New Z Axis Design

[twforeman]

I've been noodling around with a new Z axis design for my v1 'oko.

Here's a teaser pic:


You can see more pics on my blog along with some design notes.

http://timf.anansi-web.com/wp/possible- ... -shapeoko/

[cozmicray]

Are you concerned about the torque created because offset
of leadscrew and center line of mechanism?
Difference between drive point and resistance point.

Do you plan to use an Acme / ball screw?

Will the leadscrew be over constrained with bearings
at top and bottom?

I can't see how it is MORE rigid?

[cvoinescu]

Are you concerned about the torque created because offset
of leadscrew and center line of mechanism?
Difference between drive point and resistance point.

The offset is in the plane the V-wheels perform best. The carriage twists in the plane of the V-wheels at least an order of magnitude less than in the two perpendicular planes. There will be some additional error, but it should be very small.

I can't see how it is MORE rigid?

I can. There are no spacers on the V-wheels, which makes a big difference in rigidity, especially important in the weaker planes (see above). Also, the spindle is closer to the Z V-wheels, and overall closer to the gantry, so its weight and the cutting forces have less leverage over both the Z and the X V-wheels.

Will the leadscrew be over constrained with bearings
at top and bottom?

Yes, but that's the normal way of running a leadscrew. A longer leadscrew will be even more constrained (e.g. a block with two angular contact bearings, spaced well apart and preloaded, at each end -- four bearings total), to prevent whipping.

[Improbable Construct]

If I rember correctly Edward has played with and implemented a side drive screw like that a couple of different times. I am not sure if there were any problems with it or not.

[twforeman]

I have the Acme z-axis upgrade already installed on my 'oko and will re-use the drive screw and bearings from that. You need bearings at both ends to reduce the backlash effect.

cvoinescu is correct, removing the spacers removes one source of flex and bringing the spindle closer to the gantry reduces the lever arm distance reducing twisting forces.

Additional rigidity is added by the 90 degree flange bent into the motor plate to stiffen it up. This should offset the fact that the plate hangs below the v wheels.

I've seen photos of Edward's offset design and it's close to mine.

My main design criteria for this project are:

1. Reduce flex in the Z

My current spindle mounts are HDPE, the new one will be aluminum. That right there should help a ton. Changing to the v2 configuration with two motor plates and the makerslide bolted to the motor plate should also remove a significant amount of flex. I'm also currently running the v1 Z configuration and I can physically flex the carriage plate with my hand and see it move. It's smaller and aluminum and not stiff at all. The v2 motor plates are substantially more rigid.

This is my current Z axis setup:



2. Change to the v2 configuration without losing too much Z travel.

I was disappointed in the amount of Z travel I could get with the stock v2 configuration and a large spindle motor. This new spindle mount plate addresses that issue. I should have about 3.75" of Z travel when it's all done.

[cozmicray]

cvoinescu Yes, but that's the normal way of running a leadscrew. A longer leadscrew will be even more constrained (e.g. a block with two angular contact bearings, spaced well apart and preloaded, at each end -- four bearings total), to prevent whipping.

By what "Normal" spec?

If you rigidly constrain a component at more places
than are needed, you will start a fight between these places. This
is overconstraint. As an example, three bearings on one shaft
do not work. It is not luck you need in trying to fit the shaft
through three bearings, it’s sympathy — it won’t go! -----

Understanding how to transform basic physical principles
into working concepts with predictable behavior is the key to
achieving high accuracy, high speed, and high reliability.
http://multimechatronics.com/images/upl ... 0Fight.pdf

I am not a Mechanical engineer but I can read / listen to what the say?
and
Have seen simulations on spacecraft instrument drive mechanisms
and the REAL thump -- thump -- thump that permeates thru entire spacecraft
when the instrument is scanning with leadscrew drive?

IMHO

[cvoinescu]

cvoinescu Yes, but that's the normal way of running a leadscrew. A longer leadscrew will be even more constrained (e.g. a block with two angular contact bearings, spaced well apart and preloaded, at each end -- four bearings total), to prevent whipping.

By what "Normal" spec?

I'm not an engineer, so I can't quote specs at you, but http://blog.helixlinear.com/lead-screw-end-mounts and http://web.mit.edu/people/kripa/publica ... dsmc04.pdf and http://www.cncroutersource.com/leadscrew.html are just a few examples that come up with a quick search.

I am not a Mechanical engineer but I can read / listen to what the say?

Well, do that then. I got my information about how to mount a leadscrew the same way.

Have seen simulations on spacecraft instrument drive mechanisms
and the REAL thump -- thump -- thump that permeates thru entire spacecraft
when the instrument is scanning with leadscrew drive?

Wha...?

IMHO

More H, and the willingness to admit O may not be sufficiently informed, would be nice.

Ignorance more frequently begets confidence than does knowledge.

[twforeman]

I'm not an Engineer, because I don't have the piece of paper, but I was a mechanical designer for over 20 years, so I do know what I'm talking about to some extent.

I also worked in a machine shop for those 20 years with both manual and CNC machines. They all had the lead screws captured on each end. I think that CNC manufacturers probably know what they are doing.

To effectively use a lead screw, you need to capture both ends, usually with radial and thrust bearings. When the screw is turning in one direction all the axial force will be taken up by the thrust bearing on one end. When the screw is turned the opposite direction the thrust changes direction and is taken up by the thrust bearing on the other end of the screw.

There are thrust bearings available that will take load in both directions, but you also need the bearings on both ends to support the end of the lead screw and prevent it from whipping when the screw is all the way down at the other end.

You also need to read that article a little more carefully. He's talking about over-constraint of the motors, not the lead screws.