PDA

View Full Version : Motors and leadscrews



Tom
13-05-2009, 10:56 PM
It's time to start thinking about motor power and leadscrew pitches for my machine.

What's the best way to approach this?
I've seen people talk about target speeds, target accelerations (0.5g? 1g?), gantry weights, cutting forces, motor curves, etc.

I should say that I'll be using trapezoidal leadscrew, which will have a much lower efficiency than ballscrews. Before now I've been thinking about 3Nm motors, but perhaps I don't need all that power.

Practically speaking, having never done this before, where is the best place to start?
Acceleration seems more important than top speed for the relatively compact routing I'm going to do (many direction changes), but what should I aim at? What do you lot use?

Speed is much easier - I'd ideally like to achieve a jog speed of 120mm/s (~280 IPM), which is 900rpm on an 8mm pitch leadscrew. Can steppers do that speed? I'm thinking cutting speed will be less closely linked to motor power, more to router power, chip load, and machine rigidity.

Cutting force seems important too, but what figures should I use? Or should I worry about it less and just bolt some stuff together?

Kip, Smiler, Lee,
What do you think? How do you do it? :question::biggrin:

Lee Roberts
13-05-2009, 11:13 PM
Hi, the other more qualified guys will be along to give you the right methods’ of doing the above.

What I can tip you with however is, motors with a lower inductance communicate faster/better with the driver cards, what this means is you get better speeds from the motor when changing direction and so on.

The trade off seems to be finding a motor with low inductance that has a high holding torque.

Tom
28-05-2009, 09:58 PM
Thanks for your help Kip,

The build's coming along nicely (so more pics soon)...

I'll fudge the motor power together using 3NM motors I think... At least I'll have some decent parts to move over to machine #2!!

It really is time to think about leadscrews now. What about leadscrew whip? My X axis screw will be 1600mm long (63"). I was hoping to get away with 16mm (or 18mm) diameter. What do you recon?

Robin Hewitt
28-05-2009, 10:48 PM
Accelerations are a consideration when you slam it in to reverse.

I'd say find the top speed where you can reverse without losing steps, halve it, then apply accelerations/decelerations above that point.

You'll probably find you only need them for the G00's a stepper motor makes a pretty good brake :biggrin:

Tom
29-05-2009, 05:54 PM
Robin, thanks for your info. If I understand your approach it's more aggressive than I imagined would be possible - I was planning to use accel and decel values for all starts and stops (can be set in EMC2, not sure about mach3). Just to check my understanding, you only find it necessary to apply accel and decel above a certain speed, and below this speed you can go from forward to reverse with no ramping?

I might try for 19mm or 20mm then. But belts, - I haven't designed the machine for those at all... :nope:

Kip, is this the "patented" design of which you speak? (http://www.bell-everman.com/servobelt.html) :naughty: Seems quite clever, but can you really patent that?

I did a search for Mike Everman actually. He seems to be everywhere. One of the threads lead me to this (start of geek voice) very interesting (end of geek voice) looking book. (http://www.amazon.com/Engineering-Formulas-Metalcutting-Edmund-Isakov/dp/0831131748/) Would probably answer all my questions... :geek:

Tom
29-05-2009, 06:02 PM
Would probably answer all my questions...

Hmm, maybe not. Having had a read online its more about the cutting load than the inertia...

Gary
29-05-2009, 06:06 PM
To work out critical speed use this calculator.
It works very well.
All you need is the root diameter, length of the screw and how you are supporting it and it will work out the max speed.

http://www.nookindustries.com/ball/BallCalculators.cfm#CriticalSpeed

Tom
29-05-2009, 06:25 PM
Thanks Gary, that's really useful.

Just for my future ref:
RPM Root Dia (mm)
600 12.7
700 14.8
800 16.9
900 19.1
1000 21.2

John S
29-05-2009, 07:35 PM
Problem I see is one of resolution.
Take the smallest pulley you can use without undue belt wear, say 20 teeth for round figures and you are running on 5mm pitch belting so one rev moves 100 mm.

Now a standard stepper moves 200 steps per rev so one step will equal 0.5mm or 0.0197" say 20 thou per step.

Dropping to 10 microsteps which is about the max usable without shedding oodles of torque will reduce this to 0.05mm or 0.00197" or 2 thou per step.

Without another belt reduction in the system somewhere before the 20 T pulley or using servos with high count encoders [ which will also reduce speed ] this system has limitations, patent or not.

.

Gary
29-05-2009, 09:04 PM
You mentioned using leadscrews, is this for price?
A leadscrew is not that efficient and you should expect between 40-60% for a lead screw and up to 90% for a ball screw.
This means that with a leadscrew you will need up to 50% more power to get the same amount of force.
So, because of this you may be saving on the leadscrew but will need to spend more on the motor and driver.

Gary



It's time to start thinking about motor power and leadscrew pitches for my machine.

What's the best way to approach this?
I've seen people talk about target speeds, target accelerations (0.5g? 1g?), gantry weights, cutting forces, motor curves, etc.

I should say that I'll be using trapezoidal leadscrew, which will have a much lower efficiency than ballscrews. Before now I've been thinking about 3Nm motors, but perhaps I don't need all that power.

Practically speaking, having never done this before, where is the best place to start?
Acceleration seems more important than top speed for the relatively compact routing I'm going to do (many direction changes), but what should I aim at? What do you lot use?

Speed is much easier - I'd ideally like to achieve a jog speed of 120mm/s (~280 IPM), which is 900rpm on an 8mm pitch leadscrew. Can steppers do that speed? I'm thinking cutting speed will be less closely linked to motor power, more to router power, chip load, and machine rigidity.

Cutting force seems important too, but what figures should I use? Or should I worry about it less and just bolt some stuff together?

Kip, Smiler, Lee,
What do you think? How do you do it? :question::biggrin:

Tom
29-05-2009, 10:37 PM
Hi Gary,

I hear what you're saying. This build needs to be pretty cheap, so in the short term I'll go with leadscrews, and just take what performance I can get. My budget means I don't really have the option to upgrade the motors. As Kip suggested - just build it and see what the results are. If it's slow, that'll help me justify the extra cash next time around.

Since its only an MDF build it's unlikely I'll be using the router and screws years from now, whereas the motors will definitely find another job to do....