Classic screw and nut(or sleeve?) system 4

[pawanhv]

hi ,
I am a bit confused here and i would like your help here please. I have a classic screw and nut(or sleeve?) system and I have attached a pic of it for your reference. the torque is applied to the screw and causes the translation of the sleeve. I have derived the relation between the torque applied on screw and tensile force generated on the sleeve. Now I want to find the minimum value of tensile force on the sleeve required to overcome the frictional force generated due to the sliding of this sleeve in a cylinder. I just want to know how do I calculate the friction due to sliding of sleeve in the cylinder? Thanks. In the pic the thing in green is supposed to be the sleeve. sorry for the image I dont have a CAD software.

 

[pawanhv]

yes exactly its a one shot application. Once you are done with it ! its finished .. what you mean to suggest is two projections on the exterior of the sleeve which mates with the two female slots on the body (or even a square sleeve) would be better than the cylindrical sleeve with a anti rotation flat to move it laterally? thanks again rb1957..

 

[rb1957]

either would react torque better than a flat

 

[pawanhv]

if I think the other way around , like I said you about the second configuration where the ideal case would be the pitch of thread on the ext surface of sleeve to be greater than at the internal surface ,turns out if the pitch is very very high it is infact a straight line . so your suggestion is infact the ideal case for me second configuration. thanks a lot rb1957.

 

[rb1957]

not sure i follow that logic ... if the sleeve is threaded on both faces, i see problems if the threads aren't the same pitch ... one thread will be driving a displacement based on a rotation; if the other thread is a different pitch it'll want to displace more (or less) for the same rotation. maybe the rotation of the outer face isn't exactly the same as the inner face (as the wall of the sleeve strains against these two different displacements). i'd expect the sleeve might displace the average of the inner and outer displacement; whatever, i expect it'd be complaining like a sob !

 

[hydtools]

A sleeve with a threaded ID and OD will not work. The stationary screw will prevent any translation pending due to the OD thread. Likewise, the stationay outer housing will prevent any translation of the sleeve due to the effort of the screw. The sleeve, when threaded on both ID and OD is captured axially between two fixed elements.

Ted

 

[rb1957]

i'll buy that ... 'cause the sleeve isn't turning it can't travel down the external thread.

 

[IRstuff]

The proper approach is some sort of twisting constraint, i.e., making the sleeve square, or keyed with a flat or slot, or even just having a single wall constraint, as is done in typical lead screw stages:

http://www.edmundoptics.com/mechani...tages/high-performance-lead-screw-stages/2440

http://search.newport.com/?q=*&x2=sku&q2=TSX-1D

In the latter cases, the speed of the motion is strictly limited by the screw pitch and turning speed.

p.s. I agree that outside threading will not work; if anything it's more likely to jam in place, since there's insufficient torque on the sleeve to cause it to rotate, since the inner screw drag will cause it to want to rotate in the same direction as the screw, but it will actually need to rotate in the opposite direction

TTFN
faq731-376

 

[pawanhv]

well all my arguments above about the external sleeve thread was based on the calculations shown in the attached file. I found by kinetic analysis alone that if the pitch in the interior of the sleeve is less than twice the pitch on the outside i could arive at what i am looking for i.e. quick displacement of sleeve .

 

[IRstuff]

Why make this so complicated? A simple flat key will work better than threading the outside.

TTFN
faq731-376

 

[pawanhv]

I heard this concept has been used before and I became interested in that . But I did not find where it is used.

 

[Engdoitbetter]

I don't understand how could the sleeve move: for a given rotation, in a general case, i.e. if the thread pitches are different, the axial movement would be different for the inner thread and the outer thread, and this would be against the basics of the translation movement.
Second, If the sleeve does rotate due to the torque applied on the screw, there will be no relative motion (or a slower one) between the screw and the sleeve, so a slower (or no) axial movement.

 

[rb1957]

external tread won't work (in conjunction with an internal thread). the sleeve is not rotating, the screw is being torqued, the sleeve is prevented from rotating, so how will it translate along the external thread ? at the internal thread the screw is rotating, and the sleeve is fixed; what is allowing the body to rotate with respect to the sleeve (which it is fixed to) ??

the pitch of the thread controls the displacement due to applied torque, coaser thread, more displacement per rotation.

you could thread the sleeve/body interface (what we've called an external thread) and it'll work If the screw is fixed to the sleeve, so that twisting the screw rotates the sleeve.

clear as mud ?

 

[hydtools]

The internal and external threads so constrain the sleeve such that the velocity of the sleeve relative to the body cannot be anything but equal to the velocity of the sleeve relative to the screw. You have not accounted for that constraint. Displacement and velocity of the screw relative to the body are 0 by the screw to body constraint.

Build a physical model and let us know how it performs.

Ted

 

[zekeman]

"....well all my arguments above about the external sleeve thread was based on the calculations shown in the attached file. I found by kinetic analysis alone that if the pitch in the interior of the sleeve is less than twice the pitch on the outside i could arive at what i am looking for i.e. quick displacement of sleeve .

http://files.engineering.com/getfile.aspx?folder=95c8ca86-138e-470a-b54f-2c"

Where did you get this "file" ?

I wasted 20 minutes reading this nonsense.

I can't believe that any credible kinematic analysis would allow the frame to rotate and in the end he/she gets an "amplification" directly dependent on the ratio of the screw rotation divided by the "fictitious" frame rotation. Fatally flawed and wrong conclusions.

Perhaps a PHD candidate??

Same guy probably can prove 0=1.

Bottom line
OP-- don't believe everything that's written especially when it is not published.

 

[zekeman]

Still trying to make sense out of the paper, I believe the author was using the frame as a reference to capture the screw and complicated a simple problem, so I thought doing it as follows:

Let the screw rotation be N1 and the sleeve rotation N2
The pitch of the internal and external threads are P1 and P2.

I get the translation to be
P1(N1-N2) which must be equal to -P2N2, the translation along the external thread.
Then you can write

P1(N1-N2)=-P2N2

In order to amplify the motion, N2 must be negative. For example, if the amplification= 2,
(N1-N2) = 2N1 , or
N2=-N1
and
2N1P1=P2N1
To satisfy this, the external pitch is
P2=2P1
If the amplification is 3,
then N2=-2N1 and
3P1N1=2P2N1
P2=1.5P1
Now, the trick is to get a small muscular little man to rotate the sleeve accordingly.
That is the problem with all this. I just doesn't work. Moreover, what has the external thread done for the system? Nothing.
Once the little guy has done the spinning, the amplification is accomplished.

 

[IRstuff]

The whole notion of "torque" on the sleeve is essentially bogus. The only torque comes from the frictional drag from the lead screw turning. That's not enough to make the sleeve turn against the outer threads. If one wants to change the linear speed of the sleeve, the two possible approaches are faster rotation of the lead screw, or a larger pitch on the lead screw.

TTFN
faq731-376