Jan 31, 2020:
Revision: v1.0

Tiny 64 x 32 OLED Display for SugarCube Si5351a VFO

A further option for builders and users of the ZL2PD SugarCube VFO - An even smaller display. It reduces the overall volume very significantly, ideal for those aiming at adding the SugarCube to portable transceivers. And all the other regular features remain available albeit in a smaller size. Interested? Read on.  


My SugarCube VFO is a tiny dual VFO controlled by a small 8-pin ATtiny85 that can be built in a volume not much larger than a sugar cube.

It connects to a small graphical display, the widely available low-cost 0.91" 128x32 OLED LCD. All that's required to control it is a standard rotary  encoder for tuning, and three pushbuttons.
The software also supports a graphical bar meter
for signal strength, RF power, battery volts etc.
Figure 1: OLED displays: 64x32, 72x40 and 128x32 alongside some 5mm and 3mm LEDs and a 16-pin DIL IC for size reference

About  6 weeks ago, I released improved software for the SugarCube VFO with user-programmable functions including:

–    VFO A Rx
–    VFO B Rx
–    VFO A Tx, and
–    VFO B Tx

My Sugarcube VFO uses the 0.91” 128 x 32 OLED display. That's the largest of the three OLED displays shown above. It uses an integrated SSD-1306 controller chip. While small, the resulting display is perfectly clear and easy to read from halfway across the room. This has allowed me to squeeze a VHF Sugarcube VFO into an old VHF FM receiver, described elsewhere on my website, and into a new compact HF SSB transceiver I am building. Many others have also made extensive use of it in their transceivers.

Despite the considerable success of the already very small Sugarcube VFO, I received another request recently. Could I modify the software to permit the use of an even smaller 64 x 32 OLED display? These measure just 0.49” from corner to corner across the display’s diagonal. (It's the one on the left-hand side of the photo at the top of this page)

I was a little reluctant to try this display. They cost considerably more, roughly double that for the more widely available 128 x 32 OLED displays. I also believed the smaller 64 x 32 display would make it very difficult to present, or read, the VFO information.

However, after a bit of to-and-fro discussion over these points, I decided to at least try the display with the VFO. In addition, I thought I would also test the similar and apparently even smaller 0.41” 70 x 40 OLED displays which I noticed were becoming available, albeit at a similarly higher price. That's the display in the centre of the photo at the top of this page. And, yes, it's not really smaller, is it!

Small Displays – Big Problems

While both 64x32 and 70x40 OLED display modules were ordered at the same time, the 70x40 displays arrived first. A brief test proved that there were problems – The display looked as if it had zoomed in on a tiny section of the original 128x32 display. It looked most unusual.

After some (considerable!) effort, the problem was traced to two factors. Firstly, as might be expected, the integrated SSD-1306 chip uses the full 128x64 pixel display memory. Logically, only a part of that memory is actually displayed on the smaller 64x32 or 70x40 OLED.

That was fine, and easily understood. However, it also quickly transpired that the stated 70x40 specification was incorrect – Not the first error I was to encounter. These modules are labeled “70x40” on the display's PCB. However, they actually display 72x40 pixels. This also makes perfect sense. The display memory and operation is very much byte-oriented. The ‘72x40’ result resolves to a display which is effectively 9 bytes wide (i.e. 9x8=72), and 5 bytes high (i.e. 5x8=40).

Next, the display is horizontally centred on the internal memory, displaying data from column 28 to column 100 of the memory’s 1 – 128 columns. Vertically, the display shows the lower 5 rows, from row 3 to 7.

The other issue I encountered was initialization. The various datasheets for the 72x40 and 64x32 OLED displays all provided (different!) recommended initialization configuration sequences. None of these worked. As a result, establishing the correct setup data sequence took some time (!!) but it was worth it. It completely resolved the strange ‘zoom’ display effect. Usefully, the same revised config data string worked with both smaller displays. That allowed quick progress with the 64 x 32 displays when they arrived a few days later.

Figure 2 : My assembled and operating 64x32 OLED SugarCube Plus VFO. I've put a small
16-pin IC alongside the SugarCube to illustrate the really compact size of this version.
The 72x40 OLED display board is slightly wider than the red VFO board while the
display itself is actually about 20% smaller than the 64x32 OLED visual area

Reformatting the SugarCube Display Data

With the displays now working, the visual clarity of the VFO information could finally be properly evaluated. This 64x32 OLED SugarCube version supports a display with 100 Hz resolution for frequencies up to 100MHz with digits in a XX.XXX.X format e.g. 14.109.5 or 3.525.6 etc. (The standard SugarCube VFOs provide 5Hz tuning and 1Hz display resolution)

All of the previous SugarCube Plus VFO features including dual VFOs and BFOs, EEPROM storage of the VFO configuration etc have been retained. The “VFO A” and “VFO B” selection icon is still shown, but now it is located on the lower line alongside a smaller 48-pixel wide signal strength meter.

 In addition, the ‘Lock’ icon is also shown when the feature is enabled.

Figure 3 : The photo above shows the prototype assembly
with SugarCube VFO PCB (red) at the back of the assembly,
with the 64x32 OLED module wired directly on top of the VFO
PCB with its ATtiny85 acting as the "filling in the sandwich".

What’s missing in this version of the Sugarcube? Well, a couple of minor items. First, the transmit “T” icon is no longer shown. Also, the “MHz” characters which are normally shown after the frequency are also no longer displayed. Neither of these items are likely to be of much concern, I suspect.

Perhaps of greater importance are the tuning step sizes. These now start from 100Hz steps rather than the 5Hz steps available with the standard (128x32 display) SugarCube software. Since this 64x32 display VFO is aimed at the most compact of transceivers, perhaps for those venturing out into the mountains and the bush, climbers and trampers etc, this is unlikely to be a problem. 

One thing to note: The 128 x 32 displays operate with the long 128-bit wide axis oriented horizontally across the panel with the four pin connections on the left hand edge (using the software for that display available on my other webpage). Makes perfect sense. Similarly, these smaller displays also arrange the longer 64 or 72 pixel axis horizontally, but, by contrast, the connection pins for the display will now lie horizontally along the upper edge of the mounted display. If you are designing a new PCB with integrated display, switches, and encoder, you will need to take note of this difference.

PCB for the Switches

In the process of building this new prototype, I decided to design a little PCB for the switches and resistors to ease the wiring and panel mounting work. I’m not really sure it provides a significant improvement in the construction process with this PCB - everyone's application tends to be different - but anyway, here it is, just in case someone else wants to use it.

Figure 4 : PCB layout diagram for front panel pushbuttons and the resistors


To help ease the wiring process, here’s a diagram of the wiring required for this version. The circuit diagram is identical to that of the original SugarCube and the enhanced SugarCube (with EEPROM programmable values). It is actually much less complicated than it looks, as you can see from the photo of the prototype at the top of this page.

 Figure 5 : Wiring details for the 64x32 OLED version and the switch PCB. Right-click on this figure with your mouse to see a full-scale diagram on your screen.

The photo of the actual wired prototype which follows this diagram is shown in Figure 2 above.


The HEX file for programming the ATtiny85 for this version of the SugarCube is available for download below. In that ZIP file, I have also included a sample EEPROM file (*.EEP) for testing and for later user amendment, as described on the SugarCube Plus page.

The details for programming the SugarCube, the EEPROM (for your specific requirements) and the spreadsheet for calculating the data to store in the EEPROM, and the fuse settings for the ATtiny85, are given on THIS page of my website. The schematic and details of the PCB component layout are available on the original SugarCube page.
As noted on the other SugarCube pages, the Gerber files for the SugarCube VFO PCB is available for your personal non-commercial use on request. Just email me. (My email address is shown on the main page of this website) Similarly, if a club wants to build a bunch of these, email me and we can chat about the arrangements. There's nothing onerous. I just need to ensure I retain copyright.


As you can see, lots of people are using my SugarCube design. I now agree that this latest version does offer an enhancement to the original design – It certainly occupies a great deal less space on the front panel. Since I’ve managed to keep the graphics design for the font used identical to the standard SugarCube VFO, I think it retains much of its readability.

I do hope you find this latest addition to the SugarCube VFO family equally useful.


Software – The HEX file for this version of the SugarCube Plus along with a sample EEPROM file for testing and later user amendment is available here.

Switchboard PCB – The set of Gerber files for the little switchboard PCB is available here

Questions? Send me an email. I'll try my best to answer as soon as I can. Just remember my (paid) work takes priority so it may take a couple of days to reply....or longer if I'm travelling or working in some weird out of the way location.

Want to go back to the main page? Click here to return directly.