[R-390] 5963 vs 5814

[Charles P. Steinmetz]

Gordon wrote:

>I am confused as they both cross to a 12AU7 ?

One should always take cross-reference guides with a grain of 
salt.  There are different degrees of "acceptable substitute," 
ranging from "same tube but sorted for a certain parameter" through 
"intended to work the same but with some internal change that is not 
supposed to make them work differently," and "similar," all the way 
to "different tube design but should work in the socket if you have 
nothing else and you're desperate."  Sometimes the guides tell you 
which is which, sometimes they don't.  And sometimes they're just 
internet compilations put together by someone who doesn't know what 
he's talking about but once heard both tubes mentioned in the same sentence.

Take the 5814, for example.  It is not exactly the same as a 12AU7 -- 
it draws 175 mA per section of heater current, rather than the 150 mA 
the 12AU7 draws.  If you plug a 5814 into a socket designed for a 
12AU7, and the heater supply can't handle the additional load (or, if 
the heater is wired in series with another 150 mA tube), you have a 
problem.  (The series-heater problem would also bite you if you 
installed a 12AU7 into a socket designed for a 5814.)  So, the 5814 
is generally listed as "interchangeable with" the 12AU7, but it is 
not exactly the same and the differences could matter in some 
applications.  Consider also the 5751, which is often listed as 
interchangeable with the 12AX7 (it is sometimes called a "mil-spec 
12AX7"), but is in fact a quite different tube.

Jim gave a good account of why the "computer tubes" are 
different.  There is no guarantee, however, that nothing but the 
cathode coating is different -- manufacturers may have made other 
small changes to make them better suited to computer duty, as long as 
they were getting a new type number.  I do not know specifically why 
the plate voltage rating is lower on the 5963, but I would not assume 
it is for no reason.  [Perhaps the low-silicon cathode slurry is more 
sensitive to ion bombardment, or it boils off easier and contaminates 
the grid, or the electrode geometry was changed a little, and the HV 
must be reduced to compensate (each of these is pure speculation, but 
shows that there is a range of possibilities that could lead to a 
lower plate voltage rating).]

All in all, perhaps it is better to think of the 
industrial/military/computer variants as "based on a XXXX" rather 
than as "a XXXX," at least until perusal of the datasheets indicates 
that they are substantially identical.  Generally, the variants will 
work in a socket intended for the prototype, but the datasheets must 
be consulted to see if any ratings are different and the particular 
application must be vetted to make sure any differences in ratings do 
not cause problems.  As noted above, true even with the 5814, which 
most people think is "the same as" a 12AU7.

Always go back to the data sheets when this sort of issue 
arises.  And these days, it's just as easy to find the actual data on 
the internet as it is to find substitution guides.  Treat the sub 
guide as an index of which datasheets to pull if you don't already 
know which tubes you want to compare, then read and digest the datasheets.

Impressive collection of tube datasheets (there are many other 
collections on line, so search):
http://tubedata.tubes.se/

Impressive collection of tube manuals for download (again, there are 
others, so search):
http://www.bunkerofdoom.com/tube/man/

Best regards,

Charles