Exciting Bitx40 Modifications

Make your Bitx40 Work FT-8 & WSPR!

Happy New Year ~ 2020!

[The following describes two methods for adding USB capabilities to your Bitx40 and further adds an alternative to the Raduino controller.]

So OK I built my first Bitx20 shortly after VU2ESE announced to the world his exciting project -- that was over 10 years ago. Even then W6JFR (now N6QW) hacked that project by converting the inductors to ferrite core, shifting the IF to 9 MHz and adding an LCD Display complete with a EI9GQ VFO stabilizer.

So it goes to reason that I would still be hacking this amazing design --today! Bill N2CQR provided two "seed" Bitx40 boards and I am happy to report that both boards have now been "hacked" so that they are USB capable and working the 40 Meter Digital Modes such as WSPR and FT-8. 

But here is the icing on the cake the hacks involve two completely different approaches and this experimentation has provided some new ideas for me for use on other rigs.

1. The first hack was to design a controller other than the Raduino. You see I got two Bitx40 boards from Bill and only one Raduino --which doesn't work. I now have a working controller which has been expanded into two working controllers to support the two boards. They are not identical...
   

   

   
   
2. The first approach was to create two BFO frequencies (Upper and Lower Sideband) using the 5 MHz LO so that you could select USB or LSB. Much experimentation was required to find these two frequencies. That done we added two VFO's in the code and also a TUNE Tone. That rig has been in operation for several weeks using WSPR and FT-8. A bit of history here -- not all Bitx40's are standard as there are as many as three variants depending on the crystal sorting process and the Center Frequency of the 4 pole filter. So be aware --not all is plug and play. You will have to find the Cf and BFO frequency of your particular Bitx40. This mode involved unsoldering one end of L5 and applying the BFO signal through a 10NF Capacitor which is fitted to the base of Q10.  
3. 
4. So now to the second Bitx40 board -- here the BFO frequency is actually about 400 Hz higher so you have to account for that in the code. This also means that the Filter Center Frequency is different! In the  SolderSmoke Podcast #216, Bill described his adventures with digging into the innards of a uBitx. One comment just sort of passed through my head and it wasn't until I was listening to #216 on my morning walk -- I had a Shazam Moment. Boom another way to do it. Bill mentioned that the uBitx being dual conversion used a technique to place a conversion oscillator with either the clock above 45 MHz by an amount of the IF or below 45 MHz by the same value of the IF. 
5. There it was... The Normal LO of the Bitx40 is in the 5 MHz range and with a 12 MHz IF a mixing process puts the display at 7 MHz and LSB. Now suppose you put the LO at 19 MHz and down mixed so LO - IF now gives the Display at 7 MHz but with sideband inversion and USB. Boom it works; but more must be done to have it work on the Bitx40 board. Read ON!
6. The code has to be modified so that for LSB you have a term that is software selected such that ( bfo - rx ) will result with the 5 MHz LO (rx). But when you change to the 19 MHz LO (rx) a new term must be called up in the software so that you now have (rx - bfo). This also causes in one case a CW rotation of the encoder makes the display go higher in frequency. When you switch to the other LO then it is a CCW rotation that will make the display go higher. For those who cannot take this quirk the LO select switch instead of a simple SPST could be a DPST switch where the 2nd half controls a small DPDT relay to reverse pins D2 and D3 so that it will always be CW increases the LCD reading. So for those of you who have trouble seeing this arrangement.
   

   

   In essence we have a cross connected DPDT relay that when not energized connects the D2,D3 pins on the Arduino to the Rotary Encoder. Upon being energized the D2, D3 pins are now switched to the opposite direction from the former state. the Encoder will now tune in the same direction for either LO. The triggering circuit would be to have a spare Arduino Pin go HIGH which would drive the base (through a 1K resistor) of a 2N3904 wired to the field coil of the relay. This would mean keeping the USB LSB Select to just the single SPST switch.
7. There are those who would maintain that juggling the software could do the same thing -- well friends a hardware solution like this is simple to accomplish and you will not spend countless hours debugging software. When you know stuff you can do stuff. 
So I rewrote the code such that with one position of a SPST switch you have the 5 MHz LO and LSB which is displayed on the LCD. The engagement of the switch shifts the LO to 19 MHz AND changes the math so that the BFO frequency is now subtracted in this case and the LCD reads USB. A further refinement would be to have the USB Boot Up be at 7.074 MHz, which is a simple code change.

I then loaded the software and connected my controller to the Bitx40 where now the original BFO crystal is retained and works for either USB or LSB. I simply plugged Clock 0 into the two pin port and fired it up. All worked well for LSB and the sound is really crisp and I have indeed installed the correct offset -- the sound is on the mark and the frequency is on the mark!

A switch over to USB and while signals were heard -- they were weak, very weak. Hmmm my new theory has been shot to hell! 

It is always best to put things aside and "noodle" over what happened. The frequency scheme does work (super good for LSB); but marginal for USB. I printed out the schematic for the LO injection and saw (after sleeping on it) a possible reason for the USB issue. 

In the original Bitx40 the frequency control was by means of a varactor tuned VFO which shortly was replaced by the Raduino. Essentially the varactor tuned VFO is a Colpitts Oscillator and in place of the tuning network the Raduino is installed. There are a couple of very Large caps -- 1000 PF in the Colpitts circuit that work OK at 5 MHz but when you place a 19 MHz signal in there -- closer to a short circuit! If you do the math assuming  a 500 PF load at 5 MHz the load is 63 Ohms and at 19.2 MHz the load is 16 Ohms. So it must be a loading issue. (Two 1000 PF is series = 500 PF)

So my first thought was to bypass the Colpitts oscillator transistor and introduce the LO beyond that stage. Boom -- when you know stuff you can do stuff!

That did the trick and the above schematic shows the where to inject the signal. We now have a way to add LSB and USB to the Bitx40 while retaining the same BFO Crystal. Listen to the SolderSmoke Podcast #216 and you will hear me discuss how Ten Tec did a similar trick to use one BFO crystal.

There is a bonus here in that you even end up with two VFO's and the memory function holds.

So dig out that Bitx40 and put it on FT-8. If you email me at craponthebench@gmail.com, I will be happy to send you both sets of code.

Here is the 2nd Bitx40 on WSPR today:

That is Europe, Africa and South Africa! The Modification definitely works. Pout = 5 watts! The antenna is an inverted V with an apex of 35 Feet. 

Caveat Emptor -- there are three different BFO frequencies used with the Bitx40 so you might have to "diddle" a bit to, get yours to work. If you don't know how to do that --don't even start my modification!

73's & Happy New Year

Pete N6QW

The Whaddon Mk VII - Paraset Clandestine Radio

What is a Paraset?

12-12-2019 40 Meters is wide open!

This is just the transmit part of my Bitx40 using my N6QW Controller. First time to be heard in the Philippines and Singapore from this QTH. Maybe this could be done with a Paraset?

20 minutes later --really open. (Transmit only)

Yes haywire and chewing gum; but can also do FT-8.  *****************************************************

No I didn't say parasite but Paraset! 

Bottom Line the Paraset was a clandestine radio set built in England and supplied to agents operating behind enemy lines in the European theater during WWII. Most went to the French Resistance; but some found their way into the Scandinavian countries. 


It is believed that Torstein Raaby, the radio operator on the Kon Tiki voyage used the Paraset while a Resistance Fighter in Norway during WWII.


The whole radio was stuffed inside an innocuous looking scruffy suitcase. The standard issued radio even included some spare tubes right in the case.

What is up here?

At first I thought it would be nuts to rely on such a radio as your primary communications link with your headquarters. Then I forgot that the British had 10,000 National HRO Receivers parked about 100 miles away whose only job was to copy your signal and at the same time some very powerful radio stations sending RF in the other direction. So perhaps when the whole system is evaluated not all bad.

It was quite an interesting radio receiver / transmitter that I previously mentioned was stuffed into a small suitcase so that it would be disguised. 

There were several options for powering this rig including from the mains as well as from a DC power pack (vibrator supply). 

The basic scheme was a two tube (6SK7's) regenerative receiver and a one tube (6V6) CW transmitter crystal controlled operating in the 3 to 8 MHz range. The receiver tuned the same range. The power output of the transmitter was at best about 5 watts. Keep in mind friend Bill, N2CQR has made 20 contacts on 40 Meters running a one transistor regen and one transistor transmitter at 100 Milli-watts. Some DX is like 1000 miles away --so 100 miles with 5 watts is a cake walk!

The transmitter tuning used a two light bulb system (Load & Tune) for indicating maximum RF Juice to the antenna. For antennas don't think of 40 Meter dipoles strung between two trees at 100 feet. Often a chunk of wire was thrown on the floor or perhaps strung around a room.

Many hams world wide have taken up the challenge to replicate this amazing radio and the internet abounds with many replica examples of the Paraset.

I tried my hand at building just the transmitter section and found it to be somewhat underwhelming. I also did some preliminary work on the Dial Drive Mechanism which was friction drive controlling the main receiver tuning capacitor

73's

Pete N6QW

Bitx40: ~ Replacing the Raduino with the N6QW controller!

Say What: Replacing the Bitx40 Raduino?

12-07-2019 ~ Pearl Harbor Day

Let us not forget what happened 78 years ago today. On a sad note there are but three remaining survivors of the original USS Arizona crew and only one will be attending the commemoration ceremony.

Many have sacrificed to make this the Land of the Free and the Home of the Brave! 

*****

I have now added switching of an linear amplifier to the PTT circuit so I can run 600 Watts with my "reworked" Bitx40.  Two 1N4148 diodes and a 12 VDC SPST relay make it all happen. Running a bit more power does help.

My latest project now on my website...

http://www.n6qw.com/The_Paesano.html

12-04-2019:

I call my Bitx40 "Maggie May" since she has been passed around a bit and before transmitting I specifically looked at the Bitx40 documentation. Shazam -- The bias was at Zero and not set to 100 Ma, and there was no drive as such. Fixing those two provided 10 Watts of solid power output on 40M. I cranked it back a bit since it would see some heavy duty cycle operation on WSPR. 

My offer stands: any one wanting to dump their Bitx40 -- a crisp $10 Bill and shipping costs from a US address is my offer. C'mon guys rid yourself of that former play toy and let N6QW work his magic. 

Send me an email to the address on the masthead if you would like the sketch code. 

N6QW, Pete the Radio Genius

12-03-2019 ~ Digital Modes with the Bitx40 and the N6QW Controller.

Raspberry Pi3 and (not seen) N6QW Digi Controller being used with the Bitx40 and N6QW's Raduino Replacement.

A couple of key comments 

1. This shows I must be close on the USB offset as both FT-8 and the WSPR were dialed in on their normal frequencies
2. I did not transmit on either WSPR or FT-8 as you must modify the Bitx40 Microphone input circuit. The Bitx40 uses an electret microphone which typically has a DC bias voltage on the input pin. As I did on The Paesano, I added a Data Port -which is nothing more than a 3.5 mm stereo jack that has a 10 NF cap in series with the microphone port to provide DC Isolation. That has yet to be done. Works perfectly on transmit see last chart.
3. So for those of you who have "put away" your Bitx40, with a few changes you can do "The Digi Dance" with your rig.
4. My offer is still open if you would like to dump your Bitx40 -- a crisp $10 bill plus shipping and I will take it off your hands.
5. 

   WSPR on the Bitx40 with N6QW Arduino Controller.
6. 

   

   
   
7. After three hours both transmitting and receiving
8. 

The Bitx40 being heard In Brazil and Sout Africa! Nice!

Stay tuned to this space for more exciting Bitx40 Modifications -- may be even 6 Meter SSB.

Pete, Radio Genius

73's

Pete N6QW

12-02-2019

Is this like blasphemy, heresy or on par with the Ukrainian Bribery scandal? No, it is none of those just a practical matter. 

In a recent Soldersmoke Podcast #215 the fact was made known that the Bitx40 was no longer in production. I then said if anyone wants to sell me one for about $10 contact me. That offer is still on the table!

Well my good friend N2CQR sent me one that had been passed around a bit (you know like Maggie May and the high school football team). Well when I connected all I soon discovered that Raduino was not providing any RF. The pot tuning also left me wondering even if it did work would I be happy.

So I took some parts, took some software and when you know stuff you can do stuff!

I started 1st with a 5 MHz PTO from a Drake TR-7 and that was to quickly establish that the main circuit board worked and was the Raduino the only problem. That was accomplished. 

Next was to get a basic Nano and Si5351 to produce the LO signal which was also to prove that my LO was correct and that a first stab at the BFO frequency was at least in the ball park. Keep in mind that not all Bitx40's are exactly alike. I think initially depending on Uma's crystal sorting there could be as much as 3 BFO frequencies.

Finally after passing the LO test I added the BFO. This test proved that the LSB BFO is very close but I may need to tweak the USB -- next phase. To introduce the BFO signal (since the LO is done through a two pin header) I had to uncouple the crystal and inject the BFO into the base of the former oscillator transistor. I am using a custom made for Pete Si5351 PLL's that has isolating caps on the outputs built into the board. Your may not so include a 10 NF on your output other wise you may smoke the transistor or your Si5351. Add the BFO signal at the junction of R102, C104.

Recapping my Raduino Replacement has 2 VFO's, USB/LSB and a Tune Tone. Did I mention the cool Green LCD?

The long term plan is a 6 Meter SSB transceiver using an outboard Transverter. Stay tuned and try to catch up.

Pete N6QW

A letter sent to DX Engineering

Are we being manipulated by the Ham Radio Manufacturers?

Here is a letter that was sent to DX Engineering?

If after reading this you have similar views --why not drop K3LR a note at the DX Engineering website "Contact Us" tab. I am sure he would love to hear from others in the ham homebrew community who like me see $4K for a Appliance Box as being out of step with our true ham roots which are now over 100 years old!

10/26/2019 ~ PPS (Pre Post Script). 

When you have the innovative and experimental mindset, then it is possible to add some new technology to fix problems with older radios. One does not learn that by knowing the 3000 menus in the IC7300. A Ten Tec Model 540 had a completely broken dial cord mechanism. Yes. radios used to have dial cord mechanisms. 

The message here is "when you know stuff, you can do stuff"! Working Contests and Operating may not give you the skill sets to pull this off. Watch the video...

October 23, 2019

Dear Tim, K3LR

Thank you for sending me the latest DX Engineering catalog which clearly displays radios and equipment that are simply not affordable for an OT like me who lives on a fixed retirement income.  Spending $4K for a radio is simply not in my budget; but that does not mean I am not on the air with some very modern equipment!

Your lead piece on page two sums it up quite nicely –endless possibilities. Recently I “homebrewed” a fully functional SDR transceiver for about $150. The reason this is possible is that there is so much amazing and cheap technology that makes this SDR rig a reality. Notice I didn’t say radio as that has implications of an appliance box! 

For a computer I am using a $35 Raspberry Pi3, the Sound Card was $15, and the USB Frequency Controller was $30. A lot of the critical hardware was obtained as free engineering samples and includes two ADE-1’s. Yes, it’s on a bread board.

The display can be an older LED TV that has an HDMI port and of course a keyboard and mouse. The driver stage is a 2N2219 and the final is the IRF510. The SDR Software is a free download from N2ADR known as Quisk.

Above is the typical Quisk spectrum and waterfall display just like the Big Boys with the $5K radios; but mine cost me $150

.

This now leads me to the remainder of your piece about Elmer’s. Seem like things have really shifted over the 60 years I have been licensed. Today’s Elmer seems to serve the function of what is the best antenna to purchase from your company, or how to navigate the 3000+ menus on the IC7300. In my day Elmer’s were the technical resource for knowing each and every component in a circuit and how it is applied, and the “why” it is there. Oh, that’s right now you send in a box top and get an extra class license and then spend thousands for a rig you have no clue how to operate.

It is with interest that your Elmer goals seem to follow what to buy, how to operate in contests and a friendly voice that says buy a FTdx101D. No where do I see a fostering working with and/or building your own radio using the latest technology. I know you have done marketing analysis to identify the demographic with the most $$$ and the new crop of rigs are priced accordingly. 

In a recent vintage ARRL Bulletin it was a position of the IARU that hams are hams for two reasons: Contests and Operating. Certainly, that is good for your business; but that position ignores what hams of my vintage did and do today. They were innovators and experimenters. Just look at Gerald Youngblood who took the concept of SDR and now he along with Raytheon are developing the next generation of communications systems (SDR Based) for the US Air Force. Suppose if K5SDR only wanted to operate and do contests? Side note, Raytheon used to be a ham equipment manufacturer. If you have ever happened to look at the nameplate on a late Model of the famous SBE transceivers – yes a unit of Raytheon.

My real point is that you might want to consider offering more “homebrew rig” kits so that we continue to foster the innovation. Besides you might actually hook a few more customers for the big box appliance stuff.

Now below is an example of where I built a solid-state version of the Collins KWM-2 which I call the KWM-4. Not every one could pull this off – but here is another endless possibility.

This is a multiband, dual conversion (10.7 MHz and 455 kHz) all solid state QRP rig. It involved some critical thinking and with the exception of the remote keypad and the Digi VFO the whole design and construction was done by me. As you can well guess I spend lots of time on the fabrication side, little time operating and no time contesting.

Since DX Engineering has become such a Big Gun in our hobby and regrettably the ARRL only wants to sell advertising for unaffordable radios, perhaps this is a time to be a true Elmer and foster and reincarnate the original “homebrew” spirit of our hobby. I invite you to visit my QRZ.com webpage and my three websites. www.jessystems.com , www.n6qw.com and www.n6qwradiogenius.us. Hopefully you will see I have built more than one radio.

73’s

Pete N6QW

A Bit Of Wizardry -- When You Know Stuff, You Can Do Stuff. My Extra call goes back to 1977 in case any one wondered.

There is just Something about Innovative Mechanical Engineering

Mechanical Engineering in our Radios, Rigs and RADIGS

Being hams, we often think about the electronic hardware in our rigs. Does it have an IRF510 in the Output or a Real RF Device like a RD06HHF1? Is it being "Run" with a digital VFO or one of those drifty, shifty Analog boxes.

BUT we often forget that some of these beloved radios rely heavily on our brother Mechanical Engineers to make it all play. You only need to look at James Millen who did all of the ME on the HRO receivers. Or pop the hood on that Collins R390A and savor the mechanical tuning mechanisms. Perhaps you overlooked the same in your KWM-2 or Drake TR-4. The ganged tuning of various networks is a marvel. Hey don't forget the SBE-33/34 and the "Geneva" Turret tuning arrangement.

But there is a special place in my shack for those really "cool guys" at Ten Tec. If you perhaps have some earlier Ten Tec radios like the 505, 509 or 540 that string driven dial system was pure Mechanical Engineering as was their PTO's. Can you imagine -- you could rebuild your PTO right at your kitchen table -- pretty clever.

But at times some of the Mechanical Engineering was hidden behind the panel and we simply used it without giving it a second thought. The Model 540 has a band switch located in the lower left hand corner of the radio. Actually it is THREE band switches with two visible on the front panel. When you put the main band switch on 10 meters you have an additional 4 position switch that is engaged to give you four band segments. But there is a third hidden band switch that is engaged as you change bands. Here is that Mechanical Engineering stuff at work

Before you gulp -- this is a Model 540 that was upgraded by me to a Model 544 and was the subject of a QRP Quarterly article in 2013. 

When we think about the Model 540 and the principal band switch you must think about what that switch is really doing. It has many jobs/functions. 

The way the frequency scheme of the  Model 540 works is to have a singular fixed VFO range operating at 5 to 5.5 Megahertz --yes you guessed it a 9 MHz IF. But to work the various bands this basic VFO signal is mixed with various crystal frequencies to produce the LO Signals that results in the 9 MHz IF --it is a single conversion radio.

Now we have all seen the 9 MHz IF and 5 MHz VFO that gives you two bands -- add the IF and VFO and it is 20 Meters. Subtract the VFO from the IF and you have 80 Meters. But you do have to pay attention to any possibility of sideband inversion.

But Ten Tec used  for all bands except 20 Meters a VFO signal that is crystal mixed to give an injection frequency above the incoming signal on 80/40 Meters and then for 20 Meters and above the LO injection is below the incoming with no crystal mixing on 20 Meters.

So this mysterious hidden band switch is THE mechanism that "on's" the crystal mixing for the various bands. 

Another Mechanical Engineering problem -- packaging. The Ten Tec Model 540 and Model 544 were very "desk top" friendly having a very small compact footprint. Thus a very long shaft could engage many band switch segments to engage multi-function circuits. But you are faced with the trade off of long band switch size and compact footprint.

Another function of that band switch was to switch in various LOW PASS Filter Networks following the final amplifier "brick". Along the way other circuits are switched into play with changing of the band switch.

Did I mention yet another Mechanical Engineering problem -- circuit shielding and isolation. That crystal oscillator and mixer circuit were shielded--yes in a shielded box there is also a MC1496 DBM that mixes the Crystal Frequency with the VFO to produce the proper injection frequency. Another wafer on that third band switch adds some filtering to the output of the driver stage ahead of the final amplifier brick. Believe it or not -- that wafer is mounted on the back side of the shielded box, right adjacent to the driver board.

Oh more ME stuff -- all of the major circuit boards have through board pin connections that on the bottom end plug into chassis mounted sockets and actually protrude through the top side -- this give you access to literally hundreds of test points. Smart Mechanical Engineering!

Back to our third hidden band switch. So if this switch is not on the same shaft as the main band switch how is it engaged. Enter a piece of brilliant Mechanical Engineering which I call "a slider crank mechanism". Two mechanical linkages affixed to the main tuning shaft essentially clock the third band switch so that when you place the main band switch on 40 Meters the Crystal Oscillator puts in line a 11 MHz crystal so the output when mixed with the 5 MHz VFO output a 16 MHz injection signal results. When that injection signal is mixed with the incoming signal at 7 MHz the result is 9 MHz at the IF. 

BTW if you look at the Model 540 BFO, a single BFO frequency will result in the "Normal" sidebands for all of the ham bands and with a bit of capacitor padding will shift the BFO frequency to the "Reverse" sideband. Ten Tec does not call them USB and LSB. I keep digressing

Here is a photo of that slider crank mechanism. Essentially what you have is two pieces of metal coat hanger with a loop formed at each end. These loops fit around a drum like spool assembly so that as you clock the band switch the coat hanger material is free to move around the spool; but the push pull action moves the third band switch in unison with the main band switch. If you don't know that the spool assembly comes apart and you somehow get the metal, bars out of the spool -- they simply cannot be fed back in place without removing the cap

The reason I mention this is that I have a project on the bench where the "slider crank mechanism" was in shambles, the dial cord was chewed up and it was obvious some non-engineer maybe one of those BTE's tried to "fix" something without knowing what they were doing. Pretty obvious who they voted for in the last election.

The metal bars were pretty bent up and I did a bit of straightening. The interesting engineering aspects of the nylon assemblies that fit on the two shafts they are molded so that the fit only one way on the half moon shafts of the two band switches and the bonus --there are really small set screws that further lock the nylon pieces to the two shafts. 

Good thing I have acquired many small Allen wrenches. BTW the same Allen wrench that works with the main tuning knob will work with these set screws on the nylon assemblies. This is a long term project and you will be amazed at the final product.

Pete N6QW