Monday, April 2, 2018

Stand-alone Simulated Analog Meter

Now that I have the latest QQ article finished, I have a little more time for some of my other projects.  One of them is working on the graphic display for the uBITX, I came up with a simple simulated analog meter that worked quite well.  Only problem is with the Raduino, board there is only one available analog pin. Thinking about ways to get around this I looked at some of the inexpensive I2C ADC baords available on e-Bay.  That would do for the uBITX, but I also wanted something I could use with the BITX 40 and a couple of other projects I have been thinking about doing.  I started to wonder how hard it would be to build a stand alone simulated analog meter with several input channels and control lines to select what values are being read and displayed. I wanted it to be small, and simple to construct.  Looking around I found aYou-Tube video showing a Arduino Nano based VU meter using one of the small .96" OLED displays.  Since I had several of those displays' from another project I thought I would try something based on it.

 The video comments had links to the source code and instructions for building the meter.  I thought this would make the basis for a nice little meter that could be added to almost any transceiver.

Thinking about what values I would like to display, I came up with three basic items.  A S-meter when in receive, and a power output display when in transmit.  In transmit, I would also like to have the capability of measuring VSWR.  Thinking about the switching functions required for this I will need one control line that monitors  transmit/receive, this can come from the PTT or key line in the transceiver.  Then  I use a second control line to select either power or VSWR when the T/R line is in transmit.  Another control line can do the same for the S-meter or some other display when in receive.  Since this is based on a VU meter, I will use that for the secondary function in receive.  Now looking at the signal lines I need to measure, they are the AGC line for S-meter, audio signal for VU meter.  And in transmit, the forward and reverse power levels will take care of power and a computed VSWR reading.

I can easily get the 4 analog and digital lines I need from my favorite Arduino for embedding in systems, the Pro-mini.  I like it because of its smaller size since it does not have the USB interface built in like the Nano.  This also means that it takes a little less current than the Nano.  Looking at the physical layout of pins on the Pro-mini, I can get 4 digital, and 4 analog pins in a row that I can use for my control and input signals. 

 I have some pre-made 8 wire cables with a connector, and matching jack that fits the pin spacing on the Pro-mini.  And since I only need 4 wires for I2C OLED display, it was easy to solder a short cable directly from the Pro-mini to a 4 pin connector for the display.
I will be able to supply 5 volts to power everything through the programming pins at the end of the Pro-mini.  If needed I can build a small board with a 5 v regulator and possibly some filtering, so I can run everything off the 12 volts in the transceiver.

After everything was wired up, I loaded the VU-Meter sketch and was very happy with the response of the meter.  Since the display is basically monochrome, it only takes a 1 kilobyte image file to completely redraw the screen.  This allows for a very fast update, and a realistic looking needle movement. The OLED display I had has a two color effect, that is created by two different colors of plastic in its construction.  I have ordered several additional displays that have only the blue overlay. Following the procedure mentioned in the original article, I drew a set of the meter faces I would need for the basic functions of the meter.
 For the power meter function, I decided on a maximum reading of about 15 watts.  The uBITX is rated at about 10 watts on 12 volts, and a little bit more if a higher supply voltage is used on the final amplifier stage.
 Not sure what I plan on using for the S-meter, for now I will probably just rectify the audio output of the preamplifier stage and read that directly.  Later I will see about some sort of an AGC circuit that I can use for the S-meter reading.   The display function I wrote for all except the original VU reading, have provisions for using simple constants to set the full scale movement of the needle, and any DC offset in the circuit at the being measured.

The VSWR display goes up to about 1:3.8 which should be more than adequate for most use.  I think I can use the same directional coupler, without its amplifier that I have in my QRP power/SWR meter. Since I am measuring both forward and reverse power, I can compute the VSWR without having to worry about the power output. 

As far as the signal levels being measured, I can take anything up to 5 volts without having to add in a adjustment pot or voltage divider.  The software should be able to manage full scale and offset values with simple defined constants.

I still have one extra control signal I can use for selecting the display function.  Now what else can I do with this display? Possibly an audio frequency spectrum display, so I can join those on 40M with their SDR rigs that are quick to let everyone know exactly how too wide your signal is.  Lets see what I can come up with, stay tuned.

UPDATE  4/4/18
Finished up the software to select the desired display screen, and modified one of the Audio spectrum analyzer sketches I found on line to fit in the program space available.  It works, but after playing with it, I don't know how useful it would be.  With the memory available, it is only possible to do a FFT that will cover to about 5 KHz. , and I would probably have to add a preamp to get enough voltage for a usable display.  
But It works and feeding a fairly large tone in from a signal generator, I could get a reliable display of the spectrum of that tone.  Trying it on a normal SSB signal did not give anything I found to be usable.  I guess if you were doing CW and had a narrow audio filter in line, you could use the FFT display to see what other signals are around you.  Here is a photo of the response with a 1200Hz. signal from a signal generator. 

Monday, March 5, 2018

Sweeprino Jr revisited Update 3/16/18

Well after three weeks of feeling terrible and not having the ambition to do anything, I finally got over what ever version of the 'flu' I had.  This has put me quite a bit behind in working on the next article in the series of Arduino building blocks for QQ.  Over a year ago I had started on a version of Farhan's Sweeperino project. at  I knew this would make a great basis for the next article

Since this was  to be a stand alone instrument or connect to a PC to use Farhan's  'Specan.exe' software, I made some changes to my Arduino/display building block.  I changed from a pro-mini to a Arduino Nano.  This gave me the USB interface to a PC, brought out the I2C lines that do not have dedicated pins on the pro-mini.  The Nano also has a heavier duty voltage regulator on the 5 volt line.  This will allow me to run a Adafruit SI5351 board with out having to add additional voltage regulation. I also added a rotary encoder with push button to use as the interface to the software.

Doing the board layout for these changes was very easy.  Mainly because except for the form factor, it is very near to what I did to make the stand alone VFO/BFO board for Pete's Simpleceiver project.  Instead of making my own boards this time, I took advantage of a promotional offer by one of the Chinese board houses.  For a limited time they had offered 10 boards for a little over $5 and free shipping. 
 I couldn't even buy the raw pcb material for that price.  After they arrived and I built one up, I found a small error I had made in the board layout.  One end of a resistor was grounded that should not have been.  After discovering the error, about 30 seconds with a hobby knife took care of the problem.  Also modified the 3D printed case I had designed for earlier projects in the series to fit the new board

I pulled a copy of Farhans sketch, and made changes to use a tft display instead of the 2 line LCD he had in his original unit.  The first cut of the software worked with the PC interface, and a simulated two line  LCD display on the TFT display.  

After making sure everything was working I decided to use the graphic capabilities of the TFT display.  In previous projects, I had used simple bar graphs to give a analog representation of values, to make tuning easier than just a digital value.  I was real happy with the simulated analog meter I came up with for the uBITX, so I incorporated it in the new display screen for the manual operation mode.   I made a few changes to the basic code  so I could also display the actual value as a number along with the analog meter.    Still trying to decide if I want to add a sticky needle in a slightly different color to indicate the peak reading for the last few seconds.  Will only take a couple lines of code, but not sure if it will be necessary.  For the frequency setting display, I decided to show the digit that is changed by the rotary encoder in a different color than the rest of the value.  So a short push on the encoder button moves the active digit around in a circular manor.

Need to do a little more checking of the software so far. Then the only other thing I want to add is a graphic sweep waveform very similar to what I have on my SNA Jr.  Just going to be very basic, so everyone can add any additional functions they want.

Will keep you informed, but I also have to hurry up and write the article for QQ to have it ready for the next issue.  Maybe this time they can get the author correct.

Update 3/16/2018

Finally finished the QQ article on the Sweeperino Jr so far.  I decided to make this two articles instead of just one.  The first one will have the Sweeperino Jr with the same basic functionality as the original.  In the next article, I will add a graphical sweep function similar to the SNA Jr.  I have the latest Eagle file Gerber files , and Arduino sketch in a dropbox folder for any one interested in building one of these.

It works quite well, if you understand that there are going to be some issues with harmonics because of the square-wave output.  I am working on a method to get rid of that problem, and mostly finished with the preliminary design for the SNA Jr. version III.  Hope to get around to finishing that in the near future.

Saturday, January 27, 2018

SWR / Power Meter QQ article

I caught a case of the 'flu' and for the last week have been just sitting around without the energy to do anything.  About the only thing I accomplished was to open the envelope with the latest issue of QQ.  I saw that there has been a mix up and my second article on Arduino building blocks was credited to Glen Popiel KW5GP.  Since reading Glen's book on Arduino projects for the ham was one of my inspirations for learning the Arduino, I didn't  feel too bad.  And, if every one hates the article, Glen gets the blame not me.
As with the last article I have build notes for those interested in a dropbox folder along with the Eagle CAD files, Gerbers and Arduino sketch  at
Once I get to feeling a little bit better, I will update that folder.  For now it is time to take another batch of cold/flu medicine and pretend to be a vegetable for a little while longer.

Thursday, December 28, 2017

uBITX TFT display update 1/10/2018

I had a little time over Christmas weekend while everyone else was watching football to work on the TFT display for the uBITX.  I grabbed a 2.4" TFT display from another project and connected it to the Nano I am using as a test bed.  Just to be sure about the logic levels the display wanted I fed the control lines through a voltage divider made up of a 4K7 and 10K resistors.  This display gave me a much larger area to work with, and I could add some of the features that several other people suggested.  One of them was the display of the A: and B: VFO frequencies.   I decided to display the active VFO frequency in a larger font, with the alternate in a smaller font and a different color. 

 Another suggestion I uses was, when the RIT function was active, I displayed the offset from the transmit frequency.  I also changed the RIT tuning steps to 50 Hz. instead of 100.  I still have several other settings I need to display on the screen, but it is starting to take shape.

One other thing I wanted to do was add was an analog meter.  Mainly to use as a s-meter, but possibly as a power meter that will switch between modes on receive or transmit.  I looked at and tried the code from several versions I found, but they were fairly large in code size and required quite a bit of processing to compute the needle movement.  I decided I wanted top try something a little simpler.  Instead of using sin and cos functions to compute the angular position of the needle, I would treat  everything as a straight line value.  This lets me just draw a line from a pivot point for the needle to some point near the top of the meter face.  The line length would vary, but the effect would look close enough to an actual meter.  This resulted in two very simple functions, one to draw the meter itself and the other to plot the needle.  To move the needle, first it is drawn in the face color of the meter to erase the old value and then redrawn in the desired needle color.
The function to draw the meter is very simple 

void drawMeter(int x, int y, int w, int h, int face_c, int bar_c, char *txt) {
  tft.fillRoundRect(x, y, w, h, 5, face_c);
  tft.fillRoundRect(x + 1, y + 1, w - 2, h / 5, 3, bar_c);
  tft.drawRoundRect(x, y, w, h, 5, YELLOW);
  tft.setCursor(x + 2, (y + (h/5))-12);

Draw a  filled rectangle with rounded corners at the desired position and size, then draw another that will serve as the background color for the meter scale.  I also added another empty rectangle around the main one to highlight the meter.  Then position the cursor and print the meter scale.  I did not add auto sizing of the meter, so the actual size of the main rectangle, will have to be determined manually from the desired text on the meter scale.  To draw the basic meter without needle is to give it a position, size , colors needed, and scale text.

// define some values for the meter position and size
#define  MTR_X   160
#define  MTR_Y   108
#define  MTR_W   155

#define  MTR_H   100

drawMeter(MTR_X, MTR_Y, MTR_W, MTR_H, WHITE, RED,"0  2  4  6  9 +20 +40 +60");

Plotting the needle was simpler than I had expected.  Since I was treating the meter range as a straight line value.  I pass the function the same parameters for position and size as the drawMeter function, and add the value, max_value and color to draw the needle.  
void plotNeedle(int x, int y, int w, int h, int value,int max_value,int color)
  int tw;
  int w2 = w/2;
  tw = map(value, 0, max_value, 3, w-3 );
  tft.drawLine(x + w2 ,   y + h - 4,  x+tw ,    y + (h / 5)+4, color);
  tft.drawLine(x + w2+1,  y + h - 4,  x+tw+1  ,  y + (h / 5)+4, color);
  tft.drawLine(x + w2+2,  y + h - 4,  x+tw +2 , y + (h / 5)+4, color);
Then I used the Arduino map() function to return a value that represented the position along the meter scale based on the value read and the maximum value.  Then just plot a straight line from the pivot point for the needle to just below the meter scale, which I compute depending on scale size.  I draw this three times with a slight offset to make the needle more visible as it moves. 

Since the only available Analog pin is A7, wrote a little code to read that value and display the meter reading. 
  plotNeedle(MTR_X, MTR_Y, MTR_W, MTR_H,last_meter,1023,WHITE);
  plotNeedle(MTR_X, MTR_Y, MTR_W, 1MTR_H,meter_reading,1023,BLACK);

Next I will see about building something to get a signal from the uBITX to display.  I might start with a simple voltage doubler off the top of the volume control just to see how well it works.  Later I will probably tap into the IF chain to get an actual S-Meter .  Also thinking about adding a pickup off the antenna lead to measure power output, I can probably use the PTT signal to allow switching between meter readings from receive to transmit.

Update 1/10/2018

After playing around with several different versions of the display screen, I have come up with something I like.  Following a couple of suggestions from N6QW, I now show options available in Cyan and any active settings in Yellow.  On the screen you can see the second VFO frequency, and that RIT is optional.   

Below the frequency read out, I have an area to display the command prompts from the uBITX, such as band change, VFO selection, RIT on/off. 

When RIT is turned on the offset from transmit frequency is displayed. I did make a change to the original code to change in 50 Hz. steps instead of 100 Hz.  

Finally I added a transmit indicator, and also checked the CW keying line and changed the mode display while a keying is present.

I am happy with the way the display looks, now to put everything in a box and see how it works.  I think I will re-write the code I have for the BITX40 and use one of the VFO/BFO boards I designed.  With that I have more Analog input pins available and should be able to implement power and SWR readings on the screen.

UPDATE 1/29/18 

I have received several questions about the display I used on my uBIX.  The ones that I am most interested in are the ones that use a  ILI9340 or ILI9341 controller.   They are available in 2.4" or 2.8 " models.  I have found the response to be very fast using the standard SPI interface.  

The main issue with them is that they usually require 3.3 volt logic levels instead of the 5 volt from Arduino, I have found a simple voltage divider does well in this application.  I usually power the LED pin through a 100 ohm resistor to 5V.  If a PWM pin was available, I would add backlight intensity control instead.  Software changes in the main sketch  ubitx_20 were fairly simple, and then I replaced the ubitx_ui  file with a modified ubitx_tft_ui file.  Changes to the ubitx_20 file are made in the declarations 

Comment out the original display driver  and replace with the Adafruit drivers
#include <LiquidCrystal.h>
  LiquidCrystal lcd(8,9,10,11,12,13);


#include <Adafruit_GFX.h>
#include "Adafruit_ILI9340.h"
#define _sclk 13
#define _miso 12
#define _mosi 11
#define _cs 8
#define _dc 10
#define _rst 9

#define CALLSIGN  "KV4QB"

// Color definitions
#define BLACK   0x0000
#define BLUE    0x001F
#define GRAY    0x3333
#define RED     0xF800
#define GREEN   0x07E0
#define CYAN    0x07FF
#define MAGENTA 0xF81F
#define YELLOW  0xFFE0
#define WHITE   0xFFF6

Adafruit_ILI9340 tft = Adafruit_ILI9340(_cs, _dc, _rst);

and then in the setup code make these changes
  // lcd.begin(16, 2);   // using tft instead of lcd

  // setup the tft display and draw basic display screen
  //tft.initR(INITR_BLACKTAB);  //ST7735 depends on exact one used
  tft.setRotation(1); // landsacape versus portrait


  //we print this line so this shows up even if the raduino
  //crashes later in the code

  printLine1("uBITX v0.20");

  tft.setTextColor(YELLOW, BLUE);
  tft.setCursor(22, 20);
  tft.print("uBITX v0.20");
  tft.setCursor(22, 20);
  tft.print("           ");

With the updated ubitx_tft_ui code I tried to keep the code in the other files as unchanged as possible.

Sunday, December 17, 2017

Santa Came Early - uBITX update 12/20/17/.........

I had not realized that Santa traded in his Sleigh for a yellow cargo van, but he must have.  After seeing that the new uBitx was available for sale last Saturday, I thought about ordering it.  Then about 2 AM Sunday morning I couldn't resist much more and pushed the button at Paypal.  

Just sitting around Thursday afternoon when the door bell rang, and there disguised as a DHL deliver driver was Santa.  I couldn't wait to see what the new toy looked like.  So after letting Santa get on his way to his next delivery, I tore into the box.  Inside the sturdy cardboard box was a plastic box with a couple of bubble wrapped boards, and a bag of parts. The uBitx board and Raduino are each wrapped in bubble wrap, and I could not see any shipping damage on either one. 

 I plan on replacing the 2 x 16 LCD display with a small TFT display, so I took a look at the wiring diagram of the display connector.  I checked the pins and they contain all pins to use hardware SPI on the Arduino to drive a TFT display.  I wired up a 1.8" display I had from other projects, and could not get it working with the pin combination I choose.  Playing around with some other wiring combinations, I was able to get it working.  Still wondering about that, because I have always been able to map anything but 11 and 13 to anything I wanted to use.  I realized that pin 12 is also used by the hardware SPI, and after changing to pin 10 everything worked as expected.  First thing I had to do was disable the LCD driver software and include the Adafruit libraries for the ST7735 and the graphics routines.  After that I replaced the LCD display routines with ones written for TFT display.  Luckily the uBITX routines map everything related to the display to a couple of simple functions.  It took very little to change these over to the TFT display.  
The only problem I found was that with the 1.8" display I only had a x resolution of 160 pixels, and that limited me to using a small size font, without the text wrapping around to the next line.  But everything should work after I redesign a display screen layout to replace the default two line layout used in the stock uBitx.  The 1.8" display could be wired in directly, but for a 2.4" or 2.8" I will need to converted from 5 to 3.3 volts on the control lines.  I have found that simple voltage dividers work well for this, as the data flow is only one way.  At last something to keep me busy for the holidays

UPDATE 12/20/17
To make things a little easier to work with on converting the uBITX display to a TFT display, I wired up a Nano, display, encoder, and a couple other parts on a wireless breadboard.  This will allow me to take everything along when I go to spend the holidays with relatives.  So far it has been fairly easy to come up with a screen layout that fits well on the 1.8" display.  I am trying to limit any code changes to the ubitx_ui module, and mainly to the updateDisplay() function.  This should make it possible to easily switch back to the LCD by just loading in  a different version of  the ubitx_ui module, or possibly use  of #ifdef depending on how complicated the whole thing turns out.
With the limited size on the 1.8" display I am trying to keep everything as simple as possible.  It looks like I will probably have to switch toa separate screen display when I switch to setup mode.  When I move to a 2.4" or 2.8" with higher resolution, I will have room for information to be displayed at one time, and may be able to add a pop-up sub menu. 

As of now I have the FREQUENCY, VFO, and MODE display finished.  Just now trying to add the RIT status display to the empty red box on the right of the screen.  Also plan on having the command prompts from pushbutton work as a pop-up item. Don't know how much I will get done before I have to pack it up and go visit relatives, but will keep you informed.

Thursday, November 30, 2017

CNC Engraver update 12/6/17

I had been planning on doing a update on the CNC Engraver purchased to use as a PCB mill, but a couple things have gotten in the way.  First is the series articles I am writing for QQ, the designing, building, and testing, along with the writing is taking quite a bit of time.  Second reason was that I was clumsy. I had the unit along with power supply and cables all wrapped up and out of the way, I thought.  Then I was moving something else and knocked the power supply off the bench.  Of course it was wrapped with the USB cable, so when it fell it tore the USB connector off the controller board.  I tried to repair the board, but too many of the traces wore gone.  I had to order a replacement board, which has now brought the price up to what it would have been if not on sale when I got it.  While I was waiting for the replacement I designed and 3D printed several things to prevent this from happening again.  I made a cover for the circuit board, and some brackets to hold the power brick in place.  Along with the new controller board, I also ordered a Panel mount USB to mini USB cable.  I printed a bracket for it and some cable management clips that fit in the 2020 aluminum extrusion that makes up the frame of the engraver.  After everything arrived, and I finished printing the parts, everything looks really nice and secure mounted on the frame.     


The first picture is of the back with the cover for the circuit board, including a cooling fan for the stepper drivers.      
A set of brackets to hold the power brick and the bracket for the panel mount USB connector.  

Next is a picture from the front showing some of the board and work piece holders I also printed.  The slots in the aluminum frame of the engraver work really well for storing accessories and work piece holders.

Along with the the repair and modification of the engraver itself, I have been spending quite a bit of time trying several different CNC  software packages.  And breaking quite a few engraver bits along the way.  To make things more interesting, I also changed my PCB design software.  I had been using the free version of Eagle, but the latest version reduces the size of board you can produce.  I looked around and found Sprint Layout, it is not free but only costs around $50.  I tried the demo version before I purchased, and found it to be one of the easiest to learn of all the different packages I had tried.  It allows you to generate Gerber files for sending to board houses, if I want to have boards made.  I can also print normal or mirrored images of the copper layer for making my own boards by 'toner transfer'.  One other thing I really like, is that I can make a complete assembly such as a bidirectional amplifier or mixer stage a single component that can be dropped onto a new board layout.  I can see where this could make several projects I am working on much easier to lay out.
With all that I finally managed to route a complete circuit board.  I think I could do a much nicer job using 'toner transfer' , but it is a start.  I will play around some more tweaking settings in the various software packages I need to use to do the entire process.  One thing I did was to make a drill file for a board I had etched previously, and it did a nice job of drilling all the holes.  I may end up routing the boards for larger less crowded projects and etching the board, using toner transfer for more dense projects. Then just use the engraver for drilling all the holes and trimming the board. That has always been the most time consuming part of making fairly complicated boards with through hole components.  And a real pain to get the alignment correct on double row header connectors.  Well still have a few engraving bits left so will have to see what I decide on doing.  

I bought a 3 lb. box of scrap pcb  material on eBay to use to play with the engraver as a pcb mill.  Was pleasantly surprised that most of the material was around 4"x6" double sided and in thicknesses from about .020" to .063".  A better deal than I had buying specific sizes in the past.  
With that I spent some time playing with the software packages I need to design and mill the boards.  First, since I switched from 
Eagle to |Sprint Layout, I had to play with some of the settings  to get the correct clearance around traces and pads.  Milling the boards seem to make that dimension a little more critical than when I etch the boards.  After getting those settings adjusted and exporting some Gerber files, next I had to work on some settings and sequence of steps I use in FlatCAM to generate the g-Code for the engraver.  Finally I built and installed a Z-axis zero probe to help adjust the Z height of the router bits in relation to the top of the pcb material.  With these settings modified, I decided to try a couple different boards.  The first was converting the LC380 amplifier board from the simpleceiver to Sprint Layout format.  This went quite well after just a little playing with the software. After using FlatCam to generate the g-Code and then making the adjustments for X,Y and Z zero in the GRBLcontrol software I use to control the engraver I milled a board. 

 I was very happy with the results, everything was nice and clear.  It did not look like there were any problems with uneven milling due to the pcb material not being level.  I was worried about that, and had another auto-leveling program I could use if needed. But everything turned out great.  The next thing I wanted to try was a SMD board with size 805 components.  I took the AD8307 power meter circuit I had used on several projects, and changed  the components from 1206 to 805 size.  After generating the g-Code and milling the board I was more than pleased with the results.  

Everything was nice and sharp, no problems with under or over milling around the pads on the SOIC 8 AD8307 or any of the other components.  The only problem was getting a decent picture of the completed board. 

I am now very pleased with the results I can get milling PC boards on the small engraver.  It is small enough and sits right next to my computer desk.  Now I can design, layout and mill a board with out having to even get up from the desk.

Saturday, October 28, 2017

AD8307 power meter QQ article additional build notes update 11/12/17

I just received the latest issue of QRP Quarterly with the first in a series of articles I am writing on using a building block approach for designing and building simple Arduino based test equipment. These articles mostly cover projects I have previously featured in my blog posts, and that I have received requests for additional information on.  I am going to post additional build information on this blog, and in a dropbox folder listed in the article.

 I will be making available the sketch, along with the Cad and Gerber files for the circuit boards. I will also have .PDF files of the copper layers that can be used to make your own boards using the "toner transfer".  For those who do not want to make their own boards, I have a limited number available that I will mail to US locations, contact me directly at for pricing.  For those who would like to order multiple boards  or are outside of the US I have created a shared project at  The direct URL for the project is 
 They have very reasonable prices for boards this size, and offer several methods of shipping depending on how quickly you want them.  Using HK post I usually get the boards in between two to three weeks, DHL gets them to you in less than a week but is more expensive.

Updated Arduino Display building block
A couple notes on the Arduino display building block.  When  I did a revised board layout to make it easier to use with a 3D printed case,there were a couple changes that are different from the schematic in the article.  The only display pins that must be connected to a specific Arduino output line are CLK (13), and SDA(11),  All  the other pins can be connected to any other Arduino outputs, as long as they are defined in the sketch.  I have moved the LED on the display to pin 9 on the Arduino.  This is a PWM pin and can be used to control the display brightness if desired. If you use this option, you can change the value of R1 to a smaller value such as 47R.  On the probe style board I used pin 7 instead of 9 to make layout a little easier.
R2 and R3 are a voltage divider used to measure the supply voltage.  The values are chosen to minimize current usage, and the values used must be defined in the sketch to correctly calculate the supply voltage input to the Arduino.  The board also has  provision for a optional reverse voltage diode.  If used uncomment the line of code in the sketch that adds this to the measured value.  The .7 volts is probably close enough to use, but you could measure the voltage drop across it and use that value instead.

Looking at the schematic for the AD8307 log amp, you will see that I use small molded RF chokes for connections to the Arduino.  In most circuits I have seen a small value resistor,  usually around 4.7 to 10 ohms is used instead.  I found that since I am using the 5 volt output of the Arduino to power both the AD8307 and the TFT display, I could get a lower noise floor with the RF chokes.  You could use the resistor instead if you want. Also when I did the board layout I only put in one 100nF bypass capacitor on the AD8307.  With good shielding of the AD8307 circuitry it did not make any difference I could see.

Here are larger parts placement diagrams for the top and bottom of the boards. 

Notice that I install all the through hole components on the bottom side of the board.  This allows the Arduino pro-mini to be installed in a 24 pin low profile IC socket if desired.  If you do I have found it works better to install the longer header strip pins  down through the Arduino and then clip off the shorter pins on top of the header strip.
 I usually just solder it to the board without a socket instead.

The placement of  SMD parts on the top layer for the AD8307 circuitry is the same for both the Probe  and  Altoids tin style boards.

 I hope to add some additional notes on the software later this weekend.

UPDATE 11/12/17

I have been fighting the small 1.44" 128x128 TFT display used in this project.  Received a order I had placed a while ago, and the ones I received were a different revision than I had used before.  I know there were two different versions and had the sketches setup so I could use either version.  Now there is another new version and I can not get it working with either version of driver I had used before.  I have tried several different configurations with no success.  Along the way I found a Arduino forum that covers this type display
It seems like there are more than a few problems with some of the newer versions.  I think I will switch over to the 1.8" display I used in my SNA Jr. instead.  Slightly larger, but I think it will fit in the same size case. I have only found one set of drivers for it, and that might eliminate some of the problems that have been reported with the 1.44" display.
Have a couple on order and when they get here I will rewrite the sketch to reflect the changes.