Topics

refurbishing TM500 and TM5000 hardware


Richard Erlacher
 

It pains me to say I've not used this TM500 and TM5000 hardware for quite some time and I fear that if I try to power it up, it'll just go poof, mainly due to the electrolytic caps in the power cicrcuits. I have a basic idea about how to reform the electrolytics that aren't deformed, and they shouldn't be, since no power has been applied for quite some time. I'm interested in any recommendations regarding bringing this stuff back to service life.

I have a variac, and constant-voltage AC power supply available, and various scopes that have been more recently powered up. I also have a wide range of passives and rectifiers, etc.

Any suggestions?

Uli


Richard R. Pope
 

To whom it concerns,
The first thing that I would do would be to replace all of the caps that show any signs of being deformed!
GOD Bless and Thanks,
rich!

On 10/27/2020 12:23 PM, EDick@... wrote:
It pains me to say I've not used this TM500 and TM5000 hardware for quite some time and I fear that if I try to power it up, it'll just go poof, mainly due to the electrolytic caps in the power cicrcuits. I have a basic idea about how to reform the electrolytics that aren't deformed, and they shouldn't be, since no power has been applied for quite some time. I'm interested in any recommendations regarding bringing this stuff back to service life.

I have a variac, and constant-voltage AC power supply available, and various scopes that have been more recently powered up. I also have a wide range of passives and rectifiers, etc.

Any suggestions?

Uli





Tom Lee
 

You'll find lots of recommendations advocating the use of a variac in (re)forming electrolytics. I disagree with almost all of them; it's just too crude a method, with too-loose criteria for success ("it didn't blow up, so it must be ok").

Some recommend reflexibly replacing all electrolytics, but I disagree with that, also, as the probability of damaging something in the process is nonzero. Plus, I'm a cheap, lazy bastard.

What I do for mains filter caps is to follow a reforming process that is modeled after what capacitor manufacturers themselves use: An electrolytic is considered healthy if the leakage current at the rated working voltage corresponds to no more than 1% charge loss per second. So, compute C*V, with C in farads and V in volts, then multiply by 0.01, to calculate the maximum allowable leakage current in amperes. Form with a current-limited supply, set to perhaps 10-100x the allowable leakage current, until the current drops below that value. If the current never goes that low, the capacitor might still pass the "it didn't blow up" test, but the dielectric is not healthy, so lifetime may be short. I've used this procedure for many years.

Others will chime in with their favorite procedure, I'm sure.

-- Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 10/27/2020 10:23, EDick@... wrote:
It pains me to say I've not used this TM500 and TM5000 hardware for quite some time and I fear that if I try to power it up, it'll just go poof, mainly due to the electrolytic caps in the power cicrcuits. I have a basic idea about how to reform the electrolytics that aren't deformed, and they shouldn't be, since no power has been applied for quite some time. I'm interested in any recommendations regarding bringing this stuff back to service life.

I have a variac, and constant-voltage AC power supply available, and various scopes that have been more recently powered up. I also have a wide range of passives and rectifiers, etc.

Any suggestions?

Uli




Tom Lee
 

reflexively, not reflexibly.

Aargh.

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 10/27/2020 20:09, Tom Lee wrote:
You'll find lots of recommendations advocating the use of a variac in (re)forming electrolytics. I disagree with almost all of them; it's just too crude a method, with too-loose criteria for success ("it didn't blow up, so it must be ok").

Some recommend reflexibly replacing all electrolytics, but I disagree with that, also, as the probability of damaging something in the process is nonzero. Plus, I'm a cheap, lazy bastard.

What I do for mains filter caps is to follow a reforming process that is modeled after what capacitor manufacturers themselves use: An electrolytic is considered healthy if the leakage current at the rated working voltage corresponds to no more than 1% charge loss per second. So, compute C*V, with C in farads and V in volts, then multiply by 0.01, to calculate the maximum allowable leakage current in amperes. Form with a current-limited supply, set to perhaps 10-100x the allowable leakage current, until the current drops below that value. If the current never goes that low, the capacitor might still pass the "it didn't blow up" test, but the dielectric is not healthy, so lifetime may be short. I've used this procedure for many years.

Others will chime in with their favorite procedure, I'm sure.

-- Cheers,
Tom


Chuck Harris
 

I just turn them on and see what happens. They are modern enough
capacitors that they will either be totally bad... eg. open circuit...
or will come up just fine.

If I am at all concerned, I will turn them on for a minute or two,
and then shut down and check for hot cans using my calibrated, though
somewhat scarred finger.

-Chuck Harris

EDick@... wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for quite some time and I fear that if I try to power it up, it'll just go poof, mainly due to the electrolytic caps in the power cicrcuits. I have a basic idea about how to reform the electrolytics that aren't deformed, and they shouldn't be, since no power has been applied for quite some time. I'm interested in any recommendations regarding bringing this stuff back to service life.

I have a variac, and constant-voltage AC power supply available, and various scopes that have been more recently powered up. I also have a wide range of passives and rectifiers, etc.

Any suggestions?

Uli






-
 

Regarding any electronic equipment that hasn't been used in a while, I
bring it up slowly using a variac and I've noticed that the odds of it
working has improved from about 50% to about 90%. But also I'm now working
with much older equipment than I did before I started doing this so the
results are probably better than they sound. I start it out at about 1 VAC
and watch the ammeter and as long as it doesn't appear to be drawing
excessive current, I leave it there for an hour or so, then take it up to
about 6 VAC, again watching the current. Leave it for a time, them take it
to about 24 VAC and repeat, then to about 50 VAC them to about 75 VAC. At
this point I usually start seeing signs of life. Then to about 90 VAC. At
this point most pieces of equipment start working even if not entirely
correctly. After a wait, then I take it to about 105 VAC. Finally up to 120
VAC. *Knock on wood!* So far I've never had anything let out the magic
smoke while I bring up the voltage on it. A few items still don't fully
work at this point due to internal problems but at least I haven't let out
any magic smoke yet and I've done this with perhaps 50 pieces of electronic
test equipment, some of it dating to the 1950s and a lot that I know
haven't been used in over 20 years.

I did this just last week on two OLD GR megohhmeters and two HP 428B
Milliamp meters made in 1969 and all four pieces work and while they're not
completely within spec, they work reasonably well. I tried it on an old
Fluke 873 Differential Voltmeter made in about 1964 and it did come back to
life but it's not measuring anything and the needle just flicks back and
forth. However I can see two caps in it that LONG ago leaked their
electrolyte, not to mention that when I found the meter it had been sitting
outside in a junkyard for so long that the handle had rotted off of it and
the paint had faded. I really didn't expect to work at all but it turned
out to be remarkably clean inside so I decided to get it a shot. All of the
thumbwheels in it had rusted to their shafts and I had to use PB Blaster to
free them up.

I mounted a 4 amp GR Variac in an old re-purposed ventilated rack mount
chassis and added two good quality Simpson 6 inch scale meters to it, one a
150 VAC or so voltmeter and the other a 6? Amp ampmeter so I can watch the
voltage and current closely. I would not trust a digital meter for this
application, they react too slowly. I added a line cord with an insulated
toggle switch and an internal FB fuse and coupled the variac output to
standard duplex AC outlet mounted on the back.

On Tue, Oct 27, 2020 at 10:42 PM <EDick@...> wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for quite
some time and I fear that if I try to power it up, it'll just go poof,
mainly due to the electrolytic caps in the power cicrcuits. I have a basic
idea about how to reform the electrolytics that aren't deformed, and they
shouldn't be, since no power has been applied for quite some time. I'm
interested in any recommendations regarding bringing this stuff back to
service life.

I have a variac, and constant-voltage AC power supply available, and
various scopes that have been more recently powered up. I also have a wide
range of passives and rectifiers, etc.

Any suggestions?

Uli






Tom Lee
 

I would say that the variac method is best used (if used at all) with vacuum-tube or otherwise classical analog gear. Modern equipment tends to react rather badly to slow ramps in voltage. Many switching supplies hate it to the point of committing suicide, and microprocessor-controlled equipment can similarly wake up on the wrong side of the bed. I either go with the "flip the switch and take my chances" method with an incandescent lamp in series with the ac line, or with the tedium of the "disconnect individual mains caps and reform them" method described in my previous post.

--Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 10/27/2020 21:33, - wrote:
Regarding any electronic equipment that hasn't been used in a while, I
bring it up slowly using a variac and I've noticed that the odds of it
working has improved from about 50% to about 90%. But also I'm now working
with much older equipment than I did before I started doing this so the
results are probably better than they sound. I start it out at about 1 VAC
and watch the ammeter and as long as it doesn't appear to be drawing
excessive current, I leave it there for an hour or so, then take it up to
about 6 VAC, again watching the current. Leave it for a time, them take it
to about 24 VAC and repeat, then to about 50 VAC them to about 75 VAC. At
this point I usually start seeing signs of life. Then to about 90 VAC. At
this point most pieces of equipment start working even if not entirely
correctly. After a wait, then I take it to about 105 VAC. Finally up to 120
VAC. *Knock on wood!* So far I've never had anything let out the magic
smoke while I bring up the voltage on it. A few items still don't fully
work at this point due to internal problems but at least I haven't let out
any magic smoke yet and I've done this with perhaps 50 pieces of electronic
test equipment, some of it dating to the 1950s and a lot that I know
haven't been used in over 20 years.

I did this just last week on two OLD GR megohhmeters and two HP 428B
Milliamp meters made in 1969 and all four pieces work and while they're not
completely within spec, they work reasonably well. I tried it on an old
Fluke 873 Differential Voltmeter made in about 1964 and it did come back to
life but it's not measuring anything and the needle just flicks back and
forth. However I can see two caps in it that LONG ago leaked their
electrolyte, not to mention that when I found the meter it had been sitting
outside in a junkyard for so long that the handle had rotted off of it and
the paint had faded. I really didn't expect to work at all but it turned
out to be remarkably clean inside so I decided to get it a shot. All of the
thumbwheels in it had rusted to their shafts and I had to use PB Blaster to
free them up.

I mounted a 4 amp GR Variac in an old re-purposed ventilated rack mount
chassis and added two good quality Simpson 6 inch scale meters to it, one a
150 VAC or so voltmeter and the other a 6? Amp ampmeter so I can watch the
voltage and current closely. I would not trust a digital meter for this
application, they react too slowly. I added a line cord with an insulated
toggle switch and an internal FB fuse and coupled the variac output to
standard duplex AC outlet mounted on the back.

On Tue, Oct 27, 2020 at 10:42 PM <EDick@...> wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for quite
some time and I fear that if I try to power it up, it'll just go poof,
mainly due to the electrolytic caps in the power cicrcuits. I have a basic
idea about how to reform the electrolytics that aren't deformed, and they
shouldn't be, since no power has been applied for quite some time. I'm
interested in any recommendations regarding bringing this stuff back to
service life.

I have a variac, and constant-voltage AC power supply available, and
various scopes that have been more recently powered up. I also have a wide
range of passives and rectifiers, etc.

Any suggestions?

Uli







Bill Perkins
 

If I am at all concerned, I will turn them on for a minute or two,
and then shut down and check for hot cans using my calibrated, though
somewhat scarred finger.
-Chuck Harris
Damn that's funny, I have one as well, a couple of 'em in fact...

Bill


Tom Gardner
 

How does equipment with an SMPS react?

SMPSs are designed to deliver constant output power. Hence if the input voltage is lower then, by design, the SMPS will draw a higher input current.

Semiconductors don't like too much current.

On 28/10/20 04:33, - wrote:
Regarding any electronic equipment that hasn't been used in a while, I
bring it up slowly using a variac and I've noticed that the odds of it
working has improved from about 50% to about 90%. But also I'm now working
with much older equipment than I did before I started doing this so the
results are probably better than they sound. I start it out at about 1 VAC
and watch the ammeter and as long as it doesn't appear to be drawing
excessive current, I leave it there for an hour or so, then take it up to
about 6 VAC, again watching the current. Leave it for a time, them take it
to about 24 VAC and repeat, then to about 50 VAC them to about 75 VAC. At
this point I usually start seeing signs of life. Then to about 90 VAC. At
this point most pieces of equipment start working even if not entirely
correctly. After a wait, then I take it to about 105 VAC. Finally up to 120
VAC. *Knock on wood!* So far I've never had anything let out the magic
smoke while I bring up the voltage on it. A few items still don't fully
work at this point due to internal problems but at least I haven't let out
any magic smoke yet and I've done this with perhaps 50 pieces of electronic
test equipment, some of it dating to the 1950s and a lot that I know
haven't been used in over 20 years.

I did this just last week on two OLD GR megohhmeters and two HP 428B
Milliamp meters made in 1969 and all four pieces work and while they're not
completely within spec, they work reasonably well. I tried it on an old
Fluke 873 Differential Voltmeter made in about 1964 and it did come back to
life but it's not measuring anything and the needle just flicks back and
forth. However I can see two caps in it that LONG ago leaked their
electrolyte, not to mention that when I found the meter it had been sitting
outside in a junkyard for so long that the handle had rotted off of it and
the paint had faded. I really didn't expect to work at all but it turned
out to be remarkably clean inside so I decided to get it a shot. All of the
thumbwheels in it had rusted to their shafts and I had to use PB Blaster to
free them up.

I mounted a 4 amp GR Variac in an old re-purposed ventilated rack mount
chassis and added two good quality Simpson 6 inch scale meters to it, one a
150 VAC or so voltmeter and the other a 6? Amp ampmeter so I can watch the
voltage and current closely. I would not trust a digital meter for this
application, they react too slowly. I added a line cord with an insulated
toggle switch and an internal FB fuse and coupled the variac output to
standard duplex AC outlet mounted on the back.

On Tue, Oct 27, 2020 at 10:42 PM <EDick@...> wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for quite
some time and I fear that if I try to power it up, it'll just go poof,
mainly due to the electrolytic caps in the power cicrcuits. I have a basic
idea about how to reform the electrolytics that aren't deformed, and they
shouldn't be, since no power has been applied for quite some time. I'm
interested in any recommendations regarding bringing this stuff back to
service life.

I have a variac, and constant-voltage AC power supply available, and
various scopes that have been more recently powered up. I also have a wide
range of passives and rectifiers, etc.

Any suggestions?

Uli







Jean-Paul
 

Have used metered variac for decades, drill:

Set 25% line, monitor input current
If it settles normally, increase to 50%, then 75% and finally 100%


The cap reforming is more of an issue on very old vacuum tube (valve) equipment eg hifi amps made in 1950s..1960s.
The 50v and under caps in TM500 5000 mainframes seldom fail, but dry caps have higher ESR and lower C, so ripple increases.

Most failures on these mainframes are blown pass transistors, rather than lytic caps.

In case you have unmetered variac, use an AC current probe to monitor input current.
Another solution: Wire a 100W incandescent lamp in series in case of shorts or failuers, it will limit the inrush current during reforming and debugging.

Good luck,

Jon


Tom Lee
 

It is optimistic to assume that an SMPS can operate correctly over an arbitrary range of input voltages. What one would hope is that the SMPS would shut down, or drop out of regulation, gracefully when the input voltage is insufficient. Many -- perhaps most modern ones -- do. However, what one not infrequently sees is that the controller misbehaves when the input voltage is held at a value less than some minimum voltage, and then mayhem results.

Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 10/28/2020 01:03, Tom Gardner wrote:
How does equipment with an SMPS react?

SMPSs are designed to deliver constant output power. Hence if the input voltage is lower then, by design, the SMPS will draw a higher input current.

Semiconductors don't like too much current.




On 28/10/20 04:33, - wrote:
   Regarding any electronic equipment that hasn't been used in a while, I
bring it up slowly using a variac and I've noticed that the odds of it
working has improved from about 50% to about 90%. But also I'm now working
with much older equipment than I did before I started doing this so the
results are probably better than they sound. I start it out at about 1 VAC
and watch the ammeter and as long as it doesn't appear to be drawing
excessive current, I leave it there for an hour or so, then take it up to
about 6 VAC, again watching the current. Leave it for a time, them take it
to about 24 VAC and repeat, then to about 50 VAC them to about 75 VAC. At
this point I usually start seeing signs of life. Then to about 90 VAC. At
this point most pieces of equipment start working even if not entirely
correctly. After a wait, then I take it to about 105 VAC. Finally up to 120
VAC.  *Knock on wood!* So far I've never had anything let out the magic
smoke while I bring up the voltage on it. A few items still don't fully
work at this point due to internal problems but at least I haven't let out
any magic smoke yet and I've done this with perhaps 50 pieces of electronic
test equipment, some of it dating to the 1950s and a lot that I know
haven't been used in over 20 years.

     I did this just last week on two OLD GR megohhmeters and two HP 428B
Milliamp meters made in 1969 and all four pieces work and while they're not
completely within spec, they work reasonably well.  I tried it on an old
Fluke 873 Differential Voltmeter made in about 1964 and it did come back to
life but it's not measuring anything and the needle just flicks back and
forth. However I can see two caps in it that LONG ago leaked their
electrolyte, not to mention that when I found the meter it had been sitting
outside in a junkyard for so long that the handle had rotted off of it and
the paint had faded. I really didn't expect to work at all but it turned
out to be remarkably clean inside so I decided to get it a shot. All of the
thumbwheels in it had rusted to their shafts and I had to use PB Blaster to
free them up.

     I mounted a 4 amp GR Variac in an old re-purposed ventilated rack mount
chassis and added two good quality Simpson 6 inch scale meters to it, one a
150 VAC or so voltmeter and the other a 6? Amp ampmeter so I can watch the
voltage and current closely.  I would not trust a digital meter for this
application, they react too slowly.  I added a line cord with an insulated
toggle switch and an internal FB fuse and coupled the variac output to
standard duplex AC outlet mounted on the back.

On Tue, Oct 27, 2020 at 10:42 PM <EDick@...> wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for quite
some time and I fear that if I try to power it up, it'll just go poof,
mainly due to the electrolytic caps in the power cicrcuits.  I have a basic
idea about how to reform the electrolytics that aren't deformed, and they
shouldn't be, since no power has been applied for quite some time.  I'm
interested in any recommendations regarding bringing this stuff back to
service life.

I have a variac, and constant-voltage AC power supply available, and
various scopes that have been more recently powered up.  I also have a wide
range of passives and rectifiers, etc.

Any suggestions?

Uli











Jean-Paul
 

Tom the SMPS we have designed, most controller ICs and TEK PSU have undervolt primary sense, so startup is inhibited till mon line eg 85V in USA or 170V in EU is reached. Then startup is sudden.

Any SMPS that has "misbehavior" at low V input "brownout" is not a production design and bound to have failures.

One reason is that switch transistor drivers must have a minimum output or the transistors are apt to fail.

Bon journée,

Jon


Dave Daniel
 

I do the same. Of I am concerned, I will look at the power supply rails woth a ‘scope to see of the supplied voltages ate well-regulated. If I see too much ripple on any rail I will I will investigate, starting with the electrolytic capacitors.

WRT to the discussion regarding using a variac with an SMPS power supply, that is a bad idea. The only reason to do that is if one is testing the supply to see if it meets it’s specifications for minimum and maximum input voltage levels.

On Oct 27, 2020, at 23:46, Chuck Harris <cfharris@...> wrote:

I just turn them on and see what happens. They are modern enough
capacitors that they will either be totally bad... eg. open circuit...
or will come up just fine.

If I am at all concerned, I will turn them on for a minute or two,
and then shut down and check for hot cans using my calibrated, though
somewhat scarred finger.

-Chuck Harris

EDick@... wrote:
It pains me to say I've not used this TM500 and TM5000 hardware for quite some time and I fear that if I try to power it up, it'll just go poof, mainly due to the electrolytic caps in the power cicrcuits. I have a basic idea about how to reform the electrolytics that aren't deformed, and they shouldn't be, since no power has been applied for quite some time. I'm interested in any recommendations regarding bringing this stuff back to service life.

I have a variac, and constant-voltage AC power supply available, and various scopes that have been more recently powered up. I also have a wide range of passives and rectifiers, etc.

Any suggestions?

Uli









Chuck Harris
 

The problem with using a variac to reform capacitors in this
way is a variac is a "stiff" source of power. Even though it
is able to make, say 6V, it can still produce enough current
at 6V to damage the very thin oxide dielectric on a capacitor.

Further, when used on switch mode power supplies, it will very
often burn out the supply. This is because the supply's regulator
turns the current on the supply up full when it is run at brownout
input voltages. Often, the switcher's parts cannot sustain that
sort of abuse.

Newer switchers are better able to handle bad power, and often test
for inputs that are outside of safe operation, but many older
switchers can't and simply will go up in flames.

-Chuck Harris

- wrote:

Regarding any electronic equipment that hasn't been used in a while, I
bring it up slowly using a variac and I've noticed that the odds of it
working has improved from about 50% to about 90%. But also I'm now working
with much older equipment than I did before I started doing this so the
results are probably better than they sound. I start it out at about 1 VAC
and watch the ammeter and as long as it doesn't appear to be drawing
excessive current, I leave it there for an hour or so, then take it up to
about 6 VAC, again watching the current. Leave it for a time, them take it
to about 24 VAC and repeat, then to about 50 VAC them to about 75 VAC. At
this point I usually start seeing signs of life. Then to about 90 VAC. At
this point most pieces of equipment start working even if not entirely
correctly. After a wait, then I take it to about 105 VAC. Finally up to 120
VAC. *Knock on wood!* So far I've never had anything let out the magic
smoke while I bring up the voltage on it. A few items still don't fully
work at this point due to internal problems but at least I haven't let out
any magic smoke yet and I've done this with perhaps 50 pieces of electronic
test equipment, some of it dating to the 1950s and a lot that I know
haven't been used in over 20 years.

I did this just last week on two OLD GR megohhmeters and two HP 428B
Milliamp meters made in 1969 and all four pieces work and while they're not
completely within spec, they work reasonably well. I tried it on an old
Fluke 873 Differential Voltmeter made in about 1964 and it did come back to
life but it's not measuring anything and the needle just flicks back and
forth. However I can see two caps in it that LONG ago leaked their
electrolyte, not to mention that when I found the meter it had been sitting
outside in a junkyard for so long that the handle had rotted off of it and
the paint had faded. I really didn't expect to work at all but it turned
out to be remarkably clean inside so I decided to get it a shot. All of the
thumbwheels in it had rusted to their shafts and I had to use PB Blaster to
free them up.

I mounted a 4 amp GR Variac in an old re-purposed ventilated rack mount
chassis and added two good quality Simpson 6 inch scale meters to it, one a
150 VAC or so voltmeter and the other a 6? Amp ampmeter so I can watch the
voltage and current closely. I would not trust a digital meter for this
application, they react too slowly. I added a line cord with an insulated
toggle switch and an internal FB fuse and coupled the variac output to
standard duplex AC outlet mounted on the back.

On Tue, Oct 27, 2020 at 10:42 PM <EDick@...> wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for quite
some time and I fear that if I try to power it up, it'll just go poof,
mainly due to the electrolytic caps in the power cicrcuits. I have a basic
idea about how to reform the electrolytics that aren't deformed, and they
shouldn't be, since no power has been applied for quite some time. I'm
interested in any recommendations regarding bringing this stuff back to
service life.

I have a variac, and constant-voltage AC power supply available, and
various scopes that have been more recently powered up. I also have a wide
range of passives and rectifiers, etc.

Any suggestions?

Uli










-
 

"It is optimistic to assume that an SMPS can operate correctly over an
arbitrary range of input voltages."

No intelligent person should expect a SMPS to operate at 1 VAC, or 10
VAC or anything close to that. That's why I start with a LOW voltage, it
will test and hopefully reform any capacitors in the input ahead of the
switcher section of the SMPS. Then as I ramp up the voltage to the 70 to 90
VAC range the SMPS will start to operate and then and only then will it
apply power to the remainer of the circuit. I really wasn't sure how SMPSs
and didital ICs would handle the low voltage but so far they have been
fine. However I don't think I would try this on anything like an HP 9845C
that I knew that had to have voltages applied in a certain order.

Another reason that I start with low voltages is to test the RIFA caps.
Those are the single highest failure item that I have encountered and I
would much rather have one burst with 10 VAC applied to it than 120 VAC
applied to it. However, surprisingly, I have not had one fail since I
started using a variac. Or perhaps they have failed but not spectacularly
enough for me to be aware of it.

Something that a lot of people here seem to overlook is the fact that
every electronic component has some resistance and will draw proportionally
less current at 1 VAC than it will at 110 VAC. Therefore I feel that this
approach makes current limiting using a light bulb or other added
resistance unnecessary.

Even worse, few seem to realize that POWER not current is the major
concern here. Applying 1 VAC over any given resistance will decrease the
Power in a shorted cap or other component by a factor of 110^2 or 122,100
so in the event of say a shorted cap, you won't burn up the associated
rectifiers, current limiting resistors, PCB traces, etc. That's also why
you need to watch the amp meter. If you get to say 10 or 20 VAC and the DUT
is already drawing as much or more current that it should at 110 VAC then
you need to stop and find out why. But at this point you've only applied
(10/110)^2 or less than 1% (OR (20/110) ^2 or 3.3%) of the Power that you
would have applied if you had just turned it on with 110 VAC applied to the
DUT so the chances of damaging *additional components* are essentially nil.

On Wed, Oct 28, 2020 at 4:47 AM Tom Lee <tomlee@...> wrote:

It is optimistic to assume that an SMPS can operate correctly over an
arbitrary range of input voltages. What one would hope is that the SMPS
would shut down, or drop out of regulation, gracefully when the input
voltage is insufficient. Many -- perhaps most modern ones -- do.
However, what one not infrequently sees is that the controller
misbehaves when the input voltage is held at a value less than some
minimum voltage, and then mayhem results.

Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 10/28/2020 01:03, Tom Gardner wrote:
How does equipment with an SMPS react?

SMPSs are designed to deliver constant output power. Hence if the
input voltage is lower then, by design, the SMPS will draw a higher
input current.

Semiconductors don't like too much current.




On 28/10/20 04:33, - wrote:
Regarding any electronic equipment that hasn't been used in a
while, I
bring it up slowly using a variac and I've noticed that the odds of it
working has improved from about 50% to about 90%. But also I'm now
working
with much older equipment than I did before I started doing this so the
results are probably better than they sound. I start it out at about
1 VAC
and watch the ammeter and as long as it doesn't appear to be drawing
excessive current, I leave it there for an hour or so, then take it
up to
about 6 VAC, again watching the current. Leave it for a time, them
take it
to about 24 VAC and repeat, then to about 50 VAC them to about 75
VAC. At
this point I usually start seeing signs of life. Then to about 90
VAC. At
this point most pieces of equipment start working even if not entirely
correctly. After a wait, then I take it to about 105 VAC. Finally up
to 120
VAC. *Knock on wood!* So far I've never had anything let out the magic
smoke while I bring up the voltage on it. A few items still don't fully
work at this point due to internal problems but at least I haven't
let out
any magic smoke yet and I've done this with perhaps 50 pieces of
electronic
test equipment, some of it dating to the 1950s and a lot that I know
haven't been used in over 20 years.

I did this just last week on two OLD GR megohhmeters and two HP
428B
Milliamp meters made in 1969 and all four pieces work and while
they're not
completely within spec, they work reasonably well. I tried it on an old
Fluke 873 Differential Voltmeter made in about 1964 and it did come
back to
life but it's not measuring anything and the needle just flicks back and
forth. However I can see two caps in it that LONG ago leaked their
electrolyte, not to mention that when I found the meter it had been
sitting
outside in a junkyard for so long that the handle had rotted off of
it and
the paint had faded. I really didn't expect to work at all but it turned
out to be remarkably clean inside so I decided to get it a shot. All
of the
thumbwheels in it had rusted to their shafts and I had to use PB
Blaster to
free them up.

I mounted a 4 amp GR Variac in an old re-purposed ventilated
rack mount
chassis and added two good quality Simpson 6 inch scale meters to it,
one a
150 VAC or so voltmeter and the other a 6? Amp ampmeter so I can
watch the
voltage and current closely. I would not trust a digital meter for this
application, they react too slowly. I added a line cord with an
insulated
toggle switch and an internal FB fuse and coupled the variac output to
standard duplex AC outlet mounted on the back.

On Tue, Oct 27, 2020 at 10:42 PM <EDick@...> wrote:

It pains me to say I've not used this TM500 and TM5000 hardware for
quite
some time and I fear that if I try to power it up, it'll just go poof,
mainly due to the electrolytic caps in the power cicrcuits. I have
a basic
idea about how to reform the electrolytics that aren't deformed, and
they
shouldn't be, since no power has been applied for quite some time. I'm
interested in any recommendations regarding bringing this stuff back to
service life.

I have a variac, and constant-voltage AC power supply available, and
various scopes that have been more recently powered up. I also have
a wide
range of passives and rectifiers, etc.

Any suggestions?

Uli

















Dave Wise
 

Sometimes the mechanism behind failure is not the drive but the transformer. To minimize cost/size/weight, designers use the smallest transformer possible. Usually the limiting factor is how much core is required to avoid saturation. Saturation can be fatal since it turns the primary winding into, basically, a short circuit. Turn on the transistor, current ramps up, core saturates, current instantly rises towards infinity, boom. The supply comes closest to this when mains voltage is low.

I learned this while burning out several BD912's in a test jig measuring how many amp-turns a particular core could take.

Dave Wise

OT: Turns out my pulse generator had a cracked solder joint on the pulse width control. Wavetek 145 owners beware.
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Jean-Paul via groups.io <jonpaul=ix.netcom.com@groups.io>
Sent: Wednesday, October 28, 2020 6:12 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] refurbishing TM500 and TM5000 hardware

Tom the SMPS we have designed, most controller ICs and TEK PSU have undervolt primary sense, so startup is inhibited till mon line eg 85V in USA or 170V in EU is reached. Then startup is sudden.

Any SMPS that has "misbehavior" at low V input "brownout" is not a production design and bound to have failures.

One reason is that switch transistor drivers must have a minimum output or the transistors are apt to fail.

Bon journée,

Jon


Roy Thistle
 

On Wed, Oct 28, 2020 at 01:22 AM, Jean-Paul wrote:


The cap reforming is more of an issue on very old vacuum tube (valve)
equipment eg hifi amps made in 1950s..1960s.
Certainly... and thus so in large measure to those old caps obsolete technology, and obsolete construction/production.
Certainly, there is no lack of opinions about capacitor reforming. (... whatever that might mean.)
Below are some short quotes, from a Rubycon document. (Rubycon, whom I trust, is a respected Japanese producer of high quality aluminum electrolytic capacitors.)

The leakage current of aluminum electrolytic capacitor is rather larger than other types of capacitor. This value will be influenced by temperature, applied voltage and applying time of voltage. Particularly, brief leakage current level which does not reach to specified time such as 2 minutes after applied voltage is unstable. (Rubycon)

Leakage current of Aluminum Electrolytic Capacitors may be increased after storage for a long time. Conduct electrification treatment for such capacitors before use.(Rubycon)
Electrification Treatment
Connect a 1kΩ resistor in series with the subject capacitor, and apply the DC voltage as high as the Rated Voltage for 1 hour.
Discharge the capacitor through a resistor of about 1Ω/Volt after the electrification. (Rubycon)


 

This highlights the problem with not running any form of stabilised supply at its intended voltage. The habit of putting a light bulb in the feed or a variac is fine in the scenario of old valve radios with simple psus based on dropper resistors or similar. A stabilised or SMPS is being invited to have a coronary if you do that to it

Robin

On 28 Oct 2020, at 16:01, Roy Thistle <roy.thistle@...> wrote:

On Wed, Oct 28, 2020 at 01:22 AM, Jean-Paul wrote:


The cap reforming is more of an issue on very old vacuum tube (valve)
equipment eg hifi amps made in 1950s..1960s.
Certainly... and thus so in large measure to those old caps obsolete technology, and obsolete construction/production.
Certainly, there is no lack of opinions about capacitor reforming. (... whatever that might mean.)
Below are some short quotes, from a Rubycon document. (Rubycon, whom I trust, is a respected Japanese producer of high quality aluminum electrolytic capacitors.)

The leakage current of aluminum electrolytic capacitor is rather larger than other types of capacitor. This value will be influenced by temperature, applied voltage and applying time of voltage. Particularly, brief leakage current level which does not reach to specified time such as 2 minutes after applied voltage is unstable. (Rubycon)

Leakage current of Aluminum Electrolytic Capacitors may be increased after storage for a long time. Conduct electrification treatment for such capacitors before use.(Rubycon)
Electrification Treatment
Connect a 1kΩ resistor in series with the subject capacitor, and apply the DC voltage as high as the Rated Voltage for 1 hour.
Discharge the capacitor through a resistor of about 1Ω/Volt after the electrification. (Rubycon)





 

Hi Uli,
In each slot of a TM500 and TM5000 mainframe there is an NPN and a PNP power transistor. They must be checked to be sure they haven't been blown out by a defective plugin.
Tek changed these power transistors at least 12 times over the years these mainframes were in production. In some cases they did not use replacements with the same pin-outs. So you need to be sure the E, B, and C leads of the replacement match what is in the mainframe.

The pass transistors can be so confusing that Jerry Scheltgen (RIP) and I compiled a list of TM500/TM5000 pass transistors Tek used, what the selection criteria was for each part number, and which mainframes they were used in.

Our list of TM500/TM5000 pass transistors is in the files section of TekScopes at
https://groups.io/g/TekScopes/files/TM5xxx%20Pass%20Transistor%20List.pdf

Dennis Tillman W7pF
TekScopes Moderator

On 10/27/2020 12:23 PM, EDick@... wrote:
It pains me to say I've not used this TM500 and TM5000 hardware for quite some time and I fear that if I try to power it up, it'll just go poof, mainly due to the electrolytic caps in the power cicrcuits. I have a basic idea about how to reform the electrolytics that aren't deformed, and they shouldn't be, since no power has been applied for quite some time. I'm interested in any recommendations regarding bringing this stuff back to service life.

I have a variac, and constant-voltage AC power supply available, and various scopes that have been more recently powered up. I also have a wide range of passives and rectifiers, etc.

Any suggestions?

Uli
--
Dennis Tillman W7pF
TekScopes Moderator


Roy Thistle
 

On Wed, Oct 28, 2020 at 01:03 AM, Tom Gardner wrote:


SMPSs are designed to deliver constant output power.
They can be... though that might actually be an approximation. The term "constant output power" is somewhat ill-defined, I'd claim.
Regardless (irregardless if you like)... most OEM SMPS are designed for/to a rated/nominal power output.
Is this/that what you mean by constant output power?