Stiffness of the spindle carriage plate

[WillAdams]

From the thread: http://www.shapeoko.com/forum/viewtopic ... 140#p54121

I added (2) .625x1.25x7" long aluminum rectangles as stiffeners to the Y axis plate to reduce flex, which seems to help a lot with the chatter. It's still there, but I would say reduced about 60%

Other people doing this sort of thing:

- http://www.shapeoko.com/forum/viewtopic ... =30#p51914 --- Edward made a plate w/ two bends in it
- http://www.shapeoko.com/forum/viewtopic ... 746#p53740 --- LTEPM added T-track

It seems strange to me that the plate is now the weak point --- I wonder if this is caused by simultaneously moving to a single point of attachment for the spindle --- would adding a second spindle mount help? Would a taller mount which is a single, non-flexing structure help this?

Could we make room inside the Z-axis channel for a second set of Z-axis rails which would simultaneously engage the wheels and make things stiffer?

Other options (which are actually workable)?

[Metropolicity]

I am going to try the t-track method but I also had lofty ideas of milling my own 3/8" aluminum plate for it.

[Gadgetman!]

I would say that the 'single mount' is the biggest issue. The point where it touches the plate functions as a fulcrum. On a two-mount system, the pendulum isn't allowed to move because one end is tied off, but here...
It gets worse if the mount only has a single horisontal row of bolts holding it to the plate.

[TomDChi]

I would say that the 'single mount' is the biggest issue. The point where it touches the plate functions as a fulcrum. On a two-mount system, the pendulum isn't allowed to move because one end is tied off, but here...
It gets worse if the mount only has a single horisontal row of bolts holding it to the plate.

Another big aspect of the flex is the fact that the mount is well below the lower set of v wheels - that distance is part of the fulcrum arm that's flexing. If the mount was between the wheels, there would be a lot less flex (but that would create a lot of other limitations and problems).

In another thread, I ran the rudimentary structural engineering numbers on a couple of stiffening options. A thicker plate should be a great option (though I don't thin anyone has fully dealt with the issue of the eccentric nuts). For those who don't want to or can't fabricate a whole new plate, side rails should also work well.

One big thing here is that we aren't trying to do hundred thousandths of an inch precision, so because there are several sources of "flex" added together by the time you get to where the end mill is cutting material, we only have to stiffen the spindle carriage plate somewhat - not "near perfect". I would think that mounting the router in the mount and applying lateral force in the X and Y directions, and comparing how much the bottom of the plate moves, versus the top of the plate (particularly in the Y direction) would give a "seat of the pants" indication of how significant the flex of the bottom part of the plate is compared with the rest of the sources of flex.

[cvoinescu]

If you worry about the eccentric nuts, you can replace them with eccentric spacers with longer bosses (plus a washer and a regular nut). Inventables sell two versions of eccentric spacer with a 3 mm boss: a compact one (2 mm tall) and one as tall as their standard eccentric spacer (6.35 mm, or 0.25"). Ideally, the boss (the round section) of the eccentric spacer should be nearly as long as the plate was thick, but 3 mm works reasonably well with plates up to about 6 mm thick.

[Neppo1345]

I've got a plan in the works for some stiffeners. I butchered one last night (and ruined a brand new endmill) due to some rushed CAM work (stupid stupid stupid)...but hopefully I'll finish the second one up when I get home from work today.

I'll post some photos tonight, and eventually put together a full writeup.

[twforeman]

I have also added some stiffening pieces and bought a couple of corner braces to bolt-on.

I bought a pair of these and plan on mounting them on the top of the router mount and fastening them to the plate.

http://www.ebay.com/itm/370128534767?_t ... EBIDX%3AIT

This should remove some more flex caused by the lever arm of the router mount.

Photos and write-up eventually - I'm doing a lot of upgrades. I should start a thread and just start posting photos.

[WillAdams]

As cool as the spindle mount is, I wonder if we shouldn't just use a taller one to begin w/.

Really bummed that the 65mm version of this mount: http://www.amazon.com/gp/product/B0152Y ... VR9BOE1NNT vanished from Amazon (I tend to accumulate Amazon gift cards, so stuff like that is essentially free if I'm patient enough to wait).

[Neppo1345]

So here's the idea that I came up with: 1/4" aluminum plate, running the height of the Z-Axis carriage plate. Haven't gotten the other side completed yet, but Once I do I'll pull the Z-Axis apart, mark and drill holes, transfer the holes to the stiffeners, tap 'em, and fasten it all together.


It'll be attached through the Z-Axis carriage plate with 3x #6 stainless screws (countersunk in the backside of the plate), it's got groove that the spindle mount sits in (which should take the majority of the load), it'll be fastened to the the spindle mount with 2x more stainless countersunk screws (the screws are really just there to keep it seated and shouldn't take any shear).


Once I get everything completed I'll do a full write up and (if it works as well as I hope) make some cad files and drill templates available if people are interested.

[wunderaa]

Very compelling solution! Please keep us updated.