Anti Whip solution for long screws?

I posted this on another thread in the Open Source forum... but I don't think it is getting much visibility there. I am really curious about peoples thoughts on this. Do you think it would work?.....

I'm going to be using a 1/2inch acme which will be 55 inches long. I'm worried about whipping as well. I devised an idea to stop the whipping, not sure if someone has used it before.

Basically... the screw will have several spring loaded doors on each side.

The hinges will be double action such as this... they bring the door to the "closed" position, but let it swing both directions.

http://www.hardwaresource.com/Store_...ts.asp?Cat=118

Planning to use a 1/2ID teflon washer of some sort in between and cut in half.

Not sure if it will work, and is just a contingency plan, so I may never know.

nice idea, but there's more than just the nut on the screw, there's the thing that bolts to the nut and the structure surround. makes it all much harder to get this all happening. the easiest way to avoid whip of the lead screw may be to change to the solutions that dont use one.

I'm wrestling with the same idea. I'm using ground 3/8-10 acme screws on a 44" long router. I'v ebeen thinking along the same lines. Maybe a "single sided" saddle on a swing-away lever, spring loaded to keep it engaged? It would solve some of your clearance/get-out-of-the-way issues but I'm not sure it would be very effective since it would only be supporting the screw on one side (the gravity side).

Either use a screw that is big enough that it wont whip or turn the nut and have the screw fixed.
In your case (using small screws) best way is to put in big screws.

I believe you can also tension the screw by putting an end nut on the opposite side of where it is mounted, which will increase your effective RPMs in the screw.

Quote:

---

Originally Posted by georgeburns

I posted this on another thread in the Open Source forum... but I don't think it is getting much visibility there. I am really curious about peoples thoughts on this. Do you think it would work?.....

I'm going to be using a 1/2inch acme which will be 55 inches long. I'm worried about whipping as well. I devised an idea to stop the whipping, not sure if someone has used it before.

Basically... the screw will have several spring loaded doors on each side.

The hinges will be double action such as this... they bring the door to the "closed" position, but let it swing both directions.

http://www.hardwaresource.com/Store_...ts.asp?Cat=118

Planning to use a 1/2ID teflon washer of some sort in between and cut in half.

Not sure if it will work, and is just a contingency plan, so I may never know.

---

You would want to use a rub block, anything threaded or thin enough to get into your all thread would have a tendancy to want to travel in the direction the threads are winding. A rub block would just glide on top of the threads providing pressure to counteract the whipping. Also if you had a gated arrangement that you are depicting i'm guessing if you didn't leave alot of play it might bind when the drive nut comes back in the opposite direction or worse yet if you didn't drive completely past the gate it might totally jam.

Quote:

---

Originally Posted by wyldesyde007

You would want to use a rub block, anything threaded or thin enough to get into your all thread would have a tendancy to want to travel in the direction the threads are winding. A rub block would just glide on top of the threads providing pressure to counteract the whipping. Also if you had a gated arrangement that you are depicting i'm guessing if you didn't leave alot of play it might bind when the drive nut comes back in the opposite direction or worse yet if you didn't drive completely past the gate it might totally jam.

---

Very true... didn't think of that. If the nut went halfway through the saloon doors and then changed direction it would really screw things up.

Perhaps using rounded edges on the doors as opposed to right angles would resolve this. By rounding the edges... the nut would either be not through or completely past the doors. And hopefully never in between.

Here's a thought for you (hope it's clear enough I did this quick)
it's simply a moving support with veeery few parts!;)

Dear georgeburns,

You can buy yourself an awful lot of mileage if you make each end of your screw rigidly "in-line" by using dual bearings (spaced apart) at each end. If you can tension the screw as well, you get even better performance.

Nook know all about this...

http://nook.texterity.com/nook/linea...ide/?pg=5&pm=1

The good stuff is on page six and after.

Best wishes,

Martin

I agree with Martin. I'm using 1/2 inch Acme threaded rod on my JoeCNC 4 x 4 Hybrid mill that is 60+ inches long. I ran into whipping problems at speeds over 240 ipm. I modified the end supports to use 2 bearings to create a fixed end condition and now I can traverse at the maximum speed my stepper motors will drive the machine (over 300 ipm).

http://www.cnczone.com/forums/attach...1&d=1213370524
This graph shows the speed at which whipping starts (1st critical speed) for a 1/2" solid steel shaft of various lengths. This is similar to the ACME screws on Joe's design. The rpm's you can run at drastically reduce with shaft length. The black line on the chart is for a shaft that is simply supported like Joe's. The blue line is for a shaft that has fixed supports on the end. When you add fixed support the rpm's required to reach the 1st critical speed are greatly increased. At a length of 56 inches, the critical speed for the fixed support shaft is more than double that for the simply supported shaft.

I modified the screw supports on my Joe's by adding an additional flange bearing on one end and a Dumpster coupling between the motor shaft and the ACME screw on the other end. This modification makes the ACME screws behave like the blue line in the graph. Here is a movie showing the screw at high speed before and after modifying the end support: [ame="http://www.youtube.com/watch?v=dDk88vqzuQY"]YouTube - Shaft Critical Speed Demonstration[/ame]


Here is a picture of the dual flange bearing support at the end of my y axis.
http://www.cnczone.com/forums/attach...1&d=1213370524

Thanks for posting picture, video and nice chart..please till me what do u mean by fixed end shaft and free end shaft?

A free end shaft has 2 degrees of freedom plus rotation. For instance, if you insert one end of an Acme screw in a flanged end bearing and hold the other end of the shaft you will be able to freely move it in 2 directions.
http://www.cnczone.com/forums/attach...1&d=1213378362

A fixed end shaft only allows rotation. If you add a second flange bearing as shown in the photo above, you will no longer be able to move the free end of the screw without bending the shaft.
http://www.cnczone.com/forums/attach...1&d=1213378362

Dear erniebro,

It's probably worth having a similar type of arrangement where the lead screw passes through the moving gantry. For example, you could have a two leadnuts on the gantry spaced a few inches apart. The two could form an anti backlash system at the same time as keeping the screw in line.

BTW, I'm sure it's been done by loads of people..

Best wishes,

Martin

I'd think it would be less work to start over with a larger screw than design a fix to doctor a problem. I have found the fix itself comes with its own set of problems.

I had an opportunity to see a very, very larger Thermwood 5 axis router with a very similar screw support design that supported the 2"+ x 30-40'(?) high helix ball screws. The screws were stationary and the servos ran with the gantry turning the nut. The bottom of the gantry structure had a cam shape where the servos were mounted. The screw supports had cam followers that rode on the cam shape to handle a reversal in direction without interference. I get the sense this worked for this application for 2 reasons. First, the screw did not turn and second, there is far more flat surface on the high helix ballscrew than a standard threaded rod.

The problem you may gain is wear from a screw cutting into the support surface.

DC

erniebro :) thanks for explaining the concept:)..I am visualizing another idea..I will use C type half nut of teflon mounted with some retractable mechanism at the bottom of my Y-slide at a distance of 6" from my lead screw nut...As the fake nut arrived at the motor end plate it will automatically retract..as it goes back it will engage again...

Another idea is to mount the same in the middle of gantry, so when the nut mount plate come closer it retract and when it goes further it reengage...

Sorry for bad english

Quote:

---

Originally Posted by erniebro

I agree with Martin. I'm using 1/2 inch Acme threaded rod on my JoeCNC 4 x 4 Hybrid mill that is 60+ inches long. I ran into whipping problems at speeds over 240 ipm. I modified the end supports to use 2 bearings to create a fixed end condition and now I can traverse at the maximum speed my stepper motors will drive the machine (over 300 ipm).

http://www.cnczone.com/forums/attach...1&d=1213370524
This graph shows the speed at which whipping starts (1st critical speed) for a 1/2" solid steel shaft of various lengths. This is similar to the ACME screws on Joe's design. The rpm's you can run at drastically reduce with shaft length. The black line on the chart is for a shaft that is simply supported like Joe's. The blue line is for a shaft that has fixed supports on the end. When you add fixed support the rpm's required to reach the 1st critical speed are greatly increased. At a length of 56 inches, the critical speed for the fixed support shaft is more than double that for the simply supported shaft.

I modified the screw supports on my Joe's by adding an additional flange bearing on one end and a Dumpster coupling between the motor shaft and the ACME screw on the other end. This modification makes the ACME screws behave like the blue line in the graph. Here is a movie showing the screw at high speed before and after modifying the end support: [ame="http://www.youtube.com/watch?v=dDk88vqzuQY"]YouTube - Shaft Critical Speed Demonstration[/ame]


Here is a picture of the dual flange bearing support at the end of my y axis.
http://www.cnczone.com/forums/attach...1&d=1213370524

---

You dont have a bearing journal for those flanged bearings do you?
I assume that your bearings located around the screw itself??