[R-390] Advice on Replacing Components
2002tii
bmw2002tii at nerdshack.com
Tue Jan 11 13:35:40 EST 2011
Jerry wrote:
>When you have a suspected leaky mica or ceramic cap or an out of spec
>resistor - do you always remove it or do you clip one end and test it? To
>remove all suspected components seems like a major time consuming task. I
>have heard many clip one end to test especially in tight places. My question
>focuses on the techniques used to resolder If the component turns out OK.
>Do you juxtaposition the cut ends and just put a glob of solder? Most
>passive components don't have sufficient extra leads so I am very interested
>in learning from the wisdom of the group.
The first thing to do is learn to really troubleshoot problems so you
have a darn good idea of what is bad before you start chopping.
Particularly with tube equipment (where you don't have to worry about
semiconductor junctions screwing up your resistance measurements),
you should be able relatively easily to figure the nominal
point-to-point resistances and voltages. The hardest thing to get
people to learn as troubleshooters is to use all of the information
they collect -- they measure all sorts of things, then quickly tire
of the deductive reasoning that is required to understand what the
measurements mean and instead resort to random replacing or snipping.
Don't be that guy (or gal). Figure out what you need to know, then
figure out how to deduce it from measurements that you can make
without removing components (except tubes, which are easy to
remove). Very often, an experienced troubleshooter can narrow a
problem to one or two components using information that a less
experienced troubleshooter has already collected but had not fully
processed mentally.
If everything is nominal, the equipment works. Find out what is not
nominal, then figure out what it means. (Where is the current
going? Why is that voltage too high/low? Why is there no/too much AC there?)
Also, note that many circuits operate fine with component values very
far away from their design values. Don't confuse yourself by chasing
what are essentially "blueprinting" tasks (resistor values off by
+100/-50%, for example) when you're troubleshooting. You're looking
for what is making the equipment not work. Fix that, THEN blueprint
if you want.
Best regards,
Don
More information about the R-390
mailing list