeview PVR. Food for thought :)
     
  15. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    I went and picked up a Telequipment D67 today for a bank-breaking £15 :D

    I can only use the calibration signal as a reference at the moment. Y amplifier/attenuation seems OK but the timing is a little off. The manual mentions how to check the timing, but then refers to mysterious section for setting the timing. Hmm.

    Anyway, I managed to get the attached shot of the 12V feed at 10mV/div. 5V seems fine with a 2mV ripple.

    What's the next step? :)
     

    Attached Files:

  16. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,404
    2,777
    Jan 21, 2010
    The obvious thing is to replace the filter cap on the 12V output. They will be fairly low value (say 470uF) but they are low ESR capacitors.

    Replacing them is often the only thing you need to do (I would tend to replace all the low ESR caps because it's easier than having to open the device up later and replace more).

    [​IMG]

    Capacitor A seems to have a domed top and almost certainly should be replaced

    Capacitor E seems to be swollen on one side and if so, should be replaced.

    The others B, C, D, and F either appear to be OK, or the image is not good enough to say.

    I would tend to replace A to F.

    On the high voltage side, I rarely need to replace these. However if either are deformed, replace them.

    If you have an ESR meter, you can check the capacitors (make sure they're discharged!).

    I'm actually somewhat concerned about the diode above the IC. The resolution of the image is insufficient to see if anything's going on there, but I think you should take a closer look.
     

    Attached Files:

  17. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    Thanks for your advice.

    The diode above the IC seems to be behaving normally (has a resistance similar to that of the diode down from the IC in just one direction, the other passes no current).

    None of the capacitors are what I'd call swollen, so I think what you're seeing there is a artefact of my low quality camera.

    I have just now spotted a leaking capacitor though. I'm pretty sure it wasn't this way before, because I looked several times! I'll replace that one and see what happens.
     
  18. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    The leaky capacitor that I spotted was the one you labelled F.

    But following your advice to replace all capacitors I then spotted another leaky one, indeed the last one on the 12V output (B).

    I think I've somehow made the situation worse though. Because not only will the machine still not power up (after replacing all capacitors and then stepping back until reverting all capacitors to originals), but the ripple is now 1V high on the 5V output and 2V high on the 12V output (3-4 times per second).

    The PSU also makes a quiet buzzing noise every time the ripple peaks!

    Scope snapshot attached (0.5V/div)

    Has this board had its day or am I still in with a chance? :D
     

    Attached Files:

  19. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    A ripple of 15mV p-p doesn't seem like much. Was it loaded or unloaded though?
    Spotting a deficiency requires a load. You can use the PVR, or incandescent bulbs instead if you have some of a suitable voltage & power.
    I replace the small primary side cap's (H: CD288 47µF 25V) more frequently than the bigger secondary side cap's.
     
  20. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    It seemed to me to be more like 20mV, which was borderline as per your tolerance level.

    Getting my meter in-line to measure current is somewhat involved, but I can do so if you suggest it. The PVR has been on, but the hard drive won't spin up. I'm not sure what else, if anything, uses the 12V supply.

    I've replaced that one too, just to be sure. But there were definitely leaking caps on the secondary side, although they only manifested themselves recently.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-