next topic »

[on:]

hello, this is kind of a large question and i would appreciate if any one who answered it could be as specific as possible.

I'm a freshman college student and for my job my boss asked me to make a small motor array made up of 24 servos.  For this project i need to be able to control each motor individually (via a laptop of something along those lines) and make them turn both forward and backwards.  Also, i want them to only rotate at about one revolution a second or less, and i want to try and keep the prices relatively low even though i do realize that this will cost probably a couple hundred dollars.  I was thinking that i would need an h bridge, an arduino micro controller, servos, wires, and some batteries.  Unfortunately, i don't have a whole lot of experience with this kind of thing and i realize i probably don't know of all of the things i will need.  Also, I'm not exactly sure How many motors i can hook up to an arduino micro controller and h bridge or even if i necessarily need the h bridge.

I would really appreciate it if someone could give me a list of good products to buy to build this array at an affordable price( and the quantity of each part) while also helping me determine how best to set this up. 

(shortened version: i need to control 24 servos individually and wirelessly while having the ability to adjust the speed of rotation forward and backward and i need a list of components that i will need and preferably some info on good products, vendors, and manufacturers)
Also, i need to waterproof it but i feel like that's something I'll have to do separately.

If you have time to help me out thank you very much.

[Reply #1 on:]

Do you really mean servos, they usually only rotate 180 degrees ?
Do you need full rotation or just, less than 180 degrees back and forth movement?

H bridges are usually used for DC motors not for servos. If what you really need is DC motors you will need geared motors to get 1 RPS, an you will need one H bridge for each motor if you need individual control over speed and direction. Geared motors that go down to 1RPS = $$$

Do you need individual speed/direction control over the 24 motors?

Wireless control over what distance ?

When you say "waterproof it", will it be submerged?, because that will have an influence on wireless options.

Have you considered your powersupply needs?, if what you really need is DC motors, you are going to need a beefy powersupply to run 24 of them.

[Reply #2 on:]

Go back to your boss and find out what this thing is in tended to do. At the moment your trying to get us to help you implement a solution (24 motors with Wifi).

My first pass at a design to do this is 6 Uno's linked using the built in I2C with each Uno controlling two dual motor controller shields (4 motors). One of the Uno would be the master and would have a wireless/SD or XBee shield with a WiFly bug used in place of an XBee.

so if I where to give your boss a quote for the hardware it would look something like

12 x Dual motor controllers @ 26 GDP each = 312
6 x Arduinos UNOs @ 22 GBP each = 132
1 x wireless/SD shield @ 21 GBP
1x WiFly @ 32 GBP

total so for 479 GBP
plus 24 motors @ who knows what as we don't know what your trying to do. Plus software design.

plus and extra 20% for bits and bobs.

Mark

[Reply #3 on:]

Why not just this,http://www.ebay.com/itm/USB-UART-32-Channel-Servo-Motor-Controller-Driver-PWM-For-Arduino-Robot-A130-/251192458425?pt=LH_DefaultDomain_0&hash=item3a7c3cbcb9

Its a little expensive but it will do what you want it to.

[Reply #4 on:]

I do know all of the specifics of the project that i should be able to answer any specific questions as you're trying to help me.

so the array is going to be placed in an high grade incubator(due to really high humidity water accumulates on the parts)  meant for cell cultures and things like that.  So it has to be able to transmit through a metal housing but past that only about 5 feet

Also, as i said before i need it to be able to rotate really slowly while being easily controlled.  These motors have to be very small (servo was the only thing i could find that was small enough but a dc motor that size would be fine.)  these motors will be rotating a rod with a ball bearing at the end in order to roll along a prefabricated rail and all 24 motors will be going along their own rail,i only need to to be able to maybe turn somewhere around 720 degrees in either direction since the rails are pretty short.  This is the reason why i need pretty precise control over the motors and also why i have to be able to control each one individually otherwise they  would shoot off the rails.  

Also, i was wondering how many motors i could hook up to an arduino, i saw one guy do 10 by using a breadboard somehow and i think minimizing the number of arduinos could help cut back on costs.

As you can see i don't really now a whole lot about servos and motors but i thought i would have to just wire some batteries in there to help run it.

That 32 servo controller also looks like it could be a great help if it turns out that i can use servos thanks.

[Reply #5 on:]

@HazardsMind That looks good! Much more so, as sniperguy455 now tells us that its for a high grade incubator!.

But now we know what it is we are trying to do.

So why 24 motors and not just one and a few cogs, shafts and gears!

So it has to be able to transmit through a metal housing but past that only about 5 feet

No can do! A metal housing forms  a Faraday  cage through which wireless signals cannot pass. So the WiFi part need to stay outside the box!. But that should not be a problem.

Mark

[Reply #6 on:]

Do all of the servos need to run at the same time or could motion be limited to a few at once?  This matters a LOT in terms of selecting a power supply. 

Using standard hobby servos is going to save you LOTS of work with separate motor controllers, so if I were you I would try to find a suitable continuous or 720 degree hobby servo. 

Pololu sells three options near the middle of this page

http://www.pololu.com/catalog/category/23

ServoCity also has some decent options (One would think they should):

This is a 3.5 rotation positional servo.
http://www.servocity.com/html/hs-785hb_3_5_rotations.html

This is a factory modified continuous rotation servo.
http://www.servocity.com/html/hsr-1425cr__continuous_rotatio.html
Unfortunately its out of stock, but if you planned on buying a lot of them they might try to fix that.

If you find a good positional servo that you want to make continuous rotation it can be done, but its not trivial.  Servo city has some info on that here.
http://www.servocity.com/html/continuous_rotation_informatio.html

Lastly while you are at ServoCity take a look at their other hardware.  Their servo gears, hubs, and sprocket chain is quite useful stuff.

[Reply #7 on:]

jroorda, that seems to be exactly what i will need for the servos thanks I think SpringRC SM-S4303R Continuous Rotation Servo looks good.

also, going back to holmes4 it is 24 servos because they will be moving eight different objects in an xyz area using the aforementioned rails therefore with the basic design i need 3 motors per object, one for each axis of movement.  I could use less servos but my boss wants each object to be modular so he could add more if he wanted so i can't use less motors for the sake of efficiency. 

When using this array i only need to be able to control one servo at a time when it comes to the power so that shouldn't be much of a concern.

Going back to the transmitting out of the incubator it looks something like this though I'm not sure of the exact model (http://www.priceit.in/instruments/co2-incubator-price-features-reviews/) what are your thoughts as to getting the controls from the array in the incubator to the laptop outside 5 feet away.  Holmes4 you said something about a Faraday cage and I'm wondering if this would pose a large problem or if i could get around it.

Also, to review it looks like I'll need 24 servos, 1 of those 32 channel servo controllers, an arduino, the wireless thingamajigs.  Also, do i need any h bridges for these or is that only for dc motors.  also do you think a breadboard might help out as I'm building this thing.

[Reply #8 on:]

You don't need an H-bridge to drive a servo - just a means to generate the appropriate PWM pulse stream. That's the big difference between a servo, and a plain old DC motor.

[Reply #9 on:]

Holmes4 you said something about a Faraday cage and I'm wondering if this would pose a large problem or if i could get around it.

A metal box stops radio waves in your case you can deal with the problem by take the wirs from the motors through as hole in he case to the control electronic outside the metal case. Then seal the hole (think grommets/or even putty).

You can make a "faraday cage" from tin foil. Try wrapping your phone in tin foil and have some one try calling you.

Instead of saying that you want to move the shelf in all 3 axis find out what patten of movement is required. You may be able to use just 1 motor and a cam or two.

Mark

[Reply #10 on:]

well now that we got most of the motor related things figured out, what exactly do i need to wirelessly send signals back and forth from a laptop to an arduino? Also, if anyone knows of a video on how to hook that kind of thing up I'd really appreciate a link.

[Reply #11 on:]

There are a number of options, look at the products section!.

1. Wifi Shield (Warning this is not compatible with the standard motor shield or with the standard servo lib.

2. Wireless/SD shield with a WiFly module (Google). If you use this with a motor shield you can't use the the SD card feature.

3. XBee shield (Zigbee) with WiFly module.

4. 2/3 above using XBee. To use and XBee you need a second attached to the PC.

Note that the motor shield and the standard servo lib are not compatible.

Options 2/3 above allow you to go to remote operation from control over the USB without code changes.

Mark

[Reply #12 on:]

You can try Bluetooth, wifi, RF, IR or Xbee. They all have tutorials.

[Reply #13 on:]

alright thanks for everything