Between getting ready for the Holidays and a couple things i need to do around the house, I have not been spending much time working on the BITX. Except for the drift issues I still have not been able to get resolved, I am very happy with the performance so far. The simple AGC seems to be working well, and makes listening much more comfortable. I took a look at the output of the detector on the AGC board, and it looks like it will be usable for a signal strength meter.
With the drift problems, and desire to add a signal strength meter I decided to go ahead and build up a digital VFO. I have a 9850 module left from the SNA Jr. and a couple Adafruit SI5351 modules from earlier projects. I had ordered 2 of the BITX boards, and had planned on making the second one into a multi-band rig. With this in mind I decided to go with the SI5351, because I could use a second clock in the multi-band version for side band selection.
Over a year ago I had built a 5351 based VFO for use with Frog Sounds 40 Meter CW transceiver. I called it the Canned Frog, because I put it in a canned meat can.
http://kv4qb.blogspot.com/2015/07/canned-frog-part-two.html
I had used a separate clock for the receiver and transmitter LOs, so had about everything I needed in the board I had laid out for it. This VFO used a Arduino Nano, and a small OLED display. This will be fine for the single band version, but want a larger display when I get around the the multi-band version. I used the basic layout I had, and brought out most of the Arduino pins for use with a different display, and other options I am thinking of adding. When I did the original VFO. I found the regulator on the Nano got quite warm when running on 12 volts. So, I added a 9 volt regulator to the board to keep the Nano cooler.
I also changed most of the passive components on the board from leaded to SMD versions, that will be mounted on the back. I have started to try to use SMD components where ever possible. Drilling holes is the most tedious problem of making my circuit boards. Only problem with home made PCBs and SMD components is watching out for solder bridges. I have been playing around with UV curable solder mask, and have gotten the process down to something fairly easy with consistent results. So after etching and tinning the board, I added a solder mask before drilling. Everything looks very nice, now to build the board and modify the existing sketch to take care of a IF offset.
12/18/2016 board and software mostly finished
Finished getting most of the board populated except for the filter and attenuator circuits, and the basic software working. Now I can try setting different levels and filter values on the board to see what values work best. The software is basically what I had used for the 'Canned Frog', with the CW keyer and CW offsets, and RIT removed. Since this used a direct conversion receiver, the signal output was on the transmit frequency. The BITX is a super-hetrodyne, and has a IF frequency around 12M Hz. Therefore the VFO signal must run at a different frequency. Because you are receiving the Lower Side Band the exact frequency used is the BFO frequency. Basically the VFO frequency is the BFO frequency minus the operating frequency.
For the 7 Mhz. band this is around 4.7 Mhz. to 5 Mhz. The original software used the actual operating frequency for display and setting the si5351. I measured the frequency of the BFO and used that value in computing the required VFO frequency. This value was used to set the si5351 clock 0 output.
After making the changes to the software, I ended up spending an hour or so trying to find out why the frequency was not changing when I turned the rotary encoder. I finally looked at signals with a scope, I found I had a bad rotary encoder. After changing that everything worked the way I expected it to.
Update 1/2/17
I spent
most of the holidays visiting relatives, and did not have much time to
work on any of the projects I am working on. I did bring my laptop
along and had some timeto work on a BITX front panel
that
will work with the SI5351 VFO and Display. The original front panel
was designed to work with a small counter and had a grill opening for a
small speaker. After trying the BITX receiver, I found that even with
an AGC circuit added, the audio sounded much better with a larger
external speaker. This version has a built in bezel for the display,
mounting posts for the circuit board.
I also had to include a recess to
provide room for the stacked Display and Adafruit SI5351 board. To
center the display, I had to move the microphone and volume control to
the opposite side of the panel.
Update 1/22/17
I 3D printed the new design front panel and tested the VFO out when connected to the BITX board. I found some of the same kind of receiver noise I had at first with the frequency counter. It looks like the BITX board is very sensitive to any noise on the DC supply line. I had a reverse polarity diode on the board, that I replaced with a 10MH. choke and added a 100 uF. capacitor across the input to the regulator on the board. With this I could not hear any noise unless I turn the volume all the way up and disconnect the antenna.
The recess I put in the back of the front panel helps keep the display and si5351boards in position, along with reducing the overall depth in the case.
Just a little more work on the software to make sure the RIT is working correctly and possibly add a S-Meter
12/2/16 Added link to Eagle files for making AGC Board
8/1/17 I uploaded the eagle file to oshpark.com For those who want the boards they can be can be ordered directly $11.70 including shipping for 3 boards.
https://oshpark.com/shared_projects/6qRXvzZR
A couple of things came up and I was delayed in building Farhan's SMD BITX 40. With the availability and price point on the board, I decided I will stop work on the version I was designing and building. I will instead spend my time on trying some modifications to enhance the board from Farhan.
For those who are building with this board, there is a website devoted to different mods for it.
http://bitxhacks.blogspot.com
Since I haven't gotten around to build a si5351 VFO yet, I installed a 10 turn tuning pot. This made it much easier to tune, but I found there was quite a bit of drift during warm up. Looking over the bitxhacks blog, one of the first mods I made was to change some of the resistor values in the bidirectional amplifiers. Changing them from 100 to 220 ohms reduces current in those stages, should reduce heating of that part of the circuit board. I also coated the VFO coil with several coats of clear nail polish to prevent coil winding movement.
Another change I made was to cut the short trace going from switched 12 volt to the U2 (VFO 9 volt regulator) input pin. Then I ran a wire from the input pin of U2 to un-switched 12 volts. I also added a .1 uF capacitor from the regulator input pin to ground, and a 47uF and .1uF from the output to ground. This keeps the VFO circuit on at all times, and hopefully further reduces warm up drift.
After all of this it is much better, but I still have some drift, up and down in frequency even after warm up. Ordering some NPO capacitors to use in the VFO and will see if this helps.
With much of the drift problem solved, I next looked at the audio. Since the BITX does not have AGC, having to adjust audio level when going from station to station can be aggravating. Farhan posted a simple AGC circuit on the bitxhacks blog, and I decided to add it to mine.
I took his hand drawn schematic, and input it
to Eagle and laid out a simple board that can be built as mostly SMD or Muppet style.
After etching a couple boards, I built up one of each to give a try. For the SMD version I also added a solder mask to make assembly easier. Adding the AGC board to the BITX took care of leveling the audio, but the added circuitry reduced the audio level. Replacing the 1 uF. capacitor from pins 1 - 8 of the LM386 with a 10uF. brought the gain back up to where it was.
Link to Eagle files and a .pdf with mirrored top layer image for making toner transfer boards.
https://www.dropbox.com/sh/og5oifvqprldp7j/AACpE5Wqs1kmKePrfw825ONDa?dl=0
Next to add the frequency counter. With a little change to the gain on the counter pre-amp I had it reading correctly. Using the setup function on the counter, I measured the BFO frequency, and set the counter mode to subtract the VFO for proper frequency display. There was some noise introduced on the audio output. A 220uF capacitor across the counter power supply leads, and a 10mH. choke in series with the positive supply lead took care of that.
Except for a small amount of frequency drift the receiver portion of the BITX is working. Now to wire up the microphone and test the transmitter. It is starting to look like a transceiver.
I have been following Paul M0XPD "Shack Nasties" Blog for quite some time. This year at FDIM, Paul gave one of the seminar presentations. I was pleasantly surprised when he mentioned my early version of the SNA Jr. in his presentation. Later, I had a chance to speak with him about some of our common interests and projects. I also gave him one of the extra SNA Jr II boards I had had made.
A week or so ago I received an e-mail from him with a picture of the SNA Jr. that he had just finished building. Today I saw that he had updated his blog with a very nice write up about the SNA Jr II. I want to thank him for his kind words.
http://m0xpd.blogspot.com/2016/11/sna-junior.html
If you have not read his blog before, it is one you should follow. I know I always find something interesting and informative in each post.
I have been gone for a while visiting family, and have not had a chance to work on the projects I have on the bench. One of these is a SMD version of the BITX 40, that was only available in India. Just before I left, I saw that a newer SMD version was available for sale world wide. And, at only $45.00 including shipping I decided to order one. A couple of days after I returned home, I received a little box from India. Inside was a very nice assembled SMD BITX board, and a bag of components needed to build a working transceiver. Since the local QRP club was going to have a table at a hamfest next week, I decided to quickly build this up for something to display.
I want to make a few minor changes to the kit, first I wanted to use a 10 turn pot instead of the one included in the kit. Along with that I will add one of the re-packaged DL4YHF counters I designed a couple years ago. I also will use an external hand microphone with PTT instead of the electroet mic. that came with the kit.
Next was to come with a case to put it in. This is exactly what I bought the 3D printer for. I printed some 6" corner pieces I had designed right after I got the printer, and cut up some double sided circuit board material to use for the sides of the case. I have switched to using .032" PCB material for chassis
panels. It is very easy to cut accurate size pieces with a simple paper
cutter. The back panel is very simple just two mounting holes for the power and RF connector, and 4 screw holes for mounting the panel.
The front was a little more difficult. I needed mounting holes for the tuning pot, volume control,a separate power switch, along with a hole for the display. I also decided to add a grill opening for a small speaker to mount behind the front panel. I went with a panel .2" thick to give room to build in a bezel for the display, and allow enough depth to add a recess to hold the 7 segment display for the counter.
After a couple hours to lay out the design and a few more to do the printing I came up with a very nice looking case for the project. I cleaned up the printed parts, assembled the top and bottom of the case, and gave everything a coat of paint.
Now to put everything together tomorrow, and see how it works.
I have been exchanging e-mail with Caglar TA2UH , concerning compile problems with the SNA Jr. he is building. Only problem seems to have been the version of the 'rotary' library he had been using. After sending him the correct one, he now has his Junior working. . It still needs to be put in a box and some accessories built, but from the picture he sent me it looks like everything else should not be a problem.
I just copied the correct version of library to the SNA Jr version II dropbox folder to help eliminate this problem for builders.
https://www.dropbox.com/sh/kw7c14euqqi28pn/AABc384tePRDZBoqo6s4YDCRa?dl=0
12/3/16
Just got some pictures of Caglar's finished SNA Jr with some accessories. Looks great, a lot of work with some hand tools to get it fit into a cast aluminum box.
And another of it in use testing a crystal filter
ANOTHER ONE 10/14/2016
Just received an e-mail from Vincenzo IZ5GVP with some pictures of his version of the SNA Jr. He also included an Arduino sketch with some of the additions he has made to the software. I took a quick glance at it, and it looks very nice. I will take a longer look (with the help of Google Translate) and see what I want to add to my version of the sketch.
When I did the latest version of software, I was in a hurry to get everything ready for FDIM. I did not do much in the way of calibration, Vincenzo's sketch has some cal routines that I will defiantly have to look at.
And another one from John VK5COR, he made his own 2 layer board. He said it worked but didn't look very nice. then he had a couple boards made. Said they look a whole lot nicer. He is now working on a box to put it in.
It is very gratifying to receive e-mails with pictures of other peoples build of some of my projects. Sometimes when you write a blog, you wonder if you are just wasting your time. When you get pictures of other people building your projects, you know that what you have written is of interest to more than just yourself. So please if you have built any of my projects please send a picture and I will put them in a Dropbox folder for all to see.
https://www.dropbox.com/sh/u9axday3qj1lxsu/AADnBO1Txj-kDRSaof_lG3tza?dl=0
Update 3/4/17
Just found a Blog entry by Gerry with some pictures of his SNA Jr. build
https://gerryk.com/node/57
Update 3/30/17
Just received a em-mail from John KG9DK with a picture of his finished SNA Jr. I have been exchanging e-mails with John for over a year now about the SNA and other topics. Nice to see he has his SNA finished, really like the work he put into the design of his front panel.
Update 3/3/18
I updated the link in the dropbox. Also adding an additional link to ones that have been built.
Tony http://www.fishpool.org.uk/snajr.htm
I was adjusting the gain on several broadband amplifier stages using a noise generator and my AD8307 power meter. It worked well, but was a little inconvenient without a small test probe. A couple months ago some one had brought a RF probe from QRP Guys for show and tell at our local QRP club meeting. I liked the format, but I preferred the direct reading in dBm. that I have with my existing meter. I guess it was time to see about doing a repackage.
It was very easy to move the components in the layout, and change from an Arduino Nano to a Pro-mini. Switching to the Pro Mini allowed me to stack the display directly over the Arduino. This reduced the width of the board, and gave me a nice probe like form factor. I etched up a board and assembled the new format meter. I had one small error in the board, but a simple jumper took care of that. After changing some pin assignments in the sketch, I had everything up and running. Now it was time to see about packaging it.
I have been playing with the CAD software I use with the 3D printer. I tried several ways to make a housing with a top and bottom piece, but couldn't get one to print that held together tightly.
 |
| Probe Cross Section |
I had made a set of corner pieces for a cabinet that had in a built in PCB holder that had worked well. So I did a design that was basically a rectangular tube with built in slots to hold the PCB. display board and centering for the display itself. It also kept the 9 volt battery from moving around. I printed a long tube with this cross section, and checked that everything would slide in easily. It looked OK, but I wanted it to look more like a probe. 
With a little cutting and slicing of the design I had a nice looking probe housing with cutouts for the display and on/off switch. Simple pieces closed up the open end and provided a removable cover for the 9V battery compartment.
I kept the SMA connector for the input to the meter. That allows me to use a cable to check output levels of modules with connectors, or install a probe tip for in-circuit measurements. One thing I decided to add to the software was to display the difference between the last peak and current reading. This allows you to get the stage output level, then go back and take the stage input reading. The displayed difference should be the stage gain/loss in dB. I took some readings across several known value in line attenuators, and found it very easy to measure stage gain/loss.
I also changed most of the passive components on the board from leaded to SMD versions, that will be mounted on the back. I have started to try to use SMD components where ever possible. Drilling holes is the most tedious problem of making my circuit boards. Only problem with home made PCBs and SMD components is watching out for solder bridges. I have been playing around with UV curable solder mask, and have gotten the process down to something fairly easy with consistent results. So after etching and tinning the board, I added a solder mask before drilling. Everything looks very nice, now to build the board and modify the existing sketch to take care of a IF offset.
Finished getting most of the board populated except for the filter and attenuator circuits, and the basic software working. Now I can try setting different levels and filter values on the board to see what values work best. The software is basically what I had used for the 'Canned Frog', with the CW keyer and CW offsets, and RIT removed. Since this used a direct conversion receiver, the signal output was on the transmit frequency. The BITX is a super-hetrodyne, and has a IF frequency around 12M Hz. Therefore the VFO signal must run at a different frequency. Because you are receiving the Lower Side Band the exact frequency used is the BFO frequency. Basically the VFO frequency is the BFO frequency minus the operating frequency. 
For the 7 Mhz. band this is around 4.7 Mhz. to 5 Mhz. The original software used the actual operating frequency for display and setting the si5351. I measured the frequency of the BFO and used that value in computing the required VFO frequency. This value was used to set the si5351 clock 0 output.
that
will work with the SI5351 VFO and Display. The original front panel
was designed to work with a small counter and had a grill opening for a
small speaker. After trying the BITX receiver, I found that even with
an AGC circuit added, the audio sounded much better with a larger
external speaker. This version has a built in bezel for the display,
mounting posts for the circuit board. 
I also had to include a recess to
provide room for the stacked Display and Adafruit SI5351 board. To
center the display, I had to move the microphone and volume control to
the opposite side of the panel. 
I 3D printed the new design front panel and tested the VFO out when connected to the BITX board. I found some of the same kind of receiver noise I had at first with the frequency counter. It looks like the BITX board is very sensitive to any noise on the DC supply line. I had a reverse polarity diode on the board, that I replaced with a 10MH. choke and added a 100 uF. capacitor across the input to the regulator on the board. With this I could not hear any noise unless I turn the volume all the way up and disconnect the antenna. 
