An attempt at an upgrade of a Heathkit HW-32A SSB Transceiver.

 

For the past couple of months, I was seeking a donor HW Monobander. But in truth, I actually had one in a storage area. We all know the huge issue of finding something in storage. Well, after several months of plaintive pleas for a donor radio, I was left empty handed. Thus, a sojourn to the storage area to retrieve the HW-32A. Yes, I had to go through all of the boxes, and it was the very last one.

Physically it is OK, but it has a significant circuit issue that after chasing it for some time I gave up. Thusly banishment to the storage area. So now I have to fix it. Yesterday I wrote the code for the ST7735 display and that is done. (More on this.)

The plan is to get the HW-32A working as designed and then the treatment with the JF3HZB dial.

I have done a fit check of several sizes of displays, but that semicircular hole is an odd duck. The panel space is 12X6 inches and I have a chunk of aluminum that exact size. 

A potential possibility is to create a whole new front panel and cut a rectangular hole in the new panel for a Color TFT display. This approach would enable the use of a large ILI9341 display. The original panel would now be a template for cutting most of the holes in the new panel. A Juliano Blue or Canary Yellow color front panel would assure a notice.

But wait! This is a complex project because of what Heathkit did! That blog readers is the challenge. At times I just do things because I can but often because it is a real technical challenge. The Circular Analog Dial with the HW-32A is a Technical Challenge. Part of that challenge was highlighted in the past postings dealing with the 40M Channelized radio that was fitted with the same Heathkit crystal filter frequency.

Today, especially with the advent of Digital LO's and BFO's the single conversion topology is the circuit of choice. The real ham technical experts (the non faux wannabe's) will give full throated testimony to the soundness of the single conversion.

But the HW-32A is a Single Conversion Design and that is the challenge in the required engineering evaluation to make that work. Let's take a shallow dive into the issue. The Heathkit 4 Pole Filter is at 3.395MHz and the heterodyne frequency is at 18.275 MHz (simple Crystal Oscillator) and is added/subtracted to the VFO frequency to form the LO. 

Shazam, Heathkit used a VFO operating in the range of the broadcast band. In fact, I think that is one of the operational checks to see if the VFO is working -- listen for it on your AM radio. This low VFO range when mixed with the fixed heterodyne signal made for a far more stable radio than a VFO operating at 10+ MHz. 

Now I earlier said I got the JF3HZB code working for the HW-32A. Well, that was only to see how it would look on various displays and how the front panel will look with the various sizes. It is not the final design!

So here is an array of technical challenges. Throw it all away, keep some or only use the JF3HZB as the VFO. Each of these involves a significant modification to the radio and the sketch. The change the VFO only option involves a bit of code trickery because of the heterodyne VFO conversion.

Let me explain the last option. You want the display to show as in the second photo. But the original JF3HZB code is based on single conversion with a subtraction of the 3.395MHz IF. This has to be accounted for in the code and since we have a subtraction it will be a sideband inversion and the frequency actually subtracted is not 3.395 but 3.396500 MHz, the USB BFO so that the dial reading is accurate.

Heathkit did make it somewhat easy as the VFO and Heterodyne Oscillator (V13 and V14) have their filaments series connected. Simply pulling those tubes out of their sockets is all that is needed to null those functions. A new LO from CLK0 can be introduced into L5 and that solves the problem of a single conversion.

Still another waypoint decision, the BFO. The HW-32A has selectable sidebands and so do we keep that function or disconnect the crystals and simply feed CLK2 into the grid of the Carrier Oscillator. The former USB/LSB select switch could now be an input to the ESP32 to provide the proper Carrier Frequency. Is there a possibility of generating a USB Carrier and the other a CW carrier on transmit and use the HW-32A on CW? More complexities

So, the real thrust is to get the HW-32A working as stock. My past foray on getting it proper was L2 and L3 as the signals (both on transmit and receive) were in serious need of a shot of Viagra. 

L3, which is in the plate circuit of the 12BY7 Driver Tube has a link coupled winding from the antenna relay and thus is also in the receive path. A limp signal on Rx and Tx would sure point to that area of the circuitry. Faint signals and but a few watts on transmit are the symptoms.

This could be a serious overall technical challenge and another adventure at the N6QW Laboratories.

Them that know can make things go!

73's

Pete N6QW

A Mate for the ARC-5 Receiver ~ A blast (of RF) from the past!

 

A bit of propagation sure helps, and two things are evident: An ARC-5 sounds better on CW and the band spread is awful. Just imagine relying on this radio as if your life depends on it.

The transmitter shown above uses a 6L6 built using a bread board, more or less on a whim, but does show this could have been OK for a station rig. At 12 watts output it would certainly be the bottom level for consistent reliable operating as a main rig. I always suspected that the design would support perhaps as much as 20 to 25 watts output. 

The shortfall I reason likely lies in the low plate voltage at 300 VDC and a non-optimized output network. The Variable cap in the upper right-hand corner of the bread board was an attempt to provide additional "Loading" in the output network. 

Featuring front panel selection of "rocks" the FT-243 crystals sound much better than the plated HC6-U type. There are three frequencies 7.010, 7.028 and 7.o58 in the rig. These were found in the junk box but certainly ideal for operation today. Not quite a Si5351 but a bit of QSY wiggle room.

Plug In Tank Coils facilitates operation on 20 Meters using two of the crystals doubling their frequency to 14,020 and 14.056 MHz which are OK for 20M. Thus, part of the tank circuit on a plug-in coil form.

We have reached the reader gag level so time to move on with more exciting adventures from N6QW's junk pile.

Them that know can make things go.

73's

Pete N6QW

Have you ever used an ARC-5 Receiver?

For a majority of the current US Ham population, to that question we likely would hear a resounding NO. Most wouldn't even know what it is.

A 3 to 6 MHz ARC-5, BC-454

My 40M ARC-5

Way before the US formally entered WWII there was a push to modernize our military communications hardware. The ARC-5 components were actually a system of small sized radios that covered a portion of the LF and HF bands (up to 9 MHz). Obviously, the intended use was in aircraft.

At this time the physical size of the ARC-5 radio sets were small in comparison to a SX-28 or HRO-5. The receiver portion of the system used 6 tubes and depending on the band used different IF frequencies. The BC-453 covered the LF band and had an IF of 85 kHz. The BC455 (6 to 9 MHz) had a higher IF in the MHz range. 

The Transmitters were a bit physically bigger and had 4 tubes including a magic eye tube. The output stage used a pair of 1625's (12V version of the 807's) good for 100 watts on CW.

ARC-5 Transmitter

Two seconds of thought should make clear that by choosing the variable IF a very stable LO could be built in a single conversion approach. The Receivers could receive AM and had a BFO for CW. The transmitters could do CW "as is" but there was an accessory Modulator Box (BC-696) needed for AM.

In normal operation in an Aircraft the ARC-5's operated from 28VDC using a strap on Dynamotor for the Plate and Screen voltage and the filaments were wired series parallel to operate from 28VDC -- aircraft power.

Post WWII the Receivers and Transmitters were highly prized by the ham community. In 1955, ten years after WWII, I bought a NIB BC-454 (3-6MHz covering 75/80M) for $3.50, and a purpose-built power supply (plug in for the Dynamotor) was another $7.50. So, for $11 I had a ham band receiver.

The units were built like a battleship with gear drives on the VFO's/LO's. The build included gluing down all VFO wiring using "Glyptal" so the was no component movement which would affect frequency. Surprisingly the dials were linear and repeatable!

[Glyptal has been used in ARC-5 radios primarily as an insulating and protective coating. It helps seal and protect electrical components from moisture, corrosion, and vibration, which is crucial for maintaining the reliability of these vintage military radios. Some enthusiasts and restorers use Glyptal to recoat transformers, coils, and chassis interiors to preserve their integrity and prevent electrical shorts. Courtesy of Microsoft copilot AI.]

The BC-453 (the LF variant) tuned at 455 kHz, suddenly became a very hot item. It was known as the Q5'er.  The trick was to take an output from the IF of a commercial receiver at 455 kHz and feed that into the antenna of the BC-453. You now had a dual conversion receiver with a second IF at 85kHz and a very narrow bandwidth (Q5). A really enterprising conversion by two California hams made the BC-453 into a SSB Transceiver

A discussion of the ARC-5 system would not be complete without a mention of Tony Vitale, W2EWL who converted an ARC-5 Transmitter into a Phasing Transmitter in a seminal QST article entitled "Cheap and Easy SSB'. In the early days of SSB, Tony's rig was found in many Mobile SSB installations. Read about it here on N2CQR's blog.

My very 1st station was the BC-454 and a 6V6 built on a wooden chassis at 3709 kHz (then a valid Novice CW operating frequency). For a TR I used an "open" big old DPDT knife switch. More than once I got RF burns a result of touching the switch while I was still transmitting.

When I headed off to Penn State in 1959, my BC-454 was on the bookshelf along with the textbooks. It was a great way to keep a connection with hobby while I was away at school.

Et Al, the ARC-5 was a pretty rudimentary system and consider that they were mass produced, and their life expectancy was in terms of months not years. So not much in the way of frills. 

For long haul communications in the Bombers, we would see the BC-348 or HRO for the receiver and of course the Collins ART-13 with autotune for the transmitter.

I have always had weird ideas about commercial or military radios. The 1625 transmitting tubes used in the ARC-5 were really cheap like 25 cents for a NIB and so there was much interest in building transmitters using these tubes. 

One issue with the 1625 was that the Cathode was internally connected to one of the grids. Hams being hams simply sawed into the base of the tube and removed that connection. You ask why? The answer is SSB operation using the Cathode Driven (grounded grid) topology which wanted to see only the Cathode. A pair of 1625 suddenly was a 100-watt Linear Amp. 

I went one step further and converted an ARC-5 Transmitter into a complete linear amp including building the power supply inside the case.

At the time I had a Yaesu FT-7 which is a QRP radio and with the ARC-5 Linear it was now an all band 100-watt Transceiver. When I sold the FT-7, the amp went with it. What a hoot.

So now you know a bit more about the ARC-5 radio system. Don't bother to look on eBay -- a decent ARC-5 is upwards of $200, and I paid $3.50 in 1955.

Finally, many conversions of the ARC-5 receivers involved rewiring the filaments for 12VAC. Actually, if you were the clever sort, you could have a twofer. If you purchased two 25.2 VAC Filament transformers one at 3 amps and another at 1 amp you had the makings of a complete power supply for the ARC-5. 

Step one made the 3-amp version as the side connected to the mains. Off the secondary you would light the filaments as well as provide an input to the second 25.2 volt trans former. Essentially the secondary sides were connected back-to-back. 

So now what was the primary on the 2nd transformer was now the secondary. That voltage was fed into a bridge rectifier and instantly about 160VDC for the plates and screens. Thus, a twofer with two transformers connected back-to-back.

Them that know can make it go.

73's

Pete N6QW

Acorn Tubes. QRP with tubes that are only a bit bigger than fat Transistors

 

Acorn Tubes are really small tubes that were designed during WWII for use at UHF frequencies. I got my 1st Acorn Tube when I bought a WWII surplus Radiosonde Transmitter. 

Needless to say, long ago, when hams would build things and not just do contests, POTA/SOTA or plain operate, these Acorn tubes found their way into homebrew QRP CW transmitters. Several years ago, I even saw a radio set using an Acorn Tube CW Transmitter complete with a matching Regen Acorn Tube Receiver. What a hoot.

Rex Harper, W1REX, during the virtual 2020 FDIM Buildathon (you know that COVID 19 Crap) used this schematic as the project foundation. I think it got close to a watt output with this lash up.

I didn't participate in that event but did build my version using scrap stuff I had lying around the garage. The tube filaments are wired in series so I could simply supply 12 VDC through a three-terminal regulator. 

Since Cycle 25 has been sub-par (a highly touted but major disappointment) and given the MUF yesterday was no higher than 30M then this jewel would be FB on 40M.

Before anyone gets their underwear in a knot about high voltage DC and the need for a special power supply... Think 20 Transistor Radio Batteries in series. The current draw is less than 10Ma so as they say in Italy -- Non è un grosso problema.

Or in plain English -- "Ain't no big deal". I checked Amazon and twenty 9VDC batteries cost less than $15.

An overwhelming resource on simple vacuum tube receivers can be found here thanks to K3UH. You could get lost on his site for weeks at a time.

Band conditions are supposed to be better today so maybe we can test the ZIA. The MUF is better but unsure about the other one.

Them that know can make things go.

73's

Pete N6QW

Looking for a Heathkit Monoband Transceiver.

Since Cycle 25 sucks, I thought I would get serious about revamping one of the Hot Water rigs. Please check your storage lockers, in the attic space or under the workbench. It doesn't have to be working just all there.

[Checked the MUF (Maximum Usable Frequency) chart for today for Southern California -- if we are lucky maybe 3oM. Strange times as checking the MUFF used to mean something entirely different.]

Perhaps some blog readers belong to clubs that have donated gear for sale and so could you possibly check that for me as well.

Another bust yesterday with the rebuilt ZIA as 20M was dead as a doornail. I would hope we get some sort of an opening so I could at least evaluate why it was in the back of the closet. It puts out about 4 watts, and I certainly can hear weak signals from my signal generator -- but a couple of on the air QSO's would open the door to the mystery.

I keep looking on eBay and some of those who post on there must be smoking hemp rope. I saw an Atlas 215X listed for $1700 and most of the Heath Monobanders are around $200 with shipping and many of those are pretty beat up.

Many other vintage 1960-1980 rigs are listed for more than they originally cost. Sold as is, untested and/or for parts or repair, that is a real pig in a poke.

This pricing situation with beat up Boat Anchors must lead to some interesting marketing decisions by current manufacturers. If the prices on old junk are high that opens the door to overpricing new stuff. Just Think $1700 for an Atlas or $1200 for an ICOM IC7300 that could be priced at $1100 and still make lots of money.

Hans Summers is selling a built version of his new QRP SSB transceiver for $185. This product the QMX Plus was showcased several weeks ago at FDIM. His Company QRP Labs has been churning out products that are well designed, work well and by comparison cost pennies. 

I would have to seriously ask why I personally would want to buy a single band 20-year-old MFJ SSB transceiver that cost more ($205) and has less capability and whose market value is less after it has received the N6QW customization. It is all about economics.

In case you didn't believe me here is that $1700 Atlas Radio.

Again, you have some great options such as the QMX Plus or even homebrew a rig. 

In my case I have a penchant for doing the weird like installing a Digital Display in a Heathkit HW-101 or converting the Model 540 Analog Ten Tec Triton IV to the Model 544 Digital Triton IV or building a rig inside a Champagne bottle. I just have a vision of a Monoband Heathkit SSB transceiver with the JF3HZB dial (and digital LO) in place of the drifty stock Analog VFO. Help N6QW achieve that goal.

Them that know can make things go.

73's and TT

Pete N6QW

The ZIA is rebuilt -- a Big Job!

We got the ZIA rebuilt just in time for all the bands to go DEAD! That is that amazing Cycle 25 that we all know just plain sucks. 

A quickie review on the ZIA which was built a decade ago and put in the closet. I have simply forgotten the why of the closeting and in the interim I have been stealing parts from that rig. 

Step one is to make it whole and fix a few things along the way. One of those fixes was to install a larger Color TFT and the problems cascaded from there. I had to do a serious rework of the front panel and move a bunch of controls. 

My 1st thought was to just make a new front panel, but the existing one was a custom size and so I resorted to a small overlay on the existing panel. That involved a lot of machining and squeezing 10 pounds into a 5-pound bag.

Love Those Drake Knobs

In the photo above you see the back side of the original panel and the machining I did for the moved controls which are on the new panel overlay. 

I actually found a way to anchor the color TFT, so you don't see the mounting holes. In some cases, panel controls have a short shaft and so the original panel was cut out, so those controls are mounted directly on the new front panel. In the case of the Mic jack and volume control their shafts were long enough so they could penetrate both the old panel and new panel with a bit of room to spare.

A few stations were on late yesterday afternoon, but one of the very stations I heard, N8OO, commented that all the bands were dead. So, I still don't know the original issue with the ZIA.

When I first powered it on, the receiver was DEAD. Oh, Oh -- but then I saw an empty LM380N-8 socket. I must have filched that for another project. A replacement installed and I heard lots of noise but only one or maybe two signals!

This is the third and last posting on the ZIA.

Them that know can make it go.

Plan View the Zia

This is but one or two of my rigs that do not use an IRF510 but an MRF260. Wish I had a bunch more of those devices.

73's

Pete N6QW

An Opportunity Lost.

 MFJ 94XX Series SSB/CW Transceiver

Shown above the 10M Version

Perhaps now close to 20 years ago, MFJ (now out of business) introduced a line of single band SSB / CW (optional kit) QRP Transceivers. These were small, a single conversion design and good for about 10-12 watts. 

I jumped in with both feet and bought the 20M variant. I went whole hog and bought the optional CW board and the 12VDC Wal Wart. That set me back well over $300. I should have not jumped so quickly! It suffered from VFO drift but otherwise was OK. This was a great idea as there were models covering HF as well as 6 and 2 Meters. Too bad it had an analog VFO.

Were there the Arduino and Si5351 at this time they could have gotten $500 for a 20M version.

The used market has treated these units well and often sell for well over $200 even the hardly working units. So, last week when I spotted a half a dozen units being sold on eBay by the same person an idea hit my shrinking brain. [I am told our brain shrinks as we get older.] There were three in that lot all on 75M and the starting prices were $70, $90 and $110. I decided to bid on the middle one.

So, what was my idea. Upon looking at the 75M unit, I found that there is actually a sub-panel behind the front panel which is held in place by the controls and one bolt. The plan was to remove the front panel and to make a new front panel. In place of the Fine Tune control, I would put a toggle switch for either two frequency range control or USB/LSB select. 

In the upper right-hand corner (printing area), I would install a smaller version of the ST7735 color display which is less than 1-inch square. The piece de resistance would be the JF3HZN circular dial. All other controls would remain as original.

The CW capability in the original was a separate board anchored inside the case and this a place to install the ESP32 and Si5351.

I printed the schematic of the 94XX Series, and it is a single conversion design using a 6-pole filter and lots of NE602's. 

There is a ton of surface mount parts and if you purchased a 75M version to put it on 20M you would need to change out the Band Pass Filter and the Low Pass Filter. Since the Si5351 would supply the LO and BFO, a simple removal of the BFO crystal is all that is needed. 

Or you could keep that crystal and do like I did on a Bitx 40 to add USB to a LSB only rig. A switch changed the LO range for the other sideband to create a sideband inversion thus the opposite sideband with one BFO crystal. 

The Band Pass Filter as far as I could see, in the small photo I had, is all SMD where I suspect the inductors could stay and with LT Spice you could find new capacitor values for 20M. But the LPF used discrete through hole parts so new parts could be installed for 20M.

The auction ended last night before midnight an up until three hours (9PM) before there were no bids. So, I bid $105 on the listed $90 unit. That bid was my gag level as I figured about $30 for additional parts and $20 for shipping. I was outbid near midnight and that 75M unit sold for $205. 

I am glad I lost as I could build one hell of a rig for $255 which is what I would have spent in total were I the winning bid.

So, I am still looking for a commercial radio where I can install the Circular dial from JF3HZB. A Heathkit Monobander or an MFJ 94XX would gladly be accepted and I will pay for the shipping. Check your garage stash and look under the work bench and see what is lurking there. Your contribution would be greatly appreciated.

Them that know can make things go.

73's

Pete N6QW