Maker Pro
Maker Pro

When a vacuum tube fails

R

Ryan

Jan 1, 1970
0
My friend's ~1960's model corn moisture meter (accurate!) recently
failed. Inside he found a blackened RCA 12AU7A vacuum tube. I want to
help him out. This is the only tube it has and he says there is not
much stuff inside.

I've found these about on the internet.

Is it reasonable to expect that merely replacing this tube will render
the unit functional, or must I consider the possibility that a failed
tube can damage of other parts?

For example, in modern day stuff, it is my understanding that the
failure of a tranistor or mosfet, may kill other parts around. (Such as
a shorted gate-to-drain killing a driver transistor upstream in an
amplifier, or one burned mosfet in the power supply leading to the
demise of the whole bank.)

Does this same thing happen with tube era machines? How likely is it?

Among the same part numbers, how is one tube with a silver top different
than a clear tube? A¹Íthere other considerations to note before
ordering?

Thank You.
 
T

tempus fugit

Jan 1, 1970
0
I'm not too sure about the other parts around the tube, but you could easily
check them with a meter to see if they are OK. Before you order off the
internet, I would check at a local music store. The 12AX7 is a very common
tube used in guitar amps, and it may be OK to replace the AU7 with an AX7 (I
think the AX7 has higher gain).

As usual, the more experienced are welcome to correct me here.
 
J

John Popelish

Jan 1, 1970
0
Ryan said:
My friend's ~1960's model corn moisture meter (accurate!) recently
failed. Inside he found a blackened RCA 12AU7A vacuum tube. I want to
help him out. This is the only tube it has and he says there is not
much stuff inside.

I've found these about on the internet.

Is it reasonable to expect that merely replacing this tube will render
the unit functional, or must I consider the possibility that a failed
tube can damage of other parts?

For example, in modern day stuff, it is my understanding that the
failure of a tranistor or mosfet, may kill other parts around. (Such as
a shorted gate-to-drain killing a driver transistor upstream in an
amplifier, or one burned mosfet in the power supply leading to the
demise of the whole bank.)

Does this same thing happen with tube era machines? How likely is it?

Among the same part numbers, how is one tube with a silver top different
than a clear tube? A¹Íthere other considerations to note before
ordering?

Thank You.

The silver coating on the inside of the glass is a very thin coating
of a very active metal that grabs oxygen. This is intended to collect
any stray atoms that are still floating around, after the tube is
pumped down and sealed. If the tube has an air leak, this silver
coating will turn to a semi transparent dusty oxide. So tubes with a
nice silver coat that does not have dusty looking edges is a sign of
still good vacuum.
 
D

Don Bruder

Jan 1, 1970
0
tempus fugit said:
I'm not too sure about the other parts around the tube, but you could easily
check them with a meter to see if they are OK. Before you order off the
internet, I would check at a local music store. The 12AX7 is a very common
tube used in guitar amps, and it may be OK to replace the AU7 with an AX7 (I
think the AX7 has higher gain).

I think that advice is bad. Well-intended, but bad, nonetheless. With
something like a moisture meter, you're going to want the same gain,
otherwise, you're likely to get bad readings. For a guitar amp, more
gain from the same tube socket is no big deal. It might even be
desirable (if your'e into distortion, I guess). For a calibrated
instrument (especially an older item which may not have enough
information available to recalibrate to compensate for the different
tube) like this meter, extra gain may literally translate to "umpty-nine
thousand head of livestock dead" due to the meter reading false.
Especially if it reads false low... In corn, that's especially bad,
because it's subject to a specific form of blue mold (name forgotten
right now) that produces a mycotoxin known to kill and/or permanently
cripple horses and cattle that eat just a few (less than a dozen, in one
case I know of) contaminated kernels.
 
D

Don Bruder

Jan 1, 1970
0
John Popelish said:
The silver coating on the inside of the glass is a very thin coating
of a very active metal that grabs oxygen. This is intended to collect
any stray atoms that are still floating around, after the tube is
pumped down and sealed. If the tube has an air leak, this silver
coating will turn to a semi transparent dusty oxide. So tubes with a
nice silver coat that does not have dusty looking edges is a sign of
still good vacuum.

I can buy PART of that...
Care to offer an explanation for the mirror-ish spots that form on the
inside of some tubes after they've been run for a while? As in, they
start out completely clear, but after a varibable amount of time,
develop mirrored and/or blackened, or even both in the same spot, areas
on the inside fo the glass?

Some kind of atom-by-atom "spray-coating" of metal being applied to the
glass by the electron flow, perhaps?

One of those mysteries I've never heard an explanation for...
 
Ryan said:
My friend's ~1960's model corn moisture meter (accurate!) recently
failed. Inside he found a blackened RCA 12AU7A vacuum tube. I want to
help him out. This is the only tube it has and he says there is not
much stuff inside.

I've found these about on the internet.

Is it reasonable to expect that merely replacing this tube will render
the unit functional, or must I consider the possibility that a failed
tube can damage of other parts?

For example, in modern day stuff, it is my understanding that the
failure of a tranistor or mosfet, may kill other parts around. (Such as
a shorted gate-to-drain killing a driver transistor upstream in an
amplifier, or one burned mosfet in the power supply leading to the
demise of the whole bank.)

Does this same thing happen with tube era machines? How likely is it?

Among the same part numbers, how is one tube with a silver top different
than a clear tube? A¹Í there other considerations to note before
ordering?

Thank You.


It's likely only the tube.

Tubes should have a nice silver lining inside the top. This is a very
reactive element that immediately "removes" any traces of air that
might leak in. If the vacuum is lost, then all of it reacts and it
goes black instantly.

Robin
 
K

Kevin Aylward

Jan 1, 1970
0
Don said:
I think that advice is bad. Well-intended, but bad, nonetheless. With
something like a moisture meter, you're going to want the same gain,
otherwise, you're likely to get bad readings. For a guitar amp, more
gain from the same tube socket is no big deal. It might even be
desirable (if your'e into distortion, I guess). For a calibrated
instrument (especially an older item which may not have enough
information available to recalibrate to compensate for the different
tube) like this meter, extra gain may literally translate to
"umpty-nine thousand head of livestock dead" due to the meter reading
false. Especially if it reads false low... In corn, that's especially
bad, because it's subject to a specific form of blue mold (name
forgotten right now) that produces a mycotoxin known to kill and/or
permanently cripple horses and cattle that eat just a few (less than
a dozen, in one case I know of) contaminated kernels.

I think its unlikely that this would be an issue. I would guess that any
decent instrument will use the tube with a cathode resistor to set the
gain, relatively independent of the tube mu. It just wouldn't make sense
to rely on a tubes open loop gain, there is too much variation.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
D

Don Bruder

Jan 1, 1970
0
Kevin Aylward said:
Don said:
I think its unlikely that this would be an issue.

Unlikely enough to stake your future on it if it turns out wrong? Get it
wrong, kill or paralyze a few hundred (or thousand) head of cattle, or
maybe a few dozen million-dollar race/show horses with contaminated
corn, and kiss everything you have or ever will have goodbye.
Experimenting on something like this just wouldn't be worth the risk to
me. Especially since I have a strong interest in (and, unfortunately,
"up-close-n-personal" experience with) the kind of crisis "wrong
moisture level" feed can produce in livestock. Take my word for it... A
prize horse or cow having seizures and ripping itself up due to blue
corn mold toxin ain't a pretty sight to see.
I would guess that any
decent instrument will use the tube with a cathode resistor to set the
gain, relatively independent of the tube mu. It just wouldn't make sense
to rely on a tubes open loop gain, there is too much variation.

Perhaps true, perhaps not... <shrug> As of right now, neither of us has
enough info to go on as far as making that decision. I still say the
best advice is replace the blown tube with the same kind the circuit was
designed around. Whatever it might be, there's a reason that particular
tube was used in the original design. And since it's a piece of
equipment that could conceivably affect livestock on a global scale (No,
this isn't hyperbole - A ton of contaminated corn sold to a feed mill
might get spread literally around the world, affecting thousands of
animals) I'd be *EXTREMELY* wary of taking any chance of messing it up
through tube substitution. This is one of those situations where the
proverbial ounce of prevention (replacing with the proper tube) is well
worth whatever cost may be involved to do it right. Taking a quick look
via google spots at least a dozen sources for the exact tube needed, at
prices ranging from $7.50 to $18.95. So we're not talking about a
hard-to-find or particularly expensive item.

And back to the OP's original question:
Tube gear was (and still is) generally a LOT tougher than today's rather
fragile semiconductor-based hardware - Basically, you have tubes, caps,
coils, and resistors to work with. If the tube goes, unless it does so
in spectacular fashion, the rest of the components that make up the
device it's in are pretty near bulletproof. I wouldn't be particularly
worried about any "collateral damage" caused by a tube failure, since
there is almost certainly nothing connected to it that's susceptible to
the sort of "chain reaction" damage that can happen as a result of a key
component dying in multistage solid-state circuitry.
 
J

John Popelish

Jan 1, 1970
0
Don said:
I can buy PART of that...
Care to offer an explanation for the mirror-ish spots that form on the
inside of some tubes after they've been run for a while? As in, they
start out completely clear, but after a varibable amount of time,
develop mirrored and/or blackened, or even both in the same spot, areas
on the inside fo the glass?

Some kind of atom-by-atom "spray-coating" of metal being applied to the
glass by the electron flow, perhaps?

One of those mysteries I've never heard an explanation for...

I think that high power tubes (that run plates cherry red) do move
some metal atoms around and produce dark shadowy coatings, but I have
never seen one that produced a mirror film. Anyway, the intentional
films are boiled off a metal hoop that has a groove to hold the
reactive metal, which is heated to incandescence by an induction coil
after the tube is evacuated. If there is no sign of the boiling ring,
then the coating has some other source than intentional gettering.
 
M

Michael A. Terrell

Jan 1, 1970
0
Don said:
And back to the OP's original question:
Tube gear was (and still is) generally a LOT tougher than today's rather
fragile semiconductor-based hardware - Basically, you have tubes, caps,
coils, and resistors to work with. If the tube goes, unless it does so
in spectacular fashion, the rest of the components that make up the
device it's in are pretty near bulletproof. I wouldn't be particularly
worried about any "collateral damage" caused by a tube failure, since
there is almost certainly nothing connected to it that's susceptible to
the sort of "chain reaction" damage that can happen as a result of a key
component dying in multistage solid-state circuitry.

You better keep in mind that a lot of tube equipment was built with
paper, or oil impregnated pare capacitors that have a limited life. The
become leaky with age, and it only take a little leakage to upset the
bias on the tube. I have a military audio generator on my bench right
now that was built in the early '40s. It was built with the best
capacitors available at the time. (Oil filled bathtub capacitors) Every
capacitor in the unit is leaky, yet the equipment looks unused. If they
want to return the unit to service, they should completely rebuild it,
not patch it.
 
D

Don Bruder

Jan 1, 1970
0
Michael A. Terrell said:
You better keep in mind that a lot of tube equipment was built with
paper, or oil impregnated pare capacitors that have a limited life. The
become leaky with age, and it only take a little leakage to upset the
bias on the tube. I have a military audio generator on my bench right
now that was built in the early '40s. It was built with the best
capacitors available at the time. (Oil filled bathtub capacitors) Every
capacitor in the unit is leaky, yet the equipment looks unused. If they
want to return the unit to service, they should completely rebuild it,
not patch it.

Ahh, but the OP was wanting to know if the failed tube was likely to
"take out" anything else, as sometimes happens with modern devices - A
failed transistor causing transistors in the next stage of an amp to
cook off, ferinstance.

*ASS* *U* *ME*-ing that the tube alone puked (perhaps the unit finally
got dropped one time too many and cracked the tube, allowing it to leak,
or vibrated the filament into breaking, or the impact mooshed a grid out
of shape or into contact with another internal part of the tube causing
a short, or some similar "purely physical damage" scenario that I can
envision being *REALLY* common failure modes for hand-held bits of farm
equipment) I can't see any reason that a simple tube swap wouldn't
restore the beast to 100% functionality.

On the other hand, it could be, as you say, a cap failure, or a resistor
gone south, or any number of other things that won't be repaired by
replacing the tube...

Not really enough info in the original post to be sure either way...
 
D

Don Bruder

Jan 1, 1970
0
John Popelish said:
I think that high power tubes (that run plates cherry red) do move
some metal atoms around and produce dark shadowy coatings, but I have
never seen one that produced a mirror film. Anyway, the intentional
films are boiled off a metal hoop that has a groove to hold the
reactive metal, which is heated to incandescence by an induction coil
after the tube is evacuated. If there is no sign of the boiling ring,
then the coating has some other source than intentional gettering.

Right. That follows. Getters that are put there on purpose are usually
pretty obvious. But the film I'm talking about probably ain't caused by
getters...

The most pronounced instance I saw was in a monster of a tube that
apparently had something to do with a commercial AM radio station's
transmitter - can't be much more specific than that, since there was no
useful labeling on the box it was in - however, the box that it was
cabled to was tagged very clearly as "Modulator", and had other info on
it that indicated the beast was from an AM station - very possibly the
one about a mile up the road from the abandoned, partially collapsed
shed I found the gear in. Nobody at the station at the time knew
anything about it, though, so I'm guessing if it WAS theirs, it was last
in service sometime just before "ancient history". :)

This particular tube was a serious piece of work... Memory is foggy, but
I want to say it had more than 20 very large, thick (probably an eighth
of an inch diameter, maybe somewhat larger, and nearly an inch long,
except for one that almost looked as if it had been nipped off at about
half an inch, if you ignored the fact that the end of it was rounded
like all the rest - a keying pin, perhaps?) pins hanging out of the
bottom. It was about as big around as a baseball at the base, flaring
out to nearly softball sized at the top, and as I said, about 10 inches
high, with a large connector on the top. (FWIW, the wire going to this
connector was *HUGE* - About 3/4 inch thick, wrapped in at least 6
layers of varnished/shellaced/doped cloth on top of the rubber-like
insulation. I'd guess it was close to 6 gauge wire inside the various
layers of insulating material.

Aside from three spots, most of it was crystal clear glass. The three
spots (one covered most of the top of the tube in a somewhat irregular,
but generally round, "splatter", the other two were smaller patches near
the base that were, allowing for the shape of the tube, pretty close to
perfectly circular) seemed to be related to some sort of pieces inside
the tube. The top one wasn't visible (the film was a near perfect
mirror, except near the center, where it was black, and any other view
was blocked either by the film, or the hardware inside the glass) but
the bottom two could be "looked around" to see that there were two
ring-like pieces of metal, each one connected to a separate pin,
directly in line with the patches. Precisely what those two rings were
is obviously debatable, but from the "spray pattern", it seemed pretty
obvious that they had something to do with the film being there. In the
center area of each "patch", the finish was a nearly perfect (albeit
curved to conform with the shape of the tube's glass) mirror. The
further from the center of the patch you looked, the thinner the
coating, until it finally petered out at about 1.5 inches from the
center, leaving perfectly clear glass.

Care to take a whack at the "WTF-izzit!?!?" on that one? :)
 
J

John Popelish

Jan 1, 1970
0
Don said:
Right. That follows. Getters that are put there on purpose are usually
pretty obvious. But the film I'm talking about probably ain't caused by
getters...

The most pronounced instance I saw was in a monster of a tube that
apparently had something to do with a commercial AM radio station's
transmitter - can't be much more specific than that, since there was no
useful labeling on the box it was in - however, the box that it was
cabled to was tagged very clearly as "Modulator", and had other info on
it that indicated the beast was from an AM station - very possibly the
one about a mile up the road from the abandoned, partially collapsed
shed I found the gear in. Nobody at the station at the time knew
anything about it, though, so I'm guessing if it WAS theirs, it was last
in service sometime just before "ancient history". :)

This particular tube was a serious piece of work... Memory is foggy, but
I want to say it had more than 20 very large, thick (probably an eighth
of an inch diameter, maybe somewhat larger, and nearly an inch long,
except for one that almost looked as if it had been nipped off at about
half an inch, if you ignored the fact that the end of it was rounded
like all the rest - a keying pin, perhaps?) pins hanging out of the
bottom. It was about as big around as a baseball at the base, flaring
out to nearly softball sized at the top, and as I said, about 10 inches
high, with a large connector on the top. (FWIW, the wire going to this
connector was *HUGE* - About 3/4 inch thick, wrapped in at least 6
layers of varnished/shellaced/doped cloth on top of the rubber-like
insulation. I'd guess it was close to 6 gauge wire inside the various
layers of insulating material.

What a beast.
Aside from three spots, most of it was crystal clear glass. The three
spots (one covered most of the top of the tube in a somewhat irregular,
but generally round, "splatter", the other two were smaller patches near
the base that were, allowing for the shape of the tube, pretty close to
perfectly circular) seemed to be related to some sort of pieces inside
the tube. The top one wasn't visible (the film was a near perfect
mirror, except near the center, where it was black, and any other view
was blocked either by the film, or the hardware inside the glass) but
the bottom two could be "looked around" to see that there were two
ring-like pieces of metal, each one connected to a separate pin,
directly in line with the patches. Precisely what those two rings were
is obviously debatable, but from the "spray pattern", it seemed pretty
obvious that they had something to do with the film being there.

Any ring structure (especially if it connects to only one pin) is
almost certainly a getter induction heating loop.
In the
center area of each "patch", the finish was a nearly perfect (albeit
curved to conform with the shape of the tube's glass) mirror. The
further from the center of the patch you looked, the thinner the
coating, until it finally petered out at about 1.5 inches from the
center, leaving perfectly clear glass.

Care to take a whack at the "WTF-izzit!?!?" on that one? :)

I doubt I have ever seen any thing like this.
 
F

Fred Abse

Jan 1, 1970
0
Aside from three spots, most of it was crystal clear glass. The three
spots (one covered most of the top of the tube in a somewhat irregular,
but generally round, "splatter", the other two were smaller patches near
the base that were, allowing for the shape of the tube, pretty close to
perfectly circular) seemed to be related to some sort of pieces inside
the tube. The top one wasn't visible (the film was a near perfect
mirror, except near the center, where it was black, and any other view
was blocked either by the film, or the hardware inside the glass) but
the bottom two could be "looked around" to see that there were two
ring-like pieces of metal, each one connected to a separate pin,
directly in line with the patches. Precisely what those two rings were
is obviously debatable, but from the "spray pattern", it seemed pretty
obvious that they had something to do with the film being there. In the
center area of each "patch", the finish was a nearly perfect (albeit
curved to conform with the shape of the tube's glass) mirror. The
further from the center of the patch you looked, the thinner the
coating, until it finally petered out at about 1.5 inches from the
center, leaving perfectly clear glass.

Those rings are how the getters are deposited. If you look closely, you'll
see that they aren't solid rings, but a channel section, a bit like a car
tire cut down the middle of the tread. The channel is filled with a
mixture of barium oxide and magnesium powder. After the tube has been
evacuated, a high frequency induction heating coil is passed over it. This
"fires" the getter. The magnesium "burns", using the oxygen in the barium
oxide, giving magnesium oxide and metallic barium, which either evaporates
or sublimes, I can't remember which, and condenses on the relatively cool
glass envelope. This is the "mirror" you see. Because barium has an
affinity for oxygen (though not as great as magnesium has, obviously),
plus it has an affinity for hydrogen, which is also slowly evolved from
the internal structure of the tube ("outgassing"), it acts as a "chemical
pump", soaking up residual gas molecules. If the tube fractures, or
otherwise "goes down to air", the barium oxidizes rapidly, and the
"mirror" turns white.

BTW, barium and some of its compounds are nasty if ingested. AFAIK, the
oxide is relatively safe, but don't take chances with busted tubes, there
are other nasty things in there, too, like the cathode coating material.
 
F

Fred Abse

Jan 1, 1970
0
Tubes should have a nice silver lining inside the top. This is a very
reactive element that immediately "removes" any traces of air that
might leak in. If the vacuum is lost, then all of it reacts and it
goes black instantly.

No - it goes white. (metallic barium -> barium oxide)
 
D

Don

Jan 1, 1970
0
12au7 tubes cost less than $10, so don't replace with a 12ax7. A music store
or old-timer tv repairer shop may also have 12au7 tubes in stock. It also may
make sense to have a tv repair shop replace any electrolytic capacitors in the
unit, as these are the most likely components to fail. In fact, the tube may
have failed due to cap failure. Tubes also fail after many hours of use.

http://thetubestore.com/12au7types.html
 
L

Luhan Monat

Jan 1, 1970
0
Don said:
12au7 tubes cost less than $10, so don't replace with a 12ax7. A music store
or old-timer tv repairer shop may also have 12au7 tubes in stock. It also may
make sense to have a tv repair shop replace any electrolytic capacitors in the
unit, as these are the most likely components to fail. In fact, the tube may
have failed due to cap failure. Tubes also fail after many hours of use.

http://thetubestore.com/12au7types.html
Yo,

I used to fix TV and Stereo's all running on tubes. Tubes burn out (the
filiment) so you just plug in any that dont light up. The 12AX7 is pin
compatable with the 12AU7 but has a higher gain. As stated in other
postings here, the gain of any tube is never used to set the gain of the
ciruit.

So put the tube in and see if that fixes it. Unlike transistors, tubes
do not 'short out' and take other ciruit elements with it.
 
M

Michael A. Terrell

Jan 1, 1970
0
Luhan said:
Yo,

I used to fix TV and Stereo's all running on tubes. Tubes burn out (the
filiment) so you just plug in any that dont light up. The 12AX7 is pin
compatable with the 12AU7 but has a higher gain. As stated in other
postings here, the gain of any tube is never used to set the gain of the
ciruit.

So put the tube in and see if that fixes it. Unlike transistors, tubes
do not 'short out' and take other ciruit elements with it.

So, you've never seen a burnt cathode resistor? How about a smoked
plate resistor.
 
D

Don

Jan 1, 1970
0
[email protected] says... said:
Yo,

I used to fix TV and Stereo's all running on tubes. Tubes burn out (the
filiment) so you just plug in any that dont light up. The 12AX7 is pin
compatable with the 12AU7 but has a higher gain. As stated in other
postings here, the gain of any tube is never used to set the gain of the
ciruit.

So put the tube in and see if that fixes it. Unlike transistors, tubes
do not 'short out' and take other ciruit elements with it.
That's cool! So I can use 12au7 in my phono preamp instead of 12ax7?
And 12ax7 in my HHScott tube tuner where 12at7 are used? Changing the µ doesn't
effect the gain. Hmmm, how about oscillation? Why did they make more than one
tube type?
-Don
 
B

Baphomet

Jan 1, 1970
0
Michael A. Terrell said:
Luhan said:
12au7 tubes cost less than $10, so don't replace with a 12ax7. A music store
or old-timer tv repairer shop may also have 12au7 tubes in stock. It also may
make sense to have a tv repair shop replace any electrolytic capacitors in the
unit, as these are the most likely components to fail. In fact, the tube may
have failed due to cap failure. Tubes also fail after many hours of use.

http://thetubestore.com/12au7types.html




So, you've never seen a burnt cathode resistor? How about a smoked
plate resistor.

Wassa matta? You've got something against fried screen resistors? ;-)
 
Top