Amateur Radio Station VE6BC

This next item is not just a piece of radio history, it is a piece of local radio history. Local being Edmonton, Alberta, Canada.

It’s a crystal oven for a broadcast transmitter. I don’t know what years it was in use but given the frequency of 930 kHz it is a virtual certainty that it was used at a local AM broadcast station, CJCA, which has operated on that frequency from 1935 to this day.

CJCA was, in fact, the very first radio station here in Edmonton starting in 1922. And there is a personal connection for me as my grandmother Hazel worked for them in 1922 (they were on another frequency then) when they were starting up operations in an unheated shack on the roof of the Edmonton Journal building in downtown Edmonton where she worked as a stenographer.

I picked this unit up at an estate sale of a long term and avid collector of vintage radio equipment. It’s in perfect condition. It was made by Bliley Electric Company of Erie, PA and bears a model number of BC-46 on the metal plate on the top and BC-46T on the label affixed to the side of the unit.

Here’s a picture of the top plate showing the frequency of the crystal inside and the specified over temperature of 52C.

The next picture shows the front of the unit where you can see the thermometer showing the oven temperature and a label indicating the oven power supply specifications of 10V @ 1A.

When I stumbled across this device and noted the frequency I knew that I should acquire it and preserve a piece of local radio history. I was curious to find out if it still worked and the first step was verifying that the oven still worked.

I hooked it up to my trusty HP power supply and dialed up 10V DC. I noted that the unit was drawing 900 mA of current. After a few minutes the mercury in the old thermometer (and it’s silver, not red, so that may help to date this thing) started to rise and the ceramic housing felt warm to the touch! Several minutes later it was sitting at 50C.

There’s the test set up. Amps on the left meter and volts on the right. I noted that the temperature rose to 50C and then the thermostatic control circuitry inside this little gem kicked in. I observed the heater cycling on and off to maintain that temperature. The first functional test had passed!

The unit has a base that plugs into the old five pin tube sockets. The same type used by the 807 triode and others. At some point I shall have to build a test platform for it. I am envisioning a small metal box with an oscillator circuit, terminals for external power and RF out with a socket on top to plug the module into. If and when I do that I will document it here.

When I was a young boy in the 1960’s my grandmother always had CJCA on her Simpsons-Sears table top vacuum tube radio on her kitchen table. I ate many meals at that table with CJCA tuned in and warm audio coming out of the old radio. As an adult I listened to CJCA after they switched to a talk radio format. In the mid-1990’s they switched to broadcasting religious rubbish and they are no longer fit to listen to.

But there was a time when CJCA was the king of the AM airwaves here in our good city and so I shall keep and preserve this device. I love the old technology and also the cutting edge. It’s all wonderful stuff!

Last year I acquired a very nice Heathkit SB-620. It was complete, clean and very good cosmetically. I thought that it would make a fun addition to use with my vintage receivers so I undertook the project of getting it working properly. Here is the how the interior looked when I received it (other than the big electrolytic capacitor that I replaced before taking this photo):

These units are basically a narrow bandwidth spectrum analyzer designed to take a feed from the IF of a receiver and display a pass-band of a width selected via the controls. When the kit was assembled the user would choose the frequency of the IF to be used. The frequency chosen affects a number of different components in the circuitry. Heathkit provided all the components needed for a wide variety of Intermediate Frequencies and the builder would use the ones needed. The rest could be put into the old junk box.

The unit I purchased was built for an IF of 3395 kHz which was used by Heathkit equipment of the era as well as some other receiver manufacturers. As I intended to use this unit with my Collins 51S-1 oy maybe my Collins R-390A I’d have to do something about the incompatible Intermediate Frequency. But that was a problem for later.

Right now the goal was to get the unit into proper working condition. To that end I ordered a capacitor kit from Hayseed Hamfest and added the optional resistor kit. That seemed prudent given the small additional cost and give that 60 year old carbon resistors could easily be out of tolerance.

While I awaited the arrival of said kit I checked the tubes and replaced any that were suspect. Then I gave the wafer switches a good cleaning with DeoxIT and MG Chemicals Nu-Trol contact cleaner. Bringing the little unit up on a Variac showed that there was a nice trace on the CRT and the CRT itself looked to be in excellent condition. Based on that and the excellent condition of the face and cabinet it appears that this unit had seen not a lot of use and had led a soft life in someone’s ham shack somewhere.

Once the capacitor/resistor kit arrived I set about replacing all those components. This was a relatively straightforward and enjoyable endeavour. Who doesn’t enjoy the smell of a hot soldering iron and flux vapour in the shack, punctuated by the occasional “thwap!” of the ol’ Soldapullt?

There’s the underside of the chassis after the capacitor and resistors had been replaced:

Here you can see the nice chrome-plated multi-section capacitor can provided by Hayseed Hamfest. The can contains four electrolytic capacitors in this unit.

Once the new R’s and C’s were installed and an initial smoke test showed no problems and a nice display it was time for the alignment process. The manual offers two methods for this, one with test equipment and one without. As I am fortunate enough to have good test equipment I used that method.

As this little spectrum analyzer is basically a super-heterodyne receiver the alignment process is similar. I normally do two passes through the alignment process to get it spot on but I just did one in this case as I may end up modifying it for a different IF anyway. I just wanted to make sure it was working properly and it seems to be.

I am pleased with the result and now I need to figure out how I will get it hooked up to the 51S-1 which has an IF 0f 500 kHz. The 51S-1 does have an “IF out” jack on the rear apron so that part is solved.

Here is mine running during the alignment process.

The cabinet, knobs and yellow display filter are all in excellent condition. The SB-620 is not nearly as common as it’s transmitter mate, the SB-610, is so I was lucky to get one and in such good condition.

The CRT is a 3Rp7 which is no longer available and uses a high-persistence phosphor to compensate for the slow scan rate of the beam. I was doubly grateful that the CRT was in excellent shape! While a substitute could likely be found it would not be the high-persistence 3RP7 that is really needed for proper use of this machine.

The P7 phosphor gives the high persistence display needed for spectrum analyzers and radar displays. P7 has two colors giving two different persistences: there is a short-to-medium one via purple-blue fluorescence and a long one via yellow-green phosphorescence. By using a color filter after the screen either color, and therefore either persistence, can be selected. In the case of the SB-620 they wanted to use the longer persistence and so you have a yellow filter in front of the CRT.

There are also two NE-83 neon lamps used in the circuitry. One you can see on the front panel. But it is more than a power indicator. It’s main use is actually as a voltage regulator for the local oscillator. The other NE-83 can be seen on the underside of the chassis. It is part of the relaxation oscillator circuit in the horizontal sweep section. Because neon lamps degrade over time and because they perform important functions in this unit I’d like to replace them. That is also on the to-do list for the rejuvenation of this fine little Heatkhit Scanalyzer.

More to come on this project as it progresses.

I recently attended a sale of equipment from the estate of a local ham, Sig Preuss, VE6SP. At the sale I picked up a Hallicrafters SC-38C that was in decent condition cosmetically and that was complete. No, that is not a typo in the model number. More on that later. It’s a very interesting story!

There was also an S-38 offered at the sale. I asked my XYL which one she preferred the looks of as it was my intent that this stylish little radio would go in the living room so we would have a radio on the main floor of our home to listen to the news on. She preferred the look of Hammertone finish on the SC-38C to the black of the S-38. So that’s the one that came home with us.

I eased the little Halli slowly back to life on my Variac and found that it did indeed work, which was a great start to the project! Naturally it had some problems, as could only be expected in a 70 year old radio. It was not very sensitive and the audio quality was poor.

The case was not in bad shape. There was no damage to it, just some small scuffs and scratches here and there, nothing of significance. While not perfect it was certainly good enough. My XYL liked the patina and originality of the radio and didn’t feel that it should be re-finished. I agreed with her. The knobs were filthy but undamaged. The dial windows were dis-coloured and no longer clear. All things that were readily fixable.

The worst thing, for me, was that a prior owner had drilled a 1/4″ hole in the front panel, below the model number, and installed an RCA jack which was wired to the headphone connector on the rear apron using #18 bell wire. That give a clue as to when it was done. Perhaps 1960’s or 1970’s? Something would have to be done about that and I had an idea. The good news was that whoever did it had chosen a good location and centered it under the model number. This turned out to be fortuitous.

I pulled the knobs and set them soaking in hot, soapy water. The radio was taken out of the case for a full inspection. This revealed that one tube was the wrong type (12SH7 in place of 12SG7 IF amp), the chassis was naturally dirty including the plates of the main tuning cap, the dial lamp was out and all the original wax/paper capacitors were still in place, as was the multi-section electrolytic in the power supply. The main tuning dial cord was also broken.

A fascinating detail that had gone unnoticed now came to light. The bottom cover had the usual Hallicrafters labels affixed to it. They were worn but perfectly legible. The main label, however, held a surprise. It said two things of interest. One, it said “the hallicrafters CANADA ltd” and the other was “MODEL SC-38C“. This radio was built in Canada and was given a model number that reflected that! Who knew? Certainly not I.

Also, on the top left corner of the front panel where it would normally say “S-38C” it said “SC-38C”. On the top right of the front panel it said just “hallicrafters” rather than “the hallicrafters co.” as US-made radios did.

This was a new and interesting facet of Hallicrafters history to me. Some searching revealed that Hallicrafters had opened a factory in Don Mills, ON in 1954. In 1955 they sold it. As the S-38C was made from 1953 to 1955 then my SC-38C was made in either 1954 or 1955. I can find no reference anywhere to the SC-38C but I did find references to the SC-40B and the SC-77A which were the Canadian variants of the S-40B and S-77A respectively. They would have also been built in the Hallicrafters Canada factory in Don Mills, ON. Interesting Canadian radio history, to be sure.

If you are reading this and you have any knowledge of Hallicrafters in Canada, their Canadian-made radios or anything related, please shoot me a message. My email is good at QRZ.com

The inspection being complete a plan was thus formed. I ordered a re-cap kit from Hayseed Hamfest http://hayseedhamfest.com and new dial windows from Retro Radio Repair http://dialcover.com. Tubes were tested and weak ones identifed. The 50L6 AF output amp was shot. The 12SG7 IF amp had the wrong tube in it’s socket, a 12SH7. The 12SQ7 AF amp/detector/AVC was weak as was the 12SA7 converter. Only the 35Z5 half-wave rectifier was healthy.

Good tubes were sourced (thanks to my pal Neil, VE6TCK/VE6JW) and while I waited for the caps and dial windows to arrive I started the cleanup. The main tuning cap plates were very dusty so those were cleaned. A basic chassis cleaning was done. The dial cord was re-strung, a mercifully simple path on this little radio! Thanks again to Neil who provided a length of the right dial cord. The lone dial lamp was replaced (#47). The knobs were pulled from their soapy bath and given a good scrubbing with a nylon brush and put back into another hot, soapy bath for round two. They were not quite clean enough yet.

Next the old dial windows and speaker were removed and the case was cleaned inside and out. The original speaker had to go. The problem there was that the spacing of the mounting studs was not compatible with anything I could find off the shelf. After some days spent searching for a compatible speaker, and half in desperation, I disassembled a Harris mobile speaker I had as it looked like it might be close. And it was!

The stud spacing was tight but it fit. All that was needed was to tweak one of the corners of the speaker with my Klein Grips so that it didn’t interfere with the top lip of the case. The Harris speaker was even the right impedance. Hallicrafters specified a 3.2 ohm speaker and this one was 4 ohms. I’d always like the sound of these speakers with my VHF/UHF mobile radios so I knew it would sound good in the little Halli.

Now I needed to do something about the 1/4″ hole in the front panel. My pal Neil came to the rescue, yet again, with a vintage 5/16″ neon panel lamp that absolutely looked the part for a radio of this vintage. It had a nice red lens and all I had to do was enlarge the 1/4″ hole to 5/16″ to install it. This was a good thing anyway because whoever drilled the 1/4″ hole left a bunch of burrs on the inside of the case. Sloppy work, that.

The dial windows and capacitor kit arrived so now it was time for the electronics work. The dial windows were a perfect fit and reasonably priced, too. And what a difference they made to the appearance of the little radio! Next up was the capacitors. I always do the power supply electrolytics first and then fire the radio up and see how it is.

There were then six paper/wax caps to replace so I did two at a time and fired the radio up between rounds to ensure I hadn’t make any mistakes and to see what improvement there was. The big improvement, to my ear, came from the capacitor that coupled the detector to the AF amp. It was likely leaking some of the B+ from the detector anode into the grid of the AF amp.

Once all the capacitors were in place I considered the state of the three slide switches. Those things were cheap when they were new and they never age well. As I was never going to change them from their default positions I hard-wired them into that state. I didn’t want any high-resistance connections in those parts of the circuit. I also installed a polarized two-wire power plug. As this radio has no power transformer one side of the line is connected to the chassis at all times. I made sure that was the neutral side.

Now it was time for the alignment. For this I use my trusted HP 8657A signal generator as the signal source and my Simpson 260 VOM to measure the output. An analog meter is the only way to go when peaking levels.

It’s a very simple radio and the alignment procedure reflects that. When I align a radio I always do it twice. Once to become familiar with the process and once more to get it just right. Amazingly, the dial calibration was very close from the start. The signals did peak up very nicely as I went through the alignment. Octal tubes don’t do well above 20 MHz and the SC-38C is pretty deaf on the high band, as expected.

I put the chassis back in the case, installed the knobs, wired up the speaker and new old power indicator and hooked up a loop antenna I had from an old stereo receiver we had. Unlike other five tube sets the little Halli has no internal antenna. The final bench test after re-assembly was a success.

Once in it’s new home in the alcove above our fireplace the little radio looked good and sounded even better. The power indicator that shouldn’t be there actually looks like it belongs there with its’ soft red glow. The #47 dial lamp is dim, as they are, and the whole package just looks “right’. I am very pleased with how it came out and I couldn’t have done it without the help of my pal Neil.

I chose the S-38 series specifically because they are small and very stylish. These were the first radios that famed industrial designer Raymond Loewy designed for Hallicrafters. He went on to do more work for The Hallicrafters Co. They are a very simple design and work amazingly well for what they are. There is beauty in simplicity.

Using the little loop antenna here in the North Saskatchewan River valley in central Edmonton the local AM BCB stations are loud and clear and you can tell that the AVC is working well: 580 CHAH, 630 CHED, 740 CBX, 880 CHQT and 930 CJCA.

840 CFCW in Camrose is not as strong as the local stations but it is clear. 1440 CKJR in Wetaskiwin, which is 10kW compared to 50kW for all the others, is weak and noisy but can be received. That was during the daytime.

I am looking forward to hear what winter nights bring!

When I first built my AllStarLink node I had the choice of using the original AllStarLink software or the newer and more advanced HamVOIP software. At the time I chose to go with HamVOIP.

This was not an easy choice to make because despite HamVOIP being “newer and better” than ASL, the source code was being withheld by the developers. This was in clear violation of the terms of the GPL (the GNU General Public Licenses).

There’s a good amount of discussion available on this subject all over the place so I won’t bore you with it here. I will say that I have a low opinion of those in charge at HamVOIP.

I was therefore extremely pleased when the folks at AllStarLink recently released a new version of their software. The improvements are vast. Goodbye, HamVOIP. Hello AllStarLink v3.

I would recommend that all HamVOIP users make the switch. You will be happy that you did.

I also moved my AllStarLink node from my Raspberry Pi 4B to a Pi 3B. I wanted the Pi 4 for the DMR node and the Pi 3 was plenty powerful for the AllStarLink node. The Pi 3 fan is a 5V job but I run it at 3.3V so that it’s quiet enough to live in the ham shack and not be a bother while still maintaining an excellent CPU temperature of 42C.

I had constructed my MMDVM DMR hot-spot using a Raspberry Pi 3 because the received wisdon was that putting the MMDVM hat on a Raspberry Pi 4 would cause it to overheat. My problem was that I found the web interface to be a little slow for my liking. It was usable, I just would rather it were more responsive.

So I set out on a search to find a way to use the MMVDM hat on a Raspberry Pi 4 and provide proper cooling. The solution was found in the Argon Fan Hat. The key feature that allowed this fan hat to work was that it passed the GPIO connector through to allow for the MMDVM hat to sit on top of it.

That, coupled with a Raspberry Pi stand-off salad allowed me to make a triple-decker sandwich. Left to right in the picture below we have:

• Raspberry Pi 4B
• Argon fan hat (plugs into GPIO header and passes it through to the next layer above)
• MMDVM dual hat (plugs into the GPIO header passed through the Argon fan hat)

By using a Raspberry Pi hardware salad and selecting the right combinations nuts and stand-offs between the sandwich layers it all went together nicely. One thing to watch out for if you do this: I had to remove one of the bolts and nuts that attach the fan to the fan hat PCB. this was because it interfered with the CPU heat sink on the Raspberry Pi and wanted to unseat it.

The Argon fan hat is here:

The fan installs some software packages to do it’s thing. They don’t seem to conflict with any of the WPSD software packages that run the hot-spot. There is a configuration utility that allows you to configure the fan behaviour from on all the time to various speeds at various CPU temperatures.

Testing is in progress but initial impressions are good. The WPSD web interface is much more responsive using the Raspberry Pi 4B than it was on the 3B. I have the Argon fan set to run at 30% speed all of the time and it is maintaining a CPU temperature of 35C which is excellent.

While the fan is not overly noisy at 30% speed, my intolerance of fan noise in the ham shack means that my DMR hot-spot has been relocated to my Harry Potter data centre under the stairs along with the rest of the IT infrastructure.

I am very pleased with the overall results of this configuration.

I’ve been fooling around with DMR for a few years now and figured it was time to build a DMR hot-spot. I had a few DMR radios on hand already. Some Motorola XPR handheld radios in both VHF and UHF trim and also a TYT MD-380 for 70cm. But what really pushed me to do this was my desire to experiment on the 33cm band and the availability of the Retevis RT10 which is a 33cm radio that does DMR. And the fact that many of the MMDVM boards support 33cm.

First I acquired an RT10. Now I needed to make a DMR hot-spot. I learned that the Raspberry Pi 4B had issues with heat when coupled with an MMDVM hat and no fan so I picked up a Raspberry Pi 3B kit as the base for this project. With it’s lower clock speed and reduced heat output it was said to work good with a “hat” on it and no fan. Fans piss me off. The only noise in my ham shack should be coming from the speakers in my radios!

Next I grabbed an MMDVM “hat” (from Amazon) that plugs into the Raspberry Pi ‘s GPIO header. I chose a duplex one with a display just to keep my options open. For the difference in price it seemed silly not to. The one I got had a TCXO (temperature controlled crystal oscillator) so that eliminated the need to tune the offsets. I got lucky in that regard but given what I know now I would never even consider getting an MMDVM that didn’t have a TCXO.

For an operating environment I selected the WPSD software. It appeared to be the best choice at this time.

It all went together quickly and easily. Because I was already familiar with DMR and all that it entails I had little trouble getting it all set up. I joined my hot-spot to the Brandmeister network for my initial testing and usage. There are others but Brandmeister looked like the big dog on the block and a good place to start.

Experimentation with my DMR hot-spot continues. Once it was tested with the Retevis RT10 on 33cm I set it up for 70cm use with my old TYT MD-380, which also worked well. The MD-380 is a better radio that the RT10 in many ways but the point of my experiment was to use 33cm for something. And now I am.

Talk group 91, the worldwide talk group on the Brandmeister network, is quite active and always entertaining. Perhaps we will cross paths there.

Here’s a picture of the completed hot-spot in action:

A couple of my friends have been experimenting with AllStarLink and it looked interesting to me so I thought I would join in the fun. I’ve never had a Raspberry Pi and this gave me a reason to get one.

My plan was to build a simplex AllStarLink node for use on the 70cm band. It would look like this when done:

Raspberry Pi <–> CM108 sound card <–> Interface <–> Motorola XPR 4550

This article is about the interface between the CM108 sound card and the Motorola mobile radio as it was the only piece needed to complete the project. There are commercial options but given how simple it is home-brewing seemed the best option. Much cheaper and way more fun than buying something. Plus I had everything I needed for it on hand other than a box to put it in.

I ordered a Raspberry Pi 4B kit and a CM108 USB sound card. The CM108 needs some hardware modification to be used for this. My pal Neil offered to do the mods to it. He has the equipment for SMD work so I just said yes.

I requested a node number from allstarlink.org. Then I installed and configured the HamVOIP software from hamvoip.org. Although I am a BSD guy I have been known to slum it in Linux-land on occasion.

I already had a Motorola XPR 4550 up on 70cm so my plan was to use that radio for this project. The only missing piece was an interface circuit to massage PTT and COR between the CM108 sound card and the Motorola. Transmit and receive audio would just be passed through between the sound card and radio as those levels are compatible and can be adjusted to suit from the HamVOIP software.

I wanted status indicators for PTT and COR and a buffer between the sound card and the Motorola. The wee sound card is a on the wimpy side in terms of driving anything. My preference is to run PTT and COR both active low on the radio. But the CM108 USB sound card must run with PTT active high due to the hardware hackery associated with getting a PTT signal out of the thing.

So I needed a non-inverting buffer for COR and an inverting buffer for PTT. A simple enough proposition. I used my two go-to transistors for this. A 2N3904 NPN transistor would take a high PTT from the CM108 and turn it into a low PTT into the Motorola while driving a red LED. A 2N3906 PNP transistor would be used to take low COR from the Motorola and send it on to the CM108 while driving a LED.

The Motorola XPR has a “switched B+” output (i.e. battery voltage) available on its’ accessory connector so I also needed a 78L05 voltage regulator to knock that down to 5V to power the interface circuitry. That would give signal levels that would be compatible between the CM108 and Motorola XPR.

On the bread board it came out like this:

The wires on the left go to the CM108 on the Raspberry Pi and the wires on the right go to the Motorola XPR. The red PTT LED is on the left with it’s transistor and the green COR LED is in the middle with it’s transistor. The 78L05 voltage regulator is on the right. I didn’t put any RF chokes in the bread board circuit but I planned to when I built the prototype.

There’s the prototype board with the circuit built and now including RF chokes on the signal lines in and out (left side goes to sound card and right side goes to radio). The two-pin headers are where the LEDs will connect. The case is a Hammond unit and the proto-board is a perfect mate to it made by BusBoard Prototype Systems in nearby Calgary. A great combination!

There it is fully wired and ready for the smoke test. A metal case would be better for shielding. Maybe for the next one but this would do for now. Also for the next one some RF bypass capacitors throughout the circuit would be a good addition.

There it is done and on the air. The interface circuit can be used to connect virtually any commercial VHF or UHF radio to a CM108 sound card for use in an AllStarLink node. This was one of my goals.

Radio interface:

1. PTT active low
2. COR active low
3. De-bounce of 100 ms on both PTT and COR
4. +12 V DC power is supplied from the radio
5. Tx audio connected directly to sound card
6. Rx audio connected directly to sound card

I guess we will have to call this a latent post as it is out of chronological order. I undertook this project almost exactly one year ago but I didn’t have this website at that time. Work began in November 2022 and was completed in January 2023.

I had acquired a complete, working and cosmetically very good Hammarlund HQ-129X in the spring of 2022 at a local ham radio flea market. It spoke to me because I love Hammarlund radios, it’s the last of their octal tube based receivers and I didn’t have a vacuum tube based general coverage receiver. The picture above is it as it looked when I got it.

This was my first radio repair/restoration project undertaking in my newly re-constructed ham shack. The front panel on this radio had already been replaced/refurbished, the knobs were all there and in good shape and the chassis was very clean. That made for a great starting point. So what did it need?

Firstly, both the Main Tuning and Band Spread dials were discoloured from decades of sun shining on one side of them and the dial lamps shining on the other. Secondly, the unit was still in possession of all of its’ original electrolytic and paper/wax capacitors. Thirdly, a check of the tubes found that the 5U4 rectifier was on its’ last legs and the 6SN7 audio pre-amp was a little weak.

And of course a full RF and IF alignment would be the last step. I could also tell that the radio was out of alignment. The calibration of the main tuning dial was off and engaging the excellent Hammarlund crystal filter clobbered the signal badly. Clearly the IF and the crystal filter frequency were not in agreement.

I ordered new tuning dials from Radio Daze and they were perfectly correct save for one small detail which I will get into later. I also ordered a re-cap kit from Hayseed Hamfest. This was a matter of convenience. I could have pieced together a re-cap kit myself but this saved me a bunch of work and I found the price reasonable. I also ordered the two tubes I needed to replace plus one complete set of spares.

Now the fun began!

Vacuum tubes

Using my trusty but simple B&K 606 tube tester I checked all the tubes. The 5U4 full-wave rectifier barely moved the needle. So that was bad. The 5U4 was running awfully hot yet the radio did function. It was amazing how well it worked, in fact. The first audio amplifier, a 6SN7, was on the weak side so it had to go, too.

I acquired a full set of NOS spares plus one 6SN7 and one 5U4. I installed the new set of tubes and took the old ones and put them into inventory. I now had all NOS tubes in the radio and one complete set of spares on hand.

Capacitors

The electrolytic caps in the power supply had to go (four caps in one can) and so did the 27 paper/wax caps throughout the radio. Luckily, Hayseed Hamfest sells a re-cap kit for this radio.

Once the capacitors arrived I started with the big quad-electrolytic can in the power supply. It was a bit tight as the new capacitor was physically longer than the original by about 1″. Then it was on to all the rest of the smaller caps. I did 5 or 6 a night until that work was complete. If I got to a point in the evening where it wasn’t fun and felt like work I would stop work and pour a whisky.

The caps on the RF/IF side of the radio were easy, lots of room there to work. The caps on the PS and AF side of the radio were harder due to tight quarters, especially the ones in the BFO can. And there were several small dead insects inside the BFO can. How did they get there? It is sealed unless the 6SJ7 is removed thus exposing the hole in the octal base. That must have been it.

I found this part of the refurbishment very enjoyable. I made only one mistake. Given the amount of things I did I thought that was pretty good. Taking out the large, old paper/wax caps opened up a lot of room to install the new small polystyrene caps. All the small yellow polystyrene caps look right at home in the now seemingly cavernous underbelly of the Hammarlund chassis.

Tuning dials

The tuning dials were very yellowed in parts, even brown in some areas. I didn’t know when I bought this radio one could get new tuning dials from Radio Daze. Once I learned that I ordered a set. This radio was going to look like it did when it was new!

I removed the old tuning dials and noted that they were attached to their hubs with tiny copper rivets. I carefully drilled the wee rivets out and then set about figuring out how to mount the new dials. The holes were small and I needed something that was thin like the old rivets once installed. This was to avoid clearance issues with the dial stops.

A fellow in a forum said that he used some #2 hardware for this and it worked. Excellent! I’d never looked for #2 hardware (nuts, bolts, washers) but I figured our excellent local nut and bolt supply house could fix me up. Not so, I was told. They went down to #4 and had nothing smaller.

I located a source of #2 hardware online (The Bolt Depot) and ordered 100 each of #2-56 bolts (1/8″ long), nuts and washers (just in case). The shipping ($14) was more than the hardware ($13) and I now have a lifetime’s supply of #2 hardware. In brass. I figured a non-ferrous material (copper rivets) was used originally for a good reason so I stuck with that idea.

After bolting the new dials to their hubs and re-installing them into the radio I discovered two problems. One was that the #2 hardware was thicker than the old rivets were and they interfered with the dial stops when tuning. A thin washer on each tuning shaft shimmed them far enough away that they cleared the tuning stops.

The second was that the main tuning dial was not perfectly round and would bind against the body of the S meter as you tuned toward one end of the band. I solved this problem by attaching the dial’s hub to a 1/4″ drill bit in my variable speed hand drill and spinning it against some 400-grit sandpaper until it was rounder and ever so slightly smaller than when I started. I probably took off less than 1/64″ to make it fit.

That was the end of the mechanical challenges. Now back to the electronics!

Dial lamps

The radio has 3 dial lamps. One for each of the main tuning dial, S meter and band spread dial. These were #47 lamps with bayonet bases in sockets. Each socket had the old phenolic “button” in the base for the centre lamp contact.

One of the phenolic bases broke during disassembly, rendering the socket unusable so I decided to replace the old incandescent lamps with LED’s. This would have a couple of advantages. It would remove ~400 mA of load from the 6.3V heater winding on the transformer and it would remove the dial lamp heat from the tuning dials.

I am intolerant of 60 Hz flicker and so I designed a small half-wave DC power supply for the LEDs. Like so:

That was my first draft and not what finally ended up in the radio. I tweaked the values of the filter capacitor and current limiting resistor a bit to get it just right to deliver ~20mA of current to each LED.

Here is the wee DC power supply for the LEDs installed on top of the tuning cover inside the radio right behind the main tuning dial. I used a chunk of leftover terminal strip to build it, because, well, you know, it was on hand. The LEDs are not yet connected. The 6.3 VAC enters stage left and you can follow it from there.

To install the LEDs in the old lamp sockets I wrapped each LED in concentric layers of heat shrink tubing, shrunk it on and added layers of ever-increasing diameters of tubing to get to the right size that would allow for a good friction fit inside the lamp socket housing.

Once installed the new amber LEDs could not have looked better. They illuminated the dials and meter very nicely and were easy on the eyes. I was very pleased with the results and all using parts that were on hand.

New tuning knobs

The general consensus is that these radios have tuning and band spread knobs that are way too small. I agree, they really are. So I acquired some knobs that looked more appropriate and offered a more pleasurable tuning experience. I believe they are Raytheon knobs.

They’re a big improvement but they’re still a little too small. The newer Hammarlund radios corrected this ergonomic fault so I shall have to keep searching. In the mean time these will do. I kept the original knobs, of course, in case the next owner wants to put it back to original.

Full alignment

And now for the final and arguably most fun part of the project. All the electronics are in good shape, the power supply voltages look good, it’s time for a full alignment of the receiver.

I followed the Hammarlund manual for this. I am fortunate to have an excellent signal generator, an HP 8657A, which is also fully calibrated so I knew I’d end up with a good result. I also have an HP 5328A frequency counter, also properly calibrated, to help nail down the crystal filter frequency.

The IF comes first and there are three gain stages plus a crystal filter between two of them. The IF alignment goes like this:

• Working backwards from output to input peak each IF stage at 455 kHz
• Determine the exact frequency of the crystal in the filter (it’s very close to 455 kHz but it’s not exactly 455 kHz)
• Re-peak the IF stages, in reverse order again, to the crystal frequency learned above

The challenge with the IF is that the tightest crystal filter setting is 200 Hz and you therefore need to get your IF stages bang on to the crystal filter frequency. This is quite fiddly. Then it’s on to the RF alignment, which is quite straightforward but gets a bit fiddly on this higher bands. It gets pretty touchy above 15 MHz.

There is an adjustment for the S meter also. Like Collins and many others, Hammarlund also specifies that a reading of S9 corresponds to 50uV of signal. The BFO alignment is pretty easy, not that I will be using it, but I wanted all aspects of the radio to be proper.

A good alignment brought this radio to life again! It was operating beautifully and properly. It was sensitive, selective and the audio quality was excellent. Because it has a BFO and not a product detector it’s not the world’s best SSB rig but that’s not my intended use. This is my band-cruiser for AM signals. It excels at broadcast band DX and I do enjoy that.

I was more than pleased with the final results. Both the look of the radio and its’ performance are excellent. I should have taken more photos of the process.

The picture below is the unit after all work was completed. The blotchy appearance of the dial lamps is an artifact of the iPhone camera I used to take the photo. In reality there is just a nice, smooth gradient of warm amber illumination.

I recently acquired a Hallicrafters S-77A that needed a little work. I’ve been wanting to add a Hallicrafters radio to my collection and in particular I wanted one of the ones styled by the famed industrial designer Raymond Loewy (look him up!). I got lucky and this one just kind of fell into my lap. The photo above is what it looked like when I received it.

The S-77A is the AC/DC/Marine version of the S-40B. Because of this the S-77A units are far less common than the S-40B.

The radio had already had almost all of it’s old paper/wax capacitors replaced with just two left to be done. And all of the electrolytic capacitors in the power supply had been done except for one. The radio worked pretty well as it was. The audio quality was good, the dial calibration was pretty good and it seemed reasonably sensitive given what it is, that being Hallicrafters’ mid-level general coverage receiver in the early 1950’s. Better than the low-end S-38 but not as good as the high-end SX-42.

All but two of the tubes tested good on my B&K 606 Dynajet tube tester. The two tubes in the audio section were a little soft. The 6SC7 AF pre-amp and the 25L6 AF power amp. I didn’t have either one in my tube inventory so I acquired some.

The two dial lamps were also the wrong type. This was a problem in this radio because it is an AC/DC radio so the tube filaments are wired in series along with the dial lamps which have parallel resistors to allow the proper amount of current to flow in the series circuit (~325 mA in this case). So to maintain proper filament voltage you have to maintain the proper series current, meaning that the dial lamps need to be the correct GE type 47 which consume 150mA of current. The balance of the 325 mA flows through the resistor in parallel with the #47 lamp.

Replacing the four capacitors, the two tubes and the two dial lamps was all the electronics work that was needed. The knobs were pulled and soaked in hot soapy water and cleaned with a nylon bristle brush. 71 years of accumulated filth was thus removed. Three of the knobs had damaged set screws so those were replaced. The front panel was cleaned up while the knobs were off. Cosmetically the radio was in good shape and a little clean up made it look even better!

It’s an AC/DC radio so there is no power transformer. It’s a two-wire line cord and one side of the line is tied to the chassis, as it was made in 1952 long before the three wire grounded circuits were in use. So there is a 50/50 chance of the chassis being hot. I should probably put a polarized plug on it.

The main tuning and band spread dials were indexed and then full IF and RF alignments were performed. Actually, I aligned it twice. Once to learn the process on this radio and once more to make sure it was as good as it could be. The radio works well but one thing I noticed is that because the B+ is only half-wave rectified there is more ripple on it than there would be in a full-wave rectifier configuration. This can be heard as a 60 Hz hum on the audio when the AF gain is all the way down.

This appears to be normal for this design. I measured the ripple at ~3% which seems about right for an unregulated 1/2-wave supply. Unregulated full-wave supplies normally have about 2% ripple. And why is the PS only half-wave, you ask? It’s a side-effect of it being an AC/DC radio. It can be powered from 115V AC or DC. Or 230V AC or DC. Because of this, Hallicrafters wired the two diodes in the rectifier tube, a 25Z6, in parallel. Hence half-wave, hence more ripple.

It’s working well now and it looks great. It’s primary use will be for AM broadcast band listening in my garage shop. Here’s a picture of it all done and in the garage shop:

The purpose of this blog is to share my ham radio activities with others who may find some of it entertaining or even useful.

I am a big fan of vacuum tube radios from the 1940’s through the 1960’s. As such you will see a lot of those kinds of radios being worked on and in service in my ham shack. Brands like Central Electronics, Collins, Hammarlund, Hallicrafters and E.F. Johnson for some examples.

I am also a fan of VHF/UHF FM repeater communications so there will be some of that, too.

Antennas and transmission lines are another major area of interest here despite the fact that I am not in a position to deploy anything of any size due to the size of our small city lot.

I got back into ham radio in early 2022 after a 20-plus year hiatus. That meant building a ham shack with operating positions and a test bench, unboxing and blowing the dust off of what radio/electronics items I had kept, acquiring the needed pieces of test equipment needed to service my radios and also acquiring a bunch more radios, in particular ones I’d wanted since I was a young SWL and later ham.

The limited size of my ham shack means that while I can’t have “one of everything” I can have a small number of radios that are “special” in some way. My collection of both radio equipment and test equipment is therefore curated to a specific set of criteria and designed to meet a defined need.

I hope you enjoy reading about my radio activities and perhaps even draw some inspiration for your own. If you’re an amateur radio operator get on the air in any way you can and use the bands. There is pressure from commercial interests targeting various chunks of our allocated spectrum and we need to use it or lose it. Get involved in a local club. Show the young folks what’s cool about our hobby.

If you’re not a ham but think you might like to be one, seek out a local club or national club that offers classes to prepare for the exams.