After 22 years of service, my scope is dead :(

Discussion in 'General Electronics Chat' started by tom66, Sep 9, 2010.

  1. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    HP 54501A, digitising oscilloscope, 100 MHz / 10 MS/s. Built in USA, 1988, by Hewlett Packard.

    It's about 22 and a bit years old, and today I power it on to find to find it is power cycling. Turn it on and the fan just whirrs, the LED blinks and the screen flickers on and off. :(

    I suspect that the power supply is failing. Likely the electrolytic capacitors are starting to go. I could fix this myself, if I had the darn screwdriver! But HP insisted on using some kind of proprietary screw. I'll have a look at finding a screwdriver. Even when it finally turns on, it refuses to measure anything. All signals are maxed out on the display, even at 5V/div. And it resets every 10 seconds or so. :(

    I did obvious checks to see if the power cord was loose, and it was not. I changed it around to make certain. I'm also running a computer from the same 4-bar so I think the power to the scope is OK. I think as the scope warms up it runs for longer (because the caps warm up and start working), but not much longer. Could it be that the mains filter cap is faulty, or is this unlikely? I suspect it's an SMPS power supply, but there are probably linear regulators for the CPU and voltage references.

    Initial signs of failure were that it took about 1 minute for the voltage references to stabilise, that is, it would be 200mV out and settle down to 0 after about 1 minute, perhaps this was an early failure warning?

    Any tips? Has anyone ever tried repairing a scope power supply?

    I bought this for £150, and although I have the money to buy a new one, I'd rather not, I was saving up for some things. :(
     
    Last edited by a moderator: Sep 9, 2010
  2. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Keep it powered off until you have replaced all of the electrolytic caps.

    Then troubleshoot to find out what got burned up when one or more of the caps failed. It may not be all that hard to find.

    Why don't you post an image of the screw heads that you need a special tool for?
     
  4. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    Serial number prefixed with 3251A, apparently service manual doesn't apply, but I'll try anyway. Most likely, it will just say to replace the power supply.
     
  5. Markd77

    Senior Member

    Sep 7, 2009
    2,803
    594
    Those special screwdriver bits tend to come in packs which have loads of different tamperproof bits. I think I have seen them in Maplin, but places like Homebase or Halfords might also have them.
     
  6. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    Here is the picture of the screw head, with a DIP8 package for comparison. It's probably not a proprietary screw, but instead designed to be tamper-proof. The calibration was last done valid to 2006, but I'm not terribly worried about voiding the calibration. It's pretty difficult to get my point 'n' shoot to take good macro photos, this was the best of the bunch.

    SgtWookie, I think the mainboard caps are probably OK. I think the power supply is faulty. The reason I think this is because the fan, CRT and power LED all flicker on and off; and the CRT is, according to the service manual, powered from a separate supply, as is the fan. Not sure about the LED but wouldn't be surprised if it's just connected to +5V with a resistor or such like. (It's the only indicator on the scope.)

    The oscilloscope says 365W maximum input, could this be because there is a linear supply inside it?

    Edit: I just turned it on and it worked for about 30 seconds, the display showed the screen, but the scope trace was maxed out, then it reset. Hmm.
     
  7. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    The screw is a torx type.
    You can buy bit sets with those in them.

    Bertus
     
  8. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    http://en.wikipedia.org/wiki/Torx
    I am quite sure the name of that screw type is Torx. I have seen it on a Agilent scopes quite often. The screw is quite common in Europa, on technical equipment. And most iron ware shops have screwdriver sets. You may get them in any price range from cheap to expensive. I have cheap set at home, that I have used for many years
     
  9. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    Shows what I know about screws. Thanks for the tip, looks like I'm going down to Maplin tomorrow. I'm going to order some quality Nichicon or Panasonic caps to replace the ones in the unit at the moment, after I find the screwdriver. According to the service manual, the power supply is in a separate unit (it's in a case) and it looks small enough to be switch mode. I'll see if the caps could do with replacement.
     
  10. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
  11. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    Already reading.

    OK, this is weird. The scope is no longer power cycling. It is stable and runs normally. However, with nothing attached, it is clipping to maximum (suggests Vref has gone; could be power connector that is loose.) Sometimes, though, it will show a weird display, like 1V, I'm hoping it will recover but somehow I don't think it will, I will have to repair it myself. With Vref all over the place, scope fails selftest (A/D converter.)

    I wonder if this fault explains the weird TI regulator bug? The oscillations look very similar, maybe the TI board was fine but the scope was faulty!
     
    Last edited: Sep 9, 2010
  12. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    My favorite car behave like that sometimes. And it is about that age. It is very impractical to not be able to close the sunroof because it is electrical operated, and the control unit has decide to take some time off. At the time it starts to rain. Most of the errors I have tracked down to bad connectors. The female connector part is simply not able "grip" the male part of the connector any more. Anyway it very important to take care of this kind of error very quickly. Because connections coming and going may cause damage to the electronics as it is in most cases not designed for such stress.
     
    Last edited: Sep 9, 2010
  13. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    I got it working, but it is off by about 900mV, and it is all over the place. It's not power cycling any more, but I don't know how long it will last.
     
  14. Markd77

    Senior Member

    Sep 7, 2009
    2,803
    594
    Stop turning it on! It won't fix itself and it might do more damage.
     
  15. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
    If the caps are shorted, I think the power supply will be damaged already.
     
  16. DumboFixer

    Active Member

    Feb 10, 2009
    219
    34
    You should be able to get screwdriver bits for that from ScrewFix.
     
  17. mbohuntr

    Active Member

    Apr 6, 2009
    413
    32
    Save yourself a little time, take a light and shine it down the screw hole. It might have a "nub" sticking up at the bottom. If so, you need a security torx set.
     
  18. tom66

    Thread Starter Senior Member

    May 9, 2009
    2,613
    214
  19. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Power supplies can take other things with them. It is worth fixing the power supply and keeping the unit off if you can. Easier said than done.

    I remember working on a pattern generator for IC templates (chrome on glass). It had a blown power supply, took me over a week to find all the bad TTL chips and replace them. I just moved down the chain from one card to another.

    Then I found the eproms (which had the x/y coordinates for a circle in binary for stepping motor controls) were corrupted. I had to reconstruct the data table and burn some new ones, which wasn't too bad since this was routine for the factory floor and the R&D dept. It was a clean room piece of equipment that wasn't in a clean room. Now that was a major pain (fluff is almost impossible to remove).
     
  20. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    It's pretty difficult to tell what size the Torx head screw is from the photo, even though you have a mangled DIP next to it.

    I bought an inexpensive computer maintenance toolkit about 15 years ago, it came with a Torx screwdriver, as personal computers sometimes use Torx-head screws.

    I don't understand why you keep turning it on when I suggested that would cause more damage? It's not going to fix itself. You're just making the situation worse by cycling power on and off.
     
Loading...