[HBR] G2DAF Receiver coils

[Walter A. Hutchens]

... are still being posted on eBay.

The latest auction is at:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5854928750&
ssPageName=ADME:B:SS:US:1

These are for the Mk 1 only -- either cascode front end, or 6BZ6 -- not the 
Mk 2 with the push-pull circuit.   

Evidently quite a few chaps are serious about building this receiver because 
there've been several coil sets auctioned (seller took some time off at the 
holidays) but the last selling price was about $90.  

This is a fine design, easily the equal of the best commercial ham sets of the 
time and I think it would beat them all if you substituted a 6EH7 RF stage.   
Weaknesses: I'm not convinced it's capable of the warm-up stability the 
author claims as the LO is undistinguished and operates at 5 Mcs.   A push-
pull front end (as in the Mk 2) would handle large signals better.   A single 
conversion design would be better for large signal handling.

The only competitor I'm aware of is the W5OMZ receiver: single conversion 
(premixed) using a 7360 beam tube mixer.   But he stuck a 6AZ8 sharp-cut 
off pentode ahead of it!   Also the use of a single pentode premixer and a 
so-so mixing scheme gave him some spurs in every band.   

He used miniductor for the front end coils on the lower bands; that would 
partially make up for the poor choice of an RF stage by reducing the off-
channel signal strength.   Toroids would be even a bit better.

This one has much cleverness but also several weaknesses like that RF stage: 
Yes, it worked but performance was unnecessarily compromised.

I keep trying to combine the best of all the designs but I also keep making 
innovative serious design mistakes.   Maybe someday ...
 
The G2DAF Mk 2 used a less-than-unity-gain RF stage in order to have a 
place to apply AGC.  It's perfectly possible to apply AGC to a push-pull 
mixer.   Using toroids you can get tolerable off-band rejection with a single 
coil.   Presto!  No RF stage required, no front end tracking, accurate 
balance (for rejection of IF feedthrough) becomes easier to achieve.

These ideas are working fine in a seven tube (1 rectifier) 'transformerless' 
80-40 band imaging receiver.   It's an excellent performer, with the only 
problems being just a tiny bit of hum (really not noticible, but annoying!) 
and I haven't yet been able to balance the oscillator injection to reduce the 
spur from MY conversion scheme.

Walt
KJ4KV