Re: Harmonic performance - SSB vs CW

Howard Fidel

For my power amp for 20 meters I built a 5 pole Chebychev  filter-2 -0.9 uH inductors 270 pf for input and output caps and 430 pf for the center cap. This is a 75 ohm design, my dipoles are closer to 75 ohms then 50. At 42 MHz the filter has a theoretical  insertion loss of 60 dB. With a 2 pf cap from input to output the insertion loss drops to -57 dB. If the relay's capacitance is around 2 pF from input to output in the board, I seriously doubt this is the root cause of the non compliance issue. Something else is amiss. I am going to be working on these issues over the next few days and will report my conclusions later.


On 8/4/2018 9:10 PM, Gordon Gibby wrote:
Looking closer at what you folks (Warren the first to report...) have discovered

E.g.:  “In my testing I also found the isolation across one relay was barely 45db in one configuration and 50 in another.Using one relay to switch both ends of a filter will compromise the filter “

It may be worth it to now look at the physical construction of the relays first.   Perhaps there is a way to insert a shield between sets of contacts. I have it disassembled any of mine yet and couldn’t find an image but the construction might be similar to this: 


On Aug 4, 2018, at 20:07, ajparent1/KB1GMX <kb1gmx@...> wrote:

Tim, no on the Diz amp.

First mine is linear (class AB1 150ma per irf510), the base amp was class C. 
About all I did use was a few passives and the board itself. Used a differnt wind for
all the coils going from bipolar to mosfet  FYI the results were at 13.6V.   At 20V
it does a nice 16W and the 10M power is 10W.  Power supply runs out at 20V
and I didn't need that much anyway but had to turn it up and see.   Due to the
really good RF layout of the board it has proved to be stable and blast proof.

Years of design and build I have a lot of unobtainium and some exotica.
And a junk box with stuff that dates back to the 60s.  I lived in Grumman
LM country.


Join to automatically receive all group messages.