(Topic ID: 175100)

System 1 power supply rebuild for learning

By polyacanthus

7 years ago



Topic Stats

  • 10 posts
  • 3 Pinsiders participating
  • Latest reply 7 years ago by G-P-E
  • No one calls this topic a favorite

You

Linked Games

No games have been linked to this topic.

    Topic Gallery

    View topic image gallery

    IMG_20161204_225537076 (resized).jpg
    #1 7 years ago

    Hey experts, I have here a System 1 power supply that I'd like to rebuild just for fun. The previous owner had replaced it with an aftermarket in an attempt to get the machine working. The machine is now working great with the new power supply but I wanted to fiddle with this one to see if I could get it going again.

    Before I looked anything up online I hastily reassembled it and plugged it in just to see where it was at. The previous owner had it disassembled in pieces when I got it. I didn't realize that I was missing the insulators, so it went poof! Oops

    So after actually reading a couple guides on these I replaced Q1, Q2, and LIC1 for obvious starters (with insulators!), CR3 and CR4 because they burned, and all the electrolytic caps. Now my low voltages (5, -12, 8, 4) are 5.58V, -11.92V, 8.72V, and 4.16V so the low voltage side seems pretty good. Are the 8 and 4 reference voltages close enough? Couldn't find how accurate those should be. I realize I can trim the 5V.

    Now for the high voltage side, I was getting like 90V I think where 60V should be. I noticed R14 and R17 were pretty hot and looked cooked so those should be replaced obviously. For the cost of shipping it was cheap enough to simply order EVERY component for the high voltage side. I started with the bridge rectifier diodes but thought maybe I should stop to learn something here. So my question-

    Is there a way to go about this systematically to find out where the issue lies or will I just cause more damage replacing some components and leaving others that are possibly bad?

    My thought as an amateur was to replace Q3, Q4, and the burned resistors and see where I was at, but I'm afraid if something else is goofed I'll just smoke the new transistors.

    Obviously I have a rudimentary knowledge of electronics especially passive components and am a skilled solderer of mostly analog stuff but when it comes to theory of transistor pairs and whatnot I start to get lost.

    Thanks!

    #3 7 years ago

    The 4 and 8V supplies are approximates -- you're close enough.
    Your 5v supply is a bit high but could be due to being unloaded. I shoot for a loaded supply of about 5.05, just a hair over 5V. That voltage is adjustable using R4. Be careful, if you go the wrong way and raise the voltage - you might trigger the SCR101 overvoltage crowbar circuit. This intentionally shorts your 5V supply to shut it down to prevent damage to CPU board. Only way to clear the crowbar is turn the power off, turn R4 other direction and then turn back on.

    R14 will run hot. You have a half watt resistor with 0.3 watts continuously across him (60% of rated value). Shouldn't run resistors with more than a 25% load on them. Just another one of many bad parts choices by Gottlieb. If ever replacing R14 - I would use a metal oxide resistor. Metal Oxide resistors can handle the heat a bit better than a carbon film type.

    R17 should not be running hot. He should have roughly 0.07 watts across him continuously so should be reasonably cool. But if the regulator is putting out 50% higher voltage then your R15, R16, R17 resistor set will be running ~50% hotter than normal due to the higher voltage. I would focus on the HV section only at this point (rest is working) - don't forget the zener diodes CR10 and CR11. For replacing the entire Q2, Q3, Q4, CR10, CR11 section - you are looking at a bucks worth of parts. For a bucks worth of parts - not really worth debugging it, just pull and replace. And make sure the resistors are proper values when active parts are removed.

    #4 7 years ago

    Cool, thanks for the advice. Yeah I ordered three of literally everything for the high voltage side for $13 shipped. I started a full rebuild but thought maybe I should try to only repair what was needed. It'll be easy to just replace everything though. I didn't have half watt resistors on hand so I ordered them as metal film, sounds like those will still be better than the original but metal oxide would've had the best properties for this application?

    After the rectifier diodes I pulled C7 as it's been mentioned as a failure point as well as Q3 and Q4 before posting. Of those Q4 seems dead.

    So let's say I didn't have all components handy or they were expensive. Would it be wise to replace the cooked resistors and Q4 then bench test?

    IMG_20161204_225537076 (resized).jpgIMG_20161204_225537076 (resized).jpg

    #5 7 years ago

    OK, I'm close now.

    I replaced Q3 and Q4 (Q2 already replaced), zeners CR10, CR11, and CR12, R14 and R17.

    Voltages are all good now except that I had to trim R16 almost all the way to get 60V. Then it slowly decreases. After about 5 minutes I had R16 maxed and 60V had fallen to about 55V. What would cause this?

    I got 95V across C6 so it's in the circuit somewhere. C9 and C10 weren't replaced because I accidentally ordered 0.1mfd instead of 1mfd but I removed both and tested extremely close to 1mfd. Could it be a funky resistor failing with heat? I left most of them in because they checked fine with the transistors and zeners removed.

    #6 7 years ago

    Okay now everything on the high side has been replaced except for the two tantalum caps and I'm still losing voltage slowly. Played 3 games and was down to 56V.

    Any idears?

    #7 7 years ago

    So if anyone ever has this issue and manages to find this thread, replacing the tantalum caps solved the dropping voltage issue. I realized the two caps I didn't replace were both related the trimmer circuit and I could just replace them with aluminum electrolytic caps I had from a radio shack grab bag. So now I have a solid 60v with the trimmer pot in the middle. Huzzah!

    #8 7 years ago

    One of them must have been shorted. Normally when these fail -- they don't fail gracefully and tend to put on a bit of a fireworks show.

    #9 7 years ago

    That's what I read, partly why I put them back in place originally. I just tested them again and neither are shorted and both show 1mfd on my DMM.

    There's no reason I need tantalums there, right? They must've used them originally to save space?

    #10 7 years ago

    Tantalums are known for low-ESR (low internal resistance), long life, decent tolerances.
    Downside - they're pricey and when they do fail, they tend to die rather violently.

    Due to the amount of room on that board, I doubt they used them to save space. And since they used a pricier cap, they must have thought they had a reason. These do keep electrical noise off the adjustment voltage but not sure why they just didn't use a simple ceramic cap. One reason may have been that affordable ceramics didn't go as high as 1uF 30 years ago.

    Reply

    Wanna join the discussion? Please sign in to reply to this topic.

    Hey there! Welcome to Pinside!

    Donate to Pinside

    Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


    This page was printed from https://pinside.com/pinball/forum/topic/system-1-power-supply-rebuild-for-learning and we tried optimising it for printing. Some page elements may have been deliberately hidden.

    Scan the QR code on the left to jump to the URL this document was printed from.