DIY Arduino Controlled NiTiNOL Actuator
Вставка
- Опубліковано 16 вер 2024
- Testing out a custom NiTiNOL actuator design methods for some future projects coming up!
NiTiNOL paper referenced in the video: carleton.ca/ce...
Music Attribution:
Music by Bensound.com/royalty-free-music
License code: JBESZVSHP9NC3BRE
Music by: Bensound.com/free-music-for-videos
License code: AMGALSMBJDB0Y1TM
Music by: Bensound.com/royalty-free-music
License code: N8P1FYGDZK2KLVXG
Music: bensound.com
License code: BRKDO60BL2DY7TNM
Music I Use: Bensound.com/free-music-for-videos
License code: HS5ILBBQRR0IIVCZ
Music I use: www.bensound.com
License code: 8UQEQR5XJXBS1BYG
Music: bensound.com
License code: TPTWKLSCAHLSTDJJ
Music by: Bensound.com/royalty-free-music
License code: 4EU5X5GJB08DOHNU
I like your video because you had problems and talked about how you work through them and came up with solutions. That is teaching. Defining exactly how something is supposed to be is not teaching.
Thank you! Glad you enjoyed watching!
I really enjoyed how to showed your thinking process follow each failure.
That blue screen gag at 0:18 aged too well 0.0 hello from July 2024 lmfao
Very cool project! Wholeheartedly agree that rocketry is an exciting application for these. Hopefully some of the following electronics suggestions are helpful.
To the point of avoiding large power components in the current source:
A constant current LED driver could be an effective off-the-shelf solution.
Otherwise, a MOSFET driven by an op-amp can make a constant current source. You would connect the FET on the low side of your load, and you would get to connect a low resistance (~0.1 Ohm) current sense resistor between its drain and ground.
The voltage across the current sense resistor is connected at the (-) input of the opamp, and a voltage setpoint is connected to the (+) input of the opamp. The output of the opamp will drive the MOSFET's gate.
You may also use a hall-effect current sensor instead of a current sense resistor. That way the only power dissipating component is the driven FET.
To further reduce power consumption you might need a switching current regulator, though it would be more complex (higher component count and more difficult to design) and noisy (due to the switching, which may require filtering the output).
Regarding the ADC / measurement noise:
Adafruit sells I2C ADC modules with great resolution that may be worth purchasing. To replace a breadboard, Adafruit's perma-proto boards have the same row connections as a breadboard but connections are made with solder like a perfboard. I've found them easier to use than perfboards.
oh man i had a bunch of nitinol growing up in the 90's, including these piston/actuator things that were super fun to play with
I love this style of content! Take my sub and keep up the great work!
@Someone_Should_Make_That Interesting test! Take my like and keep up the great work!
You can use the non stretching wire as well by winding it into springs yourself and setting it's shape. Makes it easier to create designs with long travel.
Frankly, closed loop toasters did kind of exist back in the 60s! Technology Connections had a video on that
Just checked it out! Very cool, we need to go back to that system
Thank you for sharing! I'm working on some SMA stuff as well and appreciated the learning opportunity
You could just use 4 normal 1/4W resisters in series/parallel instead of your massive babies first one.
Not necessarily 'better' but you dont have to special order the big one at least.
Cool! I was thinking the same - glad someone shared that thought
I've never worked with NiTiNOL, so, sorry if it makes no sense. But...
Could you try driving it with a PWM fed MOSFET? Maybe the thermal inertia of the wire is enough to for you to have a stable control within reasonable frequencies.
Any linear power control will have big losses and need big dissipation.
Sounds like it could be worth the attempt! It’s already running with a pwm controlled mosfet so a test run to see if it overheats may be worthwhile, currently the current is limited by the extra resistance
@@Someone_Should_Make_That if that fails, you should consider a DC-DC converter. Won't be hard to integrate it to your PID. Keep us posted!
Kudos for the project!
Nice! Im also looking into nitinol for control surface actuators on a miniature glider for returning a sensor package a weather balloon. I was wondering just how controllable nitinol is, and found your channel.
I’d consider working with an STM32 of some kind or maybe an ESP32 which both have 12-bit analog to digital converters (arduino nano’s atmega328p has a 10 bit adc), which offers a 4x higher reading resolution. You should also consider using a good power source and reference voltage circuit to ensure proper precision.
There is also a component called opamp that can be used in your constant current circuit to reduce its power loss across the resistor, not forgetting about the heat sink on the ic to ensure it works properly (edit: I did forget, oopsies).
Last but not least a breadboard is terrible for any kind of precise circuitry; breadboards are meant for very early prototyping. Exceptionally on your case, the high current flowing through the circuit gets especially distorted as you have pliers-like things that grab onto the components’ legs and dissipate a lot of the power that should go to the actual nitinol wire.
I’m not an engineer myself, more like a hobbyist (on a cellphone at that) of sorts. Hopefully I could be of help.
Excellent advice, appreciate the feedback! I’ll definitely keep this in mind when I revisit
The ADC isn't that noisy and really all of the MCU's are about the same. The problem is the circuit. Or more specifically a breadboard is terrible for anything sensitive to noise. You also need to make sure the AVCC power supply has proper decoupling and you can decouple the ADC input as well. It's pretty much mandatory to average a bunch of samples together to get one "sample" if you're not already doing that. MCU's have low impedance ADC inputs so make sure the ADC is getting enough current or boost with an op-amp.
Batter cells are 1.8volts with a bigger resistor then balance the resistor load and ohms so u can set your volt amp. No need for the constant.
This comment gave me a stroke
sweet old tektronix scope!
Thank you! I find there’s something so much cooler about the old analog scopes! Even if on paper they are much less capable haha
can you tell the percentage contraction of SMA per length
it would be nice to see the schematic and put the link of the paper.
kurutoga roulette = goated pencil
really looking forward to additional content about nitinol
Glad you enjoyed! I have some rough ideas in mind so hopefully they will turn into videos!
👍👍👍
Wait you have friends? Real ones?
I wish I had friends who didn't message me back.
Everybody knows what Nitinol is?!
If by everyone you mean us nerds... yeah