Update on 720A
I can't replace the bad resistors in the A decade - they are in the oil bath and the manual says that in this case you sent the unit back to Fluke for a re-build. These days, I suspect they won't even do re-builds at all (or only at silly money), but make you buy a new one, and may well not supply any parts unless they do the repair (as happened to a friend of mine with another instrument).
I'm really not convinced that "the very act of replacing the resistors is likely to cause additional damage" - I've been around delicate and high-Z electronics for a while now and can get in and out in unsoldering pretty handily. I'm also very aware of the critical importance of not messing up the conformal coating which is there for a reason (and re-sealing it where necessary).
I accept that if the primary resistors keep on drifting then I am "in a hole" and should probably save my pennies for another one.
The problem is that the only source (that I know of) of used 720As in UK is either eBay or very occasionally a cal. lab. that is closing. "Known good" isn't normally a term that applies to either.
Dave
Hi David,
It isn't your soldering quality or ability that will degrade
the instrument, it is rather the quality of the parts you
add, or replace, and what you do to fudge things into sort of
working, that will damage/degrade the instrument.
Devices like the 720A are firmly in the category of magical
devices. Resistors, switches, pots, and even the plating,
and the solder alloy are critical to the ultimate performance
the 720A can achieve.
Fluke worked every angle they could think of to make the 720A
a precise and very stable device. If you change a resistor
type in a critical place, the whole recipe can fall apart.
The resistors they used aren't the sort that will simply drift
off value. They are wirewound using a very special alloy, and
are essentially good forever. If they have changed, either
someone has changed them, or something drastic has happened,
like they have been overloaded, or an internal contact has
failed.
The pots used as trimmers should have only been trimmed a couple
of times over the life of the instrument... hardly enough to wear
them out... though corrosion is always possible.
My concern is you seemed to be going at this repair like you
were working on an oscilloscope, rather than a super precision
instrument.
-Chuck Harris
David C. Partridge wrote:
I can't replace the bad resistors in the A decade - they are in the oil bath and
the manual says that in this case you sent the unit back to Fluke for a re-build.
These days, I suspect they won't even do re-builds at all (or only at silly
money), but make you buy a new one, and may well not supply any parts unless
they do the repair (as happened to a friend of mine with another instrument).
I'm really not convinced that "the very act of replacing the resistors is likely
to cause additional damage" - I've been around delicate and high-Z electronics for
a while now and can get in and out in unsoldering pretty handily. I'm also very
aware of the critical importance of not messing up the conformal coating which is
there for a reason (and re-sealing it where necessary).
I accept that if the primary resistors keep on drifting then I am "in a hole" and
should probably save my pennies for another one.
The problem is that the only source (that I know of) of used 720As in UK is either
eBay or very occasionally a cal. lab. that is closing. "Known good" isn't
normally a term that applies to either.
Dave
_______________________________________________ volt-nuts mailing list --
volt-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts and follow the
instructions there.
My concern is you seemed to be going at this repair like you were working on an oscilloscope, rather than a super precision instrument.
More a case of: Oh my goodness how did this end up in this condition, and is there any chance I can bring it back from the edge. No, I didn't use super precision resistors for the two padding resistors I inserted but I did take care to use metal film resistors with a low thermal coefficient (ppm/C). If the values I've used don't need changing (which I think I will know soon enough), I'll get in some really good ppm/C parts (maybe from Edwin Pettis).
Yes, I know that the two 9.896k resistors sets that have gone low will probably remain suspect until they remain stable at their current value for a long while.
I wish I could afford a new 720A ... but if wishes were horses, beggars could ride.
Aside: I'm still struggling to understand how two sets of four precision 2474 Ohm WW resistors (9896 Ohm) in an oil bath can go low as they have in this instrument, while the remainder appear to be almost dead on specification.
Dave
The bridge balance pot may not be so lucky, though. I don't think it needs to be anything very special.
As far as I can tell it is a very special pot wired with two resistance tracks. The coarse track has with two wipers connecting to the fine track which has just one wiper, all driven by a clever mechanism to provide both coarse and fine adjustments with only one shaft.
In line art like this:
x
x
x<-----
x x
x x<---
x x
x<-----
x
x
Dave
Ah, ok. I went for a quick look through the
manual, and couldn't find it...
It could always be replaced with a concentric shaft
pot that has two appropriate sections wired in series.
That would lose the "nifty" factor, though.
-Chuck
David C. Partridge wrote:
The bridge balance pot may not be so lucky, though. I don't think it needs to be anything very special.
As far as I can tell it is a very special pot wired with two resistance tracks. The coarse track has with two wipers connecting to the fine track which has just one wiper, all driven by a clever mechanism to provide both coarse and fine adjustments with only one shaft.
In line art like this:
x
x
x<-----
x x
x x<---
x x
x<-----
x
x
Dave
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.
Maybe a quality Bourns wirewound 10 turns pot would work as a replacement?
On Aug 9, 2017 9:22 AM, "David C. Partridge" david.partridge@perdrix.co.uk
wrote:
The bridge balance pot may not be so lucky, though. I don't think it
needs to be anything very special.
As far as I can tell it is a very special pot wired with two resistance
tracks. The coarse track has with two wipers connecting to the fine track
which has just one wiper, all driven by a clever mechanism to provide both
coarse and fine adjustments with only one shaft.
In line art like this:
x
x
x<-----
x x
x x<---
x x
x<-----
x
x
Dave
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/
mailman/listinfo/volt-nuts
and follow the instructions there.
Hi David,
It is amazing what gets done to equipment in the name
of repairing it...
I have trouble believing that the precision WW resistors
in the oil bath did go low. I wouldn't presume to argue
that you didn't measure them right, but there has to be
something other than their value drifting low. WW
resistors usually go up and become erratic on their way
to becoming open circuit.
[I had a 720A once that did the same thing. I got
rid of it in hopes of getting a better example, but never
did... And no, I didn't chop it up. I observed that it
wouldn't calibrate, diagnosed that it was something in
the oil tank, and decided that I didn't want to go there
at that time.]
I am thinking that the resistor leads have to leave the
oil tank some way. Is it possible that the insulator on
whatever sort of feedthru has been compromised?
Otherwise it seems like it has to be something like
misdocumentation, or misalignment of a switch or something
like that.
-Chuck Harris
David C. Partridge wrote:
My concern is you seemed to be going at this repair like you were working on an
oscilloscope, rather than a super precision instrument.
More a case of: Oh my goodness how did this end up in this condition, and is there
any chance I can bring it back from the edge. No, I didn't use super precision
resistors for the two padding resistors I inserted but I did take care to use
metal film resistors with a low thermal coefficient (ppm/C). If the values I've
used don't need changing (which I think I will know soon enough), I'll get in some
really good ppm/C parts (maybe from Edwin Pettis).
Yes, I know that the two 9.896k resistors sets that have gone low will probably
remain suspect until they remain stable at their current value for a long while.
I wish I could afford a new 720A ... but if wishes were horses, beggars could
ride.
Aside: I'm still struggling to understand how two sets of four precision 2474 Ohm
WW resistors (9896 Ohm) in an oil bath can go low as they have in this instrument,
while the remainder appear to be almost dead on specification.
Dave
I am thinking that the resistor leads have to leave the oil tank some way. Is it possible that the insulator on whatever sort of feedthru has been compromised?
They are glass (or very similar) feedthroughs, so if that were compromised, I think I'd see oil ... There is no visual sign whatsoever of trauma.
If I find anything in that area, be sure I'll shout.
Dave
I wouldn't expect physical trauma to be the issue, more something
like a metal flash due to an arc, or tin whiskers... Something
like that.
If the feedthroughs were capacitors, I would suspect a metal
migration similar to what is being seen on many older silver
mica capacitors.
I am pretty sure that everything inside of the can is isolated
from the can. You could try measuring the resistance between
the affected resistor's terminal, and the can, looking for a
high impedance bridging resistance.
-Chuck Harris
David C. Partridge wrote:
I am thinking that the resistor leads have to leave the oil tank some way. Is
it possible that the insulator on whatever sort of feedthru has been
compromised?
They are glass (or very similar) feedthroughs, so if that were compromised, I
think I'd see oil ... There is no visual sign whatsoever of trauma.
If I find anything in that area, be sure I'll shout.
Dave
_______________________________________________ volt-nuts mailing list --
volt-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts and follow the
instructions there.
Chuck wrote:
I wouldn't expect physical trauma to be the issue, more something
like a metal flash due to an arc, or tin whiskers... Something
like that.
The damage to the resistors in the unit was almost certainly caused by
an electrical overload (the "trauma" to which I referred). Or, more
likely, to a number of overloads perpetrated by one or more clueless
previous owners on more than one occasion.
The other damage (disconnected wires, etc.) was caused by someone
hamhandedly attempting to diagnose or repair the original overload
damage, or contemplating harvesting parts from a unit they knew was
damaged beyond any practical repair.
The only cure is to replace all damaged resistors with new resistors of
equal or better quality. If they are in the oil bath, the complexity of
any repair is raised exponentially. Note that there are almost
certainly other damaged resistors in addition to the ones that the OP
has identified.
As a practical matter, an instrument with damaged (overloaded) resistors
will never work properly again. The only 720As I have ever seen that
worked properly after damage of this sort were repaired by Fluke at
astronomical cost.
Best regards,
Charles