Reforming caps and CRTs

[Tony Duell]

> 
> Some of the smoke came out of my Osborne 1 this weekend.  It doesn't
> appear to have been fatal, as it was still functioning when I cut the
> power.  I'm giving the CRT some time to fully discharge before I open
> it up, probably next weekend.  It was probably a failing capacitor,

I don't know why people are so worried about the residual charge on a 
CRT. Unless you're working on the CRT or flyback transformer, there's no 
way you're going to come into contact with it.

And I am not sure what good waiting a week does. If there's a bleeder 
resistor, it'll discharge in minutes. Similarly on a _working_ device, 
the beam current will often discharge the CRT at power-down (I find it 
rare to find much residual voltage on a CRT final anode connector). If 
neitehr of these is the case, then a week may well not be long enough.

> although it didn't explode, it merely got really warm over a period of
> hours.  But since there are some similar machines that I haven't
> recently used, this question came to mind.
> 
> I've used the variac technique to reform capacitors, but thus far
> haven't done this on a machine that contains a CRT because I don't
> know what the reduced voltage is going to do to the CRT.  The next

The CRT is the least of your worries.

I am not sure what good applying a Variac to the mains input of a machine 
using regulated supplies actually does. It's noy going to do much 
reforming of capacitors deep in the circuitry. And if the PSU is a 
switcher (and a lot are), it can do damage (switchers are constant 
_power_ loads to a very good approximation, they draw more current at 
lower voltage).

if you want to reform the capacitors, IMHO you need to remove them and 
run them up on a current-limited DC supply. 

Even then I am not sure what the point is. I've found capacitors, 
particularly tantalum bead types, just fail at random. I don;t think 
reforming them would help.

-tony