Is a 31 inch diameter Indexer possible?

[Mayo]

Do any of you have an indexer capable of cutting a 31 inch diameter? If so, I'd love to see photos or get info on how you did it.

I recently had an inquiry if I could reproduce 6 foot tall styrofoam chess pieces of a specific design. I've thought about cutting them in slices but the glue up will result in visible glue lines that will not sand out because the glue will be harder than the styrofoam. Is there a better way to glue up slices in styrofoam where the glue lines won't be a sanding problem? Wood glue when dry is impossible to sand off without deforming the shape of the foam.

At a 6 foot tall height, this chess piece design would have a diameter at the base of almost 31 inches. What is the possibility of building an indexer that could handle this kind of job?

I don't currently own an indexer so I have no idea what limitations the software may or may not have. I understand 3D files of the chess pieces are required and I have already looked into obtaining those. I know I would need additional software to produce the cut file.

Any info would be greatly appreciated!

[myxpykalix]

I had asked a similar question because i wanted to carve a design on the skirt of a table approx. 30" diameter and the basic answer i got was it couldn't be done due to gearing and the larger radius but haven't heard how you might be able to rig it up with different gearing and such. If anyone knows please let me know.

Also you will have table clearance issues. Here is my table with my indexer gulley on the left in the picture.
All those rails are bolted on and can be removed. I have thought about taking the 7 center rails out, mounting the indexer in the center on one end and that would allow me to basically do exactly what you are wanting to do.
And may have to fashion some support rails like pictured (did i mention i was a master at photoshop?)
So far i haven't had a request to do anything that large so it goes undone, but you never know.


2759

2760

2761

[Gary Campbell]

Mayo...
Jim McGrew has just done this in SC. There are posts here and on the Vectric Forum showing these chess pieces. I know Jim and his work, both first class. Find his profile and send him an email.
Gary

[cabnet636]

the material we are using on the chess pieces is mdf and foam so with painting the seams are hidden, i wonder if white solid surface adhesive will work although with slicing some faux work is probably inevitable as even with wood unless the grain is in the same direction the seams are still laminations, thes are the local chess geeks!!


2762


2763

[cabnet636]

ps i would consider a foam hot wire cutter or out source to a large 5th axis, ( my next machine) jack that is one cool machine, i have another frame under construction that is down the center like you are thinking with a removable table!!


jim

[myxpykalix]

Well the answer i got was that it would lose steps (or something like that) due to the gearing in relation to the larger diameter which makes sense and my thought was couldn't you change the values in your settings to compensate for that? To what i have no clue but if anyone has any experience turning large turnings i hope they jump in.

[khalid]

I think slicing is the best option.. If you have to do surface carving ( upto 2") on the 31" pillar, the indexer will be a good choice..but any deep like James has shown his master pieces, you can't do that with indexer...

Jack, i don't think he will get loose steps on foam ( if he use powerful motors)... However, as the dia increases you will loose the resoulution ;).. just think if you have 0.1 degree deviation in 2" dia work piece it will be negligible on the perimeter..but the same will be more prominent in 31" diameter piece..

I think my bad english will make some sense to this house.

[khalid]

Jack..you have a nice table

[rb99]

At least with foam you don't have the weight to contend with...

RIB

[myxpykalix]

khalid,
"you will loose the resoulution"--that was what I really meant, not "lose steps" but i was confused. Someone needs to pin a note on me and leave me out at the dogtrack...

Say for example you have a column that is 12" diameter. And it does not lose resolution.
Now you have a column that is 36" diameter. Is it a logical conclusion that you could change your values by a factor of 3 to compensate for that loss? Or am i not understanding the issue?

I think in the longrun doing it in layers like james has done will be far less work then trying to rig up some table.
James do you have any closeups of your joints? And how much work was it to glue and sand it down to make it look good?

[cabnet636]

the better the machining the less work one has to do on cleaning up the product, we used 4" grinders with sanding disk (60 grit) and dremal type tools for cleaning up the details, always presand mdf it has a slicker finish than most would think!! i made a lot of mistakes and with a three inch material depth bits are 4" so i broke some of those too, learning to clear enough area for the collet is another challenge

2764

2765

2766

[cabnet636]

a bit more info


2767


2768


2769

[Mayo]

Awesome photos James - and awesome work on a complex project! In seeing what is really involved, I may decide to pass on the job and just refer it to you.

Rather than try and reconfigure my table, I had planned on making a dedicated machine just to do the large diameters.

Speaking of loosing steps or resolution, is that due to backlash or something else? This place makes drives with no backlash...
http://www.doughtydrive.com
and check out this machine...
http://www.doughtydrive.com/machine%20demo1.html

[khalid]

Jack.. suppose you have two concentric circles then you know the elementary mathematics..
S =R x Theta

As the Radius grown for the same angle theta, you will get bigger and bigger S...

Now think in this way, suppose you have kept the steps with increment of THETA for indexing...Where THETA will be your fixed angle for each increment...Now see the figure for smaller circle suppose you are using 3mm ball nose and it almost fit on the S of smaller circle.. Now if you use the same resolution (THETA) for bigger circle, then just imagine what will happen...

The "S" changes with Radius "R" whereas the "THETA" is fixed and to be considered in radians...

NOTE:
This Theta will depends upon the steps on the controller circuit i.e. 1/2,1/4/,1/8,1/16 etc, steps of the motor and gearing ratio... This can only be change on controller circuit only... For larger diameter you will change the controller setting to high steps mode...but this will takes more time on the work piece machining with smoother finish..

For a Flat piece you can compare it with stepover value... the greater the stepover the less resolution the smaller the stepover the greater resolution ;)

Sorry for the bad picture..hope it clears the concept...
2770

[myxpykalix]

Khalid...I know your not here in USA but you need to speak english here! lol You could have been talking serbian and i wouldn't have understood it any better. I appreciate the effort but i barely got thru "readin, ritin, rithmetic" in highschool so that was way over my head! I appreciate the effort though buddy.

I didn't realize till now that james chess pieces were hollow. I would take this project on if i were you mayo.

[Mayo]

Khalid...
I'm not following the math of your explanation (I didn't do well in math in school) but I kind of understand the drawing. However if the design and cutting file takes into account that it is wrapped onto a 30 inch diameter why does the steps matter?
Are you saying that different diameters of work will require changing the number of steps from one job to the next? I would hope if this is true, that the toolpath software would automatically make these changes or compensate for it.

Admittedly, I have little understanding of the whole indexer machining process.

[rb99]

I think he is saying the resolution degrades considerably as you get larger.

RIB

[khalid]

Yes, Richard pointed very well...Mayo, You are also right that the software takes care all these things in the toolpath..but again it all depends on the Hardware too..Fortunatly current days hardware are marvellous and state-of-the-art...microstepping is available in the controller card, the Stepper and Servos are also available in High resolution...

Now , just want to put more thrill in the discussion..consider a stepper motor with 1.8 degree steps, and your controller card can't microstep....Now your CAM program don't know about your hardware resolotuion, however CAM software gives you the toolpath of maximum, optimum, and best resolution..Now here the Hardware Fails... What you will get if your motor can move only 1.8 degree/step and the toolpath command your motor to move 0.8 degree/step????..

The same is true for S=R x Theta, Where S=Arc Length.. As the radius grows the ARC Length will increase for the same Angle...

0.01 degree upto 1meter radius will give you the ARC Length value very small (S= 0.01 x (2xPi/360) x 1) , whereas as the diameter increases , lets say 100 meters, the same Arc length for the same Angle you will Get (S= 0.01 x (2xpi/360) x 100)

The higher the radius the poorer the resolution in Indexer case...

[khalid]

Sorry Guys, currently involving in Turbomachinery vibrations issues...
http://maintenanceforums.com/eve/forums?a=tpc&s=917103451&f=3751089011&m=371109783& r=964102093#964102093

So my minds goes in mathematical way...

[joecnc2006]

Did you see Arts?

http://www.turningaround.org/

[richards]

One of the factors is the minimum distance per step. A machine that is optimized for a 6" diameter may not be able to preform properly with a 20" diameter or a 31" diameter.

When compared to a 6" diameter, a 31" diameter will have 5X more "coarseness".

If the indexer motor is geared properly to give you the required resolution, and if you have space enough to mount an indexer with a 31" diameter, you would get perfectly good results, assuming that your feed/speed/stepdown/stepover settings were correct.

[gerryv]

I think it comes down to the dia. of your piece times pi (3.14) to give you the circumference divided by the number of steps per rev. your motor takes. Therefore, 31" X 3.14 / 210 steps (a cheapie motor) = 0.46" "flats". Yuck. Thank goodness for micro-stepping, gearing and stepovers. Is this right?

That's not addressing some of the other influences already mentioned - I'm considering the forces in trying to accelerate and decelerate a 31" dia. x 10' chunk of hardwood (XXXX pounds?) through varying arcs - quickly. Wow, I'm guessing I'd worry about lost steps with an open loop setup. Just guessing though :-)

[Brady Watson]

You need a planetary or harmonic gearbox, or multi-stage belt box to have the correct resolution out at that diameter, with minimal backlash. You won't get, nor will you want, very high RPM if you free-wheel spin the indexer at that diameter. It can certainly be done, however you need the space & work for a machine like this, and unless the machine has a Y axis, you will be limited on toolpath strategies & versatility.

-B

[Mayo]

Gerald - this machine would be for styrofoam only - not a hardwood tree!

With a very low bed (if any) this machine wouldn't have to be any larger than a 4x8 shopbot - half the diameter is only going to be say, 16 to 18 inch. It would be taller but it would still fit in a normal garage or small shop.

[khalid]

Mayo..in this case your only limitations will be the tool length... your Z-axis bottom will strike to the work piece if you go deeper at some locations and shallower at other.. just imagine....

I personally want to make a machine that will only do Indexing.. The Y-gantry will be fixed to the machine base.. and centered on the center of chuck...

[myxpykalix]

khalid...like this one? Art and I along with a few others started the legacy yahoo group and from that machine he developed his first prototype (man was it ugly).
You are the europen version of McGyver, well hes the american version. Give him some hair pins, a little cardboard and chewing gum and he'll make you a cnc machine.


2771

[hfjim]

Folks,

I have a lathe/indexer I built with a 48" dia x 24' between centers cutting capacity. You can see it at www.hfmillwork.com (http://www.hfmillwork.com). It was built with Minitec components and Shopbot controls.

Mayo,

If you're still looking for someone to cut those chess pieces, let me know.

Jim Turner
hfmillwork@bluefrog.com (mailto:hfmillwork@bluefrog.com)

[myxpykalix]

Jim,
That is an indexer on steroids!! I thought i understood that there would be problems with the "resolution" due to the size?
Do you have any other pictures or any youtube video?
Explain this if you can beccause i'm confused.
Since we are dealing in 360 degrees rotations if i have a 12" diameter column the length between each degree is not as far as if you have a 48" diameter column. If having a larger diameter is a problem how do you compensate for the difference in length between each degree?
Is the shopbot software you use to run your indexer modified in any way?
What kind of setup do you use for the actual cutting?
That is a VERY slick setup!

[khalid]

Jack.you need no worries.. If you have programmed the part for 11" then cut the part with 11" workpiece..the problem will only arises if you use the 11" toolpath to cut 31" part..

Just remember the above fact and you will have no trouble..

[hfjim]

Jack,

The indexer rotational stepper has a belt reduction of about 5:1 so it won't rotate very fast,but the resolution is fine. I do have another motor (variable speed) that I can use if I do more conventional lathe work. It also turns fairly slow (20-60 rpm. You can't turn these big columns very fast as balance is a big issue.