All Things Wireless & Letterpress

All Things Wireless & Letterpress

Saturday, June 13, 2015

"Operation QSL" : Letterpress QSL Cards for the Vintage Station.

I thought "Surely . . . surely, in this world of re-invested technology from the past, in a world where the ancient Letterpress has come to almost dominate the Wedding Announcement / High End business card / Premium Stationery market . . . surely someone out there is regularly addressing the Amateur Radio community with QSL cards that correspond with the vintage/ quality/ nostalgia factor that obviously drives the "Boat Anchor" crowd.

I am amazed at the multiple thousands of vintage radio dollars which are exchanged daily either over or via eBay or Craigs List or at Hamfests or via the various News groups, FaceBook pages and Pinterest!  I myself have quite an investment in Hammarlund, Drake and vintage Yaesu equipment, not to mention a  Homebrew code station built following lines of the late 1930s QST projects.  Yes, I'm in that mix, too.  Which is why I want QSL cards and stationery to reflect that!

I already waxed lyrical over the art and tradition of the QSL card in a prior post.  In this installment, I would like to share what WD4NKA, the Staff and the Management at Paper Wren Press ( which houses WD4NKA!) is planning to do to address the "Vintage Wireless Stationery Void", with any luck and a bit of support - which is to . . .

. . . Offer a humble selection of vintage Letterpress QSL cards!  At a price that may not cover all our costs, but one that is fairly competitive with current prices. Cards that reflect not only your vintage, lovingly restored or meticulously hand built rigs, but also the Era when your fifty to sixty-pound gems were the State of the Art!  

The Project:

In general, I culled about twenty five years of QSL designs from 1919, and on into the War Years.  I identified what I believe to be primary design elements, most of which are pretty obvious.  For one, almost all are either balanced in their appearance by symmetry or by what we called in photography "Rule of Thirds", which balances asymmetry.   Normally the call letters are a pronounced complimentary colour, and more often than not, red.  Basic QSL information corresponds to what most log books of the era called for.  Unless you were an ARRL member and had the "diamond" printed, there were very few extraneous logos tacked on, although there were a few.  Just not enough to rise to the level of standard design.

They looked official.  They did not look like travel post cards like many do today.  They were not glitzy.  They were, in fact, almost spartan, save for the Otto Eppers and Phil Gildersleeve cards that were actually limited edition art cards!  (We are also contemplating an Otto Eppers design, btw.  Maybe Robert Crumb, too!)  

For those who do follow this blog regularly and are unfamiliar with Letterpress, let me share with you a video I made some time ago, showing myself printing a Wedding Announcement:

This second video shows a little about the actual Letterpress itself.  This is actually a short documentary of the Kluge that Nick Sambrata and myself got working at Mama's Sauce printery, Winter Park FL.


Needless to say, these old Iron Horses are amazing animals, and carried the bulk of the printing needs of the Western World for over one hundred years.

The Designs:

Here are the three designs we will be working with Owosso Graphics, Owosso MI, to produce in hardwood mounted copper:

This one is for the NRR folks.  But the design can be modified as a "custom order".  These cards have a pre-printed black shell containing the telegraph key design, title, and qsl info, and the morse code "nrr" that serves as the top and bottom borders.  This card will be standard post card size, 3.5 x 5.5 inches. 

This card is much more traditional, and is largely lifted from about three different pre-depression era QSL cards.  This has the traditional spaces to record your transmitter, receiver, power output and antenna (aerial) data.  This one is for the Dyed in the Wool 1929 "Post Cairo" op who designs his or her rig from the ground up.  Not for sissies!  This card will be standard post card size, 3.5 x 5.5 inches.

This card is more mid- 1930-ish, although examples exist as late as the 1960s.  There has always been a following for this card design, it seems.  My actual examples came from cards that immediately preceded the WW2 ban.  But I have seen these used even when I was a Novice in 1977!  I guess that would mean it's a "timeless" design, for a Timeless Station.  What do you think?

In Summary:

There are a number of QSL printers out there that do great work at a fraction of the prices I am asking.  That is the whole point of modern printing!  To drive down prices, and they do, for sure!  But not without sacrificing that tactile dynamic that can only be made by direct impression of a die or plate against paper.  And engraved plate, or handset metal type produces an effect that two generations have grown up never knowing.  
Probably the best way to describe the difference between modern printing and the ancient contact printing modes might to consider a high quality photocopy of a Dollar Bill.  Not that I am suggesting do this, it's illegal.  But if you held that photocopy in one hand and a real dollar bill in the other, you would know instantly which was real.

That's what we're after here.  At about sixty-cents per card for 250 2-colour cards.  $150.00, with a 2-3 week turnaround time. 
Well, this is what has been keeping our creative juices flowing at the Print Shop at WD4NKA / Paper Wren Press.  Hoping to hear back from some of you.  Contact me via Face Book or at my email box at or

More will be posted as we make progress with this project. Once again, Wireless meets Letterpress.

vy 73, de wd4nka.

Friday, June 12, 2015

Resurrecting the Drake TR-3 and AC-3 Transceiver and Power Supply.(Part 2)

Having finished the AC-3 power supply, and upon being satisfied that it was operating to specifications, it was only natural to try it out.  I already had an idea of what would happen, but I plugged the supply up anyway, attached it to the TR-3 via the big Cinch Jones plug, ramped her up, and voil'a!  

Pop, snap, fizz.  And a mild aroma of ozone filled the air about me.

There is no doubt that the TR-3 was dead.  This must be distinctly understood, or nothing wonderful can come of the story I am going to relate. The relay would not key.  The lights would not illuminate.  Smoke whisked from the audio section.  Various sparks flew from the Vox and Anti-vox potentiometers.  Nary a sound made it's presence known through the headsets.  Not so much as a hum.  The TR-3 was as dead as a door nail.

Mind!  I don't mean to say that I know, of my own knowledge, what there is particularly dead about a door-nail.  I might have been inclined, myself, to regard a stripped cabinet screw as the deadest piece of ironmongery in the trade.  But the wisdom of our ancestors is in the simile; and my unhallowed hands shall not disturb it, or the Country's done for.  You will therefore permit me to repeat, emphatically, that the TR-3 was as dead as a door-nail.

Not being one to follow a straight line, even with coloured pencils and Magic Markers, I set about to do what most geeks like me do when we don't know what else to do: hunt down all the paper, moulded and electrolytic capacitors / condensers, identify their values and replace them.  And to be sure, this is Standard Operating Procedures.  It's not that these capacitor types might be bad, the are bad!  The results were to reveal that such was no more so than it was with the TR-3.

The truth is, there really are not that many to replace!  There are five paper caps and two electrolytics, almost all types the same value.  The papers were replaced by polymer and one orange dip, and the electrolytics were replaced by new electrolytics.

All of these caps are found on small upright phenol boards mounted either near the side of the radio where the control pots and jacks are (Vox, Anti-vox, Meter Zero, Mic, Headphone and Key jacks) - or center rear.  The paper caps are pretty obvious,  they are identical white tubular types, all 0.47 uF save for one, which is .01 uF.  The above photo shows their appearance and the boards they are mounted least, in the early model TR-3s.

The ideal removal tool for these components is a small flush cutter.  The following describes the technique I used to replace them :  I cut them out by cutting their wires as close to the component itself as possible.  This leaves the two original wires soldered into place.  These wires will serve to mount the new component by providing something to solder to without disturbing the original solder.  I found over the years that things go a lot better if you avoid disturbing original terminal solder connects.  That is, if they are good solder connects!

Now, most of these boards can be removed, but go slow. If you can, keep a digital camera that can do micro-zoom handy.  Almost all of the inexpensive digital cameras like the Canon Power Shot will do nicely.  I recommend this because not only are these boards held in place by one or two bracket screws which must be removed, but they are also hard wired to adjoining circuitry at multiple points.  Most of these wires must be cut.  As you cut them, leave them in position so you can re-align them with the wires on the board when you re-install them. 
As you can see from this example, the vertical board is held in place by two angle brackets screwed to the chassis, and is also soldered into place by 20-gauge bus wire, plus insulated wire leads.  These insulated wires are not necessarily colour coded, so be observant!  Take good notes, perhaps do a drawing, or best, take a photo.  Digital camera technology is your friend here!  Digital Cameras are part of my restoration tool-box.

 Be sure to photograph, or at least make notes on the wire attachment to each side of these boards!  There will be bus wire cuts, insulated wire cuts and component (resistor/ ceramic disc capacitor) cuts that will need to be re-attached.  I might add that there were two boards whereas I found  I could leave the wires intact, requiring that I simply remove the bracket screws and lift the board a bit.  All I need to do is snip the cap out and have sufficient room to re-solder a replacement component into place.

I wanted to show this photo.  The smoke I mentioned earlier came from this 3.3k half watt resistor which bridges these two component boards.  There was an associated ceramic disc capacitor on the other side of the board, a small bypass cap, that had shorted, creating a path to ground that permitted too much current for this resistor.  In these tube type Drakes, you want to be mindful that they create a lot of heat in confined areas where ventilation is limited.  Many owners will mount muffin fans on the cabinets to force additional air exchange, which isn't a bad idea!

Convention almost implies that we be mindful only of replacing Paper, Moulded and Electrolytic caps. The tendency (and often read written opinion) is to leave the ceramics and silver micas alone.  I say look at everything!  There IS no hallowed component.  They are ALL prone to failure with age.  It doesn't mean you go out replacing all the caps, but don't assume all ceramics and micas are innocent or somehow unaffected by age, time, and heat.  Pay attention to any associated resistors.  A brown or discoloured resistor is your clue to higher currents, which may be due to either a value change in that resistor due to age, or a high current short nearby - normally a cap.  And, beware of strictly going by continuity checks alone.  Most of the shorted caps in my experience showed open continuity on a VOM.  However under high voltages and operating currents, these opens will close!  As such, do not be in a hurry, and keep your variac handy.

 BTW, I replaced that 3.3k resistor three times until I figured out what was going on.

The technique employed here involves the use of Squigs, or "Pig Tails" as we used to call them.  I already made a blog entry about the use of these items.  At one time you could buy them, but these days, I make them from 22 gauge tinned bus.  For details, visit my corresponding blog entry "Resurrecting a Drake TR-3: Pigtails!"

Here is a shot below deck showing the replacement caps having been "squigged" into place utilizing the original wires that were left soldered to their terminal points. You may have to click on the photo to get a good look at these solder connections.  Note the rear boards are not yet bolted down. I was still in the process of squigging their connections back together.

Touching on other things that were going on besides the circuit work, I pulled every tube/ valve, every crystal, one by one and gave them a thorough bath of DeOxit.  You can see the bottle in the photo with the hypodermic cap.  Relays, sockets, any place where metal contacts metal is treated to a liberal dose of DeOxit.

The white metal covers that cover the mA meters and the planetary dial were removed, sanded, repainted with white semi gloss, and replaced.  Possibly a better choice may have been flat white.  But no matter, the blue dial filter (acetate) diffuses any light bouncing around behind the dial.  Make sure there is electrical contact with the slide-on light bulb sockets. Those areas might be best sanded to ensure grounded contact with the bulb holders both for the dial and the mA meters.

Accompanying these activities was the treatment of the cabinet.  The cabinet was awful.  The TR-series Drakes use a clam-shell cabinet which is held in place by twelve machine screws - six for the top, and six for the bottom.

I opted to do the wrinkle finish on the top, and use flat black for the bottom, just to provide an interesting visual break.  I did the same combination when restoring an enlarger, and it was quite attractive!  To read more about how handle wrinkle finish treatments, refer to my prior entry concerning the rebuilding of the AC-3 Power Supply. 

It takes about twenty four hours for the wrinkles to "set" to the point where the cabinet sections can be safely handled. As such, a lot of patience is required. This patience is amply rewarded once the project is completed!  I think Drake should have used Wrinkle Finish too. Again, just my opinion.  I have a lot of those.

Here is another view of the cabinet, under natural light.  All sides wrinkled very nicely owing to the even distribution of heat due in large part to the ventilation holes.

Wrinkle Finish is like a letterpress printed piece.  You have to behold it up close and touch it to really appreciate it.  Wrinkle Finish provides a very robust, tactile surface that engages not only the eyes, but the sense of touch.  This is why I especially like to use it, when appropriate. 

Up to this point we are finished with the circuitry, we finished  the de-oxidation, we finished painting the inside and outside metal work, we also replaced the feet.  What next? her up!

I loaded her into a 60 watt bulb to see if there was anything like output.  And yes!  As it turns out, she is willing to yield about 100 watts cw, and later when I tested her on SSB, about 120-150 watts on voice peaks.  Her finals are old, but are themselves, replacements.  I suspect they were replaced sometime in the 1980s.  I have the originals, where are blown, with cracked glass. 

Here she is, all buttoned up and ready to be parked next to the 2B.  My intention from the start was to use the TR-3 as (gasp!) a cw transmitter!  This was because I had no idea that she would ever speak with a voice again!  She performs marvelously in this capacity!  Flipping her over to cw position, she immediately emits a very weak signal perfect for spotting.  Since this is in transmit mode and the changeover relay is already in transmit position, I can start sending immediately.  When I switch back to the 2B and switch out of the cw position, the spotting signal ceases.  Having grown up with home-brew transmitters and separate receivers and never owning an RF activated ATU in my life, I am used to flipping switches between transmit and receive.  Part of what geeks do!  I did the same with that Gonset shown in the video and one of the photos  which . . . is no longer there as of this writing.  It is now comfy in it's new home at AF4Ks shack.  The Drake is now filling her shoes, and quite nicely so!

A few things still need to be done with the TR-3.  I need to re-wire the filament voltage to the mA meter and dial housing pilot lamps.  The sideband selection switch lights are, however, working fine.  The mic gain is a bit crackly, that will be ultimately replaced. I would like eventually to find a new face plate, too. The TR-3 is, however, dead on frequency.  The power supply is running smoothly.  I do wish to, eventually, replace both mA meters.  It is not uncommon for the d'Arsonval movements to become magnetically gaussed and sticking.  I have to tap the S-meter to break it loose from it's internal magnetic charge.  The rig's mA meter simply isn't working.  All tuning is done using the metering on my Dentron MT-3000A.  But again, I am used to meter-less rigs.  

So, there she is.  But the story isn't over.  What about phone?  

I had thought to let that go.  But curiosity got the best of me.  I did not have the proper plug for the mic jack,  and honestly, I was tired of having to mail order everything.  Our local MilSurp store which used to have bins of these connects had none.  But since I was there, I purchased a 1/4" jack, and retrofitted the original.  Of course, I am keeping the original to replace when the time comes.  But at least I was able to wire up the Shure 444 mic and give her a spin.

And spin it did!  Wow!  ....and talk about great reports!  As of this writing I have worked quite a number of stations under sorry band conditions, all on forty meters, and all say the same thing : the voice "punches" through!  AF4K reports that this TR-3 is a "DX-machine!"   Signal reports aren't bad, either.  As I said, I am seeing 125 - 150 watts on voice peaks.  That's low for these rigs, but the tubes are old, too.  

Hey!  So long as nothing smokes!

The receiver is awesome.  I suspect the passband to be around 2kc.  Upper and lower sideband filters are performing fine.  I might address the carrier null, but it is fine as it is.  Just a wee bit of carrier heard through the HQ-170.

So, there you have it.  The "Resurrection" of a TR-3.  Satisfaction level from 1 - 10: 12.  She is QRO and Go, both SSB and CW.   What about AM?  Well . . . that's for another installment.  My coffee is getting cold.

vy 73 es God bless u es urs!

de wd4nka

(BTW: I published this soon as I was done with only cursory proof reading.  If you find blatant, glaring grammatical or factual errors, check back again.  I proof read these after publication and edit them over time  because, well, that's me, I guess.)


Thursday, June 11, 2015

Resurrecting the Drake TR-3 and AC-3 Transceiver and Power Supply.

When I was gifted by an elderly ham friend with a few parcels from his ham shack, the focus was on a Kenwood TS-420, a Dentron MS3000A and two Ham-M rotors with control boxes.  Almost incidentally tossed in was a rather neglected, somewhat dirty  Drake TR-3, and not far in the storage area where it was kept (incidentally  NOT where the other items were kept!) I found it's equally dirty power supply, the AC-3.

The storage area was the corner of a garage attached to a mobile home.  You from Florida might have an accurate picture what I mean by that.  No climate control, constant 90-100 degree heat 10 months out of the year, no climate control, and lots of crawly things.  Well . . . it wasn't quite that bad, no crawly things, but the lack of climate control was evident.  This poor Drake probably had not seen air conditioning since the 1980s!

Early Drake radios used a copper clad chassis which looks awesome when new.  Get some age and the copper stains become very evident.  Oxidation is a no-brainer.  Fortunately the chassis are heavy enough to maintain their integrity over the years, in this case, fifty of those years.  I would say this radio was decommissioned and set out to pasture about twenty  to twenty five years ago.  It had been re-tubed sometime in the 1980s, from what I might tell.  The original finals and drivers - in fact ALL the original tubes - were boxed in the boxes of the tubes which replaced them.  You can tell an original Drake valve, or tube: Sylvania.

At some point I decided to attempt to resurrect the AC-3 and the TR-3.  I figured that while I more than likely will not "restore" it, I may be able to resurrect it, for use, perhaps, as a cw transmitter to pair off with my Drake 2B.  It was worth a shot, anyway.

The first object of my attention would be the power supply.  It was an obvious disaster just looking at the electrolytics.  So the first thing I did was search and download any image I could find that might show what the below-deck wiring should at least look like.  Sure enough, Google Images turned up the perfect photo.

I had already determined that this was going to be a rebuild from parts that I either already have, or could easily procure.  There are, of course, rebuild kits for these power supplies, but the AC-3 kit is the most expensive, and requires a bit of body-work to mount.  The manufacture of these kits obviously favour the AC-4 supply, which is a direct bolt-on device with a price tag twenty dollars less!!   If I was going to have to take my tools out, then I'm gonna do it all the way, "Old School" style.  Besides, I had, actually, small hope that this project would actually be successful, but I had to try.

The transformer tested good as far as voltage went.  I expected more from the secondaries at first, but then I recalled that this was an era of Sweep Tubes and voltage doubler supplies.  So, I located a schematic to verify.  (Read: I pulled my TR-4 / AC-4 manual and schematic that I already had. There are essentially no electrical differences between the AC-3 and the AC-4)

Looking over the supply schematic, I noted that the electrolytic filter caps did not show polarity.  This wasn't a problem . . . . until I looked over the negative bias supply.  Then it dawned upon me that I only read about bias supplies, I never actually handled one with my own hands.  I might guess the filter caps have grounded positive terms, but was I missing something?   After a lot of verification I found that indeed they do mount in reverse.   And, if you look closely at the schematic symbol, the caps DO show polarity.  The symbol has two parallel lines, of course, however one is straight, the other is slightly curved.  This is consistent with the traditional printed condenser/ capacitor schematics, historically.  The curved line represents the ground side, in the case of polarized caps.  And indeed, the bias supply caps have the curved line on the top, consistent with the negative bus.  All this may seem logical and inanely elementary.  But I get like that sometimes.  Ever look at a simple word long enough and it starts looking weird?   Like it isn't spellt correctly?  Yeah, as a letterpress printer, that happens to me, too.   

 . . .Sigh . . .

Ok, so I literally gut the whole chassis, remove the transformer, caps, strip it down entirely.  Now, what to do about the caps?  There are five gaping holes in the chassis.  I would like to keep a vestige of the original "look", I already knew this project was not going to be a restoration.  It was going to be a resurrection.

Early in the game, back when I was a Novice, I was in love with gear from the 1920s and 1930s.  And back in the 1970s, there was a lot of it around.  A lot could be had cheaply, too.  Back in the 70s, the big focus was on restoring Victrolas and Gramaphones, not vintage Radios.  I was a lone wolf, or so it seemed. I learnt then that everything looks better in Black Wrinkle Finish!  Yes.  And I still feel that way.  It handles heat better than any other paint (in my humble opinion) and it covers a multiple of ills.  And it looks right crack!  (very nice.  Professional.  Sanitary.  Choose your American adjective.).  

When I first painted this chassis, which had to be done in stages  (one side at a time, and the top separately), the first thoughts of ultimate success - in some manner - began to germinate in my small mind.  Wrinkle finish paint does this, Brethren!!  It inspires success!

You can get Wrinkle Finish at many Automotive supply stores.  I got mine at Auto Zone.  It's not the cheapest can of paint on the block.  Here is my technique:

First, understand that this paint was not intended to be sprayed on primer.  It's a bare metal application.  We used this on exhaust headers back when I did motorcycles, it goes directly on valve covers, and parts that get very, very hot.  You spray it directly on the metal cold, then heat it with a heat gun.  This means that you have to do small areas at a time.  For radio cabinets and chassis, I spray one side well.  Don't spray too thick, but don't spray too thinly.  A generous coating is fine.  I do not mask off the side I am not spraying : overspray isn't a problem.

I take the piece inside the shop and begin to use forced hot air on the spray painted side.  I move the heat gun back and forth, for usually about fifteen minutes, and after a while one corner will begin to wrinkle,  You can see it in the reflected light.  Then that wrinkle will spread across the painted surface to the other side.  It's fun, actually, to watch this phenomena.   You can control areas that have less wrinkle that other areas by focusing the heat on that side.  It takes a bit to develop the touch, but after a while a feel is developed.  After you achieve the level of wrinkle that suites, the process is repeated for another side.  Again, over-spray isn't really a problem because the wrinkle dynamic hides any overlap.  After a while I found that Wrinkle finish painting can be a very intuitive and individual process.  Almost an expression, an art.  Wow.  Making an artistic statement with your transmitter.  Imagine.

In the meantime, I tested the voltages present on the transformer windings.  When I took the AC-3 apart, I tagged all the wires, but good to double check, anyway.  The HV and the LV windings checked at 280 and 125vac respectively.  Both of these windings see voltage doublers with an output of 650 and 250vdc.  The bias supply checked in at 65vac and the filament winding at 10vac.

The caps were all bad and the diodes were questionable.  I replaced them all with 1000v piv silicone diodes, my standard go-to power supply diodes.  As for the caps, I purchased several 100uF/ 450v filter caps.  These were used either singly or in series (in the case of the HV line) with their original equalizing resistors across them.  Since I had these large holes in the chassis which originally facilitated the original wafer sockets which held the above deck filter caps, the decision was made to utilize these holes, as already mentioned. To accomplish this, I used a combination of cardboard tubing painted black, black plastic 35mm film canisters, wrinkle-finish painted pill vials and a hot glue gun..

I began to rebuild the cap that would filter the 650vdc "Hi Voltage" line.  Using a 1.25" diameter cardboard tube, I cut a base from phenal to become the mounting platform for the cap's internal structure.  The photo above gives you a pretty good idea how it came together.

A close-up of the series junction.  The junction between the caps is tapped as a voltage doubler. Black electricians tape cover the two 1-watt resistors.
The wires are all fed through the bottom disk which I just sawed and cut to fit the bottom of the cardboard tube.

Another veiw of the base showing how the wiring is arranged at the bottom.  The red wire is the HV bus, the yellow is for the center tap.  The uninsulated bus is for grounding.

After the tube is slid over the caps, it is sealed to the original wafer that held the original filter cap, with Hot Glue. When the glue cools, the wires are held in position, and when this happens, the series soldered caps in the cardboard tube are braced and held firmly in place.

The larger of the three filter caps is the HV filter cap we just described above.  The smaller two are filters for the LV voltage doubler, only these are housing only one cap.  The covering for these smaller caps are 1.25" diameter pill vials.

The bias supply needed filters too, so we used the remaining two holes on the chassis to mount yet another pair of filter caps.  These were covered by black plastic 35mm film canisters.  The mounting bases are the caps for these canisters. 

Note that I did not paint the rear apron of the chassis?  That was so to preserve the original printing on the rear deck.

I opted for point - to - point hard-wiring, not only to preserve the original "look", but also because this resurrection was going to be accomplished using parts on hand or parts easy to procure.  As it was, I used the original equalizing resistors, I merely supplied all the caps and new terminal strips and diodes.

The re-built caps now mounted, the transformer in place, the wiring under-deck being complete, I now add the rest of the hardware, the bias adjusting pot, and the cabling.  Once finished, another round of voltage checking and smoke testing was undertaken to make sure that the Drake TR-3 had every good chance of having proper operating voltages applied.

. . . and here she is, the completed AC-3 in her new make-over, dressed in wrinkle finish black, sporting new caps, new diodes, fuse and fuse holder, original bias pot, transformer and cables. Total cost to me was under thirty dollars.  Thirty "year 2015" dollars.  Not too bad, I might add.  A decent "junque box", a few runs to the military surplus store (Skycraft, Orlando FL) and a sprinkling of creativity, and she's ready to roll with all voltages checking to manual specification.

Stay tuned for the next installment: Resurrecting the TR-3!

de wd4nka