[HBR] Another Receiver Project -- HBR-4, Part 2

[waltah at earthlink.net]

Gee, just *sailing along* -- considering that I only have a couple of 
hours a day to work on it. 

The three IF stages are wired and test okay -- only modest 
problems with instablity.   The plate detector for AM is fine, ditto a 
6BN6 beam discriminator tube as a product detector.   And the 
audio amp -- 6U8 driving a 6J6 -- is okay.   The function switch -- 
LSB-USB-CW-AM is working.   Probably will add an audio tuned 
circuit for CW selectivity at a later time.   

With the S-meter at half scale the audio amp overloads at about 
1/3 of full volume.   Have to use a scope to look at what's 
overloading; I probably have to reduce the overall audio gain by 
increasing the feedback.  This is the RCA tube handbook 15-watt 
hi-fi amp, adjusted to a 6U8 and 6J6, with overall voltage gain set to 
10.   

Or the problem could be too-high output from the IF stages -- 
around 100 mV is right.   If that's it I need to increase ratio of the 
AGC voltage to detector signal -- most easily done with a voltage 
divider at the last IFT output.

The output of the AM plate detector is about 2x the voltage of the 
product detector for the same input.   Adjustment of the plate load 
should take care of that.  

The 6BN6 circuit is excellent; not even close to the complexity of a 
beam tube product detector, conversion voltage gain of about 50 -- 
it's a much better match to a plate detector (for AM) than the usual 
triode or diode product detector.   The authors I followed 
(W9BIY/W9IHT, 'Some Ideas in a Ham-Band Receiver; SSB for The 
Radio Amateur, 4th edition, 1965) claim 40 db rejection of AM, 
meaning really great detection linearity.  Plus the tube limits 
sharply at not too much more than usual signal levels, meaning 
you get something of a free noise limiter.

That article is one of the best I've seen for discussing how-to with 
technical details, rather than just telling you the radio is great, be 
sure to leave plenty of room in the chassis, use small wire ... The 
receiver's not bad either -- plug in coils a la Crosby and a pretty 
straightforward *single conversion* (4500 kcs IF with a lattice filter) 
design.   Replace the AM detector (diode) with a plate detector, get 
rid of the hang AGC circuit, and substitute better tubes (6AK5 RF 
stage ...) and you'd have a most worthy HBR project with about 12 
tubes.   

Well, it needs a calibrated dial.   The writers used a NPW-0 dial as 
on the HROs.   

Anyhow, this stuff's fun when it goes well.  Maybe I'm not making 
so many mistakes as on previous projects.   (So far only one tube 
filament wired wrong ...)   Or maybe something in the front end will 
prove to be impossible?   Should know in a couple of weeks!

Walt 
KJ4KV