Image Map Image Map
Page 1 of 2 12 LastLast
Results 1 to 10 of 16

Thread: Replacement ideas for PB70 PSU that failed.

  1. #1

    Default Replacement ideas for PB70 PSU that failed.

    Okay the other issue that fell into my lap this morning was I was going through and testing the new CR2032 Battery job that I made with the 4 pin connector and a bit of solder. When I plugged in the the machine and hit the power button, nothing happened. So I checked to see if I had dislodged the connector or anything. Everything was connected. Tried again. No luck. So I took out the Power Supply and disassembled it. At first glance nothing looked suspect, but then I noticed something. There is a sticky residue on the top side of the PSU PCB that smells. It doesn't smell like what I recall electrolytic fluid smelling like, but the stain spreads from the far right side of the board, all of the way towards the left side, back end of the board. I guess that some of the electrolytic caps failed, but I know so little about PSUs that I am way too skittish to try it. So I thought I would ask for assistance. Not in repair, but replacement. Finding a legit replacement seems impossible, although I did see some on e-bay, but it was outside of my capable price range. So I thought getting one of those little PICO ATX PSU might work? Then get an ATX to AT adapter? I know based on the specs on the PSU that I need -5V and -12V, which means I have to get a specific cable, but aside from that, I should be good. So I know very little about Pico PSUs, but I know I need power for the 2 Floppy drives, one 5.25 drive, and one 3.5 drive and one IDE drive, so 3 power plugs 2 standard molex and one of the 3.5 inch berg style floppy power connectors? I apologize for all of the constant questions here, but I do appreciate the help. If anyone thinks they can fix the original PSU, please let me know. If it's within my capability financially, I would go that route, but looking at the stain on the board, it appears that the caps have obviously been leaking for a while; which is very strange as the PC worked last week when I used it. I am not surprised, though. This is the second Packard Bell 386 that used this style of PSU that has failed. So this power supply apparently can have its issues. Thank you again for your suggestions. I appreciate the assistance.

    Regards,

    Nathan

  2. #2

    Default

    Here are some pictures to showcase the discoloring I am speaking of - discolored pcb of psu 1.jpg - discolored pcb of psu 2.jpg - discolored pcb of psu 3.jpg

    The first picture is a basic picture to show the PSU from above to show the extent of the darkening discoloration of the PCB. The surface where the board is discolored is slightly tacky almost like I would expect when using flux, but all of the solder points are on the back of the board? The second picture is a zoomed in picture of an area of discoloration that appears to the naked eye to be significantly darker than the other areas. And the third picture shows an area of the PCB where there is no discoloration next to the discoloration. What is interesting is that it appears that the area of the board under the really big huge capacitors doesn't appear to be discolored at all, and it doesn't feel tacky to the touch. I don't know if there is any thing about this board that is repairable, but I know that if it is, the repair is well beyond my ability at this current point in time. But hopefully the pictures will show what I mean about the board being discolored. I hope it shows up in the cropped and shrunk pictures.

  3. #3
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    35,356
    Blog Entries
    18

    Default

    Do you know where this PC was used? In a kitchen area, perhaps? What you're describing sounds more like gunk in the air settling on the boards, than something from a component failure. I'm using my own eyes here, so take this with a grain of salt...

  4. #4
    Join Date
    May 2006
    Location
    Melbourne, Australia
    Posts
    6,909

    Default

    Quote Originally Posted by Nathanieltolb331 View Post
    Okay the other issue that fell into my lap this morning was I was going through and testing the new CR2032 Battery job that I made with the 4 pin connector and a bit of solder. When I plugged in the the machine and hit the power button, nothing happened.
    Coincidental, or could it be that the "new CR2032 Battery job" is somehow overloading the power supply.

    How did you prove that the power supply is the problem cause?

    Quote Originally Posted by Nathanieltolb331 View Post
    So I took out the Power Supply and disassembled it. At first glance nothing looked suspect, but then I noticed something.
    I did too. Is photo #2 showing that one of the two blue RFI capacitors (bottom left) is ruptured? If so, it will be unrelated to the symptoms. Open circuit results in a reduction in RFI supression. Short circuit would result in a fuse blowing or circuit breaker tripping somewhere.

  5. #5

    Default

    @modem7 - The Blue RFI capacitor is not ruptured. It's a trick of light and shadow because the legs extend out from the circular top piece. @Chuck(G) - As far as I know. The description was that it was an office home computer that was used until the owner passed away. Then it was put in storage until sold a couple years later. When I put my nose close to the discoloration and sniff it, the smell that I can identify is similar to the smell that you get from new electronics when you turn them on for the first time basically. My skills with the multimeter are very limited and I am terrible at checking things like voltages and everything. I am trying to learn but I am splitting my time 8 different ways and unfortunately some things just end up getting put off, and right now that's learning more about electronics and soldering and diagnostics.

  6. #6

    Default

    Quote Originally Posted by modem7 View Post
    Coincidental, or could it be that the "new CR2032 Battery job" is somehow overloading the power supply.

    How did you prove that the power supply is the problem cause?
    Sorry for some reason I didn't answer this question. I'm just so scatterbrained right now. Well the thing is, I removed the Power Supply, and since it is an AT PSU I tried it not hooked up. It doesn't kick. So I tried it plugged into the motherboard, and it does nothing. I took a larger, known working PSU and plugged it into the 386 and it worked as expected. With and without the CR2032 battery connector attached to the external battery header. If I plug in the PSU with the power off and I put my meter probe in the different connectors while grounded to a RF shield I have nearby, I don't get any readings on the Power Supply. But based on my lack of how to use the meter, that could all be me.

  7. #7
    Join Date
    May 2006
    Location
    Melbourne, Australia
    Posts
    6,909

    Default

    Quote Originally Posted by Nathanieltolb331 View Post
    Well the thing is, I removed the Power Supply, and since it is an AT PSU I tried it not hooked up. It doesn't kick.
    Which could be because the power supply requires an adequate load in order to operate.

    Quote Originally Posted by Nathanieltolb331 View Post
    So I tried it plugged into the motherboard, and it does nothing.
    So, at this point, power supply still not proven as good or bad (i.e. could be an overloaded good power supply, or a bad power supply).

    Quote Originally Posted by Nathanieltolb331 View Post
    I took a larger, known working PSU and plugged it into the 386 and it worked as expected.
    Which points to the first power supply as being faulty. So, just a coincidence that it went faulty at the same time as you implemented the "new CR2032 Battery job".

    Quote Originally Posted by Nathanieltolb331 View Post
    If I plug in the PSU with the power off and I put my meter probe in the different connectors while grounded to a RF shield I have nearby, I don't get any readings on the Power Supply. But based on my lack of how to use the meter, that could all be me.
    This will be the first power supply that you are writing of. Because you have the power off, you must be doing resistance measurements. I see no point.

    You have proven that the first power supply is faulty. If you did want to verify that via multimeter, you would adequately load the power supply (just in case required), then do voltage measurements (DC voltages, and if applicable, POWER GOOD).

  8. #8

    Default

    Okay, thank you for the information. I was just wondering, is it possible that the Switch is faulty? Is there an easy way to check? Also, something I forgot to include in the pictures is that there is a portion of the board attached to the roof of the PSU case that has the fuse and two other components.
    case top components.jpg
    The fuse looks okay I think, but the white rectangular component appears to have water damage and smells very much of electrical smell. I know that doesn't signify anything but I thought I would share.
    possible water dmg.jpg

    Is there any where that I may be able to read up on the different electrical components that are commonly used? I know simple things like capacitors and resistors, but I know next to nothing about so many other parts. I am familiar with names like npn and pnp transistors, and I think I have a rough idea of what they look like; 3 legs, one collector one emitter and a ground? But I really want to know more. I feel like I could do so much more if I know what the different components are and what purpose they serve. Thank you again for the assistance.

    -Edit- I was looking at the label on the power supply and realized that Packard Bell had just put a sticker with their model information over the top of the original manufacturer. I pulled it back and I can see that the original manufacturer of the power supply was Tatung and the model is called TSS-70. I don't know if that helps in any way, but I thought I would share since I thought the manufacturer was DVE.
    Last edited by Nathanieltolb331; September 21st, 2020 at 08:58 AM.

  9. #9
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    35,356
    Blog Entries
    18

    Default

    Component markings are typically solvent-based, so not subject to water. This looks more like it was hit by oil or alcohol.

    I suspect that there's a lot you don't know about this PSU's history.

    At any rate, if all you need is a +/-12, +5 PSU board in a small form factor, that should be fairly easy to find. You'll have to supply connectors, though.

  10. #10
    Join Date
    Apr 2015
    Location
    Austin, Texas
    Posts
    2,221

    Default

    Quote Originally Posted by Nathanieltolb331 View Post
    Okay, thank you for the information. I was just wondering, is it possible that the Switch is faulty? Is there an easy way to check? Also, something I forgot to include in the pictures is that there is a portion of the board attached to the roof of the PSU case that has the fuse and two other components.
    Switches can most definitely go bad, especially high voltage and/or high current switches. Over time the contacts inside the switch can get pitted and burned from arcing, which eventually causes so much buildup that the contacts no longer make a good connection. When the contacts start having a resistance to them, they heat up and the heat causes more resistance and more heat. Eventually the contacts can get hot enough to burn and have molecular changes to the metal that makes it stay bad permanently. Mechanical relays have the same issues.

    You can check if a switch is bad by measuring the resistance between the terminals while turning the switch on and off a few times. If you get more than a couple of ohms, then the switch is bad. Likewise, if the switch shows no continuity, or intermittent continuity it is also bad.

    Quote Originally Posted by Nathanieltolb331 View Post
    Is there any where that I may be able to read up on the different electrical components that are commonly used? I know simple things like capacitors and resistors, but I know next to nothing about so many other parts. I am familiar with names like npn and pnp transistors, and I think I have a rough idea of what they look like; 3 legs, one collector one emitter and a ground? But I really want to know more. I feel like I could do so much more if I know what the different components are and what purpose they serve. Thank you again for the assistance.
    Lol, there is no "ground" on a transistor. A transistor is basically a solid state switch. On an NPN transistor, the emitter is the pin the current is sunk into, the collector is the pin that current is sourced from and the base is the pin that controls the flow of current.

    https://technologystudent.com/elec1/transis1.htm

    There are plenty of websites, and even videos on youtube of electronics and how they work.

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
  •