Re: Crystal CW filter with Version 4 uBITX boards? #ubitx

Jerry Gaffke
 

Exactly how do the transformers ruin this?
The uBitx transformers are broadband.

I was imagining somehow holding both filters to 200 ohms, using only one set of transformers. 
That might be tough to do and still get the desired filter characteristics.
Though a CW filter may not need such a flat top.

Might be possible to have two sets of transformers in parallel.
The transformers don't have a fixed impedance, they just establish an impedance ratio.
So a high impedance into the filter should mean much higher than 50 ohms into that transformer.
Though all sorts of ways for this to trash performance.

I'd be fine with pluggable filters.
But am curious where dual filters would fail.

As your 45mhz-DialFreq spur shows, things can quickly go south in unexpected ways.
Measurements are  useful.
Even if it seems to work.

Jerry 

 


On Sat, Jul 7, 2018 at 04:51 pm, ajparent1/KB1GMX wrote:
The big difference is the lack of transformers it actually helps for that.  That version of the
bitx runs the amp interfaces at 200ohms.  In that case the unusd filter is not likely to
look like a rf short.  I wonder if the ubitx can be talked into running without the
transformers for the filter.  Either an L-network or none at all.  It may impact the
filter bandwidth and shape factor some.  I may test that for a single filter.

It should work but the choice of filter frequencies needs a little consideration to
avoid bleed though and spurs.  Things like non harmonically related and not
too close to the same frequency.

Join BITX20@groups.io to automatically receive all group messages.