[R-390] Fixture for finding low-noise 6DC6's?

[Roy Morgan]

At 09:48 PM 12/1/2005, Tim Shoppa wrote:
>Is there such a thing as a test fixture or something I can lash up
>to find the lowest noise 6DC6's of my collection?

Tim,

Yes, you could build one: Take the front end of an R-390A, hook up power 
and mechanicals, feed the IF to another radio and do measurements.

Of course you should not bother doing that. Just use your radio as it is.

>  Or is the only real
>test putting it in the R-390A itself and measuring S+N/N (probably
>with the GAIN ADJ for each one, I guess.).

It's not the "only real" test, but it certainly is the easiest and most 
practical.

The IF Gain Adjust should not have too much effect as you find the lowest 
noise tubes, as long as it's not set too high so the IF strip noise masks 
the RF section noise!  If when you pull the RF tube out, the noise in the 
speaker does not get lower, you have some work to do before you test for 
lowest noise tubes.

Roger has posted the way he and his fellow R-390A technicians selected 
tubes for best noise ratio.  There is also a very long collection of 
similar articles, including Rogers, posted on the net.  It's called 
sensitivity-alignment.pdf and came from Wu Li's Pearls of Wisdom page, 
which I cannot find just now.    In the past, I got it from the Frequently 
Asked Questions page, and even though I now can't find it in the web page, 
it is there:
<http://209.35.120.129/Pearls/sensitivity-alignment.pdf>

In summary, I suggest you get an AC voltmeter and resistor load on the line 
output, a signal generator with some modulation and a 3- or 6- or 10- db 
pad at the receiver input.  You are looking for the tube(s) with the lowest 
signal to noise ratio, NOT just the lowest noise. (A really dead tube will 
give you very low noise, but no signal either.)

Notes:

General Radio and others made "Output Power Meters" that are very useful 
for this job. The GR types are 583 (low power), 783(high power), and the 
most recent is the 1840A. These things present a wide range of resistive 
loads and a meter to measure power delivered.  There are a couple versions 
of a military meter also, some TS- number I can't recall.

Any AC meter that will measure the audio output level can be used (with a 
resistor for a load) - a dB scale is useful but not necessary.  An 
oscilloscope will do the job.

Pulling hot tubes out of the chassis is less painful if you have a tube 
puller of any kind.  Some rubber roller "tires" from the copy machine 
repair person work very well, the wire kind recently discussed on the list 
work fine, rubber covered chemistry lab grabbers work fine.  It would be a 
shame to drop and break the hot rf tube you just decided was the very best 
in your collection.

Those rectangular carpet samples from the rug store work well on the bench 
to help prevent breaking and losing stuff and save paint jobs on radios.

Happy Noise-Figuring, all.

Roy

- Roy Morgan, K1LKY since 1959 - Keep 'em Glowing!
7130 Panorama Drive, Derwood MD 20855
Home: 301-330-8828 Cell 301-928-7794
Work: Voice: 301-975-3254,  Fax: 301-948-6213
roy.morgan at nist.gov --