[HBR] HBR2K Chapter 7 -- More cleanup detail, some progress

[hbr@mailman.qth.net]

Previously the HBR2K project got as far as a working second mixer 
and stable IF's with easily duplicated transformers.

In the last couple of days I tackled the tunable IF.   I had chosen the 
rear gang of the two gang tuning cap to tune the VFO and had that 
working, tracking the dial, etc.   It now became clear that the rear 
gang would be best for tuning the IF so I switched the VFO to the 
front gang and went all through the dial tracking process again.   It's 
not that hard -- the error is only about 5 kcs so you can get under 
500 cps at the 100 kcs points in an hour -- but that is an hour of 
serious concentration.

I picked a coil that looked likely to be about right to tune 5520-6020 
from the archives, made a rough calculation of the required shunt 
capacitor based on delta C for the 500 kcs delta F and darned if it 
wasn't spot-on -- the junkbox coil was perfect with the slug near mid-
position.   With the parts installed, I was able to listen to Catholic 
Global Radio and World Harvest Radio with an antenna draped 
around the first mixer tube.   Oh, and the New York ATCC talking to 
international airliners and lots of foreign stations.

Cleanup time -- filament circuits were untidy and I wanted to filter the 
line to the front end, I needed to run an AGC line to the front end 
section, and I wanted to switch in another cap on the AGC line when 
on SSB.   A day's work, counting errors.   We will not say how long 
it should have taken.

About half the time now goes to redoing things that were either poor 
judgement or just mistakes.   But 90% of the wiring is done ... it 
won't get any worse.

I had leakage of the sideband injection oscillators (3181.5, 3178.5) 
into the IF, causing a small steady AGC voltage.   Poking around, I 
discovered that the leakage was due to direct coupling from the 
crystal trimmer caps to the adjacent second IF stage.   A shield cut 
from a tin can made the problem disappear.

On to the front end proper.   Had to modify a sheetmetal partition on 
the recycled FT-101 bandswitch to clear the RF stage socket.   Then 
I went one by one through the leads attached to the bandswitch, 
figuring out where each would go in the new (vacuum tube) world 
rather than the old transistor one.   The leads got extended and 
tagged.   

Next came some more coils.   Although the splendid slug tuned cup 
and core preselector will be retained, the FT-101 switches an 
additional coil in series with each of these on 80 and 160 meters.  
There really isn't space to remount those coils in the original 
location.   In addition, two traps are needed to control breakthrough 
of the first intermediate frequency -- unless I really want to hear 
World Harvest Radio along with the ham 40 meter band.   One 
existing trap could be relocated; for the other I was able to recycle 
(with retuning) a trap that is already in the bandswitch assembly and 
was used only on transmit.   The plate coil of the 1st oscillator was 
another "not quite enough room for that" item that would have to be 
wound.

I started with the 1st oscillator plate coil and went on to wire that 
circuit and get it working.  Since the crystals are itty-bitty HC-25 
units the big concern is overdriving them.   With a 12AT7 Pierce 
circuit, I tacked in a crystal and tried it out.   No oscillation with a 
330k plate resistor (26 volts on the plate) but the GDO behaved very 
strangely when tuned past the crystal frequency.  220k did the trick --
at 44 volts on the plate, the circuit oscillates nicely at 27,021.076 
kcs --just a kc high and perfectly stable, indicating that there's not 
excessive drive.  Grid voltage about -3.5 volts so there's plenty of 
signal.  

The practical problem with the 1st oscillator, mixer, and RF stage is 
that they are located in the bandswitch area and while they can be 
reached for post-construction service, access isn't good enough for 
easy wiring.   So they have to be wired and tested as far as possible 
before the bandswitch assembly goes in.  

Recycling the FT-101 coils sure simplifies things.   As things stand 
now, the hbr2k requires the winding of just four coils -- two small 
single layer 1/4" slug tuned ones, and two (probably) torroids.   
Essentially all the trimmers are part of the bandswitch assembly and 
are prewired.

I have done some of the RF stage and 1st mixer wiring.  Next steps 
are to continue on the front end.  Things have been way too easy the 
last ten days.   However, with luck I should be able to install the 
bandswitch and try out the whole signal path within a day or so! 

Walt Hutchens  
KJ4KV