[R-390] tube testers

[pete wokoun, sr.]

>From: Barry Hauser <barry at hausernet.com>
>To: Kenneth Arthur Crips <CRIPS01 at MSN.COM>, r-390 at mailman.qth.net,bernie 
>nicholson

and there can
>be errors in the tube charts.  The manufacturers did not necessarily 
>calculate the correct minimum reading for every number.  Many of the 
>entries were probably determined through extrapolation/interpolation, if 
>that.  (Or they cheated and used a conversion factor on the values from 
>some other tube tester.)


I can tell you folks how I recall we college co-ops did it back at Hickok...

When the tube manufacturers introduced new tube types, 3-5 tubes were 
normally received along with the new design specs.  Each of these new tubes 
was set up on a super lab tester at exactly the published specs to determine 
precisely how each one compared to the published specs.  For example,  one 
tube might have 104% of the rated Gm while another might have only 97% of 
the rated Gm.

I believe Hickok's testers usually had a fixed plate voltage but some could 
be set within limits.  It was the bias that was adjustable with a front 
panel control.  When one of these new tubes was set up in a particular tube 
tester, the bias was adjusted to approximate the plate current that was used 
in the manufacturers determination of Gm.  When you had about the right 
plate curent the shunt control was adjusted so that a tube with 100% of a 
new tube Gm would read exactly 2/3 full scale on a fail-?-pass type of 
scale.  The bias and shunt readings that you set up would be what appeared 
on the next roll chart.  If the tester had calibrated Gm scales, the bias 
would be adjusted until the meter read exactly what the lab standard read 
for that tube.

Note that the tube would only have to drop to about 70% of its new tube 
value before it reached the bottom of the "?" area and start to be rejected. 
  That why a lot of tubes that test in the fail or reject part of the scale 
still work in equipment.  Remember, tube manufacturers wanted to sell tubes 
and it was better to err on the side of conservatism (better to throw away a 
still usable tube than to keep a bad one).

When I was doing this back in the middle 60s, there were only about 3-4 
models that had their roll charts upgrated with the new tube types.  There 
was also that cardmatic one that got the new types.  If you wanted data on a 
tube for an older model that didn't have it on its roll chart all you had to 
do was call customer service.  I remember dragging those oldies out of the 
'morgue' and getting the settings for a customer or two.  A new cardmatic 
model was also being develolped for the military but I don't think that ever 
got pass the prototype stage.  I remember doing the layout on the power 
supply pc boards for those.  By that time tubes were on their way out.

Introducing new tester models was a nightmare...you had to go and verify 
every tube type with the new tester settings.  I spent many an hour doing 
that on their 'mustang' model.  Kind of boring but lots of overtime.

enough rambling....
pete, KH6GRT