Path: utzoo!utgpu!watmath!xenitec!zswamp!p0.f171.n221.z1.fidonet.org!Geoffrey.Welsh From: Geoffrey.Welsh@p0.f171.n221.z1.fidonet.org (Geoffrey Welsh) Newsgroups: comp.sys.cbm Subject: C64 power supply Message-ID: <999.251E4D0F@zswamp.fidonet.org> Date: 25 Sep 89 14:47:26 GMT Sender: ufgate@zswamp.fidonet.org (newsout1.26) Organization: FidoNet node 1:221/171.0 - Izot's Swamp, Kitchener ON Lines: 73 > From: scott@max.u.washington.edu > Message-ID: <7296@max.u.washington.edu> > My power supply for the C64 has broken down, and I am thinking of > repairing it myself. I was able to find all the necessary parts > to try to fix it except one, the tranformer. So I have two questions > in regards to the transformer. I hope someone will be able to > answer them. > > 1. How is it likely that the transformer is the culprit in the break down? > 2. Where can I buy a transformer that has the same specification as that > of the one used in the C64 power supply? 1. Negligible. 2. As far as I know, Commodore uses a custom-wound one (they're cheap in sufficient volume). Chances are that the failure is in the voltage regulator. This may or may not help you, depending on how old your power supply is. The first ones (in the larger cases, with vents in them) can be opened, but the newer ones (smaller, solid "blocks" with plastic fins) are poured full of epoxy, so that you'd destroy the innards while trying to get at the parts. If you have an older one and the voltage regulator is at fault (i.e. there's at least 8V going in and anything but 5V coming out), then you could replace it with part number LM323 (I'm sure there are others, but that's the one I have ued successfully). Beware of pinouts; even if the case looks the same, the in, out, and ground pins may be arranged differently! If you want to build a new supply from scratch (using cables salvaged from the old one), that should be easy using two transformers. The 10V power line requires only 1A typically, so you won't have a hard time finding a part for that. Supplying the voltage regulator is a bit more tricky. You want the input to be above 7.5V for the regulator to work, but every extra bit over that means more power dissipation in the form of heat. Buying an 8V transformer and putting it through a bridge rectifier won't work, since the voltage drop accross two diodes is enough to drop 8V below the regulator's operating input level. On the other hand, a rectifying transformer (in the form of a centre-tap transformer with one diode) has a lower voltage drop... perhaps someone with a bit more experience than I can tell you what values to look for, since my solution was brute force (expensive heavy-duty parts, BIG heat sink - but I got a power supply that ran two and even three computers at a time as other supplies died out!) NOTES: It wouldn't be a bad idea to put a 5.2V Zener diode between the 5V output and ground. Put a fuse "inside" the loop created if the Zener cuts in, so that the fuse'll go if the regulator ever dies & cranks full voltage at the C64. Don't fall to the temptation to buy one hi-current 10V transformer to feed both the 10V AC and 5VDC lines; the C64 internals do not expect the AC lines to have any voltage or signal relative to GND or +5. The effects of not observing this expected isolation range from mild (VIC chip displays off patterns on screen, even when C64 power is off; fuse inside C64 blows) to disasterous (blow most major LSI chips int he C64). Put a large filter capacitor on the INPUT side of the regulator (1,000 microFarads per amp of expected current - I used a 4700 uF myself) and a small capacitor (perhaps the exxperts can suggest an appropriate value?) on the output side. This will give a steady voltage to the regulator and help reduce noise amplification by the regulator, respectively. Hope I've been helpful, informative, or entertaining. -- Geoffrey Welsh - via FidoNet node 1:221/171 UUCP: {{uunet!}watmath!xenitec!}zswamp!171.0!Geoffrey.Welsh ARPA: Geoffrey.Welsh@p0.f171.n221.z1.fidonet.org