On 8/22/2010 12:19 PM, Richard wrote: The motherboard has both composite and digital XRGB video outputs. I've been working with a composite monitor, as I don't have a compatible RGB model. Unfortunately, I don't have a scope or anything more advanced than a multimeter to troubleshoot with. - Mike

That depends on the multimeter. Some have capacitance ranges. I wouldn't be suprised if somebody had made a comined multimeter and ESR meter either... But I do question the statement that capacitors are likely to be the problem. I power analuge circuitry (PSUs, CRT monitors, etc) I would agree that electrolytics cna dry up, or short, or.. But in digital circutiry, the only capacitors you are liekly to find are low value ceramic ones (e.g. as a load for a crystal) which are very reliable, and sppply decoupling capacitors. If the latter fail open, then provided only 1 or 2 have gone you won't notice it, and if they fail short than either the cupply rail is pulled right down (easy to spot, although finding which capacitor has shorted is nore 'fun'), or the capacitor gives itself away with smoke signals (tantalum bead capacitors are good at this :-)). -tony

A common way to repair an Apple III on its days was to raise it some 1" and let it bang on the table. It usually works =)

I don't know the Apple /// at all, so I can only give very geenral tips.. The first thing I woudl is check all the supply rails at the mainboard with your voltmeter. I know you say you've swapped out the PSU, but excessivel load on on rail could be pulling it down. And do you know that the replacement PSU is good? Is there any way you can get a cheap logic probe? A 'scope or logic analyser would be great, but you can do a suprising amount with just a logic probe and it's a lot cheaper. If you can get a logic probe, I owuld look at the sync signal(s) on the RGB socket. At least see if there are there (which would indicate the video timing circuitry was doing something). And then see if there is any activity on the CPU addres and data pins, the address and data pins of the ROMs and RAMs, etc. -tony

OK, so the replacement PSU was good when it was installed. However, this does not mean there's not a fault on the mainboard that's pulling one of the rails out of tolerance. Given that you have a voltmeter, I can't see any reason not to check those supply rails on the mainboard. I find it curious... All the books I have on electrronic troubleshooting and repair emphasisis that the only wat to put something right is to make measurements and think about them before you change anything. And yet it appears that these days you replace parts in an attempt to fix the problem without knowing if said parts have anything to do with the fault. You knwo, if I ever meet the guy who decided that board-swappign was a good method of diagnosis and repair, along with the guy who decided that soldering (or even surface-mount soldering) was impossible in the field, I am... well lets just say it has something to do with an induction coil and his reproductive organs ;-) Is the Apple III schematic on-line anyewhere? I might be tempted to take a look and suggest other thigns to measure/check. -tony

My comments on board-swapping were _NOT_ directed at you. Old-timers here will know it's a perpetual rant of mine :-). Quite simply I can't see how making random changes to a device until it appers to work again is in any way actually putting it right. However, I have been caught so often by making assumptions when doing a repair that I have learnt (I hope) that it is best to assume nothing and check everything. Even if you know the PSU is good, it's best to check it when powering the computer. YOu may well think you know a switch is working perfectly, this did not stop be being led on a merry dance one afternoon when the paper-out microswtich in an HP printer clicked perfectly but was in fact stuck _closed_. Do you have a scheamtic of the machine? If so, is it on-line anywhere. I'll fidn a way to take a look at it and suggest more tests... -tony

It never hurts to be sure of something like this :-) OK, I managed to take a look at them... Assuming you can get a logic probe, my first test would be on the reset circuit -- say on the rest pion of the 6502. Since you mention the machine does something, I would be suprised if it was being held in the reset state, but I've been cuaght by that once (a leaky capacitor in the reset circuit..). So again, it's best to check. Then look at the 'timing' schematic (there's one with a name something like that anyhow). That seems to be where the video sync signals are produced, the CPU clock, etc. Make sure they are. Another word of warning (*becuase I've been 'bitten' by this). A logic probe will tell you if a sighnal is doing something, but in general it won't tell you what it's doing -- the relative timing with other signals, etc. I onve had a fault where a 16 bit shift register (made of 4 4-bit shift register chips) had only the first 4 bits changing. So I assumed the second chip in the chain was dead, and thus not responding to the input from the first chip, and not passing anything on to the next chip i nthe chain either. Well, I changed it, only to find no change in the fault. So I tested again, this time using a logic anaylser to find it was the _first_ chip in the chain that had died, the signals from it were completley mis-timed, Replaced htat (Acutally using the chip I'd taken out when doign the first 'repair'), and everything was fine. But anyway... If you're getting the main clock signals, I would then start looking at the 'video' schematic. You know the video output is missing, let's find out why... Of course it may be a fault elsewhere (CPU/ROM/RAM, for ecample) which is causing some enable signal not to be present, but that's what youneed to establish. -tony