Unexpected Deflection Numbers for Rather Large ShapeOko

discussion of design changes / improvements / suggestions
Improbable Construct
Posts: 997
Joined: Tue Apr 10, 2012 3:21 am
Location: Fairhope, AL
Contact:

Unexpected Deflection Numbers for Rather Large ShapeOko

Post by Improbable Construct » Tue Aug 14, 2012 2:52 am

When I first expanded my ShapeOko Y axis, I measured the deflection in the Makerslide over the 1000mm rails.
http://www.shapeoko.com/forum/viewtopic.php?f=7&t=306

The deflection was about .007" not bad for use in wood.
Now that I have increased to Y axis to 1200mm, I expected that number to change, but it still seems to be right in the .007" range. My dial caliper is analog so it is hard to tell +/- .0005.

The surprising part is that the deflection on the X axis is only .003"!
I have doubled up the Makerslide and it make a big difference!
I checked it over and over and still got .003"

So I am going to put my stamp of approval on the double Makerslide X axis!
Granted, I think it was my idea in the first place, but it really is a good mod for deflection as well as twist.

So with a 800x1000mm cutting area I have about .010" of deflection. That is well within my margin of error. I would guess that the wood my table is made of, changes more than that, from the cool wet morning air to the direct sun it gets in the afternoon.
I measured the deflection of the Y axis by moving the gantry from the end stop to the center of the Makerslide with the X axis closest to the measured side. I had my DW660, the dust shoe and vacuum hose attached. I measured the deflection of the X axis by moving the spindle, dust shoe, and vacuum hose to the center of the Makerslide from the end stop.

I think if I add some sort of a center support to the Y axis there will be very little deflection at all.
Then it will be time to combat backlash... :D
Shapeoko #Classified some of the bolts may be original parts.
Shapeoko 1 # ???? Stainless plates, still in the box.
Shapeoko 2 # 3926 not stock
Shapeoko 3 # 0003
Store:
http://ImprobableConstruct.com
Twitter:
https://twitter.com/ImprblConstruct

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by cvoinescu » Tue Aug 14, 2012 8:13 am

I saw a very interesting method to brace a horizontal beam a while ago, and realized it may work with Makerslide. It resembles a suspension bridge, only it's all done within the volume of the beam. The beam has a narrow channel cut in the bottom, very deep at the ends and very shallow in the center of the beam. Viewed from the side, in section, the top of the channel is the same shape as the top of a suspension bridge between two towers. A steel cable runs through the channel and is tightened against metal plates at each end of the beam. The cable resists downward deflection of the beam in the same way as a suspension bridge does, with the mass of wood above the cable transmitting the load to the cable.

Makerslide is mostly hollow, so the exact same design won't work. One idea would be to run two steel wires, one on each side of the Makerslide, below a small number of dowels that penetrate the Makerslide from side to side and protrude a few millimeters on each side. The dowels at the ends would be near the top, the ones in the middle of the span near the bottom. The dowels have to be notched to keep the cable in place. The cable can be attached to the first and last dowel; or it could go through holes in the mounting plate. The cable would then have to be tensioned somehow.

On second thought, this is probably overly complicated and not worth it, but maybe it gives someone an idea how to do essentially the same thing but much easier.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

Digitalmagic
Posts: 123
Joined: Sat Jun 23, 2012 8:47 am

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by Digitalmagic » Tue Aug 14, 2012 11:10 am

What is the meaning of "deflection" in your measurement?
What I understand is that you are evaluating the straightness of your Y makerslide with different gantry positions, right?
In this case, you kind of assess your frame accuracy, something like a static quality.

What is more critical is the deflection (displacement under load) of your X/Y axis (mainly) when the bit is pressuring against the material to carve.
If, by example, you try to carve a square, centered on your waste board (carriage at mid gantry, gantry at mid Y axis), and increasing depth, speed and material density, you would get a clear result of your machine accuracy.
s/o #650 - Wish my digits are ... magic!

Max Metz
Posts: 240
Joined: Sun May 20, 2012 11:55 am

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by Max Metz » Tue Aug 14, 2012 1:20 pm

Its great to know the double Makerslide on the x axis has such strong effect IC!

Your suspension bridge idea sounds great cvoinescu, braced curves are very rigid - the little Arial Atom sports car made in the UK is certainly testament to that, they are extraordinary.

I like that this place is awash with ideas. Best regards. :D

edwardrford
Posts: 1250
Joined: Mon Apr 09, 2012 5:40 pm
Location: Dixon, IL
Contact:

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by edwardrford » Tue Aug 14, 2012 1:57 pm

@IC, I think adding some supports to the middle of the Y axis makerslide is a great idea. If I remember correctly, cptnPanic had posted a mockup some time ago showing a similar thought, although I can not find it right now.

The easiest way I can think of would be to simply add two more endplates to your build. One in the middle of each rail. Here's a quick mockup to illustrate:
ic_mockup_supported.jpg
ic_mockup_supported.jpg (19.11 KiB) Viewed 6320 times
I used a triple angle bracket (misumi) in the mockup and after generating the render realized it would probably be better to put that on the 'outside' of the plate (so it was under the rail and out of the way) Although it might be hard to get a screw in there to actually bolt it down because of clearance issues? I'm sure you can figure out the details on that.

To prevent the X/Z carriage from slamming into the plate during either an X *OR* Y move (it could potentially be behind it!), you may want to cut the top of the plate off, just underneath where the belt slot is, like so:
IC_supported_potential_cutline.jpg
IC_supported_potential_cutline.jpg (14.26 KiB) Viewed 6320 times
Shapeoko 1 #0 - a couple of upgrades.
Shapeoko 2 #0 - a couple of upgrades.
Shapeoko 3 #2 - Stock

Improbable Construct
Posts: 997
Joined: Tue Apr 10, 2012 3:21 am
Location: Fairhope, AL
Contact:

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by Improbable Construct » Tue Aug 14, 2012 8:45 pm

edwardrford wrote:@IC, I think adding some supports to the middle of the Y axis makerslide is a great idea. If I remember correctly, cptnPanic had posted a mockup some time ago showing a similar thought, although I can not find it right now.

The easiest way I can think of would be to simply add two more endplates to your build. One in the middle of each rail.
I was thinking the same thing. I looked at the old end plates but they just don't line up in any useful way with the new end plates.
edwardrford wrote:I used a triple angle bracket (misumi) in the mockup and after generating the render realized it would probably be better to put that on the 'outside' of the plate (so it was under the rail and out of the way) Although it might be hard to get a screw in there to actually bolt it down because of clearance issues? I'm sure you can figure out the details on that.
I have my spoil board cut to the largest dimensions that the machine can cut so there is plenty of room for the angle bracket on the inside. I bet it would look better tucked under the Makerslide.
edwardrford wrote:To prevent the X/Z carriage from slamming into the plate during either an X *OR* Y move (it could potentially be behind it!), you may want to cut the top of the plate off, just underneath where the belt slot is, like so:
IC_supported_potential_cutline.jpg
I will feel a bit bad cutting the tops off of those end plates but I will get over it. I am sure an IPA will help.
Shapeoko #Classified some of the bolts may be original parts.
Shapeoko 1 # ???? Stainless plates, still in the box.
Shapeoko 2 # 3926 not stock
Shapeoko 3 # 0003
Store:
http://ImprobableConstruct.com
Twitter:
https://twitter.com/ImprblConstruct

edwardrford
Posts: 1250
Joined: Mon Apr 09, 2012 5:40 pm
Location: Dixon, IL
Contact:

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by edwardrford » Tue Aug 14, 2012 9:06 pm

I will feel a bit bad cutting the tops off of those end plates but I will get over it. I am sure an IPA will help.
An IPA will help most anything.

Let us know when you get it put together and have a chance to re-measure deflection on your Rather Large Shapeoko.

-Edward
Shapeoko 1 #0 - a couple of upgrades.
Shapeoko 2 #0 - a couple of upgrades.
Shapeoko 3 #2 - Stock

dointhangs
Posts: 179
Joined: Tue Jun 05, 2012 7:31 pm

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by dointhangs » Wed Aug 15, 2012 5:10 am

So cool - I am wondering with these wonderful x-axis double makerslide upgrades the obvious improvement is a result of improved stiffness from having two sections of makerslide (obviously the main thing here) but is there also something you guys are gaining by having the two sets of wheels (stabilizing the carriage)? Sorry if I have missed a thread on this. So a guy I know who I sent my hello world vid too was mentioning that long travel machines like what you guys are working on have absolutely made an impact in the surf industry in shaping boards. Sorry I dont know much about it and havent had a chance to talk more with him.
Shapeoko #298, dual drive upgrade, acme z axis, open bed endplates and a slightly larger ("expanded desktop") work area!

Digitalmagic
Posts: 123
Joined: Sat Jun 23, 2012 8:47 am

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by Digitalmagic » Wed Aug 15, 2012 2:07 pm

Did you get expected results while carving a centered piece ?
Waiting for a Rather Large carved part, too! :)
s/o #650 - Wish my digits are ... magic!

Improbable Construct
Posts: 997
Joined: Tue Apr 10, 2012 3:21 am
Location: Fairhope, AL
Contact:

Re: Unexpected Deflection Numbers for Rather Large ShapeOko

Post by Improbable Construct » Wed Aug 15, 2012 2:50 pm

dointhangs wrote:So cool - I am wondering with these wonderful x-axis double makerslide upgrades the obvious improvement is a result of improved stiffness from having two sections of makerslide (obviously the main thing here) but is there also something you guys are gaining by having the two sets of wheels (stabilizing the carriage)?
When I originally made this mod, my goal was to try and remove the side load on the V wheels.
I noticed that there was a lot of slop in the bearings, and that it was causing unwanted movement in the Z axis.
By doubling the Makerslide it removes a lot of the side loading on the V wheels.
Digitalmagic wrote:Did you get expected results while carving a centered piece ?
Waiting for a Rather Large carved part, too! :)
So far all I have done is to surface my spoil board. It came out pretty good. I have some minor lifting of the bit when going from X+ to Y+. It is tiny but you can see it and feel it with a fingernail.
Shapeoko #Classified some of the bolts may be original parts.
Shapeoko 1 # ???? Stainless plates, still in the box.
Shapeoko 2 # 3926 not stock
Shapeoko 3 # 0003
Store:
http://ImprobableConstruct.com
Twitter:
https://twitter.com/ImprblConstruct

Post Reply