So, I’ve been considering trying to make a smaller proffieboard for a while now. Because of some posts on here, I started actually working on it. The result, is called the Proffieboard M2, (M stands for “Mini”, and the 2 is because it’s basically a Proffieboard V2.)
Here’s what they are probably going to look like:
Top:
It might not be obvious from the pictures, but these are going to be very small. About 5mm shorter and 5mm narrower than a Proffieboard V2. This makes the PCB 10.16 x 27.94 mm. The USB connector sticks out a little, so we could just call that 10x30mm. That’s less then half the surface area of a V2 proffieboard. These will fit in any hilt. They would fit inside the blade. In a graflex, they would fit lengthwise. I’m tempted to try to squeeze one into one of my MR .45 scaled lightsabers.
To make it this small, I had to make a few changes, compared to a V2 board:
3 FETs instead of 6.
No voltage booster, so sound won’t be as loud.
No SD card, uses a fixed 4Gb memory instead.
No “SD Power” pad.
Only one GND/Batt- Pad.
No 5V pad (no booster, duh!)
No Reset pad.
No reverse polarity protection.
CPU and amplifier uses BGA chips which are smaller.
Less used pads are surface-pads instead of through-hole pads.
Designed to work with the oshpark 6-layer PCB design, but I cheated a little and used 4/4 trace width/separation.
Everything else should still work just like a Proffieboard V2.
Unfortunately, due to the chip shortages, these boards are still entirely hypothetical. I can’t even find enough chips to build prototypes for these things, much less launch and sell them. If/When that changes, updates will be posted in this thread.
Since these are designed, but not in any shape or form made yet, it may be time to change stuff that would make this board better, so comments and suggestions are welcome.
*I’d still want reverse polarity protection. I have similar hilts w two that the battery installs backwards and was thankful it was there when I goofed up once. That said it’s not hard to put a small circuit in a hilt somewhere connected between the board and the battery.
There just isn’t any room for reverse polarity protection on the board.
The CPU is still protected because of D61, but the amp would most likely fry if you plugged the battery in backwards.
If you have room for a battery holder, you probably don’t need an mini board though.
This seems like the prefect proffieboard for my needs!
I never liked the FETs for reverse polarity protection, since they can slowly burn out at low voltages. I built two lightsabers for my children with a MCP73831 connected to the same VBUS as the Proffieboard. It all worked perfectly on my PC for several days, but three hours, with the kill switch in the off positions, on a very noisy Belkin USB charger killed Q7A.
I prefer to use a polyfuse with a diode for reverse polarity protection. It does not actually block the reverse voltage but reduces the time and voltage a device is exposed in reverse. The nice thing is most of the time no parts are seriously damaged. Everything is fine once the reverse voltage is removed and the fuse cools back on.
One question does come to mind: Since you no longer have a boost voltage regulator, doesn’t D61 reduce the voltage range the board will operate at?
I don’t think so.
V2 proffieboards don’t run the booster all the time, only when it’s doing something.
The motion chip seems to work anyways, and so does the SD card. And I know the CPU is happy to run at voltages down to about 2.2 volts.
The only thing I have to be careful about is that the FETs must have a Vgs-threshold of 1 volt to make sure that it works well without a booster.
Love this! Lower volume is the only downside for me, then again the docs recommend not putting the speaker levels very high anyway (to avoid blowing them out) so probably won’t limit realistic volume.
I don’t need something small, but I’d love something cheaper and simpler.
Question - which chips are the hard to find ones? The CPU & motion controller? If we had a working CPU alternative (like the RP2040/ESP32), then theoretically would these have been feasible now?
It occurs to me that in cases where a small board is required, people will probably also use a small speaker, which probably couldn’t handle high volumes anyways.
I’m not really sure if the M2 will be any cheaper. It does have a few less components, but it also uses BGA components which means more assembly work and possibly lower yields.
It’s mostly the CPU that is hard to find. The motion chip is also difficult, but I have a few of those already. Replacing the CPU requires large software changes. I am exploring that, but the amount of work is substantial, so it will take a while. So far, all I’ve really achieved in terms of alternate CPUs is better teensy4 support. It won’t be small, but I’m thinking that a TeensySaber V4 might be cool to make.
Ok, this is like the perfect size board for a spin saber/stunt saber.
I’ve been contemplating how to get the outside hilt diameter at or below 1" for easier grip during flow work and spins.G (so a Sub 0.875" Chassis diameter)
Very cool, will have to get a couple of these to experiment with!
I would be very interested in a TeensySaber V4, if only for nostalgia reasons. Only free time has stopped me from following the existing guides to do a V1 style build but subbing in the v4 Teensy.
V4 Teensys are also absolutely ridonkulusly fast.
600Mhz, 2 operations per cycle, UHS-I SD card read speeds and lots and lots of memory.
Many of the optimizations I’ve done for proffieboards over the last years are simply not needed for V4 teensys. One drawback is that they are not 5V-tolerant, so they may be easier to kill than a Teensy V3.2, or a Proffieboard.
So the M2 board would ultimately replace what is currently the v2 proffieboard, or, will it be available as an additional board for smaller installations?
Not sure.
Between the M2 and the V3, it may well be that the V2 will disappear, but it really depends on production costs and demand. Ultimately, each board maker is going to have to decide which boards they want to make and sell, so I’m not really sure what will happen.
I love this, and it would definitely be enough for quite a lot of simpler installs.
I guess my only comment is that I feel a fourth LED pad would be more useful than a fourth data pad, but maybe that’s just me. If this board is aimed at stunt sabers, it means a Tri-Cree LED uses all available pads leaving none free for an accent or status LED. That said, I appreciate it’s very easy to make these observations but much harder to implement them.
But I love the concept and would definitely buy and use them if they ever come to fruition.
The FETs take a fair amount of space on the board.
I would have to make the board larger to make that happen.
Also, you could still wire up some always-powered neopixel accents.
A small change that might make sense would be to have usb pads rather than a USB port. That way folks can place the port else where for easier access.
It would also allow for the installer to choose usb-c or anything else really, lightning port if you wanna live precariously. also allows for installation of usb charging without the bulky mess of adapters etc.
