[Hammarlund] HQ-110 stability, or, making a silk purse out of a pig's ear,

[Kenneth G. Gordon]

I spent some time last night studying the schematic and the manual 
for the subject receiver and discovered a few interesting things, 
most of which were NOT covered in the manual.

First of all, the 0B2 regulator is connected not simply to the HF 
oscillator, but also to the screen of the 1st IF amplifier and to both 
"screens" of the 1st mixer, the 6BE6 V-2, AND is fed to the RF 
amp's AGC rectifier, providing "delayed" AGC voltage for the RF  
stage.

Thus, any problem with the regulator would also effect the mixer, RF 
amp, and 1st IF amp. Conversely, any problem with the mixer, RF 
amp, the RF amp's AGC rectifier, and/or 1st IF amp tubes, could 
also cause a problem at the HF oscillator.

Secondly, the BFO is NOT regulated at all, but receives its plate 
voltage directly from the output of the second HV filter cap. I 
suppose that due to the NEXT issue, the BFO output must have to 
be at a fairly high level, and couldn't run effectively from 108VDC 
regulated. It also probably draws too much current for the 0B2 to 
handle.

Thirdly, and to me most interesting, is that the 2nd IF amp, the 
pentode section of a 6AZ8, is also used as a "linear detector" (since 
when is a detector or, in this case, a mixer "linear"?) for CW/SSB, 
the BFO output being fed to this tube's grid. 

Audio from the plate of this stage, when in the CW or SSB condition, 
bypasses the last IF transformer's two tuned circuits, and is fed 
directly to the grid of the first audio amp.  This reduces CW/SSB 
selectivity, and makes it impossible to use the AGC and "S" meter 
on CW or SSB.

This also reduces the available audio level, a fact mentioned in the 
manual. The manual says to "turn up the audio gain a bit..." on 
CW/SSB.

The 2nd IF stage obviously functions as a sort of product detector 
for CW or SSB.

Lastly, the AGC is tuned for AM, being a fast-attack, fast-decay 
type, which is NOT suitable for CW or SSB.

It is very interesting to me what Hammarlund engineers did with this 
receiver, accomplishng a lot with a little.

However, I feel that there can be considerable improvement made 
to the operation of this receiver with a little effort.

I am considering doing the following:

Install a REAL dual-triode product detector which will take its IF 
signal from the output of the last IF transformer, thus preserving as 
much selectivity as possible, and enabling the use of AGC and "S" 
meter in CW/SSB. 

I would use a 7963 subminiature very hi-mu dual-triode since it can 
be mounted on a terminal strip below deck.

I would remove the BFO signal being fed to the 2nd IF stage and 
connect the BFO to the new product detector, adjusting the BFO 
output for the proper level. 

Then I would disconnect the audio signal to the 1st audio amp from 
the 2nd IF, connecting the audio signal from the separate product 
detector in its place at the mode switch.

I would leave the AM detector and the two AGC rectifiers in place.

I would change the MAN-AGC switch to one with at least one more 
position so that I could enable fast and slow AGC, adding a 
capacitor to the AGC circuit where appropriate.

The regulator circuit must be looked at carefully. 

I may use some strategically placed zeners of the proper values. For 
one thing, I think the BFO voltages should be regulated, and I hope 
to be able to separate the regulated voltage sources from one 
another to preclude interaction.

Well, so much for that.

Ken Gordon W7EKB