TCD230040 died, power supply issues..

Discussion in 'TiVo Help Center' started by punkin, Dec 15, 2011.

  1. punkin

    punkin New Member

    6
    0
    Dec 15, 2011
    I've got one of the dinosaurs that has served me well, going on 10 years and it's been great (lifetime box).

    Last week, it said it overheated and powered down. On reboot a few hours later, stuck on Welcome, Powering UP. I've got plenty of hardware around, so I read a bunch, downloaded InstantCake and made up a new drive. Same problem.

    A buddy has a similar unit, so we started playing and measuring, I finally swapped power supplies with him, and my new BakedCake drive functioned flawlessly. So...we tested, pin by pin voltages. BOTH power supplies had nearly identical voltages loaded and unloaded... roughly 11.77v and 5.06 v on HD cable...the 20 pins we went down, and BOTH were nearly identical on every pin, loaded and unloaded.

    WTF? I can't measure current draw, but I see no sign of swollen caps, the power supply ckt board is dark (like it got hot) around the heat sink and main 2200mfd caps, but nothing is swollen, nothing appears to be leaking, measurements are the same.

    What am I missing? Yeah, I know I can order a new PS for $70, and I'd be willing, but I'm trying to figure out what I'm missing at the component level. I'm an electrical engineering dropout from 20 years ago, so I know just enough to be dangerous...but now I want to defeat this challenge!

    Help?

    Thanks guys...you've gotten me this far!
    --Scott
     
  2. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    How similar is his unit (what's the model number)?

    Is your power supply working okay in his unit?


    If you've got scorching on the circuit board, something got hotter than it should have. Are those 2200uF caps in parallel on the +5 or +12 rail?

    Is it possible that the hard drive that was in it when it started up had something go wrong with it that caused it to draw too much current? That could explain the scorching.

    You can probably replace every cap on the board (except the big one on the input side, which is probably still fine) for around $10, which would be cheap insurance since you don't know for sure what might have gotten overstressed by that drive, if it is the culprit.
     
  3. punkin

    punkin New Member

    6
    0
    Dec 15, 2011
    His unit is a TCD240 series. I didn't try my PS in his--I guess my logic was if his PS ran mine, and mine didn't, mine had a problem.

    My surprise was measuring EVERYTHING, and it seemed ok. Can Caps affect current delivery? I'm perfectly willing to drop ten or twenty bucks (locally) to source new caps, and replacing them would be a breeze.

    I guess my logic says if all my voltages are good, something is creating a problem with current delivery. I just don't know where to go from here...
     
  4. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    Ordinarily if you said "Series 2, strange symptoms", or "Series 2, power supply problem", my immediate answer would be "capacitor disease" or capacitor plague as wikipedia calls it.

    In your case, however, you have the scorching that indicates an ability to supply too much current rather than an inability to supply enough, and you seem to have replaced both power supply and hard drive instead of trying original hard drive with new power supply, so I'm thinking original hard drive developed fault causing it to draw too much current, and since your original supply was able to supply it enough to burn the board, it's probably still in pretty good shape solid state devices wise, but who knows how badly the caps were stressed, so I'd recap it on GP.

    Just be sure to use Low ESR, 105 degree rated, same uF and same WV or next notch higher, maybe 2 notches, and of course, observe polarity.

    Check out the forums at badcaps.net to see what brands to avoid.
     
  5. punkin

    punkin New Member

    6
    0
    Dec 15, 2011
    Thanks Brother, I appreciate the insight! I'll recap it and go from there. Your logic makes sense, so I'm gonna roll with the advice!

    Much appreciated, results to follow...
     
  6. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    That's often the best kind.:)

    Which is not to say that mine always does.:D
     
  7. punkin

    punkin New Member

    6
    0
    Dec 15, 2011
    Ok, maybe this is another clue. I replaced the 2200 uf caps, and decided to stop and test again. By stroke of scientific genius (accident), I bumped the heat sink that Q1 Transistor is attached to, and was quite shocked. (literally) Throwing the meter on the heat sink and ground, I've got about 80v coming through that heat sink. It got my attention! Now I'm trying to think what is shorting to that heat sink? There is some corrosion on the back side of the board, but no obvious sign of anything shorting.
     
  8. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    80V AC or DC?

    Your schooling was recent enought to cover the basics of switching supplies, wasn't it?

    Maybe that heat sink is supposed to be electrically hot?

    Unplug and use the ohmmeter part of your meter to see what it's connected to.

    Is Q1 on the input side or the output side? That is, is it electrically part of the input side of the supply or the output side.

    (I'm not going to open up the one machine I have with what may or may not be a similiar supply, I've got too much stuff on the bench already)

    Have you checked the 4 DC output voltages with the drive and motherboard attached?
     
  9. punkin

    punkin New Member

    6
    0
    Dec 15, 2011
    Schooling? I barely passed Network Analysis, and then I realized EE wasn't for me, and changed my major...that was almost 20 years ago! Knowledge of switching power supplies? Magic smoke is still in / Magic smoke let out. That's about it for me. I know enough to be dangerous without killing myself.

    In all seriousness, I swapped the power supply with a craigslist one. Still curious about my power supply, the craigslist one did NOT have voltage on the heat sink. My old one did have--as previously stated, and yes, it was AC voltage. I don't think it was supposed too, as none of the other ones have had voltage present.

    I ran continuity tests from the heat sink, and it seemed to be connected to everything on that side of the ckt. board. I don't think that's right. I'd be surprised if the little bit of corrosion present could cause that.
     
  10. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    When you say that side of the board, do you mean the AC input side or the DC output side?

    Are you talking about the big heat sink that comes up and folds over 90 degrees above several capacitors, or a smaller one closer to where the AC cord plugs in?
     
  11. punkin

    punkin New Member

    6
    0
    Dec 15, 2011
    Not that heat sink.. there is the one you describe that has three transistors attached, and hovers over the main set of caps. Not that one.

    There is the other one that is more like a 90 degree "fence" and has one transistor attached, and one really large cap centered in the middle of the "fenced area" right by where the cord plugs in.
     
  12. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    Okay, I'm going by an S3 HD power supply that I happen to be able to look at at the moment, and on memory.

    That really large cap is the one with the wrap-around sticker with the output voltages and amperages listed, right?

    And Q1 is a TO-220 type package? 3 leads coming out the bottom of a plastic thing on a flat metal thing that extends above the plastic thing and has a mounting hole with a screw through it holding it to that heat sink?

    Is there an insulator between Q1 and the heat sink, and a plastic ring that the screw goes through that insulates it from the metal tab, or does it just bolt right onto the heat sink, so that there's 0 Ohms between the tab (and probably the middle leg) and the heat sink?
     
  13. Redoctobyr

    Redoctobyr Active Member

    313
    90
    Jun 21, 2008
    punkin, I just happened to notice your thread. I am in a similar situation. TCD24004A, with a failing power supply, it hangs at "Powering up". But the lights all still come on, the fan spins up, the video output works, etc.

    I installed the power supply from my other Series 2, a TCD540040, and now it will boot up.

    While doing some work previously on the machine with the failed power supply, I also got shocked by one of the heatsinks, and measured ~60V on it, as I recall (this was about 6 months ago). I think it was AC. I don't recall which heatsink it was, sorry. I can check if it's helpful to know.

    Its power supply still *works*, and it puts out around 5V and 12V to the hard drive (I think the 12V was a little low, maybe more like 11.5V?). I did try running the hard drive on a separate power supply, to see if the built-in supply could at least feed the rest of the TiVo. It still would not boot.

    Now I'm trying to get a different power supply, I may just try to buy a cheap complete Series 2 from Craigslist. A whole machine for ~$25 is more appealing than a power supply for $15 :)
     
  14. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    The 540 supply and the 240 supply have the same output rails on the same pins of the ribbon cable, but the 649 (dual tuner S2) supply doesn't have a +3.3V rail, so it's physically but not electrically interchangeable.

    Also the 540 supply might be a little strained trying to run a 240 that's has a second drive added.

    Look at the capacitors under the heat sink overhang on the DC output side of that 240 supply and see if at least one of them isn't just a little bit not perfectly flat.

    See this thread

    http://www.tivocommunity.com/tivo-vb/showthread.php?t=479176

    for steve614's excellent picture of how subtle the difference can be.

    Wouldn't you rather have a power supply where the problem capacitor has already gone bad and been replaced with a new good quality one, than another used supply where another one of those caps from the same manufacturer as the one that went bad is sitting there waiting to go bad without warning?
     
  15. Redoctobyr

    Redoctobyr Active Member

    313
    90
    Jun 21, 2008
    unitron, I am only running a single disk in the 240, so that should hopefully make the load more manageable for the 540 power supply.

    Absolutely, if I could fix mine, that would be cool, and doing it for maybe $2 of parts at Radio Shack would be nice too. I checked the picture in the thread you linked do, that was a great illustration.

    I just grabbed a good flashlight and looked over all the capacitors, especially those under the heatsink overhang. I could not see a bulge on any of them, unfortunately. When viewed from the side, each one has a nice flat top. Did a bit of reading on "capacitor plague", and I also can't see any visible stuff coming from the bottom of them. That said, most of them have some white goop/glue spread around on them, which helps obscure the bottoms of the ones that are clustered together. So it's tougher to check for leaks. But nothing jumped out at me, certainly.

    There is a resistor that looks kind of brown, and the white goop stuff around it is also brown, pic attached. But I think that's just from it being warm. I measured it at 99 ohms, and from looking at the color bands, and making my best guess as to what each color is supposed to be, I think it's a 100 ohm resistor.

    Edit: punkin, my apologies, I did not mean to try and derail/hijack your thread.
     

    Attached Files:

  16. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    Not to be speaking for punkin (or anyone else), but I think of it as more of a contribution than a hijack. Any thread that gets long enough is going to fork a little.

    You and punkin are going to make me open up my 240 and pull the power supply for reference, aren't you?

    Good picture, by the way.

    I'm wondering if the discoloration on the resistor is either from the capacitor next to it, or maybe the enamel insulation on the otherwise bare wire with which that coil is wound has partially vaporized and deposited on it, which might account for the difference in appearance between the left and right hand sides of it in that picture.

    That could mean that the coil has been getting hotter than it should.

    You don't have the fan in backwards, do you? Blowing into the case instead of out?

    One end of that coil should connect to either the red (+5) or yellow(+12) wire and the other should trace back to a leg of one of those capacitors probably.

    See which capacitor(s) has(have) its(their) positive lead connected to one end of the coil, probably the same end that connects to the red or yellow wire. That, or those, is/are the one(s) to suspect, most likely.

    You can carefully bust off the white goop, it's just glue used to hold stuff together during soldering and shipping.

    Look for signs of scorching.

    Radio Shack, if the one near me is anything to go by, isn't going to have the right quality capacitors at the right values.

    Where are you guys geographically?
     
  17. steve614

    steve614 what ru lookin at?

    10,722
    0
    May 1, 2006
    Dallas, TX
    That resistor definitely looks stressed from internal heat.
    Aside from the scorch marks, you can also see that the outer covering is bubbled up and cracked in some places.
    If the heatsink is energized that could explain it. There could be a short somewhere on the board that causes more current to flow through that resistor which results in the resistor getting hotter than it would under normal conditions.
     
  18. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    Actually, your idea makes more sense than mine.

    The coil didn't overheat, the resistor overheated and that's what discolored/crazed the varnish/enamel on the coil wire on that side only.

    Now I'm going to have to get my 240 off the top of the fridge and open it up and check the power supply to see if that smaller heat sink near the AC input is supposed to be "hot".

    Boy you guys are a lot of work.:)
     
  19. steve614

    steve614 what ru lookin at?

    10,722
    0
    May 1, 2006
    Dallas, TX
    The heatsink should not be "hot" (energized) under any condition. If it is, I would suspect a short somewhere on the board.
    If an exposed piece of conductive material is energized by design, that is just a lawsuit waiting to happen.
     
  20. unitron

    unitron Well-Known Member

    16,576
    41
    Apr 28, 2006
    semi-coastal NC
    Technically, it's not exposed, as you (the owner/user) aren't supposed to open the case.

    I just checked an HD supply, which has no mica insulator between that transistor (or whatever that semiconductor is) and the heat sink.

    With the meter set on 200V AC, I get 60V between that heat sink and chassis ground (which is supposed to be transformer insulated from the incoming AC line), I get 60V between the sink and the AC "hot" (the black wire attached to the smaller slot on the wall socket), I get 60V between the heat sink and what people call the neutral but what's really the "grounded conductor" in a 120V AC circuit, i.e., the white wire--known as "the identified wire", and I get 60V between the heat sink and the "grounding conductor--the bare or green wire.

    With the meter set on 200V DC I get negative 60 between the heat sink and all of those points.

    I suspect low current, high impedence capacitive coupling explains it, but I ain't gonna grab hold of it to test that theory.:)
     

Share This Page