[R-390] Lankford Filter Mod

[Charles Steinmetz]

Perrier wrote:

>I would like to see that with your skill set you would design a plug 
>in DBM first mixer replacement using the LM 1496 for the *A*.  I 
>think that it would be a quantum leap improvement.

I'm not sure a "plug-in" replacement is practical.  The first 
challenge with adding any SS circuitry to a 390/390A is getting a 
"ground" reference into it and providing low-voltage DC power for 
it.  We have seen any number of designs that create LVDC supplies 
from the resources already available inside the radio, but none of 
these is a solution I'd adopt for various reasons (hum, excessive 
loading of power transformer, etc.).

I would specify an additional power transformer to run the DBM, 
making it not strictly a "plug-in" replacement.  There is also the 
question whether it could be packaged (complete with any necessary 
heatsink) in the space available over and around the mixer tube 
socket.  The power and ground could go to a connector at the top of 
the unit, so if it would physically fit the available space it could 
be "plug-in plus."

There would still be things that could be done better if some further 
mods to the radio were allowed.

Note that high dynamic range active mixers and post-mixer amplifiers 
draw quite a bit of current and can be sensitive to noise on the 
power supply, so the power supply needs to be beefy, well regulated, 
and low-noise.

I would certainly not use an LM1496 as the mixer -- the best you 
could do with that at the levels required would be at least 20dB 
noisier than any good design.  A passive diode DBM looks attractive, 
but it would need a post-mixer amplifier, so we're back to the "not 
really plug-in" issue.  Adding a diode DBM and using the existing 
mixer tube just for post-mixer gain is a possibility, but it's not a 
"plug-in" solution.

>Please remember, the roofing filter mod was never stated as a cure 
>for other R390A problems. It maybe that to achieve its best 
>performance enhancement other mods need to be implemented.

My concern is that it doesn't cure ANY problem that simply using the 
4kHz filter doesn't cure (or replacing the 8 or 16kHz filter with a 
narrower filter).

>There are over 10 authors that I'm aware of whom over the last 20 
>years who have made design improvements.  Some major, some minor.

Well, changes, yes.  Not all "improvements," by any means (judging by 
the random selection I've seen, not very many are net 
improvements).  Anyone contemplating making such changes should do 
their own careful design evaluation to see if the mod being 
considered is likely to solve any real problem(s) without creating 
any worse problems.  But not everyone who plays with 390s has the 
engineering skills to do this, so when I see a mod that, based on my 
design evaluation, (i) would likely solve no real problem, (ii) would 
likely cause worse problems than any real issue it may help, or (iii) 
is not the best way to attack the problem, I think it is appropriate 
to speak up.

>I have already released a two part schematic that incorporates most 
>of those upgrades for those who want to join the dark side of 
>excellent performance.

Again, anyone contemplating making changes should do their own 
careful design evaluation to see if the mod being considered is 
likely to solve any real problem(s) without creating any worse 
problems.  If they have questions about the wisdom of any mod that 
they cannot determine for themselves, they should feel free to ask on 
the list.  Making a change and observing how it works "on the air" is 
a useful process, but it takes thorough and systematic testing to see 
what other aspects of performance may have been compromised or 
degraded by the change -- things one could easily miss just by using 
the radio on the air in average conditions.


Best regards,

Charles