Reflecting new addition to the project

A vintage ham radio station needs a matching vintage SWR-meter. That is exactly what I got today. It is beaten up and untested. Like its new housemates (DX-60 and HR-10) it could use new knobs. I plan to make new knobs for them all. White knobs may be the closest to the originals. Image shows prototype of black knob on the DX-60. I kind of like it. The orange one is half a joke. What do you think?

HM-11 SWR meter shown along DX-60.

The front is quite nice but the case is dented and and scratched. It would need a paint job to get up to the same standard as the DX-60 it will be connected to. One suggestion though is that I leave it as it is but clean and protect it with Turtle Wax.

You can see slight differences in the nuance of the colors of the paint. Heathkit used several shades of the green color. This has been a source of confusion and frustration for restorers.

Internal components may need replacement, but I hope the galvanometer works.

The meter comes from the estate of LA8KJ, kindly administered by LA3RK who I had the delightful experience of meeting for the first time.

HM-11

It is a very conventional SWR meter with forward and reflected modes and a sensitivity adjustment so that the reflected mode can be calibrated in SWR. The intended frequency range is from 160 meter to 6 meter ham bands. Power levels up to 1 kW. As is common for such meters, don’t expect to use it for QRP levels.

The HM-11 is actually a new appearance of the AM-2. The circuitry is the same but the casing is styled to match the DX-60 and other equipment at the time. I think the technical term for this type of upgrade is SSNW (Same Shit – New Wrapping). Newer versions are HM-15 and HM-102 which are basically newer wrappings still.

The builder can select to build it for 50 ohms or 75 ohms by the choice of a couple of resistors.

LA8KJ

The meter was owned by LA8KJ, Carl Daniel Haaøen who went silent key 2nd of November 2015 at the age of 86. He was licensed from 1965 but was an active participant in his local group long before that, including a period as QSL manager.

As a young boy he worked at two different radio manufacturers but later found a career outside radio. His sense of humor shines through in this piece he wrote for “Hallo Hallo” (search for LA8KJ).

Thanks!

Thanks to LA3RK, Olaf for excellent service and also some parts he gave me for another project coming up!

Comparing old receivers

I am not finished with the HR-10 yet. A performance test is a little premature. Still, that is just what I did today. With the LA2OLD net running on 80 meter I listened using both the HR-10 and my old Grundig Satellit 2000 from about 1975. I should say, the Satellit 2000 was an expensive receiver at the time. It’s versatility and performance justified the price. HR-10 from the early 60’s was marketed as a “basic receiver” for amateur use only.

Grundig Satellit 2000 with optional SSB 2000 unit in foreground. Note that the uppermost of the three tuning knobs on the right have a fine tuning adapter on. I made this to ease tuning when I need to be extra precise. It slids easily on or off any of the three knobs.

Listening, first on AM then on SSB, the Satellit won hands down. I switched the 80 m halfwave dipole between them to compare. That dipole is of course much better than the telescopic antenna built into the Satellit. Still, the Satelitt with its own antenna did better than the HR-10 on the big dipole. Strong stations sounded much better on the Satellit. Weak stations that were unreadable or even below the noise floor on the HR-10 could be read on the Satellit.

Frequency stability of the Satellit is also much better. Something that is very noticeable on SSB.

Is the HR-10 a lost cause?

NO. I enjoy playing with it. The tuning dials are nice because they are made for amateur bands only.

I think there is room for improvement. Except for realignment that I have mentioned before, I think all electrolytic capacitors need a check and possible replacement. I am postponing that until I get working on the DX-60 so that I can borrow an ESR meter and order many capacitors at the same time.

Also, I think the preamp should be made switchable. One way or the other.

Instability may be curable with zener diodes in the power supply.

Never heard of Grundig Satellit 2000?

In brief:

150 kHz to 30MHz in 20 bands, except for 400-500 kHz. That’s AM with selectable bandwidth (2.5 or 5.5 kHz). 8 of the bands are bandspread BC bands, two of which also includes ham bands (40 and 15 meter). Also 87-108 MHz FM.

Optional SSB 2000 unit containing BFO and a product detector I actually understand (unlike this one). Also has a switchable audio filter that is useful for SSB and good for CW. A switch to disable AVC and an MVC potmeter which we would usually call RF gain.

Twisted progress

I have made a significant improvement. I will get to that but not without first boring you with some peculiarities from long time back.

You may remember the paper material I got. One reader called it a “time capsule”. Included was the book “Radiobyggeboken” (The radio construction book). Look at what I found in that book:

V2 on the left is last IF amp and detector. V3 to the right is the BFO. But what is the schematic symbol on the connection between the two transformers (top between MF2 and MF3)? Text explains it as a pair of wires twisted together to form a capacitor

May be capacitors were expensive in the 60’s. May be the builder was frugal. Or maybe this was common practice. I don’t know. The idea was new to me. Though I have to admit I once made a variable capacitor using cardboard and Aluminum foil.

It’s fun reading about almost forgotten tricks. This one hit home extra because I have been working on BFO injection to the IF in the HR-10 receiver. See my mostly failed experiments!

Product detector rant and mystery

Of course, the proper thing to do would be to build a product detector into the receiver. An example can be found on this blog post about improving the HR-10.

I can hardly think of the word “product detector” without thinking back to were I was introduced to it. That was in the Danish book “Vejen til sendetiladelsen” which I read in preparation for my license exam. I passed the exam thank you, despite reading a foreign book and never understanding the schematic of that product detector. I didn’t have Internet or search engines at the time, but I have now spent some time trying to find a similar schematic and an explanation. After some effort I found a close match:

From a nice RF Cafe blog where it is fig. 11. That blog seems to be a reprint from February 1972 Popular Electronics.

This is almost identical to what I found in the Danish book. The differences are irrelevant to my lack of understanding it. The similarity undermines one hypothesis; that there was an error in the schematic.

My problem is that I can’t understand how this works, or even how significant level of signal remains on the audio side. As I read the schematic, the point between the two diodes will quickly become positively charged, stopping almost all further signal propagation or mixing. A little will get through because of tiny reverse currents and capacitance in the diodes. But that should be minimal. Nothing in text or drawing suggests the breakdown voltage of the diodes are involved.

Did I finally find the explanation after all these years? Nope. The text in the blog/original article matched the Danish text so closely I am convinced the Danish writer had this article by his side when writing.

I now rely on my readers being smarter than me to explain the workings of that detector once and for all.


Should I modify the HR-10 with a product detector? I hesitate. This is not the best receiver I have. If I wanted to build a better receiver I would not use the HR-10 as a starting point. It has a high value as a really nice vintage receiver. I have fallen in love with this charming box with all its shortcomings. Putting much effort into making it closer to a modern radio seems hard to justify. But I could still pick some of the low hanging fruit. That was the idea behind my attempts at injecting more BFO power via a simple capacitor from the BFO to the screen grid of the IF amp. The lack of dramatic success may be because I tried to inject to the screen grid instead of the control grid. Or because I didn’t use enough capacitance.

What if I (reluctantly) connected the capacitor to the control grid instead? It would have to be a much smaller capacitor. Still, the IF resonant circuit would be detuned a bit and a little power lost to ground. Schematics in this blog post. I decided to try, accepting that I would have to realign the IF again. But I planned to do so anyway once I get to improve the signal generator (SG-402).

I felt like trying the idea from Radiobyggeboken. Meanwhile I had discovered this idea was not unique at all. SM7NDX mentions doing something like that on his HR-10. Unfortunately details are sketchy.

Come on let’s twist again

I just tried randomly. No calculations, no objective measurements.

Twisted enameled wire serves as a tiny capacitor connecting BFO to IF last amp.

I have no idea if this “capacitor” is close to the sweet spot or not. My ears can tell me this was a major improvement.

OK. A little dressing is needed etc. but I have (almost) picked one low hanging fruit. I have a twisted feeling of accomplishment.

IF not entirely successful…

After half a century one would expect the IF transformers to need a realignment. The method is simple.

  • Disable the local oscillator by simply grounding the grid.
  • Tune to the lowest frequency, i.e. approx. 3.5 MHz.
  • Alternately peak each core of each of the three transformers.

To do the peaking, a signal in the IF passband is required. The IF is about 1682 kHz. With the receiver tuned to 3.5 MHz enough signal leaks through to do the peaking. In the old days I used the local oscillator of another receiver as my signal source when I needed one. The best cost saving is no cost at all.

Some time ago I got an old Trio SG-402 signal generator. Luxury. The analog IM-5284 was connected to the speaker output so I could easily watch signal strength as I tweaked each transformer core. This was kind of a historic moment as this was the first time this millennium that the IM-5284 wss powered from batteries. I’ve mostly been using it for current measurements as I have several digital multimeters that works at least as well for all other measurements.

Signal generator on top of speaker. Meterman 37R in front used for frequency measurement. IM-5248 in background for signal strength measurement.

Alignment should be easy. It wasn’t. Some problems:

  • The signal generator was hard to turn precisely and didn’t stick well to frequency.
  • My body movements around the receiver influenced the measurements appreciably. I need to find a less care free way to introduce the signal into the receiver. See thin orange wire in photo! Some proper cables may be needed.
  • The large analog meter on the IM-5284 is wonderful but a little slow. Patience is required.
  • Minute adjustments of iron cores is difficult and more so when they are old. They tend to stick while you gradually increase torque, until they suddenly loosen and travel too far. Patience, patience, patience.

What I think I learned is that the adjustments needed were very small. With the inaccuracies in frequency etc. I am not even sure I improved anything.

I will try again, but I have to work on the first problem with the signal generator. I will try to devise a larger tuning wheel. For my Satellit 2000 I made an adapter to put on the tuning wheel when very fine adjustments are needed. I think I will try the same. Also, I consider opening the unit because the tuning did not feel smooth. I kind of expect to find a capacitor that hasn’t aged well. Familiar story?

If not entirely successful, try again! 🙂

Intriguing unit

Did I find gold? Read on to find out!

The set came with a bag of tubes. I didn’t have the time to study them, but one unit stood out. In blue metal can with white text.

Tubes now in a box. Intriguing unit on top.

Y201 and Y203 looks very much like crystal component IDs. 17.0 MC and 200 KC looks very much like frequencies of the respective crystals. 6.3 V happens to be a common filament voltage, including in the HR-10. Why would it need filament voltage unless there is a vacuum tube in there? Could this be a crystal calibrator?

The socket for the crystal calibrator in the HR-10 was empty. The original crystal calibrator for the HR-10 does not look like this, but it is conceivable that LA4FK had plans to adapt this one. That could be as easy as rewiring the socket. I was optimistic I had found something very useful. I would have to research what this is. Do you think I was on to something? Make up your own mind first. I’ll make a little but relevant digression.

The Collins R-390A/URR

The 1976 edition of the World Radio and TV Handbook is were I first saw the R-390A/URR. Looking like it belonged in a James Bond movie, the receiver was big, heavy, expensive but probably a good performer. It was hard to accept that any receiver could be better than the one I owned. An ad would never lie of course.

It was also the first time I saw the term “mechanical filters”.

For a long time Collins was a highly esteemed brand in radios. When their radio production faded they still kept on producing those magic mechanical filters for others. They were not at all inexpensive.

The R-390A lived on for a long time after it was technologically old. Not the least because the military was an eager customer. Many variants were made, like a lighter version for aircraft e.g.

A search on the Internet brought me to http://www.r-390a.net/Redux/01_Jan_03.pdf which contained the text “Calibrator Y201-17.0MC,Y203-200KC”. Note that this was a forum dedicated to the R-390. My heart started beating faster. Let’s keep on digressing.

I inventoried the tubes and found that many of them were used in only one product. The R-390A/URR.

I was told LA4FK had access to and sometimes brought home decommissioned radio equipment from his workplace in the military. It is not inconceivable that an R-390A could be among them. Seems he pulled the tubes and this blue thingy out of one.

To use it I had to find its pinout and other specs. My searching failed me but I kept on dreaming that I could get this crystal calibrator to work in the HR-10 with some modifications. After a few days I realized I could search for R-390A schematics and figure out the pinout from that instead. Took me seconds to find excellent copies of the schematics. I also found the Y203 et. al. in there.

So, did you make up you own mind about what this is?

Turns out this is a crystal oven with two crystals and a temperature controlled heater. The 6.3 V is for the heater. That also explains the “75°C” on the can. The 200 kHz (or 200 kC as they said in the old days) is indeed for a crystal calibrator. But all the active circuits are on a board in the R-390A. Not in the blue can. Sigh!

The 17 MHz crystal is for the first local oscillator.

Well, I learned a few things. Also, if an R-390A comes my way some day, I have spare tubes 🙂

Dead tree resources

LA4FK appears to have been a well organized person. With the set came a ring binder with instructions, notes, receipts, copies of articles etc.

Binder with notes and more. Picture shows notes and schematics for a VFO. It is not known if LA4FK ever built that one, but apparently he went far in preparations. Background shows a radio glossary and a textbook on radio building.

A lot in it does not seem useful. A few things works as useful pointers in the restoration work. There is very little in there that is material for the blog. Except perhaps a couple of things. Something in there was a puzzle for me. I think I shall make a puzzle for you to solve. In a future blog post.

Two books were also included. There have been some development in technology since these books were published in the 60’s. Some fun tidbits found anyway. I found one curious little thing to write about there.

No trees were harmed in the writing of this blog post.

Sidetracked. Story of insufficient capacity

The previous interesting cheat got me intrigued. Could I improve CW and SSB reception with a simple change? As simple as adding a capacitor in the right place?

I was setting up for aligning the IF and front end stages, but I got sidetracked by this idea.

I thought the best place to insert the BFO signal would be the screen grid of the last IF stage. The best way to get it would be from the anode of the BFO triode. The largest capacitor I had that could take the voltage with good margin was 10 nF. The change was barely noticeable and hard to evaluate as the BFO frequency also changed. I decided to connect 8 of them in parallel.

8 x 10 nF in parallel. To the right is an attempt at making a capacitor from enameled wire. Why? That is a story for another post.

Even that did not make a convincing improvement. It is possible that a much bigger capacitor is needed. Or a different way of connecting it. Or perhaps remove the 5 nF capacitor from anode to earth as that one swallows much of the signal.

Incomplete soldering

While investigating where to connect the capacitor I found an incomplete soldering joint.

For the picture I lifted one leg of the resistor away to show that it was not properly soldered.

I don’t know how long it has been like that. Perhaps since it was built. Tension kept it pressed against the right tube socket leg. Therefore functionality was not affected. I choose to solder it in place.

So that is part of what has kept me busy instead of aligning the receiver. Which I should get around to do.

Interesting cheat

I’ve suspected for a long time that the BFO signal is weak. Today I made an experiment that strengthened that hypothesis but also encouraged me to do something about it.

While listening to a strong station on SSB I have to turn the RF gain way down to prevent terrible distortion. That makes the sound in the speaker rather low, even with AF gain at max. I can get much louder sound on AM.

While listening to a strong local station I turned on an RF signal generator and tuned it for zero beat. I could now turn up the RF gain and listen to loud and clear sound. That’s encouraging and promises that if I could get a stronger BFO signal injected I would significantly improve SSB reception.

BFO schematic. Note that there is no output path! The on/off switch is connected to the 125 V DC supply.

When I first looked at the schematic a while ago I couldn’t find what path the signal takes from the BFO to the IF chain. I suspected a misprint of the schematic. However, the newer HR-10B schematics are exactly the same. You’d expect such a mistake to be corrected. In fact, Heathkit used to deliver errata sheets and updates to their manuals.

As it happens the BFO triode is one part of a tube (6EA8) that contains the last IF amp pentode. That is where the BFO signal is supposed to be mixed in. But there is no input for the BFO signal:

2nd (last) IF amp. Screen grid gets clean DC stepped down from the 125V supply line. T3 connects to diode detector.

Best guess is that there is internal leakage in the tube that gets the signals mixed.

So what to do?

One simple idea is to connect a capacitor from the anode of the triode (pin 1) to the screen grid of the pentode (pin 3). The two .005 capacitors may swallow a lot of that signal, but there should still be more left than just internal leakage. I think.

Another idea would be to twist together wires to the anode and the screen grid so that the wires makes a simple transformer. That transformer would also unavoidably become a capacitor. I would have to think through the phasing in that case.

This is where I would like to hear your opinion.

In depth

Manual, including schematic for HR-10(B) Searchable

SB-600

As you may have noticed, my SB-600 could use some cleaning.

I did not build this one myself. It used to serve on a club station with a now long decommissioned SB-100. After years in dark and dusty storage it got a new life as a partner to my SB-104A. It is not as stylish as the SB-604 would have been, but appearance wise it was an improvement of the non-boxed speaker element I had been using so far.

To clean it I removed the internal HP-23E power supply that I do not need. It is meant to be used with many tube type transceivers. It’s condition is unknown but I didn’t want to harm it with liquids. Turned out it was secured with only three of the four screws and no washers. Not metric screws of course 🙄

I wanted to remove the speaker element and the wooden plate temporarily for the same reason. However the way it is supported is unclear to me and I couldn’t find a manual online. More precisely, I could find SB-600 manual scans online, but they were all missing the critical pages. Do you have a more complete manual? Another reason I wanted to work on it was that I would want to move the front grid back so that the screws do not protrude outside the front edge. Looking at pictures on the internet, this one seems to be modified.

I noticed it is also lacking rear rubber feet, so I will have to fix that. It could also use a paint job to cover the scratches. Unfortunately Heathkit was not very consistent in the colors used.

Anyway, with some effort (and care) it is much cleaner now.

A cleaner SB-600. In background part of HP-23 supply getting capacitors charged by IP-2728 power supply (center right). Charging current measured by IM-5284 (front right).

And it still makes sound.

Microphone candidates

“Anyone can talk, even a parrot”. That is what hardcore CW operators have said. I enjoy the challenge and exhilaration of working CW, but I know how to talk too.

To operate on AM the DX-60 needs a microphone. The previous user (LA4FK) apparently used a hand held mike, but that one is not available. The microphone input socket is of an unknown type without provision for Push-to-talk (PTT). At the time of writing this I have not yet put much thought into Transmit/Receive switching so I don’t even know if I am going to use conventional PTT.

Microphone socket

None of my candidate microphones has a plug that fits. Potential workarounds:

  • Make a short adapter cable. Seems like a PL-259 can be made to fit.
  • Replace the microphone socket on the transmitter
  • Add an extra microphone socket on the rear of the transmitter.

Primary candidate

This beauty was a gift from my friend Per Kristian. I can assure you he is much more of an expert on microphones than I am. I just like the way it looks. I think the nostalgic look is a perfect match for a nostalgic transmitter.

Old looking Shure 55SH microphone
Shure 55SH

There are a few challenges to get this one on the air.

  • Impedance is low. The tube type transmitter prefers a higher impedance. I may have to make a transformer. Potentially built into the transmitter, perhaps with its own input socket
  • I need a cable with an XLR plug for the microphone end. That’s not a plug type I have in my junk box
  • There is no provision for PTT, but there is an on/off switch that could be rewired. That one locks in either position.
  • It needs a stand, either for desktop placement or supported off the rig or wall (saving desk space). The threads for the stand are inconveniently non-metric and placed in front of the XLR socket.

None of those should be show-stoppers. Just hassles. Even if this beauty never gets on the air it serves well as decoration.

Runner up

I have a Yaesu YD-844 microphone, currently fitted with the two-pin+screen plug (Amphenol CDM 80-MC2M) that fits most newer Heathkits. It needs no work, other than matching plug and socket. Impedance is right (actually switchable), PTT is there, and the stand is part of the assembly.

Outsiders

If for any reasons the favorites fail I have a Turner 754 and a Drake 7075 which happens to look suspiciously similar. Those both have the 4-pin plug used by several rigs. According to Turner catalog from 1970 the Turner has a frequency response of 300-3000 Hz. Taylor made for common communication standards.

Before the parrot can get on the air however, we need to work on the DX-60.