[R-390] R390 progress 6AK5/6AJ5

[Dan Merz]

Roy,  I'm going the right direction now.  Things are looking better.  I did
the 31 band exercise to start the 2nd oscill alignment using the scope
(wasn't able to see much voltage on the grid, low mv range so just kept
going using the scope).  Then I went down to 14 and 7 bands and both tuned
to very nice waveforms using the trimmer caps and 20 meters really came
alive.  I'll redo all this and the other bands when I get the oscillator
reassembled and the crystal switch hooked back up to the gear train.  I'm
still looking at the high voltage on the screen grid of osc #2 and think I
may have that cause figured out.  The grid leak resistor has aged to about
60 kohms in both oscillators (orig. 47Kohms) and is about the same in both
oscillators so that doesn't account for the difference between the two
oscillators.  I'll probably replace these anyway.  However the screen
resistor in #2 is 58Kohm (68K orig) and this would tend to make the screen
voltage higher than spec.  In osc #2 the screen resistor has aged to 158
Kohm (82 Kohm orig.) and this would tend to make the voltage lower.  I put
the 6AJ5 in osc #1 and it lowered the screen there compared to the 6AK5W.
The conclusion is,  the 6AJ5 I have shows lower screen voltage than the
6AK5W's.  I expect this difference will persist even after I replace the
aged resistors which I think are the cause of the different screen voltages
of the two oscillators.  The 6AJ5 tube I'm using may be a poor example in
making a general comparison of the two tube types since it didn't test as
good as the new 6AK5W's that I'm using.  Dan 

-----Original Message-----
From: Roy Morgan [mailto:roy.morgan at nist.gov] 
Sent: Monday, May 02, 2005 2:55 PM
To: Dan Merz
Subject: RE: [R-390] R390 progress 6AK5/6AJ5

At 05:33 PM 5/2/2005, you wrote:
>Roy,  much thanks for the comments.  I concluded from your comments 
>that the multiple peaks that might occur with T402 adjusted with too 
>much inductance are due to occurrence  of various harmonics as the 
>trimmer on
>31 is swept.

Dan,

Most likely, but I would have to mess around with it for a while to be more
sure.

>The goal is to have only the right harmonic there and low enough 
>inductance that the others can't be produced,  and only one (at least 
>according to the manual procedure).  I'll go with that interpretation 
>and see what falls out when I start alignment.

Ok good.

>   The VTVM measurement in the manual is made on the grid of the mixer 
>tube - the oscillator signal is injected to the cathode.

Ah, I have not read that manual for some time.

>The Engineering Report on the 390 says about 10 to 15% of the injection 
>shows up on the grid so I guessed that's what was being measured at the 
>test point specified.

Likely.

>  However,  in rethinking this,  the VTVM measurement is a dc 
>measurement of negative voltage on the grid so probably reflects how 
>the bias on the grid peaks as the various harmonics are produced by the 
>trimmer sweep and not really an integration of the rf voltage that I had in
mind.

Yes, the mixer produces bias from the oscillator injection voltage.  It's
like a rectifier.

>I've been measuring at the cathode with the scope so far and using the 
>10x probe on the scope.

Good. that should give you good results.

>  I don't have a vacuum tube VTVM,  but do have a 1 gigohm input 
>impedance "electrometer" that I built for looking at grid voltages.  It 
>should work fine as long as we're talking about dc voltages.

Not if the lead from the end of the test probe is a shielded wire with lots 
of capacitance.  (RG-58 is something like 20 pF per foot.)   Put a one meg 
ohm resistor at the probe end and stick that into the test point to see if
you get the same results.

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 --