Over the years I've owned a number of Apple ][ computers including a ][ europlus and a number of //e's. More often than not, if the power supply hasn't failed already (due to age), then it does shortly after it is fired up. Generally I've found it to be the AC capacitors that fail - spectacularly with a fizzing sound and a bit of horrible smelling smoke. Generally most of the psu's I've come across have fired up fine once re-capped - these would have been psu's that worked fine when the unit was put into storage. After 25 years it's not surprising, and you may find many of the electrolytic caps are also starting to vent and leak either from the top (obvious), or from the bottom. So in this thread I'll go through the process of re-capping the psu, with lots of photos. WARNING : This power supply has very high voltages present when powered, and some of the large capacitors will hold significant voltage when the unit is off. Some of the heatsinks may be live when the PSU is operating - do not operate the PSU without the cover fitted. Here we have a typical Apple //e PSU - removed from the case : Under the cover we see the electrolytics and AC caps that we will be replacing : So we remove the PCB from the bottom of the aluminium case, and start the process of removing all the capacitors that we will be replacing. Note the polarity of the electrolytic capacitors is clearly marked on the PCB: The old caps : Failure quite visible on this AC cap : Next we have to source suitable replacements. The caps should be low ESR, high temperature caps, and of a good quality. I've been buying the Hitano capacitors from Worldwide Electronics in WA. I put together this spreadsheet with old caps versus suitable replacements : One thing to be aware of is that the modern day replacements are far smaller in physical size that the old ones. This can cause complications with the lead (leg) spacing required to fit the new ones to the original PCB. Apple provided a couple of alternative lead spacings on the PCB for the AC caps and some of the electro's, but for the large electroytics it is neccessary to bend the legs out such that the lead spacing is correct for the PCB : New caps fitted : Board back in the case : Job Done! My Apple ][ Thread here
nice guide i wish i understood how PSU worked and why capacitors are so fundamental to their operation!
One way to get around the smaller size is to use caps with higher voltage ratings. That way the fit straight in and you have more overhead for longer life. Also, did you glue the caps in at all? Might be worth dabbing a bit of hot melt glue or silicone around to prevent stress fractures on the legs of the caps.
AC power goes up and down in a sine wave, when you try to straighten that out to low voltage DC you still have bumps and an on-off pattern. Capacitors store up some of the power when the voltage is high, and let it go when the voltage is low, thus buffering/smoothing out the voltage. That's also why LEDs on monitors etc. stay on for a few seconds after pulling the power, and why it's dangerous to work on things like PSUs and TVs even when they're unplugged.
Very nice power supply love the old school design. Would be a little warm in there but surprisingly quite good for its age as the heatshrink on the electrolyics haven't pulled back that much. Thats what you get with real Nippon capacitors instead of Chinese crap
Great guide, cheers. The size difference in the new caps is always interesting to see, especially on the 16v 330.
Nice writeup. Those X2 supression caps really need replacing. I just did the same with a *couple* of my apple II and IIgs psus. Though I used Panasonic TS-UP, FC and AM series caps for the electrolytics (I have a small vintage audio gear business so I get them regularly for customer jobs). photo-14 by , on Flickr Just like a nice stack of pancakes. Apple II power supply pancakes!
It works! A quick thank you to aXLe. I had a problem with a IIe I bought on eBay. Looking at the voltage outputs the 5V lines were low, and the 12V ones non existant. Then suddenly as I was probing they all came alive! But alas trying again later I got the same results. So there was something wrong. Looking for the schematics I couldn't find the exact one. Only a US model which has some differencies to this. Probing around again I could see that essentially there was signals everywhere, so nothing seemed failed that I could see. Or certainly not completely failed. Making it one of the worst types of failure to find. Then I found this forum post. I was 50/50 on if this would work, but considered it would rule out some of the doubt over those components. Then it would give me a clearer indictaion that the problem was elsewhere. However, after changing all the caps as directed above everything started working again. The feedback waveforms that I monitored before were now about a fifth the frequency they were. So the caps certainly made a difference. I did find that a company is selling replacement PSUs on eBay: http://www.ebay.co.uk/itm/321027370466?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649 Being sold by the company: http://www.atlaz.com/ Thanks again!
Thanks for the info on the parts listing. Just got myself an apple 2e and on first boot the power supply fried. Just followed your parts list in the spreadsheet and ordered from the same place, hopefully I should have the parts soon and I'll see what happens from there. Cheers
Finally had a chance to recap the power supply. good news is that it fires up but the bad is that the voltages are down so not sure where to go from here. +5 -4.7 +11.5 -10.93
Those are with no load. I will test under load and see what I get. It could be okay but the computer loads to a screen with thick horizontal lines and no beep.
I just finished testing under load and psu looks good. provided an image of what is displayed on the screen on startup. -5.14 +5.00 -12.08 +12.00
Hmm, what trouble shooting tools do you have available (multimeter, scope, logic analyser, etc?). What cards are in the machine? I've seen a similar issue that turned out to be be the 74LS245N at position B2 - worth swapping to test but may be soldered in? Perhaps take all cards out of the machine and re-test. Then have a good look at the EF rom and the CPU - it's worth re-seating them. While at it, have a good close look at all the socketed chips to see whether there appears to be any corrosion. Somewhere to start
Hi aXLe, All I’ve got is a multimeter. I have already taken out all the cards but same result. Just waiting on a chip extractor to re-seat the cpu and roms as I don’t want to do any more damage. I have also purchased these chips SN74LS244, SN74LS374 and SN74LS245N. I’ll let you know how it goes. Cheers Rob
