[R-390] R390 RF Cap Confessions

[Roger Ruszkowski]

Fellows,
 
Not only have I been replacing the black or brown plastic tubular paper coupling and bypass caps
referred to as black beauties, but some times I have installed new caps of different values.
 
Many of the caps were selected as from a preferred parts list of items in the supply chain
and selected to maximize common values for procurement discounts  and simplify construction.
 
I have been using 0.001 UF for the 5000 PF by pass caps. 
I use a 0.02 or 0.022 UF for C553 the blocking cap on the 1st IF stage.

I use 600 volt caps.
I do not know why but I find the 600 volt caps have a better signal to noise ratio when I am done.
Maybe its just the luck of the caps I was comparing back when, but the few times I have had
a chance to install a 200 or 400 volt cap and then replace it with a 600 volt cap, the higher the
voltage the better the signal noise ratio was. 

The difference is not dramatic use the lower voltage if its what you have.
Use the original value if you feel the need.

But if you need to buy a dozen go for over 400 volt to 600 volt and bump the value up.
I use a factor of 20 on the by pass filter caps and a factor of 2 for the inline signal passing DC blocking caps.
I think on the inline blocking caps the higher voltage yields less popping so the meter needle has less jitter.
You do get some stage gain so you need to be careful. You can get oscillations. You do let more noise pass 
with the signal so consider if the extra noise is raising the noise floor or if you are working in a stage
where the extra noise is still masked by other stages.

You get to back the IF gain down after you increase some inline values  

All subjective, as a sample of a few is not a significant statically meaningful value.

I can not say that my final signal to noise ratio has improved because I just replaced 
a bunch of old poor caps or because I replaced a bunch of old caps with more filtering or both.

Exact before and after measurements of each cap was not done.
And the few IF decks I have done is not meaningful as statistics.

The decks would not do 27 to 1 when I started and changing tubes
did not get me to the signal to noise ratio. But after doing a change
off all the usual suspects, I have been able to get every IF deck up to 27 to 1
and the receiver up to a good solid 20 to 1 from end to end on any and every frequency.

Roger Ruszkowski AI4NI