0 Members and 1 Guest are viewing this topic.

#### muniorbust

• Jr. Member
• Posts: 17
« on: December 06, 2010, 12:35:52 PM »
This isn't for a robot, but this forum seems to be where the experts are.  I hope someone can help me.

I'm building a special treadmill that will keep a unicycle rider in the middle of it no matter his speed.  I'm starting with an existing treadmill.
Assuming I have the sensing and processing figured out, now I need the right motor.

I'm not very knowledgeable of motors, but I don't think I can use the AC motor in my current treadmill. I think I need to PWM a DC motor.  Correct me if I'm wrong.
I've used the RMF calculator but I'm not sure I have the right values.

Mass = 160 lbs    (me + unicycle)
Desired Velocity = 22 ft/s    (15 mph)
Desired Acceleration = 11 ft/s^2    (tried to make a reasonable guess)
Expected Efficiency = 75%    (just used the default)
Incline Angle = 0    (I'd actually like to have it incline, but my riding speed will drop, so I think 0 is what I want here)
Wheel Diameter = 0.5    (There's a pulley on the treadmill's front roller. That's what I put here, right?)
# of powered wheels = 1

RMF = 8220
Motor Rotation Speed = 840 rpm

A motor with these specs should work:

587 lb*ft
840 rpm

Does that even fall into the realm of motor possibilities?
Where would I get a motor with this kind of RMF?  Or how do I change my numbers to make my project work?

Thanks!
Buzz

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #1 on: December 07, 2010, 12:14:55 PM »
Looking at the values I'm using above, I can't change any except Desired Acceleration.
And changing acceleration has a big impact in RMF.

How do I determine Desired Acceleration?
Just measure the distance I can go from a standstill in 1 second?

I hope someone will offer me some help!
Buzz

#### VegaObscura

• Robot Overlord
• Posts: 153
« Reply #2 on: December 07, 2010, 07:31:14 PM »
I have a feeling your calculations are a bit off.

I guarantee you do not need 587 lb-ft of torque.  To put that into perspective, a 2005 Corvette produces 424 lb·ft of torque.  I do hope you're not planning on riding a unicycle faster than a Corvette can drive.

When I'm doing force and torque calculations I like to convert all units to metric and always round to the nearest whole number.  It just makes things a lot easier and leaves less room for mistake.  Do check this and make sure I don't make any stupid mistakes.

160lbs = 73kg

11 ft/s^2 = 3m/s^2

I wouldn't worry about your desired maximum velocity; if you can do this acceleration, then you can do that velocity easily.

Force(in newtons) = mass(in kg) * acceleration(in m/s^2)

Plugging in our numbers gives us

F = 73kg * 3ms^2

So F = 219Newtons.  This is the force we need to produce.  But what we need isn't a force in newtons, its an amount of torque.  So we multiply this force times the radius (in centimeters) to give us our required torque (in newton-cm).  Remember, you measured the wheel's diameter, so taking half of that will give us its radius.

0.25feet = 8cm

T = 219 * 8

T = 1752n-cm.  This is the torque the motor needs to produce.  Converting newtons to pounds gives us:

1752n-cm = 384lb-cm

Then converting cm to ft gives us:

13lb-ft

But of course, this is ignoring friction and inefficiencies.  I would recommend doubling it for a good number.  So the final torque you would need should be approximately

26 lb-ft

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #3 on: December 07, 2010, 07:57:45 PM »
Awesome!  Glad I don't need that Corvette engine!

Thank you so much for your help.
I'm going to look over your numbers over the next day and make sure I understand them.

As far as the existing treadmill motor, I forgot to mention one thing.
I want to be able to make the motor go forward AND backward, because I'd like to ride backward.
With that added requirement, can I use the AC motor?

(Also, my treadmill is cheap  and has a manual speed mechanism.  You crank a handle, which tightens or loosens a cable, that slides the motor along two rails.  The pulley on the motor is a special spring loaded split design, so the v-belt it drives moves deeper into the pulley or back up as the motor position changes.  This effectively changes the motor pulley's diameter, and thus the gear ratio and speed.)

#### Soeren

• Supreme Robot
• Posts: 4,672
« Reply #4 on: December 07, 2010, 08:06:37 PM »
Hi,

I want to be able to make the motor go forward AND backward, because I'd like to ride backward.
With that added requirement, can I use the AC motor?
That depends on the motor, some AC motors are reversibel (if you make separate connections to the field coil). What markings are on the motor?

(Also, my treadmill is cheap  and has a manual speed mechanism.  You crank a handle, which tightens or loosens a cable, that slides the motor along two rails.  The pulley on the motor is a special spring loaded split design, so the v-belt it drives moves deeper into the pulley or back up as the motor position changes.  This effectively changes the motor pulley's diameter, and thus the gear ratio and speed.)
Once, there was a car using that principle (I think it was the NSU Prince, but it's but a foggy memory).
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Engineering is based on numbers - not adjectives

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #5 on: December 07, 2010, 08:34:43 PM »
The motor in my treadmill has a plate on it that says:

GE Model 5KC46HR166
HP 3/4
RPM 1725
v115
CY 60
A 11.0
Code 901M
PC B1070

#### VegaObscura

• Robot Overlord
• Posts: 153
« Reply #6 on: December 07, 2010, 09:41:30 PM »
As far as the existing treadmill motor, I forgot to mention one thing.
I want to be able to make the motor go forward AND backward, because I'd like to ride backward.
With that added requirement, can I use the AC motor?
Probably not.  Since treadmills don't normally have any reason to go in reverse, its unlikely that the company that made it invested in a reversible AC motor.  But do keep in mind if you have to replace this motor, its probably going to cost you into the hundreds.

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #7 on: December 08, 2010, 12:46:57 PM »
Yeah.. I kind of expect \$100's.

This self-balancing unicycle uses an Ampflow A28-150 motor (\$300) and an OSMC speed controller (\$200).  I figured my unicycle treadmill is more or less equivalent to a self balancing unicycle...  only my motor is moving the ground, not the wheel (and of course I'll be doing the balancing)

To calculate the torque of the A28-150 motor, I googled and found Torque in ft-lbs = (horsepower / RPM ) * 63025

So for the A28-150:
T = (3 / 6000) * 63025 = 31.5 ft-lbs

That's more than the 26 ft-lbs requirement above, so that's good.

Any suggestions on where to purchase a motor and H-bridge controller?
If I can save some \$ for a motor that is NOT robot battle-ready motor I'd prefer that.

Thanks for the help thus far!
Buzz

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #8 on: December 08, 2010, 09:39:49 PM »
For the A28-150 the specs shown are:

3" Diameter, 4" long
1/2" diameter shaft with 1/8" keyway, 1.75" long
24V (can be run higher)
3 horsepower
1970 oz-in Torque
Max current 285 Amps
82% Efficiency
6000 rpm
3.8 lbs
Neodymium magnets

I calculated
T = (horsepower / RPM ) * 63025
T = (3 / 6000) * 63025 = 31.5 ft-lbs

But the torque is already shown as 1970 oz-in.
Converting to ft-lbs = 1970 / 12 / 16 = 10.3 ft-lbs

Huh?  Why are these so different?

#### Soeren

• Supreme Robot
• Posts: 4,672
« Reply #9 on: December 08, 2010, 10:03:22 PM »
Hi,

The motor in my treadmill has a plate on it that says:

GE Model 5KC46HR166
HP 3/4
RPM 1725
v115
CY 60
A 11.0
Code 901M
PC B1070
I cannot find any data on it, or pics.
Is it open, or can it be opened? And are you up for trying to convert it?

If not, at least it tells you what you need (if it is strong enough for what you want it to do).
The 3/4 hp is about equal to 775*3/4 = 581W, so you need to find a roughly 600W (output) motor going 1725 RPM to match the one you have.

Your motor is spec'd for 115*11 =1,265 VA ("W") of input, so is around 50% efficient.
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Engineering is based on numbers - not adjectives

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #10 on: December 09, 2010, 07:58:03 AM »
I cannot find any data on it, or pics.
Is it open, or can it be opened? And are you up for trying to convert it?

It is not currently open but I'm open to the idea.

I think it has enough torque (though as I showed in my previous post, I'm not sure I'm calculating torque properly):
T = (3/4 / 1725) * 63025 = 27.4 ft-lbs

So, the first step is to see if the motor can be opened?

Quote from: Soeren
some AC motors are reversibel (if you make separate connections to the field coil)
I'll have to read about this when I get a chance.  In theory, with the motor open, what am I trying to do?

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #11 on: December 09, 2010, 07:59:30 PM »
I looked at my motor.  It looks like it has 4 long bolts going straight through.  It also has a large hump-shaped cover on the side attached with screws.
It looks a lot like this one:

So, it sure looks like it can be opened.  Now what?

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #12 on: December 10, 2010, 11:57:08 AM »
Soeren, what do I need to do first?  I can open my motor and take pictures to help you determine if it can be reversed.
If you (or anyone on these forums) are going to help me, more or less step by step, let's do it.
Otherwise, I won't have a clue.

I'm generally a good DIY guy, and have opened and cleaning motors before.  I'm sure I can handle the work, if I knew what to do.

And just to make sure you understand, my goal here is to make the motor to be controlled both forward AND reverse on command, controlled by some controller.

I'm looking forward to getting started.
Buzz

#### Soeren

• Supreme Robot
• Posts: 4,672
« Reply #13 on: December 11, 2010, 06:52:20 AM »
Hi,

Busy month (socially) and my PC (laptop) suddenly dies under me without warning (about 20 times a day or more), so I don't get that much done presently.

Soeren, what do I need to do first?  I can open my motor and take pictures to help you determine if it can be reversed.
Opening it is the first step, to see if it's possible at all - also look under the bump on the motor - might have terminals for direction. Sharp focussed close-ups will help.

And just to make sure you understand, my goal here is to make the motor to be controlled both forward AND reverse on command, controlled by some controller.
Do I sound really slow?

Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Engineering is based on numbers - not adjectives

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #14 on: December 11, 2010, 07:49:35 AM »
Do I sound really slow?
Ha...  just trying to avoid a "why didn't you say so in the first place" ending.

I skimmed it but will need to read it more carefully.  Hopefully I'll find the time for pics too.
Thanks!

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #15 on: December 11, 2010, 03:13:07 PM »
Here are some photos of the motor with the panel open.  I also opened the capacitor hump and sure enough there's a capacitor.

If I open the motor itself, do I need to worry about that capacitor discharging on me?

I'll wait for further comments/instructions before doing anything else.

#### Soeren

• Supreme Robot
• Posts: 4,672
« Reply #16 on: December 12, 2010, 12:29:51 PM »
Hi,

Given it some further thoughts, I think you won't be able to use it for your purpose, as the inertia of the motor will make it slow in changing direction.
You would be able to make it reversible though.
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Engineering is based on numbers - not adjectives

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #17 on: December 12, 2010, 05:33:33 PM »
Ok thanks.  So, I'll need to buy a motor and a motor controller that my Propeller can send PWM to.

I mentioned this before:
This self-balancing unicycle uses an Ampflow A28-150 motor (\$300) and an OSMC speed controller (\$200).

Any suggestions on where to purchase a motor and H-bridge controller?
If I can save some \$ for a motor that is NOT robot battle-ready motor I'd prefer that.
Any suggestions?
Buzz

#### muniorbust

• Jr. Member
• Posts: 17
« Reply #18 on: December 13, 2010, 12:31:38 PM »
I'm eyeing the NPC Black Max (NPC-4200) motor.  It's only \$5 more than the Ampflow A28-150 Motor but has a mount already attached.
It weighs a lot more (15.5 lbs versus 3.8 lbs), but that's fine for my application.

Its torque is significantly more too.

Either way this OSMC speed control seems to be the one I'd need.

I think I've gotten used to the idea that this will be costing me some serious cash.  But if anyone has another suggestion, let me know.

Oh, and I have no clue about a power supply yet.  I guess I need an AC to DC power supply that can provide 24v or 36v and the right current.
Any help?
Buzz