Page 3 of 12 FirstFirst 12345 ... LastLast
Results 41 to 60 of 231

Thread: CubeRevo HD PSU repair

  1. #41
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Thanks stroker.
    Resistance between 5v and GND I’ve measured both at the female part of the output plug, and from the blobs of solder on the underside. Unfortunately, I cannot check directly with the HDD output because it, and the output plug is soldered in to the PCB – you can’t pull it out to get to any header pins. Tonight it’s still giving me 96 ohms.
    Between 3v3 and GND, I don’t get a stable reading to J15 nor to the output wire. It may start at around 800Ohms then steadily rise, may give a flash to a few MOhms or OL, and then come back to around 8kOhms. On the underside it did stabilise at around 860 ohms. Incidentally the SATA power cable does not carry 3.3v – only has four wires. So actually there are two 12v and two 5v connectors to the main board, one via the long plug and one via the SATA cable direct to the HDD. Also I had previously checked that I had 12v at the output end, but not 5v nor 3.3v.
    Looking at IC 41 there are only three pins on the underside, the middle one being GND. No short to either of the others.
    All the checks I’ve done when powered up were with and without the main board connected and with and without the front board connected.

  2. #42
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    I want you to try to check 3 diodes D41,D42 and D43.

    D41 is near the heatsinks and D42 and D43 are 3 legged devices one on each of the heatsinks.
    With D42 and D43 the outside legs are joined together so check between the middle leg and either of the other legs.
    These checks will be more difficult than before, there are capacitors connected as before but these diodes also have lower forward voltage drop.
    So instead of reading about 600 in one direction they will read about 100.
    You may not be able to get any sensible readings due to the capacitors and the transformer interfering.

    Note the 12V supply for the hard drive is different from the 12V on the main output connector.
    The hard drive supply goes off when in standby as do the 5V and 3V3 supplies.

    I would really like to know the voltages on J2 and R61 these are near PC91 measure R61 end furthest from PC91.
    This could help explain what is happening on the output side of PC91 which is confusing me.
    It may be easier to get at these if you remove the large heatsink, otherwise you may have turn the board upside down.

    I think we need to start thinking about getting IC91 replaced.
    Although the voltage readings above may throw suspicion on PC91

  3. The Following User Says Thank You to stroker For This Useful Post:

    jphilweybr (05-01-2017)

  4. #43
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Thanks again stroker.
    Before I reply directly, I’d better tell you of a thought that came to me in the night. I said yesterday that all the checks I’d done were with the main board disconnected and connected. True, and also of the last one linking R98 and R95 when no sign of life appeared. But with that one, and the main board disconnected, I did not re-check voltages at the output cable. I did that tonight, and hey presto, I had 5v and 3.5v at 3v3! It’s just the fan didn’t kick in, but, come to think of it, we’ve had 12v on the main output all along. Not tonight, but I can check voltages at the fan header pins, and even try it out in a computer.
    So you are on to something stroker, and sorry I didn’t think of that before.
    As to the other points :
    D41 shows infinity in one direction and 402 (0.4v) I think in the other.
    With D42 and D43, I’ve tested at the underside from the middle pin to the others. I don’t get a stable reading, but they dance around in the 70s and 80s in both directions. I don’t get an infinity reading in either.
    With the link between R98 and R95 still in place, voltage at the cold end of R61 is 3.5v, and at J2 1.9v

  5. The Following User Says Thank You to jphilweybr For This Useful Post:

    stroker (05-01-2017)

  6. #44
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Wow I wasn't expecting that - well done for your discovery.
    This proves the power supply is working apart from maybe some of the on/off control.
    I think we need to stop for a while and confirm what we have working.

    I know I suggested using the fan to indicate that power is on but for some reason this is not suitable.
    If you run any checks on the fan you need to check it's voltage rating it may be 5V or 12V etc.
    I would expect the fan is OK anyway.

    If you have a SATA hard drive available we may be able use that to indicate power is on as it will spin up.
    No need to connect the data cable just the power cable.
    As I said before the 5V, 3V3 and hard drive 12V come on together if the HD spins all 3 supplies should be present.

    If you can use a hard drive in this way we can check if it spins with/without the mainboard connected.
    And again with the main-board connected and with/without the front panel connected. This will be quicker than using a meter.
    I suggest using an old hard drive as power will be going on/off many times which will not do it much good.
    If you cannot use a hard drive just use the meter on one of the supplies 5V, 3V3 or hard drive 12V.
    Be sure to turn the main power off when plugging/unplugging power connections to any of the boards.

    If you get good results with this try removing the link on R95/R98, and unplug the main board.
    Turn the main power on and check if shorting 5VS and P/S (J7 to J9) causes drive to spin.
    If it does it will show the on/off control circuitry on the power supply board is working but unfortunately suggests the fault we are looking for is on the main board or front panel.

  7. #45
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Thanks stroker.

    I'm out for the rest of the day, so will get back to you ttomorrow.

  8. #46
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    IMG_4632.jpgThanks, stroker. I had got round to thinking it was time I took stock. So today, as well as doing the measurements you asked for, I’ve re-done a lot of the others and taken time to prepare a little do***ent. I’ll try uploading it as PDF, or failing that as a pic (you have to pay to upload pics, which I’ve done).
    A short summary of what comes to my mind, albeit I haven’t got past Chapter 1 of Teach Yourself Electronics (if there were such a thing!) :
    You’ve clearly found a way of fooling the standby switch system by linking R98 and R95. I can get all the voltages needed at the output cable and the SATA cable, which can fire up a HDD.
    When I remove the link, we lose +9v and +3.3v. and the SATA voltages.
    Nevertheless, there is no voltage at the fan header pins on the main board under any scenario.
    There is no voltage drop through R98/R96 nor through R95 – except through R95 when it is linked to R98.
    Nothing lights up the LED on the front panel.
    The front panel switch has no effect in any of the scenarios we’ve put it to the test.

  9. The Following User Says Thank You to jphilweybr For This Useful Post:

    stroker (08-01-2017)

  10. #47
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Thanks for the summary it is very useful and has pointed out something important I think we had missed.

    If correct, I would like you to double check, it shows that the front panel kills 5VS, I think this is on J7 on power supply.
    It's important to double check this so we don't chase a red herring.
    Then with front panel still connected check for a short from 5VS to GND.
    Check continuity from 5VS to the pins on the 7 pin connector on the front panel, I think the larger connector just carries display data.
    So you are trying to find out where 5VS is on the front panel, I assume it does does go there?

    Remove and unplug the front panel and check for shorts across capacitors CE1 and CE2 you can also check for shorts across the red/orange glass diodes and also ZD1 (3 connections) near the transformer. Note these devices are SMT (surface mount technology) they are soldered on the surface of the board as apposed to PIH (pin in hole) which have leads that pass through holes in the board.

    Looking at the pic of your front panel I don't see transistor Q1 is there one?
    Please post a pic of the other side of the front panel as I think I will need this.

  11. #48
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    I have split this into two posts so please read the previous one as well.

    To answer a couple of your questions.

    Wire jumpers on the power supply - the circuit board is single layer, it has copper tracks on one side only. Connecting certain points together is impossible using the tracks on the board because you need to get over/around other tracks. They are sometimes referred to as wire bridges which is perhaps a better a better name. More complex circuit boards a double layer or double sided with tracks on both sides which effectively use tracks on one side to jump or bridge tracks on the other side. Then there are multi-layer boards which have copper layers inside the board.

    Difficulty reading the marking on devices - in more recent times manufacturers have changed to laser marking to put printed details on devices. I am not keen on this as it is sometimes very difficult to read. Previously they used to use ink marking which was much clearer but who is said progress was supposed to be an improvement?

    I have looked as some other power supplies and would suggest that PC41 and PC91 are PC-17L1. I would expect just about any opto-isolator to work but you need be careful as they come in different packages, some have 4 pins like ours but others have 6 or 8 leads.

  12. The Following User Says Thank You to stroker For This Useful Post:

    jphilweybr (08-01-2017)

  13. #49
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Yes I did check the 5VS voltage several times on Dec 29th because I didn’t quite believe it. Then yesterday, the same thing – only this time I was no longer surprised. And to repeat, with the main board connected, the voltage is there, but once the main board is connected to the front board, it disappears.
    OK, now to today’s homework.https://www.satpimps.co.uk/images/smilies/07.gif
    With the front panel connected, the resistance between 5VS is about 95ohms, much the same as when no panel is connected to the PSU board.
    The front panel is connected to 5VS by Pin2 of the wider connector (i.e. the one next to the red wire)
    Resistance across CE1 was 94ohms. In hunting for continuity to 5VS, and elsewhere, I come across several resistances around the 94-96ohms range. That can’t be a coincidence, so what’s the reason? Sorry if I’m jumping to Chapter 19 of the Beginner’s Guide to Electroncis!https://www.satpimps.co.uk/images/smilies/redface.gif
    Across CE2, I measured 116ohms.
    Across D1, D2, D3 and ZD1, I’m getting resistance readings in lots of kohms and Mohms – in the case of ZD1 just OL. However, I didn’t get proper diode checks. Here, I may be running into a problem with these very small, closely-spaced components, in that my meter probes are a bit thick and my eyesight may not be up to it – despite paying through the nose for new specs just before Christmas!
    Probably like you, I found Q2 near the standby switch, and Q3 and Q4 near the transformer – but no Q1.
    Now I thought you might want to see the underside of the front panel board, and I’ve been looking at it myself. Trouble is it’s sort of navy blue with just dozens of small lines because all those little squares produce red characters in the display. So here are three. Maybe I'll try again in daylight.
    Thanks for the explanations – you must be taking me into at least Chapter 5! A little confession – at one point I nearly made a fool of myself by asking you how is the box grounded. I could only think they are relying on the Neutral being connected to the Earth busbar at the house distribution board. And then I thought – well how else, so I didn’t ask.

  14. #50
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Well looks like I didn't manage to upload a pic, so another go nowIMG_4639.jpgIMG_4638.jpgIMG_4637.jpg

  15. #51
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Thanks for the pics but cannot see enough detail.
    Try and get it all in one pic like you did on the front side and then maybe 3 or 4 sections.

    5VS is being killed by something on the front panel - I have a feeling for the transformer circuit - want to find a way of disabling it or removing power to this section.
    I have been looking for data on VFD (Vacuum Flourescent Displays) but so far only got a headache! They are a bit like valves - anodes, cathodes and grids!

    Can you check - are there devices hidden under the display itself?

    The standby switch is it momentary or latching? Does it have an LED in the middle of it? It looks strange to me!
    Is there continuity between 5VS and/or P/S and any of the pins on the standby switch (4 pins in a square)?

    Which connector on the display board is P/S on? If it's on the same connector as 5VS unplug the other connector.
    If 5VS is present try pressing standby button, perhaps check this with HDD connected to power lead - does it spin?
    If that does not work unplug front panel completely and try shorting 5VS and P/S - does HDD spin?

  16. #52
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    First a few more pics – better than last night’s I think.
    As to the other components, I could not see anything directly under the glass. I could tilt the whole display unit up a bit to peer underneath and just found the square component in the pic, together with some small ones which may have been visible before. And I notice that this board has some tracks running on the underside and some above.
    The switch does appear to have a little LED in it, but this will not be visible to the user, because it is covered by a brushed chrome button which you press. It has six pins. If I number the top row (nearest the cables in the pic) as 1,2,3 and the bottom row as 4,5,6 there is continuity between 1 and 3 and between 4 and 6. If I press the switch and let go, there is no change. If I hold down the switch, I get continuity between 1,3,4 and 6 –so the top row is momentarily connected to the bottom.
    Yes, shorting PS and 5VS does power up the HDD.
    I’ve found continuity between the main board, and pins on the front board for FL+. FL-, GND and +5VS – but nothing else – notably no PS, and no 12V or 5V. Hmmm. Should I be looking at the main board?
    Will be breaking off now – you probably won’t hear from me again until Wednesday evening.
    IMG_4640.jpgIMG_4641.jpgIMG_4642.jpgIMG_4643.jpgIMG_4646.jpg

  17. The Following User Says Thank You to jphilweybr For This Useful Post:

    stroker (09-01-2017)

  18. #53
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Thanks for pics they are much better.

    You were asking about GND.
    The term is quite often used very loosely and to my mind wrongly - I usually prefer to call it 0V.
    It does seem to suggest it is earthed but in many cases it is not and is simply a reference/common point.
    I would expect your mains lead only have 2 cores with no earth.
    It would be dangerous to use neutral as an earth because under certain (external) fault conditions this could become live.
    Also if the mains lead plugs into the box it maybe possible to plug it in the opposite way around causing live and neutral to be swapped making the chassis live!

    Now back to the front panel.
    Check for continuity between 5VS (or pin2 of the longer connector) and the middle pin of the 5 pins on the transformer.
    Assuming you have continuity I want you to do something drastic but repairable.
    There is a track running down from the middle pin of the transformer with 2 dogs leg bends in it.
    I want you to cut across the track in the middle of the longest straight section so as to disconnect the connection to the transformer.
    Use a sharp knife, scalpel or stanley type knife, do it slowly and carefully we just want to make a break the track without damaging anything else.
    Cut or scrape away at the track until there is a clear break in it. If it helps you can scrape away the blue coating (solder resist) first to expose the copper track.
    Then check you no longer have continuity between 5VS and the transformer.
    Now connect everything back up and check if 5VS is present, if it is try pressing the standby switch - a little prayer may help!
    If anything is not clear ask before doing the above.

    I think the display needs a fairly high voltage produced by the transformer. It looks like the 2 output pins of the transformer are covered in silicone?
    The display would presumably be on even in standby - showing the time?
    Transistors Q3 and Q4 are prime suspects so see if you can read anything from them - only expect a few characters - you may find it best to shine a light across them while you look down. Another closeup view pic of the area around the transformer may be of help.

  19. The Following User Says Thank You to stroker For This Useful Post:

    jphilweybr (11-01-2017)

  20. #54
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Thanks for the info.
    There is continuity between +5VS and the three middle of the five pins. There are two black lines from the centre pin. Did you mean me to cut through both? Sorry, I don’t know whether the two lines make one track or two tracks. I think I’ve displaced the wire enough to sever the longer one, but perhaps not surprisingly, I’m still getting continuity to +5VS.
    The pins from the transformer near the edge of the board feel sharp to the finger; I don’t know what the smudges near the base are.
    Yes the display was always lit up – can’t remember whether it was the time.
    On Q3 and Q4, I can see EY and then two horizontal lines – maybe arrows in opposite directions.
    Pics of my cut and of the transformer area attached.
    I thought I’d check with you before going any further.
    Thanks again.
    IMG_4636.jpgIMG_4648.jpg

  21. #55
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Thanks again for the pics.
    I'm not sure you understood me completely - I needed to explain better - sorry I forgot you are new to this sort of thing.
    The darker parts are the gaps between the copper areas, so the track is the part between the black lines you refer to, if I understand you.
    From the pic though it looks like you have almost cut it anyway, you just need to cut a little more at the edges.
    Perhaps you can have another look, a good magnifier would help you see.
    I think the board is just double sided (no internal tracks) but I am not sure so do not cut too deep.
    If in doubt just leave it for now, although if you can cut the track it could prove very useful as it will hopefully cut 5VS from the transformer.
    If 5VS is then OK I am hoping the box will come out of standby, obviously without the display working, and we can see what is working.

    Besides that can you make out the details on R17 and R18, they are probably the same but I cannot make it out from the pic.
    Good luck!

  22. #56
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Hope this isn't a giant step backwards, stroker but I've had at least a mini-disaster.
    After the last post, I was re-checking the voltages by joining J7 and J9, and something happened which caused a bit of a flash. I discovered that it had blown the fuse in the plugtop.The main fuse on the PSU BOard was not blown and I am still getting similar voltages to before between the IC91 heatsink and R95 and R96, but nothing at the output cable end.

    Sorry!

  23. #57
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    OK - these things happen
    Are you saying you've lost all outputs or just some? Was the front panel connected?

    Don't forget we have 2 power supplies, the always on one and the beefier one which should come on when you short J7 and J9.
    If you have 5VS etc. the always on supply is working.

    Do not short J7 and J9 yet until you measure the voltages on each of these first and check for shorts. In fact do not short J7/J9 unless you find something wrong.
    Check for a short to GND from 5V and 3V3, also check HDD power lead is not touching anything it shouldn't eg. chassis.
    If you find anything shorted unplug the front panel and main board and see if the short goes.

    If you find nothing wrong look for something burnt and check for smells (with power removed).
    You say there was a flash this could leave a burn mark where something has shorted out when something touched something it should not have.
    Is there a chance that any of the bits of copper from cut track has got in somewhere?

  24. #58
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Is it possible you shorted J5 and J6 by mistake - check for burn marks on the links.

  25. #59
    Verified Registered User
    Join Date
    Aug 2010
    Posts
    168
    Thanks
    82
    Thanked 117 Times in 105 Posts
    Ok let's do some checks.

    1. Measure voltage between IC91 heatsink and R82 end nearest transformer expect around 330V dc. The incoming mains is filtered by L81, C81 and C82 to prevent electrical noise getting out of the PSU. Rectified by D81,D82.D83 and D84 and then smoothed by C85, the biggest capacitor on the board, this produces over 300V dc.

    2. Leave powered up for say 10 mins, power off and check if C85 has got hot, if any of D81-84 is faulty ac could get to C85. This would usually blow the fuse but maybe worth checking. While doing this you can check if any voltages are present on the output end.

    3. If you have no output voltages take a close look at IC81, A6351 next to C85, look for cracks and/or burn marks. This is the switching device for the always on supplies. Try to check diode D23 on the cold side of the transformer next to R24.

    4. Link R95 and R98, as before, at there ends nearest the mains input, this powers up the main/beefier supply which is used when the box comes out of standby. The link also bypasses the usual power on control circuitry. Power on and check 5V, 3V3 and HDD 12V.

  26. The Following User Says Thank You to stroker For This Useful Post:

    jphilweybr (12-01-2017)

  27. #60
    Senior Member
    Join Date
    Dec 2016
    Posts
    113
    Thanks
    77
    Thanked 40 Times in 40 Posts
    Thanks and sorry to send you off on another avenue od detective work!

    Voltage between IC91 heatsink and R82 is 19.2vDC – which I guess you’re going to say shows something’s wrong. Curiously, last night I found the voltages from there to R98 and R95 were similar to previous readings namely 11v and 1.8v respectively, but today I’m getting respectively 5.1 and unstable or zero. Also, I checked the voltage to R82 three times within the space of ten minutes or so, and each one was a shade lower than the previous – 20.0, 19.8 and 19.2.
    D81, D82, D83, and D84 are all good – a shade under 0.6v in one direction and infinity in the other.
    C85 did not get hot.
    I see no sign of damage in the IC81, C85 area. Where pins enter the board, there are little patches of yellow around the holes, as there with many pins, particularly resistors. In fact, now that I’ve looked at them, the pics on the resource you found show something similar.
    No voltages at the output end. The four GNDs have continuity to one another and to the chassis.

Page 3 of 12 FirstFirst 12345 ... LastLast

Similar Threads

  1. hd tv repair
    By BOOGIBEAR in forum HD Televisions Discussion
    Replies: 6
    Last Post: 04-10-2011, 09:03 PM
  2. tv repair off c#m#t??
    By BOOGIBEAR in forum HD Chat
    Replies: 0
    Last Post: 13-09-2011, 12:24 AM
  3. echostar repair
    By micheal in forum Echostar
    Replies: 2
    Last Post: 21-02-2009, 05:11 PM
  4. BT Repair Charge
    By fergiz in forum General Chat
    Replies: 39
    Last Post: 09-01-2009, 12:31 PM

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •