Author Topic: Less noisy alternative to servo. Muscle / nitinol wire?  (Read 3772 times)

0 Members and 1 Guest are viewing this topic.

Offline perikTopic starter

  • Newbie
  • Posts: 6
  • Country: se
Less noisy alternative to servo. Muscle / nitinol wire?
« on: November 09, 2019, 10:54:14 am »
I´m looking for a less noisy alternative to using servos for a project and I was thinking of using musclewires (Flexinol / nitinol wire) as an replacement. I havnt worked with musclewires before and wonder if this is possible or if they are very slow (I know they are driven by heat so slow cooling might be a problem)

Basically my project need to pull and push a small weight with a thin ironstick as arm. I need something that goes two ways. When I applies current to it I want it to pull and when I decrease the current again I want it to gradually go back depending on the current. Im using a pwm signal from my raspberry pi to control the current.

I wounder if anyone has some experience with doing the same? Basically replacing a servo with musclewire?

For example this one:
https://www.sparkfun.com/products/12096

I guess the thinner the wire the faster it reacts since it cools off faster?
« Last Edit: November 09, 2019, 10:57:05 am by perik »
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #1 on: November 10, 2019, 08:49:43 am »
An electromagnet should work if you control the current gradually.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline perikTopic starter

  • Newbie
  • Posts: 6
  • Country: se
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #2 on: November 10, 2019, 08:56:46 am »
I had electromagnets in mind for a while but in my case the heating is a problem. As well as I need to pull and push around 3 cm
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #3 on: November 10, 2019, 09:40:21 am »
Not all servos are that noisy when you get into the really tiny ones. Do you need no noise or just really low noise? You could also take a look at things like the steppers used on CD drives or even the internals of a hard drive.

This will get you started for a look. Single gear reduction (standard servos are 5-8 reductions) only and even drop the motor voltage a touch to reduce the speed and noise. https://www.horizonhobby.com/helicopters/aircraft-servos/20-gram-performance-linear-long-throw-servo--15mm-lead-spmsh2025l
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Online artag

  • Super Contributor
  • ***
  • Posts: 1284
  • Country: gb
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #4 on: November 10, 2019, 10:24:31 am »
As usual, it's a tradeoff of power, cost, speed, noise. You can get quiet geared DC motors - try Maxon. The more expensive servos are also quieter : it's a question of how good the tolerances are on the gears, so metal-geared servos tend to be better.

Thermal actuators are quiet, slow, and take a lot of power. That's fine in some applications.

Steppers are noisy if the steps are large. Use small steps above audio frequency and they're a lot better : try the drivers from Trinamic.

CD drives, as beanflying says, are also good : you can hear a CD drive move but not usually at the individual step level. They use a very long pitch leadscrew so the step rate is low.

Pneumatic and hydraulic actuators are extremely quiet but the pumps and air-release valves are not. Try the murata micropumps or the piezo aquatic aerators for silent operation (the muratas produce surprisingly high pressure)  and pump the air both ways (as you would with hydraulics) rather than letting it escape.
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #5 on: November 10, 2019, 04:24:34 pm »
I had electromagnets in mind for a while but in my case the heating is a problem. As well as I need to pull and push around 3 cm

Depends on amount of force and power needed, acceleration, precision, etc.

3 cm is very doable with an electromagnet with a sliding nucleus. Also, you can have an arm that amplifies the travel range. Heat should not be a problem.

You could use an actuator using compressed oil.

Depending on the application you might benefit from having position feedback.

Many variables.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline perikTopic starter

  • Newbie
  • Posts: 6
  • Country: se
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #6 on: November 10, 2019, 10:16:41 pm »
Do you need no noise or just really low noise?
Ideal no noise but guess it is not possible :)

You could also take a look at things like the steppers used on CD drives or even the internals of a hard drive.
Good idea! But I need lots of servos so prefer buying instead av deconstructing :) Do you have an example of what kind of steppers this is?

Single gear reduction (standard servos are 5-8 reductions) only and even drop the motor voltage a touch to reduce the speed and noise. https://www.horizonhobby.com/helicopters/aircraft-servos/20-gram-performance-linear-long-throw-servo--15mm-lead-spmsh2025l
Sorry for stupid question but Single gear reduction - is that the same as Bearing that says single in the specs? At the moment I have got some cheap SG90 servos. They are pretty noisy though. If you know some other cheap servos in with lower noise you are welcome to recommend

3 cm is very doable with an electromagnet with a sliding nucleus. Also, you can have an arm that amplifies the travel range. Heat should not be a problem.
what is a sliding nucleus? I did some tests and to get a strong electromagnet the heat was actually a problem :)

As usual, it's a tradeoff of power, cost, speed, noise.
since in the project im working i need a lot of servos, around 100 or even more so will try to get a decent servo but with low cost. Because of the amount Im using at the same time I also need them to be more quiet. At the moment I have got some cheap SG90 servos. They are pretty noisy though. If you know some other cheap servos in with lower noise you are welcome to recommend

Steppers are noisy if the steps are large. Use small steps above audio frequency and they're a lot better
I understood first now that steppers cannot be controlled with pwm signal. Maybe also a more expensive alternative compared to servos ...
« Last Edit: November 10, 2019, 11:12:23 pm by perik »
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #7 on: November 10, 2019, 10:45:23 pm »
what is a sliding nucleus? I did some tests and to get a strong electromagnet the heat was actually a problem :)

I meant solenoid. You can get solenoids with linear-moving plunger or with rotary movement.

Again, it depends on many factors. Probably not suited if you need fine control of position where other solutions will give you better control.

https://www.electronics-tutorials.ws/io/io_6.html
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #8 on: November 10, 2019, 11:43:41 pm »
"Flexinol" wires are nice, but require a fair amount of current to operate. Yes AFAIR, the thinnest they are, the fastest they react (but obviously the less tension they can handle before breaking). You may need to experiment a bit.

There are a lot of linear actuators out there, eg: https://www.robotshop.com/en/micro-linear-actuators.html
they tend to be pretty expensive though. You may be able to find much cheaper ones on Aliexpress or the like.

Servos are a bit noisy while rotating, but once in a fixed position, they will normally not make any significant noise unless you are approaching their max rated torque (in which case they can get into the typical "pumping" state.) Of course the servos (or true linear actuators) will be more accurate position-wise (if that matters in your application, maybe it doesn't) than Flexinol.

 

Offline jmelson

  • Super Contributor
  • ***
  • Posts: 2851
  • Country: us
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #9 on: November 10, 2019, 11:48:11 pm »
I´m looking for a less noisy alternative to using servos for a project and I was thinking of using musclewires
Servo motors can be almost totally silent (even in a quiet room) such that you can only hear the bearings.
It just requires proper servo control laws that give smooth current to the motor and eliminate dithering.

Jon
 

Online artag

  • Super Contributor
  • ***
  • Posts: 1284
  • Country: gb
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #10 on: November 10, 2019, 11:53:02 pm »
You can buy the CD steppers with leadscrew on ebay very cheaply - eg https://www.ebay.co.uk/itm/392331862778 (you will find cheaper ones than that). There are others with finer leadscrews if you need slower movement. Search for 'linear stepper'.

Steppers do need dual h-bridge drivers to operate them. If you use proper stepper drivers you will get all sorts of advantages but for a low-voltage stepper like the ones above you can still do them pretty cheaply with something like https://www.ebay.co.uk/itm/173044559728

I can't suggest any _cheap_ quiet servos - I'm afraid the only quiet servos I've seen are expensive ! But there may be other mechanisms like the CD sliders. For more powerful units, headlight positioners are worth considering. They're also stepper motors with a leadscrew. Mirror motors are more often just a geared DC motor and a cam.

If you have large numbers then pneumatics might be an answer : valves can be https://www.ebay.co.uk/itm/254229792066 rather than something expensive from Festo. Actuators might be syringes, or small balloons. Pump might be a fish tank aeration pump. The thing about pneumatics is that you'll typically have one pump and lots of actuators rather than lots of motors, so it doesn't get noisier - unless you need a bigger pump, or let the air out too noisily.

Muscle wire has quite a small extension with heat, so if used in extension mode needs to be wrapped around pulleys to get enough length. If used in a bending mode they will give more movement, but then you need thicker wire.

You might learn something from http://www.osloom.org/. I haven't seen updates for ages and I guess has been abandoned but they used muscle wire to avoid patents on many other forms of heddle actuator. It looks as though it had an acceptable level of performance but I don't know what happened to it.

Most of these aren't very accurate without feedback. But they're OK for pushing from one limit to another.






« Last Edit: November 11, 2019, 01:10:57 am by artag »
 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7012
  • Country: ro
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #11 on: November 11, 2019, 12:11:43 am »
Another silent actuator is the wax piston.  Slow and inefficient, but very strong.
https://en.wikipedia.org/wiki/Wax_motor

Twisted fishing line (only some types, not any line) can actuate, too, when heated.



« Last Edit: November 11, 2019, 12:17:24 am by RoGeorge »
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #12 on: November 11, 2019, 01:46:06 am »
Do you need no noise or just really low noise?
Ideal no noise but guess it is not possible :)

You could also take a look at things like the steppers used on CD drives or even the internals of a hard drive.
Good idea! But I need lots of servos so prefer buying instead av deconstructing :) Do you have an example of what kind of steppers this is?

Single gear reduction (standard servos are 5-8 reductions) only and even drop the motor voltage a touch to reduce the speed and noise. https://www.horizonhobby.com/helicopters/aircraft-servos/20-gram-performance-linear-long-throw-servo--15mm-lead-spmsh2025l
Sorry for stupid question but Single gear reduction - is that the same as Bearing that says single in the specs? At the moment I have got some cheap SG90 servos. They are pretty noisy though. If you know some other cheap servos in with lower noise you are welcome to recommend

3 cm is very doable with an electromagnet with a sliding nucleus. Also, you can have an arm that amplifies the travel range. Heat should not be a problem.
what is a sliding nucleus? I did some tests and to get a strong electromagnet the heat was actually a problem :)

As usual, it's a tradeoff of power, cost, speed, noise.
since in the project im working i need a lot of servos, around 100 or even more so will try to get a decent servo but with low cost. Because of the amount Im using at the same time I also need them to be more quiet. At the moment I have got some cheap SG90 servos. They are pretty noisy though. If you know some other cheap servos in with lower noise you are welcome to recommend

Steppers are noisy if the steps are large. Use small steps above audio frequency and they're a lot better
I understood first now that steppers cannot be controlled with pwm signal. Maybe also a more expensive alternative compared to servos ...

The SG 90's are among the noisiest  ;)

The CD types Artag linked above are the sorts of stuff inside a CD drive so either wreck one to get a freebie or buy one for very little (fun to play with regardless). You just need a driver board to suit.

Look closely at the photo of the Horizon servo I linked it only has a single bearing at the oposite end to the drive. Not really an issue given the power but it will help it run more smoothly. I run the next size down and smaller in some of my indoor R/C toys (sub 2g servos).

A reason to go servo over geared steppers is micros like the arduino already speak servo but you need another board to talk stepper  :)
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5569
  • Country: us
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #13 on: November 11, 2019, 03:16:34 am »
Maybe put the garden variety servos in a soundproof box.  You can extend the mechanical motion out through small holes in the box.

Implementing this approach will require attention to the paths of sound transmission.  Much will come through the mounting points.
 

Offline DrDisco

  • Newbie
  • Posts: 6
  • Country: dk
  • Mechatronics technician, trying to do electronics.
    • Sensa
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #14 on: November 11, 2019, 09:19:47 am »
Some small steppers with some trinamic drivers. Should work out. The trinamic drivers have a stealth chop feature. That will silence the motor right down.

They are quite expensive, but there are Chinese dev boards you can buy in the same form factor as the conventional a4988 Polulu modules:
https://www.aliexpress.com/item/32886021442.html?spm=a2g0o.productlist.0.0.479f50c2Xsl2fZ&algo_pvid=f8e16a04-8a46-4601-8c2b-a3339f76c234&algo_expid=f8e16a04-8a46-4601-8c2b-a3339f76c234-11&btsid=5e5ae011-27f9-44c6-8dcf-356585bc0018&ws_ab_test=searchweb0_0,searchweb201602_2,searchweb201603_52
 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7012
  • Country: ro
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #15 on: November 11, 2019, 09:39:24 am »
A stepper motor does not produces any noise if driven with sinusoidal current.  Most of the cheap stepper drivers does not produces sinusoidal currents, thus the stepper is humming/buzzing.  Texas Instruments has a few motor driver chips very good at producing clean sinusoidal waveforms.

However, servos are usually driven by DC motors, and gears are noisy, unless special gear profiles and special materials are used.

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #16 on: November 11, 2019, 09:47:59 am »
Things to bear in mind from the OP when he added it.

Quote
since in the project im working i need a lot of servos, around 100 or even more

Cost and complexity are a factor. As a simple PWM from most micros will drive a commercial servo it makes more sense over a stepper unless you want to spin custom boards.
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Online artag

  • Super Contributor
  • ***
  • Posts: 1284
  • Country: gb
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #17 on: November 11, 2019, 12:17:49 pm »
Another common 'linear stepper' mechanism is https://www.ebay.co.uk/itm/401829445368

Anybody know what these are used for ? they seem very widely available.
 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7012
  • Country: ro
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #18 on: November 11, 2019, 03:05:49 pm »
Looks like the mechanism used to move the magnetic heads in floppy disks.  The force was rather small, much smaller than in a small servo like SG9.

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 3382
  • Country: gb
Re: Less noisy alternative to servo. Muscle / nitinol wire?
« Reply #19 on: November 11, 2019, 04:07:49 pm »
Someone has made an RC style servo that uses Nitinol wire.  It's faster than I thought it would be.

 

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf