High Lead screw Motor selection.

[forseral]

Hello all, long timer lurker first time poster here.

Having built a prototype machine to learn, I'm now working on my second and more permenant machine. I'd like to ask a few questions since I cannot seem to find an

answer I understand amongst all the posts.

This question has been asked 100 times before but i must ask again. How do I select a strong enough stepper motor (and kit).

My current setup is a moving gantry with the following:

-Dual X-axis lead screws - 1/2-8 (8 start)
-Y ball screw - 3/8-8tpi (1 start)
-Dumpster CNC anti backlash nut
-20mm supported linear rails from glacern (72" x 36")
-50 pound gantry
-Bosch 1617 1/2 collet router
-Probotix 4axis breakout board w/ probostep drivers and 24v power supply
-ubuntu w/ EMC


I plan on cutting wood including exotics. Given the machine size, I would like to reach rapids of at least 120ipm (hence, 1tpi lead screws) tho I'm less concerned

about the Y axis since it's so short and I have a ballscrew.

I used the at Torque and linear motion formula to find out that need about 942oz-in pullout(dynamic?) torque.

((weight)(inches/revolution))/2(pi)(efficiency).

(50*1)/2*3.14 * .75 = 58.875lb-in = 942oz-in.

Am I completely off base here? That seems a bit excessive - what am I missing? Or is it because one revolution has to move 50 pounds for an inch?


---------

Either case, I'm willing to pay for decent sized motors as I prefer to focus using the machine more than tweeking it. So, all that to ask:

1)Is my calculation correct? If not, what should I be using. If so, then is it because of the "high lead" lead screw

2)Given answers for #1, can anyone make a motor suggestion?

3)I've seen a lot of great reviews for the geckodrives. Given my need to "set and forget" (or as close to possible), is it worth purchasing the g540 or other gecko drive?

[dgates80]

If that calculation is right -- and my initial feeling is that it is -- then you are going to need some beefy motors that will pull a fair bit of current. The G540 at 3.5A may not cut it, I think other Gecko's can handle higher currents though.

Don't listen to me though, I am a CNC idot that doesn't know nothin'. I did stay at a Holiday Inn Express last night (since I spent all my money on building the machine....)

[ger21]

Given the machine size, I would like to reach rapids of at least 120ipm

120ipm isn't what I'd call rapid.

With a Xylotex, 24V, 250oz motors and 1/2-8 2 stat, I can get 190ipm on my Y axis. Whipping limits my 60"+ X screws to 150ipm. So, in theory, you should be able to get much faster than that, provided the screws don't whip.

Here's some more info.
According to this catalog, it takes 1.7 oz-in of torque to lift 1 lb.
http://www.nookindustries.com/pdf/NookInchAcmeScrew.pdf

You're not lifting it, you're pushing it, which requires much less. But, you're also accelerating it, which requires more.

Before they started using R&P, some used 1/2-8 start screws on Joe's Hybrids, with the same 300-380 oz motors they use for rack and pinion.

1" per motor rev is the same as most are currently doing with their R&P setups. The difference is you're losing about 25% due to inefficiency.

My guess is with a G540, 380oz motors and 48V, you'd get at least 300ipm on the X axis. Probably less on the Y. Just a guess, though.

[forseral]

Originally Posted by ger21 120ipm isn't what I'd call rapid.

haha, yea, true - when i posted, i forgot to revise that to a higher number. tbh, i'll take as high as rapids as I can get.

Originally Posted by ger21 You're not lifting it, you're pushing it, which requires much less. But, you're also accelerating it, which requires more.

I guess the part about which I'm confused is the fact that equation states it takes so much more torque to move an 8 start screw than a 2 start screw. What's the difference?

Originally Posted by ger21 Before they started using R&P, some used 1/2-8 start screws on Joe's Hybrids, with the same 300-380 oz motors they use for rack and pinion.

thanks for the facts. this tells me I should probably just go for it! =)

[forseral]

Originally Posted by dgates80 I did stay at a Holiday Inn Express last night (since I spent all my money on building the machine....)

yea... and I was hoping gecko would get confused w/ geico and give me 15% off the g540

[ger21]

I guess the part about which I'm confused is the fact that equation states it takes so much more torque to move an 8 start screw than a 2 start screw. What's the difference?

More threads per inch gives you more mechanical advantage.

[forseral]

Originally Posted by ger21 More threads per inch gives you more mechanical advantage.

Do the 8 starts not count for anything?

[ger21]

Yes, it does.

I'm not a screw expert, but basically, lead is more important than starts. As the lead increases, mechanical advantage goes down, but efficiency goes up.
A single start screw with a lead of .125" might be 35-40% efficient.
While an 8 start screw with a lead of 1" is about 75% efficient.

So while you might need twice as much torque to move the same load, only half as much torque is being lost to inefficiency.

[arizonavideo]

The way I look at the Acme screw is the friction is about constant but a high lead will move two or three times further for each turn. Combine that with the fact that steppers loose power at high RPM so spinning a screw fast will deliver less power to the load so high lead screws will deliver more power to the tool at speed.

I would go with the 8 start.

[jsheerin]

The calculation you show is just some approximation / simplification of reality. To really figure it out, you need to know what the cutting force will be which is tough to know without measuring yourself (but 50 pounds is a good approximation), you need to know what the acceleration and mass of your moving components is, and the efficiency of your drive system. Basically you use F=m*a to calculate how much force is required to accelerate your machine at your chosen acceleration. Then you add that to the cutting force. Then you calculate what the input torque to your screws needs to be (accounting for efficiency). From here, you also need to know what the rpm of your screws will need to be and then find a motor with a torque curve that provides you with the torque you need at the max rpm you need.

However as Gerry said, with the high lead screws, G540 and ~380oz*in motors you will probably be fine unless you want really high performance. In that case you could go for higher torque N34 motors, bigger motor controllers and power supply, rack and pinion on your long axis, and possibly a larger diameter ballscrew on your shorter axes to allow higher speed without whipping.

[Boomslang]

Originally Posted by jsheerin The calculation you show is just some approximation / simplification of reality. To really figure it out, you need to know what the cutting force will be which is tough to know without measuring yourself (but 50 pounds is a good approximation), you need to know what the acceleration and mass of your moving components is, and the efficiency of your drive system. Basically you use F=m*a to calculate how much force is required to accelerate your machine at your chosen acceleration. Then you add that to the cutting force. Then you calculate what the input torque to your screws needs to be (accounting for efficiency). From here, you also need to know what the rpm of your screws will need to be and then find a motor with a torque curve that provides you with the torque you need at the max rpm you need. However as Gerry said, with the high lead screws, G540 and ~380oz*in motors you will probably be fine unless you want really high performance. In that case you could go for higher torque N34 motors, bigger motor controllers and power supply, rack and pinion on your long axis, and possibly a larger diameter ballscrew on your shorter axes to allow higher speed without whipping.

Sorry to budge in, but does anyone have a spreadsheet they would like to share that will calculate the above? (too lazy..... )

[jsheerin]

Your laziness somehow motivated me to post some of the sheet I use. Yellow is input, green is output (generally). I don't guarantee accuracy of anything, but especially the buckling stuff towards the bottom.

Edit - old file removed - see below for a newer one.