Building a Low Power RF Attenuator and Some Nano VNA Fun

This Project's main goal is to build a RF ATTN (attenuator) to lower the RF Output level of my Baofeng UVR-5 Handy Talky that will be used as a Hidden Transmitter Hunt Fox Transmitter.

The secondary goal is to get to know and use my newly acquired Nano Vector Network Analyzer to test the attenuator.

The Baofeng TX outputs 1.2 watts on Low power.  Ideally, I want to transmit approx. 200mW to the Fox Antenna.  General License Test math says:  If I place a 6dB (Divide by 4) ATTN between the TX output and the antenna feedline- the resultant RF Level should be approx. 300mW.  Since I used to work for the Government... that's close enough.

Regarding how to build the ATTN, this particular Youtube video inspired me the most:
#91: Basic RF Attenuators - Design, Construction, Testing - PI and T style - A Tutorial - YouTube

But instead of doing lots of math in my Spiral Notebook... An online App was discovered to provide the schematic and calculate the resistor values. 
Check it out: Pi Attenuator Calculator - everything RF

Values of R1 and R2 are rounded off.

In my collection of 1-watt resistors- I didn't have the exact values.  So put resistors in parallel until the approx. desired value was obtained.

Then blob soldered two SMA-F chassis connectors to a piece of blank Printed Circuit Board.

Lastly, blob soldered the resistor combinations to the PCB, accidently in a neat fashion.

How did I do?  Is the Attenuation close enough?  Is the SWR on each side of the ATTN = 50 ohms?             Enter the Mini Vector Network Analyzer...

Important Note:  This my second time using the VNA to make a semi precision measurement.  I had to start over many times until I was satisfied that I was doing the testing correctly.

The VNA was set up and calibrated to measure a 2-port device via Port 1 to Port 2.  And SWR measured on Port 1 only. 

The SWR and Log Magnitude (signal level) displays selected. 

VNA frequency span is set:  10MHz to 200MHz- as seen at the bottom of the screen.  The Green trace is the amount of attenuation measured across the frequency spread.  There is a marker (little box with 1 inside of it) at 141.1 MHz on the Green trace that indicates 5.71 dB loss, Port 1 to Port 2.

Also, another marker that indicates the same frequency on the Blue SWR trace at the bottom of the display- which reveals it is 1 to 1 across the frequency span.   When removing the test cable from Port 2 of the VNA, which unterminated the ATTN of its 50-ohm load... the resultant SWR at Port 1 was 1.5 to 1. 

Another concern is:  Will the 1.2W of RF Power melt down the ATTN's resistors?  After testing the Baofeng HT for 5 hours with the ATTN in series with the antenna- the resistors get noticeably hot to the touch.

As mentioned above, the Baofeng's Power Out measures as 1.2W.  At the output of the ATTN, the Power Out measures approx. 320mW.   Almost as calculated.  Much too close for Government work.

If the Nano VNA sparks any interest for you, gentle readers... I received my inspiration and information from this Youtube Ham Radio enthusiast:  smoking ape nanovna - YouTube - a Smart easy to listen to presenter.    There are many Mini VNAs to choose from.  I chose this particular model because the Port 1 to Port 2 dynamic range was listed as 70dB versus 50 or 60 dB in others. 

End of Blog.  As usual- THX for reading this far.  30, 73 - Ricky KR7W.

Hidden Transmitter Hunting - Fox or Bunny Transmitter Controller

 The KR7W Hutch of Hidden Transmitter Hunt Bunnies are becoming aged.  The old timer bunnies are made from Icom handy talkies from approx. 1982, use a Ham Gadgets Repeater controller circa 2005, and contain a 5-pound Sealed Lead Acid battery.  The battery seems to double in weight for each year I grow older.   Enter some fairly new technology – that’s not just for newly licensed hams.

Since College Days- KR7W's workshop has been
dubbed "Acme" after the many projects built. Where some projects
worked as well as Roadrunner catching gizmos that the Coyote
used to entertain us with.

At Amazon.com, Baofeng UV5R handy talkies are currently priced at approx $20 and their Extra Strength 3800 m/A batteries are the same price.  For about $40 (plus ancillary parts) a long-lasting hidden transmitter can be built.  But, how to control it?  Or... how to key the FM signal plus the station ID for a certain time, then shut down for a certain time, and then start back up, over and over again?     Arduino Nano (3 for $20 at Amazon), plus inexpensive miscellaneous parts to the rescue.

 This blog article hopefully reveals my so far successful creation of a reliable, powerful, Hidden Transmitter or Fox or Bunny that will run for the duration of the hunt.

 The Baofeng part is easy– insert the charged 3800 a/H battery, set the frequency, set Low Power, let the LOCK.   The Arduino controller was challenging.  I do not know the C++ Programming language to create a program of my own.  Persistent Google searches found a Fox Controller Project that uses an Arduino Nano that has simple wiring and a well-documented sketch (the program file) that helped me to figure out what to change to fit my Fox Hunt needs. 

The Block Diagram - Schematic basically explained:    The Arduino is programmed to operate Push To Talk and play a user defined Morse Code sequence for 33 seconds.  Pin D12 outputs +5V when this happens.  +5V is divided by R1 and R2 to place approx. 2.5V on the G of the 2N7000 transistor.  A GND then appears on the D lead– which is PTT toward the Baofeng.   PTT is wired to the 3.5mm plug Sleeve lead.

 The Morse Audio appears at Pin D13.  The Audio Level is a 5V square wave that is divided (attenuated) by R3 and R4 to emulate the level of a Baofeng Speaker Mic.   Capacitor C1 blocks the DC voltage from the Baofeng mic lead.  Transmit audio is applied to the Ring lead of the 3.5mm Plug.  Note:  Pin D13 has a board mounted LED connected across it.  When the Morse Message is playing- the LED flashes along.  

9V from the battery passes through a small toggle switch that turns the Arduino controller ON-OFF.  The 9V source connects to the Arduino at the V-in pin.    GND for the Baofeng’s PTT and Mic Audio In is delivered via the sleeve of the 2.5mm Plug.

Many people have told me that the hardest part of programming Arduinos is ‘Getting the Sketch into the Arduino’.    I encourage you to watch some Youtube videos like:  Arduino Nano how to load a Sketch - YouTube... 

Or read about it at:  Getting Started with Arduino products | Arduino   Ask around at your Radio Club. The sketch I used is here:  Auto-Keyer for Radio "Fox Hunting" | Arduino Project Hub  

I’d like to tell you that customizing the Arduino to meet your Fox/Bunny Hunt needs is EASY– but, nah.  However, it’s not too difficult if the notes in the sketch are carefully read.

An example is Changing the Morse Message.  Note:  If the line of programming in the sketch is preceded by a double slash // – what follows is the explanation the programmer left for you.  Also known as 'comments' or 'commented out'- which has no effect on the Adruino Programing.  

 // the loop routine runs over and over again forever:

void loop() {

  digitalWrite(relay, HIGH);   // turn the relay on (HIGH is the voltage level)

  delay(1000);                 // wait for a second, 1000mS

   SendText("A A A A A A A A A A A DE KR7W"); 

   delay(1000);                 // wait for one second

   //SendText("KR7W FOX HUNT");// alternate text to use

   //tone(TonePin, frequency);    // send 7 sec tone

  //delay(7000);

  //noTone(TonePin);

   //delay(1000);                 // 1 seconds transmit w/o tone

 

Find the line that begins with: SendText(  The text between the quote marks is what will be sent in Morse Code.  The amount of text, between the quote marks, determines the amount of time Push To Talk will be operated.   I added this amount of As to make the PTT stay operated for 33 seconds.  Experiment to meet your needs.  Change the sketch then reload it into the Arduino.

 Here,  digitalWrite(relay, LOW);  turns OFF Push to Talk.  Then delay(cycleDelay); makes the Arduino wait for 15 seconds (defined elsewhere).  It performs this command twice for a wait time, with PTT non-operated, for 30 seconds.

 digitalWrite(relay, LOW);    // turn the relay off by making the voltage LOW

   delay(cycleDelay);           // wait for cycle time (because the largest value is 16383)

  delay(cycleDelay);           // wait for cycle time 

 

The following photos show how I went about creating a Fox / Bunny Controller that works for me and just might provide some ideas for your project.

 Electronic parts are shown on the schematic-block diagram, Fig 1, above.

 A Mint Tin ended up being the perfect size for the Arduino and the 9V battery.  My non odorous mint tins, as well as the Arduino, battery connector, small toggle switch, 2N7000 transistor, Audio Plugs, hook up wire, solder, soldering tips, resistors and capacitors conveniently came from Astronaut Bezo’s Amazon dot com.

 

(Fig. A) A piece of Printed Circuit Board with Dremel cuts creates Manhattan style solder pads for wires to attach– as well as a ground plane to easily GND a component or wire connection.    

(Fig. B) A piece of 5mm scrap plywood was cut into 2 pieces and glued together to become the 'platform'.  The smaller plywood piece became the divider to hold the battery in place.  The Arduino is glued to the platform with E6000 hobby cement.

Hot Glue would work just as well.

The wire shown- passes through small holes drilled through the plywood and printed circuit board.  The wire ends are soldered to the PCB Ground Plane.  The intent is to Bond the PCB GND Plane to the Mint Tin.   This may not be needed- but did it, as it was easy.

(Fig. C) The notch Dremel carved into the plywood platform holds it in place under the switch.  At some point the platform will need to be removed for Arduino reprogramming.  Accidental Engineering provides the platform to easily slide out toward the bottom of the tin without removing the battery for access to the USB port.

 

Got Questions?  Email me.                                                

  As Always– thanks for reading this far.  30, 73, Happy Trails, KR7W

 

 

Hallicrafters R46 Speaker - Perhaps a Useful Hack

 

The Classic little h R46 speaker nicely matches up with the SX-71 or SX-101 Mk 3.

This R46 came with a 10-inch speaker which in my opinion was intended for listening to 3875 AM, Shortwave Listening AM broadcasts which included music, and the Phono Input where the record player delivered music to the receiver's 6V6 audio amplifier.

Whilst listening to SSB or CW from either receiver- the 10-in speaker cuts the high frequencies too much. It's my opinion that the cutting of the high audio frequencies reduces the intelligibility of SSB and CW signals.  

For me- the best speaker to listen to modern day ham radio communications is with an older Mobile Radio Speaker- like GE MASTRII or Motorola equivalent.  In addition, a tuned port CW speaker is beneficial to use with older receivers that do not have modern DSP noise reduction or sharp selectivity.  

Ressonant Speaker for CW.pdf (qsl.net)

A tuned speaker for CW (k4lxycw.wixsite.com)

In this modification, the goal is to improve the audio intelligibility from my two little h boat anchor receivers.  Specific measurements and how to assemble will not be provided- but photos will hopefully give enough info if you choose to modify your Hallicrafters R46 or another classic boat anchor speaker. 

The original R46 10-in speaker was removed and stowed away for safe keeping.   The mask between the speaker and the cabinet was used as a pattern to create an interior front panel between the h cabinet and the new speakers.   3/16" hardboard / Masonite material was used.  5mm or 1/4" plywood would work just as well.

Arrow points to the original internal mask between the speaker and metal grille.

Next, a platform / deck was built to match the needed elevation of the Comm and CW speakers.  Plywood would work best, OSB was used- as it was on hand.

Height of platform that the speaker(s) sit on depends on the size of the speaker(s) being used.  The opening space where sound it emitted from is sharpie penned on the hardboard front panel.

The vertical 'stick' provides support for the DPDT switch that switches between CW or SSB speakers.  The switch is elevated to a convenient height for easy access from the front of the speaker.  

PCB material was my choice to support the slide switch that connects to the support stick.  Masonite or 5mm plywood would also work well.

Arrows point to the R46's original speaker baffle attachment points.

Squint just right and kinda see the two speakers behind the h grille cloth.

Note:  No holes were drilled in this R46 Speaker enclosure.  This modified R46 can easily be returned to Hallicrafters collector status.  

End of h R46 modification report.   Best Regards, Ricky KR7W

Adventures in the 2022 Fall Classic Exchange

 2022 Winter Classic Exchange, AKA 'CX' – CW Portion

 This is my first CX operating under my own call sign.  Previously I’ve operated many CXs at W7OS- the Doc Spike Antique Radio Museum in Tacoma, WA- before moving to Idaho.   

CX Info:  Classic Exchange CX

The following radios emerged from moving boxes, were briefly tested, and took turns on the modest sized op table:   Hallicrafters SX-101 and SX-71, Drake 2B, Heathkit HW7 QRP, Homebrew WW2 Paraset Replica.  They shared tabletop space with a WRL Globe Chief 90 and a Heathkit VF-1 VFO transmitter- which outputs 50-60 Watts. 

Hallicrafters SX-71, SX-101, Heathkit VF1 VFO connected to the WRL Globe Chief.  A MC Jones Micro-Match power meter sits atop the VF1.

The Antenna here is a HOA stealth 20M-40M fan dipole in the attic, about 25 ft above ground and surrounded by 1000+ roofing nails.  There’s a bit of manmade noise on 40M.  I am amazed that it works as well as it does.    

Starting on Sunday- the SX-101 and Globe Chief TX combination worked well.  After pairing the Drake 2B to the Globe Chief TX, it was discovered that 2B did not receive 40M.  The required 3 Qualifying QSOs were made on 20M and then the 2B was swapped for the SX-71.  Casual operating on Sunday netted a surprising 20 QSOs. 

Tuesday's game plan was to qualify a Homebrew WW2 Paraset replica and a HW7 QRP transceiver.  

Paraset in wooden box.  Russian Cold War tank radio set straight key plugs into Aux Key Jack of the Paraset.  Under the modern speaker is the Paraset's 300V Power Supply.

The Paraset only operates on 40M.  The TX is a XTAL controlled 6V6 oscillator and outputs 3.5W.  The RX is a one tube Regenerative Detector with a second tube audio amp.  It runs on a separate HV Power supply.  

Transmitting with the Paraset the involves connecting the antenna, choosing the desired crystal, then key down to peak the Tank and Ariel tuning caps for maximum brightness of the two RF Power indicator lamps.  RX involves setting the Tuning cap to match the desired freq as noted on a calibration chart or the RX frequency can be determined by listening on a separate receiver to hear the small signal from the Regenerative Detector.  The RX tuning cap in conjunction with the Reaction control adjusts the RX frequency and the sensitivity.  Each control affects the other.  The Receiver is quite sensitive and somewhat selective, but the op must listen closely as there is a high frequency audio hiss present.  The RX is easily overloaded by nearby higher-level signals. I was lucky to make 6 QSOs before a loud digital signal started up 5 KCs away.    

For a small taste of Paraset info, see:  #5 Valve QRP - Building a Paraset - YouTube

 

HW7 sits on top of its modern 14V LiFe battery.  Russian straight key and speaker.

Heathkit’s second worse ham radio product, the HW7 replaced the Paraset and 1 QSO was made before the 40M band died. Then 3 QSOs were made on 20M after the band woke up. .  The HW7’s TX outputs 1.8W.  The RX is direct conversion- which produces an upper and lower sideband signal.  On 20M the VFO tuning is doubled, so a very small movement of the tune knob changes the frequency many KCs so it’s hard to precisely tune in a station. 

It was very enjoyable to be able to make 34 Qs using no more than 60 Watts attached to a crappy antenna with ham radio sets 65+ years older than an IC-7300.  Could ‘IC’ be the abbreviation for Ice Cold? As compared to the warmth and glow produced by vacuum tube radios?  I think so, but YMMV. 

The best CW signal I heard came from The Doc Spike Antique Radio Museum W7OS's XTAL Controlled Globe Scout 65.  It possessed a distinctive chirp but was easy to copy VS other signals with too much chirp.  An ARC-5 comes to mind.  In this CX, I hit the Jackpot- to copy Jim W8KGI's many rigs all in a row.  This year on 20M Jim offered up 6 combos. Also it’s interesting looking at the QRZ pages of participating CX ops to see some photos of the unusual or interesting equipment they reported using.   

For my next CW CX event I’d like to have a second 50W transmitter- hopefully a homebrew.  Thanks to all who participated.   

Ricky KR7W

 

 

2 Meter Coaxial Dipole for APRS or 2 Way Comms

 This article was originally written in 2013.  For an unknown reason- Google's Blogspot had denied access. Now, The New and Improved Ver 2 has been edited to remove the 'fluff' for clarity and easier to access.   

My intended purpose for this higher efficiency, more gain that a 14-inch rubber duckie 2 Meter portable antenna was to slip it into an outside pocket of my hiking pack to transmit my APRS Position whilst enroute to Summits On The Air Peaks.  Once I reached the peak- then the antenna would be used for 146.52 FM contacts.  

Before I retired- I worked as a 2Way Radio Technician. One of the common base station antennas I encountered was the Coaxial Dipole... sometimes known as a Sleeve Dipole... where a metal sleeve on the lower side of the antenna acted as the counterpoise.  The coax cable feed line below the sleeve didn't affect the tuning or performance.  Motorola spec sheets say the sleeve dipole is the same gain and performance as a J-pole, a Roll-up Slim Jim- but without the extra length of the 19-inch tuning stub.  

This DIY project is to create an antenna that emulates the Coaxial Dipole.

Above drawing: The location where the coax shield ends, and the top radiator (the whip) begins is the dipole's FEED POINT, the center of the dipole.  One of the 1/4 Wavelength poles is the top section whip. The opposite pole is the coax's shield from the feed point to the beginning of the 7-8 turns of coax. The coil or turns of coax act as a Radio Frequency Choke. The purpose of the RF Choke is to stop RF from traveling any further down the coax feed line to the radio.  The RF energy stops 18 in from the feed point- which is the correct length for 1/4 Wavelength 2 Meter radiator. 

The main support or handle of this antenna is 3/4" PVC Irrigation pipe which houses the coax cable lower pole.  Sked 20 or 40 will work.   A mobile whip antenna was added as the top pole- so to survive an occasional tree branch collision when hiking with the antenna in my pack.

Finished antenna shown below: 

NMOQ (quarter wave) has no loading coil inside.

After cutting the cap it should look something like above.

I used RG-174 coax as the feedline- harvested from an Amazon purchased jumper (info below).

When I place the antenna in a pocket of my pack- the whip extends about 18" above my head.

I sometimes use another 18" piece of PVC pipe with a coupler as an extension for this antenna to elevate it higher when I think I am in a Sketchy APRS area. The extension makes it easy to jam into cracks or a rock pile on a summit.  It also makes it easier to get bonked off the trail by hitting that low hanging branch.   I have used electrical tape to secure it to my trekking pole handle stabbed into the dirt.

The piece of Perf Board supports the coax inside of the PVC pipe and makes the connection to the hacked NMO cap.  Plywood, Plexiglass, or other non-conductive material can sub for Perf Board. 

A= Heat shrink tubing is placed over the coax center insulation because when soldering the center conductor wire to the larger copper wire... the heat will melt the coax insulation.  The heat shrink keeps it from melting out of place.

B= Solder the coax center conductor to the NMO connection.  Once soldered- use wet paper towel to absorb excess heat.

C= Epoxy or hot glue could be added at this joint for rigidity.

Slide the Perf board with attached modified NMO cap down into the PVC coupler. The coax cable needs to be inserted first.  Apply epoxy glue to the board and the top of the coupler for the NMO.  A too much epoxy mess will not hurt- as epoxy is nonconductive.  


  

The 3-inch slot cut into the PVC pipe provides adjustment of the location of where the beginning of the coax RF Choke coil starts.  If the antenna needs to be longer then move the coil further down- this makes the length of coax inside of the pipe longer.   

The RF Choke coil can be temporarily secured in place with electrical tape while SWR testing takes place.  Once the coax coil position is found- a couple of wraps with quality electric tape will hold it in place for years to come.  

A standard 1/4 VHF quarter wave mobile whip antenna is used as the upper pole of the dipole. 

ADDITIONAL INFO Section...

I built another version of this antenna called the Lazy Pole- which uses a 3/4-inch wooden dowel for the support.  The article for it is at:

bark_2020_04.pdf (w7dk.org)

Page 24

[ www.w7dk.org / Newsletters / April 2020 / Page 24

A 15 ft long RG-174 coax jumper with SMA (for Japanese handy talkies) and a Reverse SMA (for most Chinese handy talkies) coax connectors on each end can be purchased here:

  Amazon.com: 

Superbat RF coaxial SMA Male to SMA Female Bulkhead RG174 15ft Cable + 3pcs RF Coax SMA Adapter Kit for SDR Equipment Antenna Ham Radio,3G 4G LTE Antenna,ADS-B,GPS and etc : Electronics

  

Note:  if the link does not work- please cut n paste it into your search engine.

Simply cut off the coax connector that does not match your radio.

Above I mention 'Quality Electrical Tape' my fav is Scotch 33, found at Home Depot.

The Flowerpot antenna is another version of the Coax Dipole Antena:

Half-Wave Flower Pot Antenna – VK2ZOI

Again, if the link does not work- cut n paste it into your search engine.

You'll need a metric to Imperial length converter.  There's an app for that.

The effectiveness of the Coaxial Dipole (named "The Home Brew Dipole") was tested as compared to other antennas typically used on a Handy Talkie.  See:

bark_2018_10.pdf (w7dk.org)    Cut n Paste if necessary.

Scroll to Page 6  

[ w7dk.org / Newsletters / Oct 2018 / Page 6 ]

End of this blog entry.  Best Regards, Rich KR7W 

Email me if you have questions.  QRZ lookup will find me.

Photo Below:  Summits On The Air expedition.  Coaxial Dipole is being held up by the brush.  KR7W has QSO with a ham approx. 50 miles away- line of sight- 5 watts Kenwood Handy Talkie.  

Rainy Day Monday - Old Radio Restoration - Philco 52-940

Memorial Day, 2020 -'Twas dark, rainy and dreary as compared with the last 10 days of nice weather.   A good day to retreat to the Covid-19 Proof Shelter in Place Bunker / Workshop for a "Now for something entirely different" project.   This old entertainment radio was picked off the shelf for restoration:

Philco 52-940 acquired at the Seaside Hamfest more than 10 years ago.

Blog Ver 2 Edit Note:  The 52-940 radio is a newer version of the "Hippo" radio from Philco:

Above:  Philco Hippo radio from 1948 (photo from Ebay)

The rainy day restoration radio- was missing its back which contains the loop antenna.  On initial power-up, using Isolation Transformer and Variac, revealed FULL AUDIO level AC hum noise from the speaker.  The volume control had no effect.  This was hopeful news that told me the series string tubes are not burned out, the rectifier and audio output tubes are working.

A sacrificial scrap wood 'cradle' was fabricated in the Acme Woodworks shop to protect the vulnerable components on top of the chassis.  Did the uncontrollably extremely loud hum cause the speaker to shatter apart?  or was the radio attacked by Speaker Bugs?  Under the chassis- the first organ removed for transplant was the DC power supply Filter Capacitor- shown under speaker above.

Under the chassis the usual paper+wax capacitors were replaced with Yellow Polystyrene caps.  Red arrow points to 3 electrolytic caps, taped together, sitting in a bed of silastic silicone compound which secures and isolates them from the chassis.  The original filter cap wires were reused to connect to the new caps.

Looking closely at the IF cans- their base is made from clear plastic- which means that they contain mica wafers  separating silver plates that create the capacitors in the IF can's 455 KC tuned circuit.  Almost always- these caps develop Silver Mica disease which causes scratchy sounding pops in the radios audio.  Since this radio plays without noise- the IF cans did not undergo the SM Cure procedure.   

At first I thought this radio was in the "Hot Chassis" category- where one side of the AC Line Cord attaches to the metal chassis- which was popular back then.  Analyzing and experimenting reveals that the NEG power supply has a separate GND path- not connected to the chassis.  Some items like the speaker common wire and tuning capacitor are GND'd to the chassis.  It was confusing where to connect the polarized AC line cord NEUT wire- which, BTW connects to the AC ON/OFF switch (backwards in my thinking).   The radio had a wax paper .047 mfd cap across the AC line to bypass noise from getting into the RF sections of this radio.  Two .022 mfd X safety caps were transplanted that now perform this function.  Note:  experimenting has demonstrated the necessity of AC line bypass caps.  Man made noise is drastically reduced in most cases.

In the foreground is the new radio back panel and loop antenna.  A piece of 2.3 mm plywood was carved, little by little on the table saw to fit into the rear of the radios case.  1.125" holes were hole-sawed.  The loop antenna organs- the dowel-bobbins and the 40 ft of cloth covered wire- were harvested from an old battery powered set from the RCT Free table. 

A same sized speaker was not found in the workshops acres of junk spare parts. The best sounding speaker that fit is from a Motorola Golden Voice radio- which sounds pretty well.  Experiments were conducted with 'tone control' capacitors- from the Audio Output Tube Plate to the Negative supply- to make sure the selected speaker didn't sound too treble-y or too bass-y.

The original crappy speaker was screwed to the radio chassis- which made it more difficult to replace.  The Motorola Golden Voice speaker is mounted to a plywood adapter plate- in the same place as a cardboard gasket resided to isolate the original speaker from the bakelite radio case.  The speaker was moved 3 times on the plywood- to get it to fit just right with the chassis placed inside of the cabinet.  The front of the plywood adapter plate is painted black to hide this modification from outside view.

The rear cover in place.  More vent holes were later drilled after it was noticed that the power supply side of the radio seemed extra hot. 

Here the radio is on it's shelf in the Bunker Workshop- not being a Shelf Queen- but playing KIXI, KOMO, KVI on a regular basis.  

The bakelite case received a bath in Dawn Dishwashing suds, then polished with automotive plastic headlight cleaner.  The brass trim got an overnight soak in vinegar then precision polished with Brasso.

After careful alignment the radio plays well, especially the Hispanic Music stations at the top of the AM band.

End of Blog Report.  KR7W