On Tue, 10 Jul 2012, Polymorph wrote:
That's because in one direction the diode is not conducting, but in the other direction it is, so in effect you see a short to ground (the other side of the transformer winding that feeds the diode is at ground). You always need to check for shorts by reversing the polarity of the meter, to see if it's a real short, or just a semiconductor junction conductiing in one direction.On 10/07/2012 8:15 PM, Polymorph wrote:On 10/07/2012 6:30 AM, Michael J. Mahon wrote:Good luck!OK, after testing I do appear to have a short somewhere. Both the -5V and -12V lines appear to be shorted - using the continuity tester, I get a continuous beep between these lines and ground. When testing the +5V and +12V lines, I get no beep at all (which is what I would expect for all lines). So I guess I'll need to trace back through the -5V and -12V lines and find whatever it is that is causing the short. I hope it's something fairly obvious, as I'm a software guy by day - and whilst I'm comfortable enough with a soldering iron and multimeter, its certainly not my area of expertise. ;-)OK... I don't think my previous testing was valid (which I'm sure someone will tell me why soon). If I test the +5 and +12V lines with the multimeter probes reversed I also get a continuous beep out of the continuity test.
I checked the voltages, both +5 and +12 were very low - about 0.8V. The -12V and -5V were about half of what they should be (i.e. -5V was measured at about -2.5V and 12V was measured at about -6.5V). So it sounds like the fault might actually on the positive side somewhere.Switching supplies are complicated. As pointed out, they often (it depends on the design) won't provide voltage, or the proper voltage, without sufficient load. I've bought surplus switching boards and they've had power resistors on the board to provide enough load, a weird method of making sure the supply works.
But at least in the cheaper ones, they rely on one feedback path. So usually they make sure the +5v line is at the proper voltage, depending on the windings and filter capacitors to keep the other voltages "right". The other voltages often don't matter so much (the +12 and -12 can usually be fairly vague, think of RS-232), while the +5v line does have to be within a small percentage or else the TTL ICs won't work.
So if the +5v line isn't working properly, the other ones can't be counted on as working properly until the +5v line is working right.
Michael > Am I right this time, or am I still barking up the wrong tree? >
-- Cheers, Mike Find my Apple II projects online at: http://apple2.sytes.net/