Jump to content

self made euc


Demargon

Recommended Posts

They are a euc instructions for DIY made years ago http://www.instructables.com/id/Electric-unicycle-Arduino-DIY/

It can be updated with hub motor and wide pedals. Would be nice to have a collaborative open source project euc made in electricunicycle.org

They are lot of features to discuss, like size, weight, power... but because is a open source project, the most important features is the scalability and compatibility.

Firs the scalability allow more people access to the product. From the basic and cheap to the expensive and fullfil. 

Compatibility means easy to find parts for repair or upgrade. Freedom to the consumer with no restrictions from patents, if you want a handle cut off switch simply put it on.

My setup for start is a light weight euc with 800w like IM v8 without shell for ease fixing and upgrade, mud guards and hardware encapsulation for water and dust protection. Because my old times with classic unicycles I like the seat with small handle, is useful for save the euc from falling, do jumps or just sitting

Link to comment
Share on other sites

I actually made a whole bunch of motor controllers I was hoping to use for an open source unicycle, I completely designed all the 3 phase inverter and hardware needed for motor control. I had it running the unicycle motors with no issue but I could never find anyone to help finish the software, I don't have enough embedded programming experience to do it. I more or less know what has to be done but getting it all together into working firmware is difficult.

 

Link to comment
Share on other sites

I have decent ability in firmware (and hardware) but unfortunately not a lot of spare time due to work.  Still, if you have a good understanding, let me know and maybe I can help.  My recollection is that the balancing algorithms are a bit non-trivial when you dig into the guts, but I know examples are available online.

It seems to me that the part to standardize is really:

  1. The motor and/or motor assembly (motor + axle + cover + hall effect sensors + mounts) - something easily available or even buildable from scratch
  2. Motor controller board (particularly with regen and able to handle high currents, and with low RFI, not sure this is simplistic)
  3. Maybe a "reference design" for the system controller board that takes in sensor data and sends out motor control commands
    1. Well-documented firmware and tuning instructions
    2. Gyro/accelerometer functionality... doesn't need to be standard hardware, but algorithm and precision/capability should be agreed
  4. Battery management system that can handle high current discharge and reasonably fast recharge, with basic protections
  5. Did I leave anything out?

Stuff like a headlight, how the pedals are attached, frame, bodywork, bluetooth capability, battery indication, batteries themselves, carry handle switch, calibration, bluetooth telemetry, etc. should be pretty easy for any designer/maker to tailor to their own vision.

Microcontroller companies are starting to come out with products targeted at the drone market, so it should be easier than ever to achieve at a decent cost.

Link to comment
Share on other sites

5 hours ago, pst said:

I have decent ability in firmware (and hardware) but unfortunately not a lot of spare time due to work.  Still, if you have a good understanding, let me know and maybe I can help.  My recollection is that the balancing algorithms are a bit non-trivial when you dig into the guts, but I know examples are available online.

It seems to me that the part to standardize is really:

  1. The motor and/or motor assembly (motor + axle + cover + hall effect sensors + mounts) - something easily available or even buildable from scratch
  2. Motor controller board (particularly with regen and able to handle high currents, and with low RFI, not sure this is simplistic)
  3. Maybe a "reference design" for the system controller board that takes in sensor data and sends out motor control commands
    1. Well-documented firmware and tuning instructions
    2. Gyro/accelerometer functionality... doesn't need to be standard hardware, but algorithm and precision/capability should be agreed
  4. Battery management system that can handle high current discharge and reasonably fast recharge, with basic protections
  5. Did I leave anything out?

Stuff like a headlight, how the pedals are attached, frame, bodywork, bluetooth capability, battery indication, batteries themselves, carry handle switch, calibration, bluetooth telemetry, etc. should be pretty easy for any designer/maker to tailor to their own vision.

Microcontroller companies are starting to come out with products targeted at the drone market, so it should be easier than ever to achieve at a decent cost.

The main thing is picking a MCU with robust motor control software ready to go, the unicycles need a decent FOC motor controller to run well. Texas instruments and ST micro both have decent motor control options, they aren't 100% open source as the sensorless motor observers are in compiled libraries but you can still have the rest completely open you can still upload it to github and anyone can build it. They're fairly barebones they can detect motor parameters and spin the motor in debug but there's a few things you would ideally need to add such as separate current limits for motor current and battery etc.

There's the open source VESC software which is quite good the downside is it's extremely complex and not really documented, the person who made it was going to add support for 6 axis IMUs but he doesn't have much time to work on it so it may not happen for 12 months or something.

I found the texas instruments motor control to be the best of the commercial options I tried, to get it working all you would need to do is get the IMU working over I2C and have the balance code set the motor torque value in the provided torque control loop. There's not too much code needed for the balancing, besides filtering you need to have offsets for the gyro as it won't be perfectly level and the balancing code needs to compensate for the decrease in torque as speed increases.

Link to comment
Share on other sites

On 3/28/2018 at 7:34 AM, lizardmech said:

The main thing is picking a MCU with robust motor control software ready to go, the unicycles need a decent FOC motor controller to run well.

Yeah, that's what I was referring to, above.  It's not exactly "portable" if you're relying on custom peripherals to do the work, but then again I imagine it's not unlike A/D libraries, etc., where the differences from one vendor's approach to another's is not too painful.  And sometimes there's nothing to be gained by getting into the gritty details, anyway.

On 3/28/2018 at 7:34 AM, lizardmech said:

There's not too much code needed for the balancing, besides filtering you need to have offsets for the gyro as it won't be perfectly level and the balancing code needs to compensate for the decrease in torque as speed increases.

If you know of some good reference designs... ;)

I've not done much work with TI processors, mostly Motorola/Freescale/NXP so I might need to do a little investing in compilers/debuggers... but if you want to discuss it further and are in no massive rush I'm game to play around as my schedule allows!

Also I think the battery management side (as far as safe/consistent discharge, as well as recharge, as well as possibly regen, as well as balancing all the while) would be another area where I don't have much personal experience but I'm happy to learn.  Usually I stick more with sensors and protocol conversion and other purely digital stuff.

Link to comment
Share on other sites

18 hours ago, pst said:

Yeah, that's what I was referring to, above.  It's not exactly "portable" if you're relying on custom peripherals to do the work, but then again I imagine it's not unlike A/D libraries, etc., where the differences from one vendor's approach to another's is not too painful.  And sometimes there's nothing to be gained by getting into the gritty details, anyway.

If you know of some good reference designs... ;)

I've not done much work with TI processors, mostly Motorola/Freescale/NXP so I might need to do a little investing in compilers/debuggers... but if you want to discuss it further and are in no massive rush I'm game to play around as my schedule allows!

Also I think the battery management side (as far as safe/consistent discharge, as well as recharge, as well as possibly regen, as well as balancing all the while) would be another area where I don't have much personal experience but I'm happy to learn.  Usually I stick more with sensors and protocol conversion and other purely digital stuff.

I just use these mostly. They come with a MCU and debugger on one board, you can get 3 phase inverter add one boards for about $50 as well. Compiler and IDE is free these days, Ti have their code composer studio thing which is a modified version of eclipse and GCC configured for their MCUs.

https://mouser.com/ProductDetail/Texas-Instruments/LAUNCHXL-F28069M

STM boards are similar with on board debugging, there's many options for compiling for those since they are just 32bit ARM code, IDEs I think you can chose between eclipse or various others. Most of the time I just build straight out of linux via GCC for the STM ARM MCUs. Pretty much any of the STM32F4 variants are fine.

This is the closest example I have found to code suitable for a unicycle this one uses brushless gimbal motors, the motor code is different but still closer than the more common slow moving geared brushed DC motor balancing robots, the TLDR more or less is you need to add compensation for reduction of torque as RPM increases, the code is pretty lightweight so any of the 32bit motor control MCUs should handle FOC + balancing without any issues.

http://www.instructables.com/id/Brushless-Gimbal-Balancing-Robot/

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...