Re: Modifications to 577 Curve Tracer Main Power Supply


Chuck Harris <cfharris@...>
 

Hi Gary,

I was tired when I read your post, and somehow took
"main power supply" to be "main transformer"... 12 hours
in planes, and airports might have had something to do
with it.

Anyway, one of the least reliable of semiconductor parts
I find are Darlington power transistors. They are hard
to test... suggesting a curve tracer, and often they will
have the output transistor develop a B-E short, and expose
the input transistor to the output terminals... This makes
them appear as a good transistor when tested with an ohmmeter,
or simple transistor tester, but a transistor having way too
low gain to do the job.

Otherwise, this circuit is trivially simple. You should
be able to troubleshoot it by making a couple of simple
measurements.

Note, the two 6.8V tantalum capacitors are most likely
underrated. They are being used in a way that commonly fails
dead short circuit. You said they were good, so maybe not
here.

Before you do anything, measure the +40V and -40V unregulated
supplies to ground. Don't get hung up on the voltages, they
must be at least 35 or so volts, and could be much more. They
should have around a volt of ripple...

Next, we must get the -30V supply to work. It is absolutely
needed for the +30V supply to work.

The grayed out components form a snubber network that is
there to slow Q772 down. If it is missing, but is now
necessary, you will find Q772 is producing AC.... your scope
is your friend, measure the collector and look for a several
volt oscillation. I truly doubt this is the problem... unless
you have substituted Q772.

Because the voltage is way too low, the entire regulator should
be working maximally hard to pull the -30V output all the way
to the -40V unregulated supply.

Start with Q788. It must be fully turned on. That means its
VCE should be less than 2V, and its VBE should be more than
1.5V. If not, it is either bad, or being driven incorrectly.

Next, look at Q786. It must be fully turned off. That means
its VCE should be the same as the -40V unregulated supply. If
not, it is either bad, or being driven incorrectly.

Next, look at Q772. It must be fully turned off. That means
its VCE should be about at the -40V unregulated supply...
probably somewhere between zero less, and the zener voltage
less... If not, it is either bad, or being driven incorrectly.

My money is on the Darlington power transistor. I don't like
them, they are mean.

That should get you started.

-Chuck Harris

Gary Robert Bosworth wrote:

Chuck: I did not find anything wrong with my transformers, thank God.
Transformers can be a real headache in many ways. I have seen transformers
that had their core characteristics changed drastically from excessive
heating. I have also seen shorted turns, and if only 1 or 2 turns are
shorted you usually cannot measure the resistance change with an ohmmeter.
High frequency switching transformers develop their own types of problems,
what with capacitive couplings and lossy shielding.

Gary

On Tue, Mar 3, 2020 at 8:30 PM Chuck Harris <cfharris@erols.com> wrote:

Have you considered that trimmer pots that are 30 years old often
misbehave?

Anyway, I would like to hear why you are so certain the transformer
is so very awfully terribly bad.

-Chuck Harris

Gary Robert Bosworth wrote:
Dennis: The parts in faded print are R773 and C773. They are not on the
circuit board.
The circuit design is not well described in the service manual. No
normal
voltages and currents are mentioned in the print.
The +40V and -40V unregulated are fine.
ALL of the components measured perfectly good including the rectifier
bridges and filter capacitors.
The power supply behaves the same whether the loading connectors are
attached or not.
All of the other supply voltages are good including +200V, +12V, -12V,
and
+5V.
I momentarily disconnected the short-circuit feedback diode CR785 when
the
loads were not attached and it had no effect.
The reason why I do not like this design is because of the fact that the
-30V output is used to bias the 6.2V reference diode VR772. This poses a
classic lock-up condition in that if the -30V does not exist, then the
reference diode is never turned on.
I have measured every circuit trace from beginning to end and there are
no
micro-cracks to open them up, and there are no solder bridges causing
short
circuits.
This is a perplexing problem that I have not seen in 56 years of working
in
electronics.

Gary

On Sun, Mar 1, 2020 at 11:53 AM Dennis Tillman W7PF <dennis@ridesoft.com

wrote:

Hi Gary,
Which parts on which schematics are shown in shaded text?

Dennis Tillman W7pF
TekScopes Moderator





Join TekScopes@groups.io to automatically receive all group messages.