Switching P.S. repair help

[George]

Hello:

Trying to affect a repair in a video system device (TBC). The symtom
is seen as slight hue shifting on the monitor evident by several
horizontal bars of various widths. I can see some slight spiking on
the 5. DC rail more or less lining up with the bars in the video.

My assumption at this point is failing filter caps in the 35amp 5v.
switching supply power supply. The unit is around 20 years old and has
seen heavy use by the looks of the dust on the fan.

Question: Would I lean towards replacing only the three 200v 330uf
caps in the primary and/or the 12 or so 1000uf caps on the 5v. output
as well?

Thanks for the advice.

George

 

[Baphomet]

Hello:

Trying to affect a repair in a video system device (TBC). The symtom
is seen as slight hue shifting on the monitor evident by several
horizontal bars of various widths. I can see some slight spiking on
the 5. DC rail more or less lining up with the bars in the video.

My assumption at this point is failing filter caps in the 35amp 5v.
switching supply power supply. The unit is around 20 years old and has
seen heavy use by the looks of the dust on the fan.

Question: Would I lean towards replacing only the three 200v 330uf
caps in the primary and/or the 12 or so 1000uf caps on the 5v. output
as well?

Thanks for the advice.

George

I would try bridging the electrolytics with some small ceramic disc
capacitors first.

 

[Jim Yanik]

[email protected] (George) wrote in

Hello:

Trying to affect a repair in a video system device (TBC). The symtom
is seen as slight hue shifting on the monitor evident by several
horizontal bars of various widths. I can see some slight spiking on
the 5. DC rail more or less lining up with the bars in the video.

My assumption at this point is failing filter caps in the 35amp 5v.
switching supply power supply. The unit is around 20 years old and has
seen heavy use by the looks of the dust on the fan.

Question: Would I lean towards replacing only the three 200v 330uf
caps in the primary and/or the 12 or so 1000uf caps on the 5v. output
as well?

Thanks for the advice.

George

It could also be a clock problem.If the TBS is one of the type that
digitizes the video then reconverts it to analog,the timebase xtal could be
on the ragged edge of gen-lock.

BTW,how expensive would it be to replace ALL the 5v electrolytics?
(compared to the time you spend troubleshooting!!)If the unit is 20 yrs
old,and it's a switcher,I'd guess they all can be marginal.

 

[Michael A. Terrell]

Hello:

Trying to affect a repair in a video system device (TBC). The symtom
is seen as slight hue shifting on the monitor evident by several
horizontal bars of various widths. I can see some slight spiking on
the 5. DC rail more or less lining up with the bars in the video.

My assumption at this point is failing filter caps in the 35amp 5v.
switching supply power supply. The unit is around 20 years old and has
seen heavy use by the looks of the dust on the fan.

Question: Would I lean towards replacing only the three 200v 330uf
caps in the primary and/or the 12 or so 1000uf caps on the 5v. output
as well?

Thanks for the advice.

George

You need an ESR meter, unless you want to shotgun all the
electrolytics in the power supply.

 

[Jan Panteltje]

Hello:

Trying to affect a repair in a video system device (TBC). The symtom
is seen as slight hue shifting on the monitor evident by several
horizontal bars of various widths. I can see some slight spiking on
the 5. DC rail more or less lining up with the bars in the video.

My assumption at this point is failing filter caps in the 35amp 5v.
switching supply power supply. The unit is around 20 years old and has
seen heavy use by the looks of the dust on the fan.

Question: Would I lean towards replacing only the three 200v 330uf
caps in the primary and/or the 12 or so 1000uf caps on the 5v. output
as well?

Thanks for the advice.

George

Scope the voltage ripple across the primary caps, if you see sharp
pulses, replace (there should be only mains ripple).
Check magnitude ripple, if too big replace.
After 20 years it makes sense to replace the secondary side caps too,
caps dry out.
Sometimes the connections get lose too, and they become inductive.
Make sure you use the right kind of caps, ripple current rating, secially
suited for switch mode etc.

 

[Ken Taylor]

You need an ESR meter, unless you want to shotgun all the
electrolytics in the power supply.

--
We now return you to our normally scheduled programming.

Michael A. Terrell
Central Florida

If you needed to replace one of the electrolytics due to aging, I'd go for
all of them as they are likely to all be the same age and thus suspect.

Ken

 

[Ross Mac]

Hello:

Trying to affect a repair in a video system device (TBC). The symtom
is seen as slight hue shifting on the monitor evident by several
horizontal bars of various widths. I can see some slight spiking on
the 5. DC rail more or less lining up with the bars in the video.

My assumption at this point is failing filter caps in the 35amp 5v.
switching supply power supply. The unit is around 20 years old and has
seen heavy use by the looks of the dust on the fan.

Question: Would I lean towards replacing only the three 200v 330uf
caps in the primary and/or the 12 or so 1000uf caps on the 5v. output
as well?

Thanks for the advice.

George

Just scope the caps and if there is any ripple, they are most likely the
problem. Even a good Fluke will troubleshoot a cap...measure the DC drop,
then switch it too AC...there shouldn't be more than a few millivolts of
hum.....good luck, Ross

 

[Michael A. Terrell]

If you needed to replace one of the electrolytics due to aging, I'd go for
all of them as they are likely to all be the same age and thus suspect.

Ken

If they all are run at the same ripple currents and temperatures, I
agree. On the other hand, some designs run one or two caps very hard,
and the others rarely fail. If the equipment is built with expensive
caps you could be replacing hundreds of dollars worth of parts that will
still work for years. It all boils down to budget, allowable downtime,
and how long they plan to keep the equipment in service.

 

[Ross Mac]

If they all are run at the same ripple currents and temperatures, I
agree. On the other hand, some designs run one or two caps very hard,
and the others rarely fail. If the equipment is built with expensive
caps you could be replacing hundreds of dollars worth of parts that will
still work for years. It all boils down to budget, allowable downtime,
and how long they plan to keep the equipment in service.

--
We now return you to our normally scheduled programming.

Michael A. Terrell
Central Florida

Ah, a man of experience! Good advice Mike......

 

[Baphomet]

Just scope the caps and if there is any ripple, they are most likely the
problem. Even a good Fluke will troubleshoot a cap...measure the DC drop,
then switch it too AC...there shouldn't be more than a few millivolts of
hum.....good luck, Ross

From the OP's description, it didn't sound like mains hum as much as R.F.
spiking; many electrolytics are notorious for developing a high frequency
impedance over time as they dry out. I suggested that he try bridging them
with small disk ceramics before going through the expense and aggravation of
replacing otherwise good quality electrolytics.

 

[Jim Yanik]

From the OP's description, it didn't sound like mains hum as much as
R.F. spiking; many electrolytics are notorious for developing a high
frequency impedance over time as they dry out. I suggested that he try
bridging them with small disk ceramics before going through the
expense and aggravation of replacing otherwise good quality
electrolytics.

Compared to the expanse and aggravation if it fails again soon after being
put back into service? After all,those caps are all 20yrs old,and there's
only 12 caps that he mentioned.

Good electrolytics for switchers don't cost all that much these days.
(compared to studio time charges)

12 1000uf/10v 105degF low-ESR electrolytics,cost maybe $2 apiece?
$24.00 plus half hour,45 min. to replace them?

(of course,at TEK's Orlando field office,I had *bins* of these caps for re-
capping TV products!)

 

[George]

Just scope the caps and if there is any ripple, they are most likely the
problem. Even a good Fluke will troubleshoot a cap...measure the DC drop,
then switch it too AC...there shouldn't be more than a few millivolts of
hum.....good luck, Ross

Ross:

I need to put the PS back together. I opened it to look at the cap
values...but...

Are you saying to place the scope on the +- of each of the three 330us
caps in the primary section to look for any ripple? Just a few
millivolts seems very low...would it be this low at this early stage?
Or are you talking about the dozen or so at the back end on the 5v.
output.

Testing with a disc cap per the first reply seems easy enough. Should
I try this as well?

What should I look for in "special caps" for a switching supply in the
Mouser catalog?

Thanks for the help!

George

 

[John Woodgate]

I read in sci.electronics.design that George <[email protected]>

Are you saying to place the scope on the +- of each of the three 330us
caps in the primary section to look for any ripple?

Well, yes, sort of, but DON'T. Neither end may be at ground potential
and you stand a good chance of a major damage incident. The high DC
voltage may damage your scope and you, as well, even if you get the
connections right. Personally, I'd replace the primary side ones with
low-ESR 105 degree types and put a 470 nF (0.47 uF) ceramic cap across
each one on the secondary side.