开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育
  1. HP-Agilent-Keysight-Equipment
  2. Topics

8711B tantalum disease

 

This would apply to all this series.

I got a 8711B with a bad power supply.? I tracked that down to an open trace caused by corrosion in one of the mag amp circuits.

The unit still wouldn't power on, traced it to a shorted tantalum capacitor on the +15 volt bus.? Replaced it, reassembled everything, the unit ran about 2 minutes then shut down again.? This time it was a shorted tantalum on the +12 bus, different assembly.

If this is like other gear I have had experience with, unless I go through and replace every single one of those suckers it will remain a crapshoot as to whether it will turn on each time.

It might be the age of these instruments.? My expectation is that every one of this series will have these failures and be unreliable.

Peter
 

Hi Peter,

It seems to be a combination of a couple of bad brands
of capacitors, and an engineering thing.

The biggest issue is capacitor rating voltage vs applied
voltage. It should be greater than 2x for best life.

The other issue is source-able current. If the power source
for the capacitor can source a lot of current into the
capacitor, it will destroy (melt down) what would have been
an otherwise self healing capacitor.

So, my philosophy is when one pops, if it is being run close
to its rating (eg. 5V on a 6V cap, 15V on a 20V cap, I just
replace it and any others similarly applied.

If it is more like 15V on a 35V cap, I replace them all.

-Chuck Harris


On Fri, 26 Apr 2024 17:21:57 -0400 "Peter Gottlieb"
<hpnpilot@...> wrote:
This would apply to all this series.

I got a 8711B with a bad power supply.? I tracked that down to an
open trace caused by corrosion in one of the mag amp circuits.

The unit still wouldn't power on, traced it to a shorted tantalum
capacitor on the +15 volt bus.? Replaced it, reassembled everything,
the unit ran about 2 minutes then shut down again.? This time it was
a shorted tantalum on the +12 bus, different assembly.

If this is like other gear I have had experience with, unless I go
through and replace every single one of those suckers it will remain
a crapshoot as to whether it will turn on each time.

It might be the age of these instruments.? My expectation is that
every one of this series will have these failures and be unreliable.

Peter
 

Peter,
I agree - tantalums are a pest.? Perhaps the problem isn't the tantalums but our fascination with equipment from the dodgy tantalum era.? We can replace tantalums, dried electrolytics, leaky backup batteries etc but who really wants to cure that fascination?
The tantalums exposed to immediate voltage at turn on are the most likely to fail. Those in timing circuits, charged via a resistor seem to be OK.
A scan of the circuit should identify the tantalums being hit at turn on - replace these as a precaution.? That solves some of the issues but obviously tedious if you have a memory or logic board with decoupling tantalums on every chip.
Sometimes it can work to your advantage: I bought a Racal RA6790 RX cheap on ebay - advertised as not working.? It was tedious - tantalums distributed all over the board decoupling power rails.? Got that radio going and then it stopped again!

Cheers

Mark
VK2WU
 

There were a lot of 3 legged tantalums which I hadn't seen before. The two outside leads were negative and center positive.? The ground plane on the processor board seemed unusually heat-sinky and I had a really hard time desoldering those leads.? It would be a massive chore to have to replace all the tantalums.

Not for an old 75 ohm 8711B, anyway.

Peter

On 4/27/2024 8:24 PM, Mark Weedon via groups.io wrote:
Peter,
I agree - tantalums are a pest.? Perhaps the problem isn't the tantalums but our fascination with equipment from the dodgy tantalum era.? We can replace tantalums, dried electrolytics, leaky backup batteries etc but who really wants to cure that fascination?
The tantalums exposed to immediate voltage at turn on are the most likely to fail. Those in timing circuits, charged via a resistor seem to be OK.
A scan of the circuit should identify the tantalums being hit at turn on - replace these as a precaution.? That solves some of the issues but obviously tedious if you have a memory or logic board with decoupling tantalums on every chip.
Sometimes it can work to your advantage: I bought a Racal RA6790 RX cheap on ebay - advertised as not working.? It was tedious - tantalums distributed all over the board decoupling power rails. Got that radio going and then it stopped again!

Cheers

Mark
VK2WU
 

On Saturday 27 April 2024 08:24:59 pm Mark Weedon wrote:
Peter,
I agree - tantalums are a pest.? Perhaps the problem isn't the tantalums but our fascination with equipment from the dodgy tantalum era.? We can replace tantalums, dried electrolytics, leaky backup batteries etc but who really wants to cure that fascination?
The tantalums exposed to immediate voltage at turn on are the most likely to fail. Those in timing circuits, charged via a resistor seem to be OK.
My impression is that the ones that fail are those that have voltage applied to them which is close to or at the part's rated voltage. One instrument that I'm familiar with used a *lot* of parts rated at 25V in a circuit that might under some conditions put 30V across them. Yes, there were failures...

--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
 

One was a 20 volt part on a 15 volt rail.? The other was a 10 volt part (!) on a 12 volt rail.

I replaced both with 35 volt parts.

Nice that the power supplies have current sensing and shut everything down when a tantalum shorts.? So many pieces of equipment don't have that and end up burning holes in boards.

Peter

On 4/28/2024 12:56 PM, Roy J. Tellason, Sr. via groups.io wrote:
On Saturday 27 April 2024 08:24:59 pm Mark Weedon wrote:
Peter,
I agree - tantalums are a pest.? Perhaps the problem isn't the tantalums but our fascination with equipment from the dodgy tantalum era.? We can replace tantalums, dried electrolytics, leaky backup batteries etc but who really wants to cure that fascination?
The tantalums exposed to immediate voltage at turn on are the most likely to fail. Those in timing circuits, charged via a resistor seem to be OK.
My impression is that the ones that fail are those that have voltage applied to them which is close to or at the part's rated voltage. One instrument that I'm familiar with used a *lot* of parts rated at 25V in a circuit that might under some conditions put 30V across them. Yes, there were failures...

1 - 6 of 6