A Simpler Throttle

Throttle Controller
#1

I saw some pictures of the throttle and it seems that its a custom made PCB that uses an Atmega microcontroller (same as Arduino).

sp140-controller
sp140-controller806×965 149 KB

My thought process is that it seems like an overkill in terms of complexity. A majority of the component count is there just to run and interface with the microcontroller; resistors, capacitors, regulators, LEDS, micro-USB, powersupply, cyrstal oscillator, etc.

What if we designed our own motherboard that only mounts components that are relevent to the throttle (barometric sensor, throttle pontentiometer, pin headers for display, etc) and off loaded the rest to a daughterboard that can be bought off the shelf (ie Arduino Micro).

A prime example of this is the kinT board for the Kinesis Advantage keyboard. The daughterboard (teensy) is bought off the shelf and soldered onto a motherboard that is custom made (by Michael Stapelberg). Kinx-Project

teensy36
teensy36800×228 74.9 KB

kint.PNG
kint.PNG671×520 45.1 KB

pcb-3d-render-front-v2021-06-26
pcb-3d-render-front-v2021-06-261912×1436 292 KB

With the current design we have everything on one single PCB. A throttle that uses a splitboard design has the following advantages are:

  1. Easier Servicing Of Hardware - With a split board design you can cheaply replace the daugter or mother board. Easily replace broken parts (ie. Linear Motion Position Sensor) on the motherboard due to its simplicity.
  2. Cheaper Manufacturing - We just buy off the shelf daughterboard (ie arduino micro) and plug it into the motherboard. The motherboard is cheap since it is extremely simple and doesn’t require extensive component placement or SMD soldering. The off the shelf arduino board is extremely cheap because of economy of scales.
  3. Improves Reliability - The off the shelf arduino board is well… off the shelf. There are millions of these arduino boards that have been tested in a variety of environments. We will just need to worry about the reliability of the motherboard (not a problem due to low component count).
  4. Faster Prototyping - You can cheaply and easily modify the motherboard. For example download a BLE chip placement plan and easily incorporate into the motherboard. Generate gerber and couple days later you get your PCB. Solder in the daughter board and the few components that go onto the motherboard PCB and you are ready to flash the daugterboard.
  5. Easier For Hobbyist - Hobbyist won’t need to be concerned with Arduino component layout. They will just need to add/modifiy footprints for the component they are adding and route it to a free pin on the daughterboard (ie arduino micro) that will get attached.

The downside is that:

  1. We might need to put a bit more effort into layout to accomodate the off the shelf daughterboard.

Hopefully I made sense and didn’t confuse you guys. If you have question or clarification don’t hesitate to ask.

I am looking to see what you guys think. (especially @zjwhitehead & @Pdwhite & @glydrfreak).

Thanks

#2

Hi @bojingles and welcome!

I think you nailed many of the advantages of using a separate daughter board design.
After all, many of our prototypes have started with an off the shelf Arduino variant and all of the components connected via a breadboard and wires.
I bought a few variations of the Seeeduino XIAO board recently to prototype with. They seem to be some of the most promising options as the formfactor is very compact and pretty cost effective. I’ll definitely be keeping an eye on those and similar boards.

From my perspective, there are a few other downsides of that approach.
Firstly, as you may have noticed, many of the Arduino boards, including the Seeedstudio ones have been either out of stock or only available in limited quantities for the last ~2yrs. This presents a problem when we already have a motherboard built out for a specific off the shelf Arduino thats unobtainable. We’d either have to switch to another board option and update the layout + pins etc. or just wait and be at the mercy of someone like Adafruit.

In the latest batch we’ve had to work around many chip shortage and supply chain issues by swapping out components and redesigning the PCB but it was worth the tradeoff as we were able to secure large quantities of chips (100s) from global suppliers like Digikey.

Speaking of companies like Adafruit, many of their Arduino designs are available for reference so there isnt a ton of reinventing the wheel design wise which helps with reliability.
Cost of manufacturing seems negligible once you hit 50+ PCBs. Yes there are economies of scale for the Arduino board manufacturers but they also have to make a profit. We take advantage of Chinese PCB manufacturers that do assembly in-house so thankfully its fairly affordable. Once you send them the completed design they can crank out boards in just about any quantity.

All that being said, we’d love if we didnt have to do any custom designs. Hardware is… hard :smiley:
But since I still havent found any decent electric paramotor throttle systems here we are haha.

It sounds like you have some electronics experience so feel free to pass along any further feedback. Also, Im always looking for more capable yet simpler and affordable Arduino boards to build off of so let me know if you find other options.

thanks for coming to my TED talk

2 Likes
#3

I too have been prototyping with Seeeduino Xiao boards. So far so good! I’m in the very very early prototyping stages.

2 Likes