Tube experts! - I T WORKS!

From: Christian Fandt <>
Date: Sun Dec 5 22:12:28 1999

Upon the date 06:07 PM 12/5/99 -0500, William Donzelli said something like:
>Here are a few comments about tubes, as there seems to be some
>misinformation floating around here and elsewhere...
>> In TVs where the tubes are low end and not always run at their best
>> operating points that may be true. In Qualtity test equipment that is
>> rarely the case.
>This is a rather common misunderstanding. In truth, the tubes from the
>1960s - the ones that most people call junk - are amongst the best. The
>quality of manufacturing (improved outgassing and getter/keepers) was far
>better than those made in the 1940s, bulbs shrunk down to submini sizes
>(Nuvistors, T3/4 bulbs, and the ceramic tubes), and the electrical
>performance went thru the roof (transconductance). Many tubes started to
>be made in Japan and elsewhere, and are very good performers, but with a
>very bad reputation.
>Color TV service is very demanding on tubes (basically the only market
>left in 1960), so the tubes were made tough. One must remember that some
>of them, namely in the sweep circuits, were essentially transmitting tubes
>with new bulbs and receiving numbers.
>> Mechanical shock and power cycling tends to ruin tubes faster. Generally
>> tubes have good lifetimes and can be considered reliable (excluding the
>> effects of heat on surrounding parts) .
>While mechanical shock certainly kills tubes, power cycling generally
>does not. The filaments do not instantly get heated like a household
>lightbulb - it is far more controlled and the filaments have time (and
>room) to expand. In some of the older tubes, this was done with springs,
>and on modern minis and Compactrons, the filament is allowed to slide
>inside the cylinder of the cathode. Even some of the tubes rated for
>mobile service with quick heating filaments to not really care much about
>power cycles.
>> That is true, usually they get filiment failures and those are easy to
>> spot. Power cycles tend to accelerate that. then again I had a tube
>> organ (some 80+ tubes) and only had one failure in 8 years.
>Most tube failures are due to loss of emmission, rather than a burnout.
>The most common failure is that the filament loses the outer coating
>(generally thorium based), and emission drops to half, maybe zero. There
>are tricks to rejuvenating filaments, but it is mostly magic and luck -
>trying to get trapped thoria to diffuse to the surface.
>Soft vacuums (gas in the tube) also is a common failure mode, as the
>glass-metal seals just are not perfect things.

Grid Emission is another failure mode less talked about but found more
often than most repair folks realize. Takes a special tube tester or one of
the very few dynamic mutual conductance tube testers marketed which will
test for this (my Jackson 658A is an example). Caused by sputtering of
cathode material onto the control grid as a result of overdriving of the
tube or the heater voltage too high or an operating environment which
allows the envelope temp to rise very high. Exhibited by unexpectedly odd
or just plain poor performance of a circuit even though the tube checks
perfectly okay on any tube tester that doesn't check for grid emission --or
grid current as Jackson names it. Grid current increases spontaneously
driving the tube out of cut-off or even harder if it operates normally out
of cut-off curve and increases cathode current into a high range which
accelerates the sputtering problem.

I feel tubes in the early computers probably suffered from this as some
were driven hard-on under normal circumstances, say, in a flip-flop circuit
where the circuit was held idle in a particular state for long periods.
Design of the circuit was likely conservative but over time grid emission
problems developed. These machines normally ran 24x7. (Actually keeps this
thread leaning more on-topic ;-) Folks who went through and tried to
troublshoot a defective circuit and saw the tubes had checked "excellent"
and didn't simply swap for a new ones would probably send themselves off
on a tangent to find some other problem within the circuit.

Anyway, Mr. Donzelli is, like myself, one of our rather quiet tube
historians in the AWA and knows even more than I do on the subject. His
clarifications above are a few of the ones I was fixing to post. Circuit
designs of the tube-type Tek scopes are usually never such that tubes will
drastically degrade in periods less than tens of years. Pretty conservative
designs they made.

Only in the last years of tube production ('70's - '80's) has quality been
observed to degrade in the TV and radio type tubes. Far less quantities
where made, because of drastically declining demand in the replacement
market, which meant that testing criteria was lowered slightly to maintain
a reasonable yield vs. cost to produce them.

Now, only a handful of the hundreds of types of tubes are made anymore
-mostly in Russia, Hungary, Yugoslavia and China and are aimed toward the
audio market or high power radio/TV/military transmitters.

>Going back into hiding...

Nice to see you again ;-)

>William Donzelli

Regards, Chris
-- --

Christian Fandt, Electronic/Electrical Historian
Jamestown, NY USA
        Member of Antique Wireless Association
Received on Sun Dec 05 1999 - 22:12:28 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:31:54 BST