Modernizing the DY-88 Power Supply

Go with a Modern Vibe

The DY-88 Power Supply is a marvelous design... for the 1940's.  It could take 3 different DC input voltages (6v, 12v or 24v) and output the 2 different filament voltages (1.5v, 6.3v), plus the low and high B+ voltages (105v, 585v).  It was designed to be as efficient as possible when the radio wasn't in SEND mode (receiver only) by employing a mechanical vibrator and transformer to power the low B+ 105v at about 1 amp current (when running off a 12v input).  




Turning low voltage to high voltage


Prior to the DY-88, I'd restored a 1960's Heathkit HP-13, which was a mobile DC to DC power supply.  It employed a multi-tap transformer and 2 transistors to create the needed low and high B+ voltages for Heathkit radios.  in the case of the HP-13, the two transistors worked in tandem to collapse the magnetic field in the transformer to create high voltage DC, which was rectified and filtered for the radio.




Prior to the DY-88, I'd never heard of, or encountered a mechanical vibrator used in power supplies; so it's been an interesting learning experience for me.   Vibrators were very common up until transistor revolution, for use in mobile applications where 12v needed to be transformed into much higher B+ voltages needed by hollow state equipment.

In oversimplified terms: the vibrator makes DC look like a square wave version of AC which can then be used by a transformer to raise the voltage required by the circuit, then rectified back to DC.  Low voltage DC enters the Vibrator, which contains a relay that makes and breaks contact rapidly, chopping up the DC.  It occurs so fast that the unit "vibrates" or buzzes.

The contacts eventually, become worn and pit making the vibrator less reliable.  Additionally, military vibrators (maybe others) used a rubberized insulator that broke down over time inside the vibrator, releasing a corrosive gas that would further affect the components and essentially, "stick" the contacts on the relay together, making it difficult to get started or completely non-operational.  Depending on how it the contacts "stick" the vibrator can act as a short and unpleasantness ensues.

The vibrators were intended to be disposable devices but they aren't being made anymore.  What to do?

Bad Vibrations

In the case of my DY-88, when I received it the vibrator power supply refused to start but eventually, after a bit of percussive maintenance (banging) and reapplying power it started.  Over the past couple of weeks it usually will start, but there are still occasions where power has to be switched off and on repeatedly to get it to start.  It is obviously entering failure mode.

There is a procedure using an AC light bulb in series with the contacts to attempt to freshen a stubborn one up, but the procedure is hard on the contacts and generally only restores it for a while.  It needs to be replaced, and vibrator power supplies have not been manufactured for many decades.  The NOS vibrators are generally not good due to the breakdown of the materials inside over the decades rendering them bad even if they've never been used.

The solution is to replace the mechanical operation of the vibrator power supply with a solid state equivalent.  

Solid State 

Entering the age of the Jetsons, we have these nifty things called transistors that are excellent and making and breaking contact of DC voltages rapidly and reliably.  I found a number of articles describing a solid state circuit I could build but most did not offer much in the way of circuit protection if a component upstream failed, or if the circuit itself failed.  There are a lot of difficult or un-obtainable components in that DY-88 that I'd rather not have to replace due to the failure of the Vibrator Power Supply.

I read an article by a GRC-9 enthusiast from 2004 (NC6AV) on using a commercial circuit that was designed for this very purpose to replace mechanical vibrators in vintage automobile radios https://www.radionerds.com/images/c/cd/Wire_VBN-1.pdf.  In the article describes the procedure for using a solid state replacement.

I purchased my solid state module from http://dodgem37.com/vibrator-conversion-module/ and ordered some high voltage diodes from antique radio supply to use in the conversion.  I cut apart a non-functional vibrator module to use the phenolic 7-pin base for the VBN-1 module.


Note


The instructions for soldering the VBN-1 module are great with one exception.  From the instructions: 

. . . The pins are numbered 1 to 7, pin 1 is the large diameter pin on the right, and pins are counted clockwise from that pin. . . . 

If you are looking at the base of the vibrator there are 2 large diameter pins.  The instructions say count them from the pin on the right.  That pin will be different if the large diameter pins are at the bottom rather than the top, when looking at the base.  Turn the base such that the two large diameter pins are facing you and located at the top (12 o'clock position).  Now the instructions will be correct.   This may seem minor, but when I first read the instructions I was looking at the two large diameter pins situated in the 6 o'clock position and all the pins were off-by one.

Results

The replacement circuit works a treat, except that I can't hear the faint humming from the DY-88 now when the radio is in Receive mode.  Of course that is immediately obliterated as I turn the radio to SEND and the Dynamotor leaps to noisy life.



Next step would be to JB-weld the can back to the base to restore the look, but I'm debating whether to do that.  I kinda like the look of the modern mixed with vintage vibe (vibe, get it?).


That's all for now,  keeping my heavyweight QRP GRC-9 rig on the air, one repair at a time.

Lower your power and raise your expectations, 

AA4OO

AN/GRC-9 aka “Angry Nine”

 AN/GRC-9 - Long lived military comms

My lovely (and radioactive) RT/77-GRC/9


I don't recall where I first read about the Angry Nine, but it captured my imagination.  I read everything I could find about them and decided it would be great fun to operate such an antique on the ham bands.   There is no logical reason to desire such a QRP radio.  The low power output on CW is indeed, 5 watts and high power is a pileup busting 15 watts.   The AM transmission are 1 watt and 7 watts respectively.  That's almost QRPp for AM mode.

I'd had some experience restoring old tube equipment; my Heathkit HW-101, Knightkits VFO and Hallicrafters keyer, and I figured I'd take the next plunge and learn to use a receiver-transmitter combination and see how mobile high-voltage power worked from Vibrators and Dynamotors.   

These radios seemed to have been more plentiful in the surplus market 10 - 20 years ago.  Now you'll occasionally see one come up on eBay or other sites, but often times they are in very rough shape or the they are foreign language versions.  I bid on a few auctions over the past couple of years and the bidding always exceeded my threshold for what I thought it was worth.  The one above was part of an auction from an individual who had actually trained on these units prior to deploying to Vietnam.  Later in life he became interested in finding one and spent time in military surplus warehouses going through pallets of equipment to find one in good shape.  This particular unit is made up of a Lewt manufactured transmitter and a Telefunken receiver.  The original owner preferred the receiver characteristics of the Telefunken over the Lewt manufactured model, so he paired the two.

Many of these old units are radioactive, due to the radium paint used on the front panels to make the lettering glow in the dark.  This particular unit is off the lower scale on the Geiger counter and must be handled with care.  Basically, I have to be careful to not touch my face with my hands after operating the unit and wash my hands thoroughly.  Radium emits Alpha particles, which are not especially strong but the resultant radioactive dust from the front panel shouldn't be breathed or ingested.

Hot receiver, in more ways than one


The AN/GRC-9 is a set of components primarily comprised of the RT-77/GRC-9 receiver-transmitter, capable of operating between 2-12 MHz in CW, MCW and AM modes.  MCW is a modulated form of CW that can be received by radios that do not have a BFO (i.e. a normal AM receiver).  

It is a mid to late 1940's design and was first documented field use in the Korean War, and was in active use through the Vietnam War and continued to be maintained in US military warehouses until 1974.  It was in use by other nations long after, most notably the Dutch military.

Out of the case, tracing a low B+ power problem



Power on the move

Designed to be used in the field, both vehicle mounted and carried by mobile infantry; there were a number of ways to supply power to the unit.  There were a few different Vibrator/Dynamotor units, that could operate from common DC voltages of the time (6v, 12v, 24v) as well as a hand cranked, field portable generator.

Keep in mind that the state of the art at the time of its design used vacuum tube technology and in the case of the RT/77-GRC/9 it required the following voltages:
  • Transmitter Plates -- 475 - 580 v @  100ma
  • Transmitter Filaments -- 6.5 - 6.6 v@ 2 amps
  • Receiver Plates -- 105 - 120 v @ 45ma
  • Receiver  Filaments -- 1.35 - 1.5 v @ 500ma
  • Keying Relay -- 6.0 - 6.9 v @ 575ma

That's a tall order for mobile and portable power supplies but designers in the 1940's were quite clever in packing power supply units.   I managed to obtain both the hand cranked GN-58 generator with the base chassis and seat for portable operations, and a DY-88 for fixed / mobile operations.  


DY-88 mobile power supply

DY-88 set to 12v powered by Amateur 12v supply

Vibrator power supply for low B+

Power filtering

I supply the DY-88 from either an RV battery or an amateur 12v power supply.  When in Standby the DY-88 draws less than 1 amp, but placing the radio in Send mode switches on the Dynamotor which draws 12 amps @12v, without key-down and up to 14 amps on high-output key-down.  It will drain an RV battery pretty quickly at that rate if the radio is left in Send mode, and works an amateur power supply pretty hard as well.  So don't expect to operate remote off a battery alone for too long if your having lengthy QSOs.  An added benefit of the DY-88 is that when the enclosed Dynamotor is running you'll have a nice extra 85 dB of generator noise to accompany your listening pleasure.


GN-58 portable field hand-cranked power supply

Generator head in carry bag

NOS Shiny




Deployed


The GN-58 is a tough workout since it has to be cranked by hand at 60 rpm continuously.  Obviously, you need a partner unless you can figure out how to crank it with your feet while sending CW. You will also want that partner to help you carry the GN-58, and the accompanying accessory bag for the chassis and seat.  IT'S HEAVY.  I haven't weighed everything, but according to the manual that came with the set, the radio / generator / accessories including antennas comes out around 120 lbs.  

If you have a BC-48 battery hooked up then your human power supply can pause cranking while your receiving.  I have a BC-48 battery enclosure that has been gutted of the original, long-dead material and replaced with 10x 9v batteries in series for the low B+ and two D-Cell batteries in parallel for the receiver filament supply.


Accessories

Bag of goodies

The radio itself has a carry bag, as well as a bag for the GN-58 legs and seat, the vertical antenna, and miscellaneous.

There's another bag (shown above) for carrying power supply cables, keys, hand mic, long wire and doublet antennas, external speaker, torture device headphones, torture device in-ear phones, as well as a box of spare tubes for the radio.

If you're traveling in a squad sized group, then many hands make light work, otherwise you're going to be making a lot of trips hauling your QRP rig up the hill.

Headphones

These Western Electric headphones clamp tightly over your ears sealing out QRM and squeezing your head like a vice.  After 10 minutes I was confessing to sins I'd never committed. 








In order to use the headphones the RT/77 receiver must be removed from the case and an impedance switch on the back, changed from 4000 to 250 using a screwdriver.  The ham I bought my set from had constructed a CW audio filter along with an impedance switch on an outboard box, that allowed the use of the headphones without switching the impedance on the receiver unit.

Homebuilt CW filter with impedance switch

Speaker


The external speaker is a rugged, high impedance device (4k Ohms), that after all these years can still output audio at high volumes without distortion.  It has a built in thumbscrew clamp that allows it to be attached to vertical or horizontal objects.

Alternately, the thumbscrew can be used in combination with the vice-like headphones to extract information from a prisoner.

Antennas

The AN/GRC-9 comes with 3 antenna systems; a multiple section, whip vertical for quick field setup and mobile use, a long wire that can be quickly deployed in a fixed station as a sloper, and a doublet for best reception, transmission in a fixed location.



For testing purposes I have my radio hooked up to my 80m Windom, which it tunes very nicely on 80m, 60m, 40m and 10m bands. 

When the weather warms a bit I will be taking the radio out for some portable use and I'll try it out with the antennas that are part of the AN/GRC-9 set.

Spares

As a military radio, it was expected that repairs should be performed in the field when possible.  The radio shipped with spare tubes for the receiver-transmitter, as well as spare tubes and vibrators for the DY-88 power supply.


More to come

In the few days I've had the AN/GRC-9 the only problems I've encountered have been related to the old DY-88 power supply.  Old vibrators cans are generally seized up, as was the case with mine.  Eventually mine became un-stuck after repeated applications of power but there are some methods to restore truly frozen ones using AC current and light bulbs (see Notes section below).

I've made about half a dozen contacts on the ham bands, including a 40m contact to a station in TX which is kinda DX for my locale.  I've received nice signal reports.  I've specifically asked stations about my "chirp" during QSOs and they've reported it as "not bad" and "charming".  When operating from the VFO (master oscillator) rather than a crystal, the GRC-9 will "chirp". It was considered an acceptable design trade-off at the time.  I've listened to the transmitter from a remote WebSDR station to hear the chirp for myself, and I agree that it isn't extreme and lends some character to the station.  The unit does drift about 200 Hz during a QSO which I also think is quite acceptable for it's age.  It's possible that if I spent more time in Send mode prior to a QSO to allow the transmitter tubes to warm up the drift might be lessened, but keeping the radio in Send mode puts quite a load on the power supply (both the 12v supplying the DY-88 and the human cranking the GN-58).

The RT-77 Telefunken receiver doesn't offer much in terms of selectivity and on a crowded band there's a lot of stations to contend with in the passband.  The outboard CW filter deals with this nicely, but it is so narrow that when shifting from Send to Standby, the resulting frequency shift often throws the station I'm receiving out of the filter's passband, so that's a bit tricky.

The receiver's tuning knob also is very coarse, in that fine adjustments are made by breathing on the knob.  However it has zero backlash, which is amazing in a piece of equipment this old.  The markings on the receiver are in 50 kHz intervals so the only way to really figure out where you are is to look at RBN for your spot.

50 kHz spacing when reading the frequency on the receiver
Note the 7.2 is 7.200 MHz in the 40m band


Images

Enjoy the pictures of the AN/GRC-9























Notes

Instructions for restoring a vibrator to operation

VB-1 and VB-7 are interchangable. I think I recall reading somewhere that VB-7 is a "lightweight" version of VB-1 but I won't swear to that.

The base is 4-pin, and the pin numbers are counted as on a vacuum tube with the same base. I wish I could post an image here without uploading it somewhere but if it's possible I've not figured out how to do it. The pins count clockwise from 1 to 4 looking at the bottom of the vibrator or the wiring side of the socket. The two large pins are 4 and 1 and the two small ones are 2 and 3.

There are two basic types of vibrators, called Series and Shunt. The Series type has a contact in series with the coil. VB-1, 7 and 16 are all Series types. I'll skip the Shunt type for now.

Pin 1 is common. Pin 4 is coil. Pin 2 is the NO (Normally Open) switching contact and Pin 3 is NC (Normally Closed). To test a VB-1/7, use an ohmmeter to check continuity from 1 to 4. If the reading is infinity, the coil could be open but this seldom happens. The problem is probably the vibrator contact. If the reading is a few ohms, connect +6 VDC to Pin 4 and -6 VDC to Pin 1. The vibrator should run. If it doesn't, most likely the contact is welded. About the only solution is to open up the vibrator, unstick the contact and try to burnish the burn marks out of the contact.

If the vibrator does run, go to the end of this screed and do the final test.

If the reading is infinity, here's how to use the two or three lamps to (usually) fix the vibrator. SAFETY NOTE: bear in mind you are dealing with either 120 or 220 VAC. If you jury rig the hookup, do all of your connections and disconnections with the "rig" not connected to the AC line. In other words, don't touch anything except the plug on the line cord or (if you go to that much trouble) the ON-OFF switch when the line cord is plugged in.

Connect the hot side of the AC line cord to one side of both lamps. Connect the ground side of the AC line cord to Pin 1 of the vibrator (or socket if you use one). Connect the other side of one of the lamps to Pin 4 and the other lamp to Pins 3 and 2. If you splurge and use three lamps, connect the "cold" side of the second and third lamps to Pins 2 and 3 respectively.

Check all the wiring and when satisfied all is OK, plug in the line cord. Probably nothing will happen immediately. Within a few minutes to a few hours lamp 1 should begin flickering and you should hear the vibrator hum. Run the test until the second lamp begins to flicker or until both the 2nd and 3rd lamps flicker.

If you are only using two lamps, when the 2nd lamp begins to flicker, wait 1 or 2 minutes then remove power (unplug the line cord). Connect the 2nd lamp only to Pin 2 and plug in the line cord. If the 2nd lamp flickers, remove power, move the 2nd lamp connection to Pin 3 and apply power. In either case (with the 2nd lamp now connected to pin 2 or 3 only), let the test run until the 2nd lamp again flickers.

For a final test, connect one lamp to Pin 2 and one to Pin 3. Connect 6 VDC to Pins 4 and 1. With the vibrator vibrating apply power to the two lamps. They should flicker alternately. Note that for this test, either use a 6 volt battery or a 6 VDC supply with both outputs not grounded. I wouldn't try to use the battery in the Jeep just in case you mis-identify which side of the line cord is grounded and which is hot.

Although a vibrator that is going to be fixed by this procedure will usually begin to work after say no more than half an hour, I have seen it take several hours. So if I have one that didn't start working fairly quickly, I'll let the test run up to about 8 hours max (or overnight) before giving up.

Robert
Gunner
USN Retired
MVPA 9480

Hallicrafters HA-1 Electronic Keyer

High Voltage Electronic Keyer from 1960

I've been interested in getting my hands on a Hallicrafters HA-1 since the first time I saw one online.  A 60 year old (as of this writing) all-tube keyer weighing in at 7.5 lbs.  Everyone needs one, right?




The HA-1 uses a mercury wetted relay capable of switching up to 500v circuits.  Wow!  It employs four 12AU7 tubes to perform the mark-space morse timing and uses an OA2 and OB2 tubes for power regulation.


They always auctioned for more scratch than I thought was reasonable or were so beat up that I assumed it had been buried in a damp basement since Eisenhower was president.

I finally won an auction for one that was in reasonable cosmetic condition and fully expected to have to overhaul the electrolytics and out-of-spec capacitors but I lucked out and received one that had been recently overhauled with new power caps and even had replacement polys.  The only issue was the the sidetone circuit was kaput because a neon bulb used in the relaxation oscillation circuit had broken its leads at some point (maybe during shipping) and was rattling around in the case when I received it.

Fortunately I had an old neon bulb in my junk box.  It was socketed rather than having the leads, but I soldered some leads to the socket and installed it where the other had broken off.  Wahlah, sidetone audio restored.


What an electronic keyer was like 60 years ago

After reading the manual and adjusting the circuit balance and element weighting, I had a go with it.  I found that it kept eating a DIT when it followed a DAH in a character like K or N.  I eventually realized that I was hitting and releasing the left paddle (dit paddle) before the current DAH was completing.  

A keyer without DIT / DAH memory

Then it hit me... this tube circuitry doesn't have an "memory".  Modern, digital keying circuits have a memory as to what paddle was pressed while the current element is being completed, but this old electronic keyer is using all its tube goodness to complete the timing for the current element.  So in the case of a DAH, it doesn't have a place to buffer that I've pressed and released the DIT key while it was completing the DAH.  I have to be holding the DIT paddle after it's completed the DAH for it to recognize and begin the DIT element. 

That really calls for a change in my paddle usage.  My brain needs to keep in mind the length of DAHs so that I don't rush and release a following DIT before the current element completes.  This is a bigger problem below 25wpm given the longer length of the DAHs.

The Hallicrafters HA-1 manual states that the keyer is different from other electronic keyers and requires "a knack" to use effectively.  I think I've discovered whereof it speaks.

I practiced for about 30 minutes and then tried some on-air CW QSOs.  They were rough.  Leaving DITS out of K's, C's, Q's and N's and other such letters really changes what you're saying... C's become M's... K's become M's (mmm... M&Ms)  N's become T's... Anyway, you get it.  Very frustrating for both the sender and the receiver.

As luck would have it, one of the stations I worked was also using an old Hallicrafters keyer.  Did I know because he was dropping DITs ?  No, I was telling him about the "new" keyer I was using and he responded that he was using one as well and went on to extoll it's virtues.  

So I plan to use it and not relegate it as a shelf queen.  I figure with enough practice it will actually improve my technique and probably improve my bug fist as well since it causes me to be far more conscious of my inter-element timing when forming characters.

Conclusion

If you have a good, modern keyer and don't need to key high-voltage tube gear, I can't really recommend that you run out and find a Hallicrafters HA-1.  But if your taste runs toward the eclectic and you want a really, large electronic keyer with no memory functions that consumes 30 watts of power whenever it's on and provides you with a wonderful tube smell, and also requires soldering to hook-up the line to connect to an external transmitter... then this is the key for you.

Plus it looks just like H.A.L. from the Space Odyssey 





That's all for now

73

So use a keyer with more wattage than your QRP station to get those extra DX contacts.

DE AA4OO

Space in Morse Code

Silence is Beautiful

The space between characters and words is just as important as properly forming the characters.




If you're rushing your characters the elements of one character will not be easily discerned from the next and the person your sending to will find it indecipherable and respond with a 73 and spin the dial.  I have a number of comments in my logs about operators who ran their characters and words together.  I tend to avoid those contacts down the log.

But, many of us, including myself, are guilty of rushing when we send, especially as a ragchew moves into the 3rd or 4th exchange.  I think the problem is that as the person sending the code, I know what I am sending and in the excitement of wanting to get out all the things I want to say and turn it over, I start to rush and begin compressing the space between my sent characters and words.  After all, it's very clear in my head what I'm sending, it must be just as clear to the listener, right?  Wrong.

RRFBCPYALLWXHRISSUNNYES88FRIGISIC756PRORUNNING100WINTODIPOLEUP66FTDESM8ERW8BJOK

The exchange above isn't far from reality and that's assuming the character spacing was good.  When the character spacing is rushed two characters become a different character or no character at all, and you sit there with your head tilted thinking "what in the world are they saying?"

Proper Space (Timing)

What is considered proper spacing?  Let's review some basics.  A DIT is counted as a single Morse Code element (think of it as a unit of time).  A DAH is counted as 3 times longer or 3 Morse Code elements (3 times the DIT time unit).  

Of course the length (time) of a sent DIT or DAH will change with the speed you are sending.  As the word per minute speed rises, the length of of DITs and DAHs decreases accordingly and vice-versa.  Unless you're using Farnsworth timing, but that's a different discussion...

Space between DITS and DAHS in a Character

Characters other than the E and T are made up of more than a single DIT or DAH.  Between each DIT and DAH making up a single character is space.  The space between each DIT and DAH making up a single character should be as long as a single Morse Code element... a DIT.  So there's a DIT's space of silence between every DIT and DAH in a character. 

"R" = DIT.DAH.DIT (where the '.' is the length of a DIT)

Space between characters within a Word

There should be 3 Morse Code elements of silence between each letter in a word, or silence the length of a DAH, at the speed you are sending.

"WORD" = W-O-R-D (where the '-' is the length of a DAH)

Space between Words

There should be 7 Morse Code elements of silence between each word you send.

"HERE<>ARE<>SOME<>WORDS" (where the '<>' is the length of a the M character)

The length of an "M" ??  Yes.  I was corrected about this in a video I made.  In that video I was counting the DITS and DAHS only, and said to count the inter-word space to be the length of the 'W' character because it is made of a DIT and two DAHS, but I was forgetting the space between the DITS and DAHS that make up the character.  A 'W' character contains 9 elements.  An 'M' contains 7 elements since it is 2 DAHS (3*2=6) plus the inter-character element that spaces them (1-element of silence) equals 7 Morse code elements.

How to Practice

If you use an electronic keyer it will take care of the inter-character spacing between the DITS and DAHS of your sent characters.  If you use a manual key you'll have to take care of that yourself.  You can practice by sending strings of DITS, listening to see if you are placing the same space between each DIT as the length of the DIT itself.  

To practice spacing letters in a word, get used to the length of a DAH (a 'T' character).  Send a T over and over making sure you have the space of the character and the space of silence equivalent.  This gets a bit more complicated with different characters.  An 'E' character is of course much shorter than a 'Z' character but you need to have the same amount of space after each before sending the next character.  I find that I tend to rush into the next character after sending a long character like an 'F 'or an 'L' and add too much space after short characters like an 'E' or 'T'.   If you use a decoder of some type it can be helpful in showing you timing mistakes.  Send into a decoder and see if it turns two of your characters into a different character (you rushed the timing), or see if it spaces the word out as if there's a word break (you're putting too much space between the letters).  It is a very humbling experience to send into a decoder.

Similar to working on letter spacing, spacing for words is potentially an even more important skill.  When we listen to Morse at speed the rhythmic sound of the characters in a word as a whole tends to tell our brain what we've heard.  If the next word is rushed then we don't process the first and miss the beginning of the next.  Practice sending the 'M' character at your preferred speed and get used to the amount of time it takes to send.  

One thing I've tried that works pretty well is setting the break-in timing of my transceiver to match the space I want between words.  At 20wpm the DIT length is between 50-60 milliseconds depending on the measurement you use.  So if I want to be sure I'm spacing properly I should have 7 time-units or 7*50 = 350 milliseconds break-in set in the transceiver.  Both my Elecraft KX3 and Ten-Tec Eagle support setting the break-in in milliseconds.  By being sure that I hear break-in occur between every word I know that I'm putting in a good minimum amount of spacing.  If I don't hear break-in occurring, it reminds me that I'm rushing my words.

When I hear break-in occur between every word I know that I'm putting in a good minimum amount of spacing

Conclusion

The silence you send is just as important as the signal.  Silence is golden



DE AA4OO


Regulated voltage for Regenerative receiver project

Mr. Regula-tor

Building a regenerative tube receiver seems to have been a rite of passage for all hams of yesteryear.  Although I built one from a kit (4-States QRP) as my first electronics project a couple years ago I thought I'd go for the real deal and build a vacuum tube regen receiver.

I'm building a design based around the 6SN7 tube. While I'm collecting parts and still locating a suitable chassis I decided to build a regulated power supply from the parts I have.  Anyone familiar with electronics could probably whip this together in no time, but being the electronics newbie that I am, it is a slow process.

I'm using a transformer from a 1950's Heathkit VTVM V-7 that I parted out. It supplies the 6.3V filament voltage from one set of windings (yellow wires)  those tested good. But the HV was an unknown as it was only half wave rectified when the transformer was used in the meter.  That meter's rectifier and power cap had gone bad so I didn't know what condition the HV side of the transformer.

Breadboarded using a full wave rectifier created with 4 diodes, buffered by a 22uF electrolytic and a 10k resistor, I saw 189 volts, with no-load out of the high voltage side of the transformer. The amount of current the transformer could provide was still an unknown. I tested temporarily with a 2.5kOhm high wattage resistor and saw 56ma of current provided with a voltage sag down to 130V but the core of the transistor started heating up.  Within half an hour it was over 120F so I discontinued that load test.

Fortunately, the regen circuit uses a ridiculously small amount of current for B+; about 4 to 5mA. Although I will likely change the audio side of the tube to deliver enough current for a speaker rather than the high impedance headphones in the current design, which may potentially double that to 10ma.  For the first incarnation I'll stick with high-impedance headphones.

The regen power supply requirements called for [email protected] and 90V@4mA B+.  The B+ voltage  was based on using 10x 9V batteries and it stated that voltage wasn't critical but shouldn't fall much below 90V, going 12% above 90V should be OK.

Generally batteries are used with regenerative receivers because they are so sensitive to power supply noise, but I wanted to give the power supply a shot first and if it proves too noisy I can fall back to battery power for the B+ and just use the filament voltage provided by this transformer.

Since I have a OB2 voltage regulation tube I want to use. The OB2 regulates at 108V so that's what I'm going with.  An OB3 would regulate at 90V, but I don't have one of those.


OB2 in action... Glow baby, Glow!

Calculating the resistor drop

A voltage regulating tube like a OB2 ionizes gas to maintain the voltage at the tube's specification.  In the case of an OB2 it tries to maintain voltage at 108V.  It requires a starting voltage higher than what it will regulate to, but ultimately can only dissipate so much current as it drops voltage.  So, a resistor must be put in series ahead of the VR tube to limit the current it will have to dissipate.  The resistor must be able to handle the current flowing through it, so that must be calculated as well.  

The calculation for the dropping resistor resistance is:
Rdrop = (Vs - Vreg) / (Ireg + Isupply)
So, in my case:
Voltage supply (Vs) = 189V
voltage regulation (Vreg) = 108V
regulator current (Ireg) the OB2 requires 5mA to do its job = 5mA
supply current (Isupply) the actual current required by the 6SN7 up to ~ 5mA

So, (189V - 108V) / (0.005A + 0.005A)  comes out to a resistor value of  10,100 ohms.  10k is the closest standard size resistor and at 108V it should be able to dissipate 1.166 watts.  So I'll need a 10k 2-watt resistor.

Parts is parts

Running the regulated power supply with a 10k Rdrop resistor for a few hours showed the transformer stabilize at 92F degrees at 70F amb.  I was using a separate 27k 2-watt resistor to simulate the ~4mA load that the receiver will draw at 108V. 

As you can see on the newly restored, trusty Heathkit VTVM; the voltage was holding steady around 108V.  With that little current, the OB2 is not visibly glowing but with the lights out the violet colored ionization is visible.



Summary

I'm going to order a larger filter capacitor.  The only one I had to test with was 22uF 360V and I'd like to use higher capacitance value of 47uF with a more appropriate voltage rating of 250V.  I will also be adding 0.01uF caps at the input and output of the filter capacitor and I may add a 0.01uF across the OB2 pins 1/7 to further attenuate any RF noise.   

With the current values I'm seeing 50mA ripple on my regulated voltage.

A bit over 50mA ripple

After I get the new capacitor and get the 0.01uF caps in play to filter out noise, I'll hook it up to the oscilloscope to check for ripple.  I'll update the post at that time.


That's all for now....

So lower your power the old fashioned way, using a voltage regulator tube.

72/73
Richard, AA4OO

Heathkit Art

Heathkit HP-23A and IM-11

I had just buttoned up a HP-23A, after testing its transformer. A fellow ham gave it to me for parts. 

This old stuff looks cool, so I took a pic with the phone and thought I'd share. 

Feel free to use as a desktop background, but if you use it in on the web or publish provide proper attribution.


Metering High Voltage on the cheap

🗲 Heathkit VTM model IM-11 🗲

I recently got a taste for restoring and using vintage vacuum tube radio equipment.  Using equipment that requires 800+ volts for making QRP transmissions is counter-intuitive to the spirit of QRP ham radio, but it's part of my journey as a ham, so I'm writing about it.  Bear with me.  Once I receive a near fatal shock I'm sure I'll move back to 12v powered equipment again.

Until then...

One of the first issues I ran into while testing the power supplies I restored for my vintage gear was how to measure voltages beyond the range of my digital multi-meter.  Most consumer grade digital multi-meters (DMMs) can only measure up to 500 volts then display an error, or stop working altogether.  Previously, I was able to measure voltages over 500V by making a voltage divider out of two 100k 1w resistors and taking my measurement from the middle of the two resistors, but it was precarious in use and added even more danger when working with this old equipment.

I'd looked at getting a DMM capable of measuring high voltage, but the recommended ones, like a Tenma 72-1055 Benchtop Digital Multimeter, start around $100.  Used Fluke meters are even pricier.  I'm sure buying a more professional DMM would be a wise investment.  As I've evidenced many times; wisdom is omitted from my DNA.


So what did hams of yesteryear use?

Behold... Vacuum Tube Volt Meter


Vacuum Tube Volt Meter

This Heathkit VTVM model IM-11 was available at my local Hamfest (Rarsfest) for 5 dollars.

Debugging

Five dollars is not a princely sum, but as with most things purchased from a hamfest, it required some attention.  

The 55 year old 16uFd@150V paper, electrolytic power filter capacitor was likely a ticking time bomb so I replaced it with a modern capacitor for safety concerns.  The closest one I had was a 33uFd@160V radial electrolytic.  I don't think double the capacitance will be a problem for a filter capacitor, it will just make the transformer work a little harder when it's first turned on. I calculated the initial charge time and it's 393ms vs 190ms for the original cap. I think the 10k resistor and transformer can handle the extra 200ms heavy load on power up.

A few wires inside the meter had come loose from some very sparse solder points and a one intermittent connection in the range switch was especially troublesome to track down.
The biggest mystery to solve was oversensitive resistance readings in the Ohms mode.  I replaced the C-cell battery in the battery cup and while I had it out I glanced at the + connection for the battery in the cup.  I appeared to have oxidized at some point in the past and was discolored.  I scraped it off until I saw shiny bits and thought all was good.  I spent more time tracing the circuit and thought I had a problem with the switch itself or the 9x resistors in the range circuit, as suggested in the troubleshooting section of the manual.  The problem turned out to be that oxidized bolt head that formed the positive battery connection in the battery cup.  Scraping it had not provided electrical contact.  In fact, when I removed the bolt (after having to disassemble the circuit board from the meter for the 2nd time), I filed down the head of the bolt  and could find no conductive metal left.  I'm guessing that a former leaky battery had converted the entire head of the bolt to a very hard, yet non-conductive material.  I've never seen anything like it before and it proved to be a useful lesson.
I had to find a replacement bolt and that lead to working on my lawn mower and then mowing the yard... not sure how that progression occurred...  Eventually I got the new bolt in the cup, the circuit board re-installed.  Ohms tested accurately, put it all back together and noticed the #50 pilot lamp had stopped working (sigh).  I removed the innards from the case one more time and got the pilot bulb settled (I think it's required to balance the filament circuit).  While I had it apart for the umpteenth time, I decided to reconnect the 1/4" plug that a former owner had disconnected while keeping their original modification allowing 1 mega-ohm to be switched in for the outermost probe when DC functions are selected but switched out when AC or Ohm functions are chosen.  I wanted to allow use of an original VTVM probe used in the 1/4" plug with its built-in 1 mega-ohm resistor.

Whew!

All-in-all, I probably spent 8 hours getting all the functions on my $5.00 meter to work, replacing old parts, undoing mods and aligning it for proper DC and AC readings.  It's a good thing I don't count my time in the cost of these projects, otherwise I could have purchased a couple Fluke meters for the cost of my time.

What's the fun in buying something that works right the first time when you get it home, huh?  Are we hams, capable of solving problems, or appliance users?  Actually, it would have been nice if it worked without new parts and repairs, but I digress.

Back to the story

This meter can measure up to 1500 Volts 🗲 
The main reason I purchased this is to measure the high voltage in my tube equipment power supplies and 1500 VDC should just about cover it.

Although this meter uses a C-cell battery for measuring resistance, it runs off service mains to power the tubes which control the meter circuitry, so it must be plugged in to be used. 

I love the look of its old "Gran Tran" power transformer inside the VTVM.  They just don't make'em like they used to.



Wiring, lots of wiring

This model was made by Heathkit, from 1961 through 1968, and used typical point to point wiring of the time, making circuit tracing loads of fun.

Point to point wiring makes for interesting circuit tracing 

Shiver me timbers, it's got decks


Fun fact:  When reading the schematic it will refer to "decks".  A "deck" is a wafer section.  When there are stacked sections as there are in the function and range controls the "front deck" is the one closest to the knob (front of the case) and the "rear deck" is the one furthest away, or in the case of working on it, the one closest to you.  When there are more than two decks, as is the case with the range control, it will refer to the "second deck".  As you'll likely guess by now, that is the second "deck" or wafer disc from the front of the instrument.

It also refers to "front half" and "rear half".  The "half" is referring to a side of the deck, so in the case of "front half" it refers to the side of a particular deck facing the front of the instrument, while the "rear half" is the side of a deck facing back (or toward you).

Clever voltage divider circuitry, what could possibly go wrong in this triple stack of wafer switches?



The left knob controls on/off and meter functions while the right knob controls the rather elegant voltage divider.

A knob for every function and a function for every knob.  NOTE: do not plug your headphones into the 1/4" jack on the front unless you want to experiment with personal electro-shock therapy.  Better, yet, don't plug your headphones in there.

Not clearly shown in this photo, voltage divisions up to 1500 volts supported

"...Weighed in the scales and found wanting"


Ok, how many of you understand that completely unrelated biblical reference?

I haven't used a meter like this since I used to plan my VFR flights using an E6B.  My modern, digital multi-meter is fairly idiot proof in terms of reading the results.  My DMM auto-ranges and displays the correct unit of measure along with the reading on its display.  It works well for a dummy like me.

A VTVM on the other hand, has a number of scales that must be interpreted based on whether you are reading DC, AC or Ohms.  Additionally, you have to pay attention to the range chosen.

Choose a reading... any reading, just use the correct scale
Note: the needle isn't at zero in this photo because I had just plugged it in before taking the picture and the tubes hadn't warmed up.  Ah, the joys of vacuum tube powered equipment

The voltage markings for the range switch refer to the full scale reading.  Resistance is the top scale, but let's ignore that because we're talking about measuring voltage...  The second scale from the top is Voltage.  Even though it appears to refer to DC for the numbers on the top and AC for the numbers on the bottom of that second scale that's not what's going on.  The second scale is for both DC and AC.  The numbers on the top are when you are using a range that is a multiple of 15, such as 0-1.5V 0-15V 0-150V 0-1500V.  The numbers on the bottom of that scale are for the ranges using a multiple of 5, such as 0-5V 0-50V 0-500V.   Clever eh?

You have to do a bit of math.  For example, if you if you're using the 1.5V range take your reading and move the decimal place one to the left.  So, a reading of 8 would represent 0.8V in the 1.5V range, while it would actually represent 8V in the 0-15V range and 80V in the 0-150V range, etc.  See, hams were smarter in the 1960s.

Always start with a range larger than what you expect the voltage to be and reduce the range for a more accurate reading if it occurs in the lower 3rd of the range.  The voltage divider set by the range knob is protecting the circuits so if you have it in too low a range and apply high voltage, bad things will likely occur.

Old school needle gauge.  There's a lot going on behind that needle.  It operates very smoothly.



Sporting some temporary probe hookups

Making probes

This meter did not come with probes. I bought another older VTVM pretty cheap, for parts from a famous auction site because it was advertised as having a full compliment of probes, but alas, they were not usable. Even the 1/4" plug from those probes was a bust. However they did come with rebuildable probe "ends"  
I used a RG-58 cable as the high voltage DC wire using only the center conductor and grounding the shield even though it is not used as the ground return. I also placed a 1/4 watt 1 mega-ohm resistor at the tip of the probe to limit any current through the probe cable.  That cable terminates in the 1/4" plug and is wired such that it is out of the circuit in the Ohms position. I secured the cable into the probe body with some glue and used two layers of shrink wrap as a strain relief.  I also put some shrink wrap near the probe tip as a bit of extra insurance.

 I used a spare DMM cable for the outer positive banana plug feed used by the AC and OHMS circuit and made a heavily insulated cable with an alligator clip for the ground probe that goes to the other banana plug. 

The outer probes are used for measuring resistance, AC and low volage DC. The center probe is used only for high voltage DC positive. Both positive probes would not be connected at the same time (as they are in the image below), and would present a shock risk if they were both connected when measuring voltage.



Summary

If you need a way to measure high voltage or are looking for a really eccentric meter to make common measurements harder than they should be get one of these VTVMs.  They seem to be commonly available at hamfests and on famous auction sites for under $10.

Dazzle your friends next time they ask you to measure something for them, by whipping this not-so-little-puppy out of your back pocket and powering it up.  As you're making your measurements quietly repeat "Mmmm, yes.  Mmmm yes, I see now".    They'll have no idea what you're referring to and be quite impressed.


That's all for now...

So lower your power and measure it with the low-range on your snazzy Vacuum Tube Volt Meter

72/73
Richard AA4OO

Subscribe FREE to AmateurRadio.com's
Amateur Radio Newsletter

 
We never share your e-mail address.


Do you like to write?
Interesting project to share?
Helpful tips and ideas for other hams?

Submit an article and we will review it for publication on AmateurRadio.com!

Have a ham radio product or service?
Consider advertising on our site.

Are you a reporter covering ham radio?
Find ham radio experts for your story.

How to Set Up a Ham Radio Blog
Get started in less than 15 minutes!


  • Matt W1MST, Managing Editor