eHam

eHam Forums => HomeBrew => Topic started by: KB1WSY on July 09, 2012, 10:31:33 AM



Title: HBR-16 Receiver Sells for $315
Post by: KB1WSY on July 09, 2012, 10:31:33 AM
A "Ted Crosby" HBR-16 homebrew receiver in fine looking shape just sold on eBay for $315 (way beyond my price range!). As I watched the auction proceed, I pondered the idea of getting hold of one of these receivers some time in the future, tearing it completely apart, and then rebuilding it from scratch i.e. using the original only as a source of raw parts -- even tossing the original chassis and building/drilling my own. This led to some additional, curious thoughts:

(1) Is it sacrilegious to tear apart a classic homebrew rig such as an HBR-16 that is in mint condition? Who is being offended? The unknown ham who built the rig? (Leaving aside the issue of whether s/he is SK.)

(2) Would it be less sacrilegious if the item was beat-up or missing some parts? (I think the answer to *that* is "yes": in that case, it counts as "restoration/rebuild" and should meet with favor.)

(3) Is it just a stupid thing to do, considering that someone went through all the effort to build the rig half a century ago? What exactly is being gained here?

(4) Plus, for some incomprehensible reason, it seemed to me to be taking the "easy way out" and inherently less satisfactory than finding all the individual parts myself. Why do I feel this way? Is it just me? Why should it be more satisfying to go through the hassle of finding these rare parts individually?

(5) Even at the stratospheric price that the item just fetched on eBay, you could easily argue that you are *saving* money (and definitely, saving masses of time) if you buy the "pre-built" item at that price and then tear it apart. Some of the parts in that set, for instance the IF cans and various mechanical parts, are almost impossible to find and nowadays -- your best bet is to rip the IF cans out of a BC-453 and rebuild them for the HBR-16. Other parts, like the Bud cabinet (the eBay one appeared to be in mint shape), confer that "period" look that would take some effort to duplicate (although at least one ham has recently done it very successfully -- see http://www.bignick.net/Morgan_Radio/Radio_8.htm (http://www.bignick.net/Morgan_Radio/Radio_8.htm).

These aren't exactly "current" questions for me, because I am busy building much, much simpler stuff. Just wondering though.

73 de Martin, KB1WSY


Title: RE: HBR-16 Receiver Sells for $315
Post by: K8AC on July 10, 2012, 05:39:53 AM
Ignoring the philosophical questions you posed, it wouldn't make much sense to me to spend the time and money rebuilding a receiver that wasn't all that good to begin with.  I built an HRB-16 from scratch not long after the article first appeared in QST.  I was able to order just about every part from the Allied Radio catalog.  Fellow hams were raving about the HRB-16 at the time and I didn't know enough to be able to form my own opinions of the design.  When I was finished, I was disappointed to discover that it wasn't a very good CW receiver - the passband was too wide for CW work, and when built as described, was too narrow for SSB.  The 100 KHz IF stages had to be detuned to arrive at a bandwidth suitable for SSB and AM.  The tuning rate with the National dial was simply too fast and the Eddystone dial was just out of the price range of most builders.  If I came across one of these receivers today, I'd replace the electrolytics and then play around with the receiver as-is. 

Anyone who believes the HBR-16 to be the zenith of receiver design  either hasn't built one, or has a very weak technical knowledge  base.   But - it was fun to build and offered a real sense of accomplishment when you finally got it to work.  By the way, I think my total cost of construction was around $135.  If I saw one today, I might offer $100 for it for nostalgic reasons.  The one you referred to on eBay looked like a very high quality build that hadn't been stored in a garage or basement for the past 45 years. 

73, Floyd - K8AC


Title: RE: HBR-16 Receiver Sells for $315
Post by: KB1WSY on July 10, 2012, 06:19:59 AM
Quote
I was disappointed to discover that it wasn't a very good CW receiver - the passband was too wide for CW work, and when built as described, was too narrow for SSB.

Very interesting. Strangely enough I had the same worry, when comparing the HBR-xx designs with other advanced superhet homebrew designs that were around at the time. Currently, I'm solely interested in CW and was puzzled as to why such a highly regarded receiver design seemed too broad for CW. I assume it lacked switchable crystal or mechanical filters because those were expensive at the time, or perhaps it is just that the contemporaneous shift to SSB resulted in CW features being neglected.

The ARRL DCS-500 is interesting. The construction article explains the compromises that were made in order to get decent CW selectivity ("Double-Conversion-Superhet-500 Hz") without spending a fortune in 1960s dollars. I started a separate thread about that design, and those who had actually built *that* receiver stressed the importance of build quality, to avoid disappointment.

One advantage of the DCS-500 is that it is (slightly) less complex than most of the HBR-xx series and therefore, I suppose, slightly easier to build. I have actually made quite a good start in squirreling away the "unobtainium" parts for the DCS-500 and hope to get to critical mass eventually. If I change my mind about building it, I can always pass on the components to someone who could make use of them.

The *dis*-advantage of the DCS-500 is that, unlike the HBR-xx series, there isn't a modern community of people who can provide advice to homebrewers. The HBR website is really good!

A bit off-topic, but for those (like me) who yearn to operate solely with homebrewed tube gear and mainly CW, the holy grail would appear to be a *transceiver* design and I haven't found any of those yet -- apart from some SSB-oriented, very complex designs in the late 1960s (for instance in RSGB manuals).

Anyway, I'm getting way ahead of myself. The current goal is to get my modest, "Novice Plus" station on the air. Finally making some slow headway in construction. Fancy high-end receivers with a gazillion tubes are way in the future!

73 de Martin, KB1WSY


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 11, 2012, 07:28:16 AM
I built an HRB-16 from scratch not long after the article first appeared in QST.  I was able to order just about every part from the Allied Radio catalog.  Fellow hams were raving about the HRB-16 at the time and I didn't know enough to be able to form my own opinions of the design.  When I was finished, I was disappointed to discover that it wasn't a very good CW receiver - the passband was too wide for CW work, and when built as described, was too narrow for SSB.  The 100 KHz IF stages had to be detuned to arrive at a bandwidth suitable for SSB and AM.  The tuning rate with the National dial was simply too fast and the Eddystone dial was just out of the price range of most builders. 

...

Anyone who believes the HBR-16 to be the zenith of receiver design  either hasn't built one, or has a very weak technical knowledge  base.   But - it was fun to build and offered a real sense of accomplishment when you finally got it to work.  By the way, I think my total cost of construction was around $135.

I think it's important to understand that the HBR series - except possibly the first one, the HBR-14 - were designed with certain goals, and fulfilled those goals quite well. Being a good CW rx, however, was NOT one of them!

The goals (maybe rules is a better word) were:
- use all-new parts available from major mail-order suppliers. No surplus, no unstocked parts. Very few parts that have to be ordered (like crystals).
- simple mechanical design
- price under $150 for HBR-16 with all-new parts; under $200 with Polar tuning cap, Eddystone dial and all the goodies
- one-box receiver in a relatively small cabinet, no separate power supply, no add-ons
- can be built with relatively few tools and aligned with minimal test equipment 
- straightforward circuits that the designer liked.
- better performance on 'phone than commercial receivers of the same price.

These were pretty much met. However, like many receivers both homebrew and manufactured, CW performance was sacrificed.

If I woke up one day back in those times and wanted to build a good CW receiver for minimal money, here's what I'd build. It would be nothing like an HBR:

- Basic receiver would cover 80 and 40. Crystal controlled converter would cover 20, 15 and 10.
- Basic receiver would use "band imaging" concept with 5.2 to 5.5 MHz VFO and 1.7 MHz IF.
- IF filter would use surplus DC-34 xtals in W6YBR circuit (these are expensive now, but back then they were less than a dollar apiece and common)
- Tuning mechanism would use ARC-5 transmitter PA capacitor with built-in gear drive and dial, or tuning cap from BC-221 or LM frequency meter
- Separate power supply, simple but solid mechanical and electrical design.

Of course such a receiver would be useless on 'phone, but that's no big loss.

73 de Jim, N2EY


Title: RE: HBR-16 Receiver Sells for $315
Post by: K1ZJH on July 11, 2012, 07:42:34 AM
The IF could be tightened by using surplus 80kHz IFTs from old ARC receivers. At least one variation appeared
in Orr's Handbook and included a passive CW filter and Q Multiplier. The goal was to design a receiver any ham
could build, and have some assurance that it work properly, providing he followed directions!

I have all the parts put together to build one, but if I ever have time to finish it I'll be using two half-lattice
crystal filters salvaged from old HR-10 Heath receivers as a roofing filter in the first IF. The coils are the biggest
pain. The original polystrene coils had good Q and extremely low loss, which gave the receiver excellent
sensitivity. Ted also used the second LO harmonic on the higher bands--that would preclude using a
balanced mixer tube.  I'd be tempted to come up with a design that used some sort of bandswitching
for the LO, along with peakable preselector using high Q coils (to avoid tracking problems), for a CW receiver
that covered only 160, 80, 40 and the 30 meter bands. A companion reciter could be built to make a complete
CW station.

Pete  k1zjh


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 11, 2012, 10:39:58 AM
The IF could be tightened by using surplus 80kHz IFTs from old ARC receivers.

Yes, but it would take at least 6 of them to do the job. The 85 kHz cans from ARC-5s are great but they're also pretty large, so the chassis would have to be a lot bigger too.

Would also violate the "no surplus" rule. However, with BC-453s at $5 back in the day, why not?

I have all the parts put together to build one, but if I ever have time to finish it I'll be using two half-lattice
crystal filters salvaged from old HR-10 Heath receivers as a roofing filter in the first IF. The coils are the biggest
pain. The original polystrene coils had good Q and extremely low loss, which gave the receiver excellent
sensitivity. Ted also used the second LO harmonic on the higher bands--that would preclude using a
balanced mixer tube.  I'd be tempted to come up with a design that used some sort of bandswitching
for the LO, along with peakable preselector using high Q coils (to avoid tracking problems), for a CW receiver
that covered only 160, 80, 40 and the 30 meter bands. A companion reciter could be built to make a complete
CW station.

The use of a bandswitching or plug-in coil LO compromises stability, particularly on the higher bands.

With 160 and a 1682 kHz first IF you may have a problem with IF feedthrough.

Making a transceiver is a bigger challenge than it may seem at first. If the LO is 1682 kHz from the signal frequency, it seems simple to just mix the LO frequency with a 1682 kHz oscillator and get the output you want.

But then you have the problem of oscillator harmonics being near the output frequency. On 160, for example, you may be transmitting on 1800 yet trying to reject 1682, which is only 118 kHz away (6.5%). On 80, when you're on 3500, the second harmonic of 1682 is at 3364. On 40, you have the fourth harmonic at 6728. On 30 meters, the sixth harmonic is at 10,092, just outside the low end of the band.

These sorts of problems are why old-school multiband HF ham transceivers had such convoluted heterodyne schemes. They're also why you see so few heterodyne-type transmitters from the hollow-state days, and so many multiplier-type.

I found a workable solution for the Southgate Type 7. But that's another story...

73 de Jim, N2EY


Title: RE: HBR-16 Receiver Sells for $315
Post by: KB1WSY on July 11, 2012, 11:02:40 AM
Quote
I found a workable solution for the Southgate Type 7. But that's another story...

Which I'd love to hear ...

73 de Martin, KB1WSY


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 11, 2012, 02:25:42 PM
Quote
I found a workable solution for the Southgate Type 7. But that's another story...

Which I'd love to hear ...

More than 20 years ago, I was at the Gaithersburg (MD) hamfest. Lots of goodies of various kinds. Haven't been back in many years but it was primo then.

I came across a fellow with a plastic milk crate full of what looked like crystal filters. The markings said they were 1400 kc. center frequency, 500 Hz pass band, made by Alpha Components. I took a chance and bought four of them.

Brought them home and through various channels discovered they were 8 pole jobs with 1000 ohm terminating impedance. Really good for a CW receiver. So I haywired together a simple 80 CW receiver from junk parts and listened in. Sounded really good, so I set out to design a transceiver using one. (First station I heard was Bob, W2LYH, but that's another story).

My favorite bands were and are 80, 40 and 20 meters. How to build a three-band transceiver with such a filter so that both tx and rx would be clean?

Lots of scrap paper and pencil stubs later, I had a plan.

The VFO would tune 4900 to 5150 kc. No bandswitching. It is built in a homemade aluminum box made of 3/16 inch aluminum plates and angle aluminum, held together by a lot of screws, nuts and lockwashers. Took a while to cut all those parts and drill all those holes by hand, but it was worth it. Tuning capacitor/dial drive is from an LM frequency meter. Be warned! Once you have used a receiver with one of those, you will be ruined for anything less.

On 80 meters, the VFO signal converts 3500 to 3750 kc. to 1400 for receiving. For transmitting, a 1400 kc. xtal oscillator signal is mixed with the VFO to give output on the same frequency as the receiver.

On 40 meters, the VFO signal is mixed with a 3400 kc. oscillator signal to give output from 8300 to 8550 kc. (This technique is called "premixing"). The 8300-8550 kc. signal then goes to the receiver to permit tuning 6900 to 7150 kc. For transmitting, a 2000 kc. xtal oscillator signal is mixed with the VFO to give output on the same frequency as the receiver.

On 20 meters, the VFO signal is mixed with a 7600 kc. oscillator signal to give output from 12500 to 12750 kc. The 12500-12750 kc. signal then goes to the receiver to permit tuning 13900 to 14150 kc. For transmitting, a 9000 kc. xtal oscillator signal is mixed with the VFO to give output on the same frequency as the receiver.

All the heterodyne crystals have APC trimmers across them to permit getting them dead-on frequency. RIT is implemented by a small relay in the VFO that switches in either a fixed silver mica or a BC variable across a low-impedance part of the VFO tuned circuit. A variation of several hundred uuF causes the VFO to move a kc. or two, which is plenty of RIT for CW.

Yes, the low end of 80 is not in the same spot as the low end of 40 and 20. But that's really no big deal.

The receiver section up to the first audio stage is built on one chassis, with buffered VFO output. This unit also has the sidetone oscillator built in.

The buffered VFO output is fed to a small chassis that holds the transmitting mixer and its heterodyne oscillator.

The higher power stages of the transmit section (driver and final) are on a third chassis.

The LC audio filter and audio output are external so they can be used for other purposes.

One power supply drives the receiver, audio output, and transmitting mixer, while another drives the transmitter itself. The transmitter power supply has an Amperite time-delay relay in the control circuits to avoid blowing up the mercury-vapor rectifiers by applying HV too soon. Total of four plug-in relays.

There's a TR box with antenna relay for antenna switching, controlled by a small switching box that also remote-switches the tx power supply.

Yes, there are a lot of boxes and connecting cables, but it does the job. 

The only parts bought new for it were some of the crystals. Everything else was scrounged, reused, or a hamfest find. There are no solid-state devices used at all except for two 1N34As in the SWR bridge of the Transmatch - which was built before any of the rest of it.

It worked so well that after a few years I did some surgery and added a second filter between the IF stages, so now there are a total of 16 poles of filtering.

Tube lineup:

Receiver section:
6EH7 RF, 7360 mixer, 6EH7 first IF, 6BA6 second IF, 6GX6 detector/BFO, 6C4 audio, 6C4 sidetone oscillator, 12AU7A VFO, 6BA6 heterodyne oscillator, 6BE6 premixer, 6BA6 VFO buffer.

Last audio section:
12BH7 push-pull audio amp.

Transmitting mixer section:
6BA6 heterodyne oscillator, 6BA7 mixer

Transmitter section:
12BY7 driver, 2 x 807 final amplifier (150 watts input)

Power supply #1:
5U4 rectifier, 0A3 and 0D3 VR tubes

Power supply #2
5Y3 bias rectifier, 0C3 and 0D3 bias VR tubes
5Y3 LV rectifier
866A HV rectifiers, 2 x 0D3 screen regulator tubes

No, it's not like any published design, and unless you have the exact parts already on hand I wouldn't recommend duplicating it. There are no schematics - no written-down ones, anyway.

Just an example of what can be done.

The filters were fifty cents each. I should have bought the whole crate!

73 de Jim, N2EY

(well, you did ask!)


Title: RE: HBR-16 Receiver Sells for $315
Post by: K1ZJH on July 11, 2012, 07:46:35 PM
The IF could be tightened by using surplus 80kHz IFTs from old ARC receivers.

Yes, but it would take at least 6 of them to do the job. The 85 kHz cans from ARC-5s are great but they're also pretty large, so the chassis would have to be a lot bigger too.

Would also violate the "no surplus" rule. However, with BC-453s at $5 back in the day, why not?

I have all the parts put together to build one, but if I ever have time to finish it I'll be using two half-lattice
crystal filters salvaged from old HR-10 Heath receivers as a roofing filter in the first IF. The coils are the biggest
pain. The original polystrene coils had good Q and extremely low loss, which gave the receiver excellent
sensitivity. Ted also used the second LO harmonic on the higher bands--that would preclude using a
balanced mixer tube.  I'd be tempted to come up with a design that used some sort of bandswitching
for the LO, along with peakable preselector using high Q coils (to avoid tracking problems), for a CW receiver
that covered only 160, 80, 40 and the 30 meter bands. A companion reciter could be built to make a complete
CW station.

The use of a bandswitching or plug-in coil LO compromises stability, particularly on the higher bands.

With 160 and a 1682 kHz first IF you may have a problem with IF feedthrough.

 
73 de Jim, N2EY

True, but a good quality band-switch should yield acceptable results up through 30 meters.  IF blowby
could be an issue, but then again I would go to a balanced mixer and IF trap on the input to reduce the
likelihood of that happening. Shielding wouldn't hurt either.

Pete


Title: RE: HBR-16 Receiver Sells for $315
Post by: KB1WSY on July 12, 2012, 03:00:57 AM
(well, you did ask!)

Very interesting, although my head is spinning.... I am beginning to understand why a fair number of commercial and homebrew transceivers from the late tube era were *single-band* solutions! At this level of complexity I find that I end up having to draw block diagrams, and indeed the ARRL transceiver projects from the late '60s often include one. I can also see that your solution of breaking things up into physically separate modules is ideal for experimenting and troubleshooting, despite the forest of cables it must entail. (It's the physical analogue of the block diagram!)

73 de Martin, KB1WSY


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 12, 2012, 07:02:39 AM
The IF could be tightened by using surplus 80kHz IFTs from old ARC receivers.

Yes, but it would take at least 6 of them to do the job. The 85 kHz cans from ARC-5s are great but they're also pretty large, so the chassis would have to be a lot bigger too.

Would also violate the "no surplus" rule. However, with BC-453s at $5 back in the day, why not?

I have all the parts put together to build one, but if I ever have time to finish it I'll be using two half-lattice
crystal filters salvaged from old HR-10 Heath receivers as a roofing filter in the first IF. The coils are the biggest
pain. The original polystrene coils had good Q and extremely low loss, which gave the receiver excellent
sensitivity. Ted also used the second LO harmonic on the higher bands--that would preclude using a
balanced mixer tube.  I'd be tempted to come up with a design that used some sort of bandswitching
for the LO, along with peakable preselector using high Q coils (to avoid tracking problems), for a CW receiver
that covered only 160, 80, 40 and the 30 meter bands. A companion reciter could be built to make a complete
CW station.

The use of a bandswitching or plug-in coil LO compromises stability, particularly on the higher bands.

With 160 and a 1682 kHz first IF you may have a problem with IF feedthrough.

 
73 de Jim, N2EY

True, but a good quality band-switch should yield acceptable results up through 30 meters.  IF blowby
could be an issue, but then again I would go to a balanced mixer and IF trap on the input to reduce the
likelihood of that happening. Shielding wouldn't hurt either.

All true, but it may still be a problem with the IF so close to the signal frequency.

73 de Jim, N2EY


Title: RE: HBR-16 Receiver Sells for $315
Post by: K1ZJH on July 12, 2012, 12:10:52 PM

All true, but it may still be a problem with the IF so close to the signal frequency.

73 de Jim, N2EY

Where do we draw the line for a modernized tube HBR?  :) 

I have sets of 40.455 MHz roofing filters in the stash, as well as matched USB/CW/LSB crystal and
Collins filter sets that could be used for up conversion roofing filters and for a variable bandwidth/PBT
scheme in a triple conversion scheme.  But how far does one's ambition stretch...

To be practical, I have the two sets of HR-10 half-lattice filters, and a  supply of ARC receiver IFTs
and all of the other parts for an advanced HBR, including a few Eddystone dials.  Most of the parts
aren't cheap, and I'm old enough to want to use up what I already have in the dungeon hoard. 
I do want try the 6EJ7/6EH7 tubes in the RF and IF stages. Heck, my SX-101A is a pretty basic
dual conversion design that does okay for what it is, and the early
versions covered 160 meters, although admittedly the extended BCB didn't exist back then :)
I'd be happy with 80/40/30 meter coverage, although a few guys on the HBR group have posted
data for 160 meter coil sets.

Pete


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 12, 2012, 02:40:56 PM

All true, but it may still be a problem with the IF so close to the signal frequency.

73 de Jim, N2EY

Where do we draw the line for a modernized tube HBR?  :) 

Wherever you want it to be. Even the Ancient Ones did odd mods, such as silicon rectifiers to replace the 5V4.

I have sets of 40.455 MHz roofing filters in the stash, as well as matched USB/CW/LSB crystal and
Collins filter sets that could be used for up conversion roofing filters and for a variable bandwidth/PBT
scheme in a triple conversion scheme.  But how far does one's ambition stretch...

As far as you want it to.

To be practical, I have the two sets of HR-10 half-lattice filters, and a  supply of ARC receiver IFTs
and all of the other parts for an advanced HBR, including a few Eddystone dials.  Most of the parts
aren't cheap, and I'm old enough to want to use up what I already have in the dungeon hoard. 
I do want try the 6EJ7/6EH7 tubes in the RF and IF stages. Heck, my SX-101A is a pretty basic
dual conversion design that does okay for what it is, and the early
versions covered 160 meters, although admittedly the extended BCB didn't exist back then :)
I'd be happy with 80/40/30 meter coverage, although a few guys on the HBR group have posted
data for 160 meter coil sets.

The first and most fundamental question is:

Are you building a receiver or a transceiver?

Because if you want the thing to transceive, there are a whole bunch of issues you have to deal with which are of no concern in a receiver. If it's multiband, that adds even more complexity.

This is the big reason homebrew transceivers are/were so much rarer than receivers.

You might also want to take a look at some designs in the RSGB Handbook. G2DAF did some really amazing stuff. G3PDM (IIRC) went even farther, designing a receiver with a tube-type PLL and 1682 kc. IF. (iirc)

Here's how the PLL worked:

The rx used a push-pull 2C51 LO driving a 7360 balanced mixer. But this meant the oscillator was bandswitched, which brings a bunch of stability issues.

So the LO was tuned by a quad of varactors and a DC control voltage. It would then be phase-locked to a tunable reference oscillator that was very stable.

The stable tunable reference oscillator starts out with a very stable VFO around 5 Mc., which was the correct range for one band (80, I think). For the other bands, the VFO signal was mixed with the signal from a crystal oscillator to provide the appropriate range, all the way to 10 meters.

You may ask why they didn't just use the mixed VFO and crystal signal directly, and forget about the 2C51 oscillator. Two reasons:

1) Balanced output without a lot of complexity
2) The reference oscillator chain output is "dirty" - various harmonics and mixer products besides the desired one. A simple tuned circuit will make the desired output stronger than all the rest, but that's not clean enough in a receiver. But the phase-locked 2C51 will only lock to the strongest signal, and will ignore all the rest.

The RSGB Handbooks of the tube era are the BEST!

73 de Jim, N2EY


Title: RE: HBR-16 Receiver Sells for $315
Post by: K1ZJH on July 12, 2012, 03:11:31 PM
Jim

Do you remember the adapters that were sold back in the 60s to permit oddly paired radios to transceive?

Pete


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 12, 2012, 04:45:30 PM
Do you remember the adapters that were sold back in the 60s to permit oddly paired radios to transceive?

The only such adapter I ever heard of is the P&H "VFO-Matic" which was reviewed in QST for March, 1958. It was designed to be used with the Collins 75A-2, -3 or -4 receiver and any transmitter that used a 9 MHz SSB generator.

It was made in two models: The 8020, which covered only 80 and 20, and the 80-10, which covered 80, 40, 20, 15 and 10.

I've never seen one nor encountered anyone who has any experience with them. The manual is not on BAMA.

Were there others?

73 de jim, N2Ey



Title: RE: HBR-16 Receiver Sells for $315
Post by: G3RZP on July 13, 2012, 03:56:27 AM
The 6EH7/EF183 really needs an unbypassed 27 ohm cathode resistor for best results. That has a marked effect on the change of input impedance with bias, and significantly improves the IMD performance. Using a low plate load impedance helps, too - you can then get no more gain than you need while being on a relatively linear part of the characteristic. That's recommended somewhere on one of the Philips data sheets.

The Marconi Marine rx, the 'Pennant' used a 6EH7 RF amp like that with a 4 diode ring bridge mixer, and had very good dynamic range.


Title: RE: HBR-16 Receiver Sells for $315
Post by: K1ZJH on July 14, 2012, 08:03:58 AM
Do you remember the adapters that were sold back in the 60s to permit oddly paired radios to transceive?
I've never seen one nor encountered anyone who has any experience with them. The manual is not on BAMA.
Were there others?

73 de jim, N2Ey
 


I've never seen one myself. I always wondered how well they actually worked!

Pete


Title: RE: HBR-16 Receiver Sells for $315
Post by: KB1WSY on July 14, 2012, 03:02:09 PM
Short of building a CW *trans*ceiver, what's the best linkup for vintage, tube based, "separate" homebrew gear? I don't mean the *simplest* linkup (for instance a knife switch for T/R), but the *best* with the aim of achieving fulll QSK operation. Here's my current plan:

--HF RX.
--HF TX initially XTAL controlled.
--HF VFO (when I have time to build it) w/ relay based keying of TX (what is the quietest relay, reed presumably? I don't like the idea of relay "clatter" noise).
--Electronic T/R switch (there are 3 or 4 designs for tube T/R switches in ARRL literature from the 1960s).

Informed by my reading of the vintage literature, this approaches the convenience of an actual transceiver, except for the need to tune the RX and TX individually. Unless you're competing in contests, this is almost as good as a transceiver.

Comments?

73 de Martin, KB1WSY


Title: RE: HBR-16 Receiver Sells for $315
Post by: N2EY on July 16, 2012, 08:08:07 AM
Short of building a CW *trans*ceiver, what's the best linkup for vintage, tube based, "separate" homebrew gear? I don't mean the *simplest* linkup (for instance a knife switch for T/R), but the *best* with the aim of achieving fulll QSK operation. Here's my current plan:

--HF RX.
--HF TX initially XTAL controlled.
--HF VFO (when I have time to build it) w/ relay based keying of TX (what is the quietest relay, reed presumably? I don't like the idea of relay "clatter" noise).
--Electronic T/R switch (there are 3 or 4 designs for tube T/R switches in ARRL literature from the 1960s).

Informed by my reading of the vintage literature, this approaches the convenience of an actual transceiver, except for the need to tune the RX and TX individually. Unless you're competing in contests, this is almost as good as a transceiver.

Couple of points:

1) VFO control is much more important than QSK.
2) Having to zero the tx to rx really only makes a difference in contests and other time-sensitive operating. For general QSOs, it's not such a big deal.
3) It is best to get experience with a good working non-QSK station before attempting QSK.
4) Keyed VFOs tend to be chirpy and have other problems. Diode noise in the final amplifier can be a problem, as can harmonics generated by electronic TR.
5) A single 3PDT relay in a small box can handle all TR functions in most hollow-state stations. First contact switches antenna, second grounds rx mute line on receive and tx VFO line on transmit, third grounds rx antenna input on transmit and can operate other functions on receive.

IOW, start simple and straightforward. Rx, tx, TR relay, VFO, antenna. QSK later.

That said, I did have a full-QSK tx/rx setup in the 1970s. It worked like this:

- Homebrew rx shown on HBR website
- Matching tx with Class C final, grid-block keyed.
- VFO for tx used heterodyne principle. Both oscillators ran all the time, mixer was grid-block keyed. Tunable oscillator around 5 MHz, xtal oscillator around 9 or 12 MHz for 80/20 and 40.
- Keyed highspeed relay did muting and tx keying duties (small surplus relay meant for keying, not a reed nor a mercury-wetted. SPDT with moving contact grounded.)
- Electronic TR by means of 6AH6 inside transmitter. Connects to final via small high-voltage capacitor. See transmitter article in QST about 1971 (The T-9er or some such was the name).

73 de Jium, N2EY


Title: RE: HBR-16 Receiver Sells for $315
Post by: G3RZP on July 16, 2012, 11:39:16 AM
I have never got on with true QSK. For me, key controlled changeover is better.

I have plans for a rebuild of a really basket case HRO I picked up - when I get another project or two sorted and first, repairs to one of the workshop benches. Getting at that is a major exercise..

The HRO rebuild is aimed at ham bands only. Set of 1.4 MHz crystal filters that I happen to have (USB, LSB, AM, 3, 1, and 0.2 kHz for CW). Then convert to 85 kHz so there's not too much gain at 1.4 MHz. 6EH7 triode connected RF amp with low plate load resistance to give just enough gain, a bandpass coupled pair, a 6DJ8 mixer, a 6C4 oscillator, the filters a 6EH7 amp, a 6EA8 mix/osc to 85 kHz, a 6BA6 amplifier at 85 kHz, a 6AU6 BFO/product detector, a 6EA8 triode as infinite impedance detector for AM, pentode as calibration osc, a 6AT6 as AGC rectifier, AGC DC amplifier and S meter driver and a 6BM8/ECL82 AF and output. Stabilised heaters for the 6DJ8, 6C4, 1st 6EA8 and 6AT6 using transistors and B+ lines of 105 and about 180 to keep heat down. Built in PSU, a notch filter and an antenna attenuator.

Should be an interesting exercise producing a long chart frame to give a 'slide rule' dial/ frequency graph, but that's work for the milling machine....More problematical will be getting the necessary machine for engraving the new dials on the HRO knobs. Not sure if I will go for nickel silver for the metal skirts or aluminium. A new etched custom name plate will be $50. but hopefully will be worth it..

A fair number of hours have gone into calculating how to get reasonably linear dial calibration. The answer seems to be a biggish padding cap across the tuning gang and then a series capacitor.

Another long term project. A good job my current plans are to retire in summer 2013...but it might depend on whether there are financial incentives to carry on working that I can't resist.


Title: RE: HBR-16 Receiver Sells for $315
Post by: K1ZJH on July 19, 2012, 11:37:24 AM
The old Johnson T/R electronic TR switch would be ideal for QSK.  You'd still need a way to mute the RX (maybe open the
RF/IF gain control?) during TX.  I tried my Johnson TR unit on my HT-32 and SX-101A setup, but since the HT-32 always
runs the driver and PA in class AB1, even during standby, the receiver noise floor was raised by several dB. The biggest problem with relays and QSK is to get the timing right. The TR relay has to be fast to avoid any hot switching. 

I'm now using an Omni VI with full QSK with a heavily modified SB-220 amp.  QSK is the cat's meow when running
spilt on CW. I could never go back to clanging relays or long delays.

Pete