[HBR] HBR Digest, Vol 89, Issue 2

Fri Jul 29 13:04:33 EDT 2011

Go Walt!!
I bought the 19JN8's a few years ago.
Chuck, W9VZR

-----Original Message-----
From: hbr-bounces at mailman.qth.net [mailto:hbr-bounces at mailman.qth.net] On
Behalf Of hbr-request at mailman.qth.net
Sent: Friday, July 29, 2011 11:01 AM
To: hbr at mailman.qth.net
Subject: HBR Digest, Vol 89, Issue 2

Send HBR mailing list submissions to
	hbr at mailman.qth.net

To subscribe or unsubscribe via the World Wide Web, visit
	http://mailman.qth.net/mailman/listinfo/hbr
or, via email, send a message with subject or body 'help' to
	hbr-request at mailman.qth.net

You can reach the person managing the list at
	hbr-owner at mailman.qth.net

When replying, please edit your Subject line so it is more specific
than "Re: Contents of HBR digest..."

Today's Topics:

   1. The long, SLOW HBR project (Walt Hutchens)
   2. Re: The long, SLOW HBR project (Bill Cromwell)

----------------------------------------------------------------------

Message: 1
Date: Thu, 28 Jul 2011 13:29:50 -0400
From: Walt Hutchens <waltah at earthlink.net>
Subject: [HBR] The long, SLOW HBR project
To: <hbr at mailman.qth.net>
Message-ID: <CA57129B.4EE3%waltah at earthlink.net>
Content-Type: text/plain; charset="US-ASCII"

A few years ago I set out to build a receiver loosely along W6TC lines but
with some changes and updates.   I reported on early progress a number of
times but as other things overwhelmed my hobby time, the project slowed to a
crawl.  I remember estimating that finishing up the next couple of circuits
might take another year or two.

I'm happy to report that things have temporarily swung the other way and
with perhaps twenty hours of time over a few weeks the project has moved
forward considerably.

This is a 'transformerless' receiver -- series string filaments, high
voltage by rectifying the AC line.  Single conversion with a 1700 kcs IF.  A
half lattice crystal filter provides selectivity.   No solid state devices
are used.  Plug-in coils.  The tube line up is:

19JN8 (pentode-triode) RF/3500 kcs band edge marker
12BZ6 Local oscillator
19J6 Mixer 
19JN8 1st IF/negative HV rectifier
19JN8 2nd IF/AGC plate detector
12AX7 Audio detector
19JN8 1st Audio amp/BFO
117N7 2nd Audio amp/positive HV rectifier

Basically everything now works.  The most interesting of the recent round of
problems/fixes was roughness on strong SSB signals caused by locking/pulling
(to the IF signal) of the BFO.  This was solved by adding a 'gimmick' of
about 1 mmf to cancel out the leakage of the signal into the BFO circuit --
think in terms of a neutralizing cap.

Another interesting one was hum modulation of strong carriers, audible with
the audio gain turned fully up and causing a slight warble on beat notes.
This was solved by AGC circuit, IF transformer, and IF bypassing changes.

Remaining problems/needed features:

1. Intermittent complete failure with a loud rushing noise replacing normal
signals.  This is most likely a front end VHF oscillation since tuning
receiver causes loud crackling/popping noises indicating high current
flowing in the tuning cap: The 19J6 mixer is a good bet since this tube is
good to 600 mcs as an oscillator and the problem isn't changed by disabling
the RF stage by flipping on the band edge marker.

2. Slight buzz audible with no signal and the audio gain fully up.  It's 60
cps + harmonics: Maybe from the line?  Switching of a rectifier modulating
stray RF?   This may take some sleuthing.

3. Slight irregular random walk frequency instability; frequency counter
points to the BFO.   Not noticed on signals but with the (crystal
controlled) band edge marker tuned in for a low beat note you can hear it.
This sort of thing is usually a bad capacitor or (equivalently) one operated
at higher than rated voltage.

(A crystal controlled BFO would avoid some problems but at 1700 kcs I think
you'd need a crystal for each sideband.   Possible option: use just one and
place the LO above the band on 40 so for normal ham practice the sideband
layout in the IF is always USB.)

4. Shift of the LO tuning by perhaps 20 cps on very strong signals.  Almost
certainly due to (unregulated) LO HV change caused by the varying load
imposed by the AGC controlled RF and IF stages.  Probably fixable by changes
to the HV rectifier/filter circuit.

(Modern 120 VAC lines are better regulated in many areas than all but the
best of VR tubes could do.  Such tubes also add substantially to the power
consumption/heat production; total power consumption now is under 30W for
filaments plus 6W for plates.   And with a line-operated set VR circuits are
problematic: There's often not enough voltage for reliable operation.

(Audio quality rates 'good' on SSB signals; fixing #s 3 and 4 should improve
it by another notch.)

5.  Needs coils for  other bands.  My (used) Eddystone 898 was already
calibrated for 160-80-40-20-15-10 and I had little trouble tracking the
calibration for 80M.   Will probably wind the 20M coils next.

6. There's a little bit of warm-up drift -- maybe 250 cps on 80M?  This will
be a bigger problem on higher bands in greater proportion than with the W6TC
designs because the LO must operate at full frequency rather than half on
the higher ones.  The layout and heat control measures are nearly maxed out
so further improvement will depend on adding temperature compensating
cap(s). 

Ideally temperature compensation is done both in the set and separately in
each coil but making up these coils is difficult enough as it is.  (Since
I'm tracking an existing dial I have to get a three point fit for the
tuning; this requires some cut and try which is a challenge the way the
HBR-type coils are put together.)  Probably I'll underdo the TC for now,
hoping to leave the possibility of adding more in the coils for 20-15-10
where it's likely to be a bigger issue.

7. Needs standby circuit/switch for use with a transmitter.  Probably a
circuit that allows monitoring of the transmitted signal.

8. Needs an audio filter for CW use.

9. Needs a cabinet and some sort of storage for coils.

One justification for obsessing over this project is that the basics
(with suitable changes to all the details)  would be the ideal back end to a
bandswitching double conversion set.   However there's between another month
and another three years of work to finish this one.

Walt Hutchens 
KJ4KV

------------------------------

Message: 2
Date: Thu, 28 Jul 2011 17:33:01 -0400
From: Bill Cromwell <wrcromwell at gmail.com>
Subject: Re: [HBR] The long, SLOW HBR project
To: HBR Receiver List <hbr at mailman.qth.net>
Message-ID: <1311888781.1966.21.camel at lancelot>
Content-Type: text/plain; charset="UTF-8"

On Thu, 2011-07-28 at 13:29 -0400, Walt Hutchens wrote:
---snip---
> 
> I'm happy to report that things have temporarily swung the other way and
> with perhaps twenty hours of time over a few weeks the project has moved
> forward considerably.
> 
---snip---
> 
> Walt Hutchens 
> KJ4KV

Hi Walt,

It must be encouraging to have signals coming through your HBR. Congrats
on getting that far. Life is what happens while we are busy making other
plans <g>. As for more work to be done - I doubt most homebrew projects
are ever completely "finished". Just one more "what if".

73,

Bill  KU8H

------------------------------

_______________________________________________
HBR mailing list
HBR at mailman.qth.net
http://mailman.qth.net/mailman/listinfo/hbr

End of HBR Digest, Vol 89, Issue 2
**********************************