Need help cutting a spiral pattern on a pedestal

[drodda]

I am trying to figure out how to make a 3d cutting file to cut spiral pedestals such as this one.
www.dkpokertables.com/junk/spiralbase1.jpg (http://www.dkpokertables.com/junk/spiralbase1.jpg)

I am wanting to build it in two halves and then attach them together to make a seamless spiral pedestal. The diameter is 20" for the pedestal. Any help would be greatly appreciated. I have a shopbot with a 12" Z axis so the diameter should not be a problem.

I realize that if I had a indexer I could do it on that pretty easy but have to get one built first to show and hopefully generate enough demand that I can justify the purchase of the indexer.

I am pretty effiecient with autocad but need a jumpstart on the brain to get this going.

[artisan]

Dave, Spirals are nearly impossible to do without some sort of indexer. If you slice the model from top to bottom into two equal halves and then lay them horizontal, you should see the problem.....undercuts. The Bot will only cut in an up and down manner and you will not be able to cut your spirals when you get to the edges. This part could possibly be finished by hand to mate with the other half....depending on your material....but it looks like quite a bit of work....D

[drodda]

Darrell,
Thanks for the reply. I see what you are talking about. Now the lights are going on in my head. So if I undestand you correctly you could do the pattern perpendicular to the pedestal but not spiral. Like doughnuts around the pesdestal. The spiral is where you get the undercut.

[drodda]

My next question: is there anyone out there who can turn a spiral like this on their indexer? It needs to be 28" tall and 20" to 24" diameter. Made out of cherry? All I need is the spiral part not the top or bottom. Please email me directly at dave at dkpokertables dot com with price and time frame to deliver. I am in Iowa.

[bleeth]

Dave-Get ahold of Gene at Marshall's Millwork-He's always messing with that kind of thing. You might consider laying up the cherry yourself. You might also pull it off pretty nicely as four quadrants glued together instead of two halves.

[davidallen]

Dave,

If you split the spiral in quarters or eights couldn't you couldn't you machine from above without undercuts?

da

[drodda]

Thanks for the replies, What software would you have to use to make a cutting file for 4 Quarter pieces to make a complete pedestal? I'm not quite seeing the forest through the trees.

[gerald_d]

Dave, the question is not so much "what software" but rather "what cutter" and "how much hand finishing". If you are happy with only a 60 deg V-cutter doing the deep part of the grooves (and the rest by hand), then doing it in quarters will be okay. But if you plan to use a big "rounding over" cutter (and little hand finishing) then even quarters won't do it for you.

For these type of questions I often think that folk concentrate too much on the magic of software rather than thinking the cutter action through in their minds..."if I came straight down with a vertical router, what type of cutter will cut the shape I need?". Software cannot bend cutters.

[bleeth]

Software wise you have to use one of the 3-d sculpture packages such as Rino or Artcam Pro. Once it is designed in the software the likely tooling strategy for a flatbed to do this would be with a small ballnose bit (1/8). The bottom line is that this is one of those projects that is doable with a flatbed cnc but can only be done efficiently and clean with an indexer or lathe.

[gerald_d]

For a simple geometric shape like a spiral, one could probably program it using no more than Excel?

[marshawk]

Wow! That's a big spiral! We would have to modify our indexer setup to turn something that wide. Anything is possible, but the best way to do this would probably be in sections on the flat and then piece together.

One of the bigger problems with turning something that wide is how much slop there is in the stepper motor. What may be acceptable when turning a 10" diameter spindle would probably be disastrous with something that wide. There are options out there for more sturdy motors, but you're talking $$$.

Gene's the expert, but when I see something like that, I have to think that maybe a non-ShopBot solution is more feasable. Maybe one of those legacy thigamajigs modified somewhat, or a Router Slider with a hand crank.

My $0.02.

Chip

[chris_booth]

Dave,

a possibility is as follows....

i) make the pedestal without the spiral (Good idea, eh?) and with a diameter just slightly smaller than the deepest part of the spiral.

Now that spiral can be "unwound" to be a straight line with the ends cut at a low angle (where one end meets the floor and the other meets the underside of the table) so...

ii) calculate how long the spiral is and what angle you'll have to cut the ends at.

ii) buy a strip (or strips)of bendywood with a cross-section large enough to make your spiral cross-section shape and a little longer than the spiral's length.

iii) cut your profile cross-section on the straight strip(s).

iv) cut your low-angled ends.

v) bend the bendywood round the pedestal.

The last time I looked Bendywood was ridiculously expensive, but then every now and again it could make a difficult job quite easy.

If you wanted to try it this way and the biggest hurdle was the maths for unrolling the spiral bit, I'll do it for you.

[gerald_d]

Following on from Chris's scheme to attach spiral strips......

How about attaching plastic or metal strips that can be a guide for a hand-held router?

A standard "indexer" is probably too small for this. But you could attach a big bicycle sprocket to the column and a small sprocket to one of your stepper motors................

(The Legacy (http://legacywoodworking.com/technique.cfm?techniqueID=7) way)

[bleeth]

You could also build the round and wrap 1/2 round flex mold around it and then finish it. The sample looks pretty dark and flex mold takes dark stain very well. That makes it both quick and inexpensive.

[drodda]

I appreciate the help but I have been trying to solve this for the last 3 to 4 months. One of the problems is that the spirals are about 2" wide and 1" tall. This makes for a quite large cylinder to start with. The thickness would have to be 1.5" to 2" if I am guessing correctly. That's an expensive test piece on the first one. I am also trying to figure out the best way to attach the indexer so the height is adjustable off the end of the table. This would allow me to cut large cylinders up to 5 ft in length.

Gerald I like your idea of gearing the cylinder with a sprocket but would need to accuratle know the travel in one revolution to syncronize the bot to the indexer. I wonder what is the strongest motor that shopbot sells? I guess a call is in store for this on Monday.

Perhaps I need to make a trip over to Morris Dovey's shop and get the dust wiped off his indexer? I guess that makes two calls I need to make tomorrow.

After spending countless hours trying to solve this I figured I would try and tap into the endless pool of knowledge of this forum. Keep the great ideas coming. I appreciate all your time.

[drodda]

I have seen that picture before and have wondered when shopbot was going to offer the female helper as an option for the bot?

I was thinking about extending my current rails forward about 2 ft which would allow me to release the x motors and fix the bot over the center point of the indexer. The indexer would be mounted below and in front of the table bed. Since the bot can only go so deep I would want to be able to raise or lower the indexer to accomodate the size of the pedestal. you would need to make it so that you can fix the carriage over the center of the work every time. I would want to adjust both sides of the indexer at the same time so that they always stay aligned together.

However I figure this out I still need to be able to quickly go back to using the bot to cut out poker tables since that is the bread and butter that purchased the bot in the first place.

I will have more time tommorrow to try and digest your suggestions about the setup. Are you suggesting removing the z motor in place of an indexer?

[gerald_d]

Dave, why bother to extend the x-rails if the bot is not going to have x-motion......? You could just lift the gantry off the current table and drop it done onto a pair of trestles. In that case, your "trestles" could be a cart/indexer that you wheel nearer to the bot table when you need it.......

I wouldn't actually remove any of the current motors because of the setup times involved to re-locate the pinion gears and set the mesh of the z. Get more motors from SB, the same as you have now, so that they can be back-ups for each other, and do your own gearing for the "indexer" or "spit barbeque rotisserie" Then you only need to swop a motor cable (an "x" port now drives rotation)........

The standard SB "indexer" is not going to work for you, so you are at liberty to think outside the box.

[drodda]

http://www.dkpokertables.com/junk/spiral-ped2-sm.jpg

http://www.dkpokertables.com/junk/spiral-ped3-sm.jpg

Here are a couple pictures that better show the detail of the spiral or rope pedestals that I am refering.

Believe me Gerald I have been thinking outside the box since the day I first saw a shopbot 5 years ago. I finally found a use that would pay for the bot last year. That and the new shop that I finally got built to hold the bot. Now I am trying to find things that would be very unique to my setup that would also pay for themselves.

Once again I do appreciate all the help. I do like your idea of a seperate cart for the indexer/lathe though. I just figured that if it was attached I would not have to lift it onto the cart. It would just roll down the track onto the indexer.

[gerald_d]

Hi Dave, permanent rail extensions would impact on a sensitive part of my anatomy.


Those are very nice pedestals! However, it looks like you could go a long way with having only a very small range of sizes, and I wonder if one shouldn't do this with a non-CNC system.......(Even the CNC side may be building a mental box around us - that is a very standard twisted column, albeit thick and short).

[marshawk]

Dave, If you want to test inexpensively, go to the Big Orange Box and get youself a bunch of that pink insulation foam. The stuff glues up nicely and carves well enough to tell you if it's gonna work.

Chip

[mikejohn]

This file
J3,83.5896,99.9817,78.0283
M3,81.8870,98.9987,79.8905
M3,80.0997,97.9667,81.6436
M3,78.2327,96.8889,83.2824
M3,76.2917,95.7682,84.8020
M3,74.2822,94.6080,86.1981
M3,72.2103,93.4118,87.4665
M3,70.0819,92.1830,88.6036
M3,67.9033,90.9252,89.6059
M3,65.6809,89.6421,90.4705
M3,63.4212,88.3374,91.1950
M3,61.1309,87.0151,91.7771
M3,58.8167,85.6790,92.2152
M3,56.4853,84.3330,92.5080
M3,54.1436,82.9810,92.6546
M3,51.7984,81.6270,92.6546
M3,49.4567,80.2750,92.5080
M3,47.1253,78.9290,92.2152
M3,44.8111,77.5929,91.7771
M3,42.5208,76.2705,91.1950
M3,40.2611,74.9659,90.4705
M3,38.0387,73.6828,89.6059
M3,35.8601,72.4250,88.6036
M3,33.7317,71.1962,87.4665
M3,31.6598,70.0000,86.1981
M3,29.6503,68.8398,84.8020
M3,27.7093,67.7191,83.2824
M3,25.8423,66.6412,81.6436
M3,24.0549,65.6093,79.8905
M3,22.3524,64.6263,78.0283

will cut this quarter of a spiral.

2605
The material is parallel with y, square with x and z.
At the end of the first cut,rotate the material around the centre of the arc, return x to its original x position,(keep the bit out of the way
) run the file again, and you get the next quarter.
Continue doing this, and you have cut a spiral groove.
Use a ball end of an appropriate size, and you have created the bottom 'valley' of the spiral.
With more work, I am sure we can come up with the cutting paths for the top 'ridge' of the spiral.
I looked at the mathematics of this (see Geralds Excel suggestion above), but a solution for all diameters, angles of spiral defeated me.
The arc is created in autocad, with many short lines.
A seperate arc for each diameter and angle of spiral would be needed for each individual spiral type.
You could create a file that did all of one face, then rotate, do all of the other face, rotate again, then again, and you have your spiral.
If the blank you are cutting is such that each quarter is greater than the depth your bit can cut, you could do it in 8 steps (or 16 or 32 even), instead of 4.
A jig that holds the blank level, parallel to y and square with z, can rotate acurately about its centre line, and can be held accurately at 90º (or 45º, 22.5º etc) would have to be constructed.
Using the big drum technique in another thread above, and restricting the angle of each cut, and swapping x,y and z around (with a horizontal mounted router,) I guess you could achieve enormous spirals.
...............Mike

[artisan]

There are soooo many things that the Bot is naturally suited to.....like adding 3D panels to your pedestal or inlays or relief carvings. I'm not sure why you would invest all this time in copying a product that already exists unless you believe you can produce it for an extremely lower price..... and competing on price is usually a poor business model in my experience....D

[drodda]

I have a customer that wants these on a poker table. He is the one that gave me the pictures. I have found two sources to purchase these pedestals. Both places are in the $400.00 a foot range. that would be over $800.00 each pedestal and the table has two on it. This would be $1600.00 for just the pedestals and then shipping. This is getting close to the price that would make me take the leap into buying an indexer. The customer is a repeat customer and He really wants me to build his table. I just can't get the better half to climb on the "buy an indexer" wagon as of yet. She still points out that my probe is still in the box and has yet to see the light of day. Yes the probe that I just had to have. I feel that I can produce these at a much lower price than I can purchase them. I just need some other products that the indexer will help with to support the purchase. I have seen so many people that purchased the indexer only to put it on a shelf and never even hook it up.

This is why I was trying to produce the spiral without the indexer purchase.

[Brady Watson]

Dave,
Does it have to be real wood? You can finish just about anything these days and have it look good. With an investement in a mold, you could cast resin in a long 1/2 round strip, heat it with steam in a long PVC tube, and bend it around a plywood core. It can then be sanded and finished like real wood. Heck...you can even cast the resin with the finish in it, then just clear it when it is done.

To do this on an indexer, you will need to make some modifications to your table to accomodate the large diameter. It isn't like you are just going to get an indexer, slap it on the table and begin cutting parts...There's a little bit of engineering and modification on your end before you can do it.

The other option as someone else pointed out, is that you can steam bend wood around a form as well. $800 may not be a bad price considering all that you would have to do to make them yourself...and as Darrell points out, pick a new design that YOU own...that has the added benefit of being easier to make and easier on you all around.

-Brady

[billp]

Dave,
As Brady points out steam bending wood may be an older technology, but it's still around because it works...Any of us who have built wooden boats know that you can take wooden frames and after "cooking" them in a steambox they will be as bendable as a piece of spaghetti for about 45 seconds after they are removed from the box. You could probably use as simple a form as a sonotube, pin one end down, and then "roll" the sonotube to get the spiral shape you want. Plenty of steam bending info on the web, and it's very inexpensive to build the setup ( a steam source could be a barbecue cooker, and we always used a plywood box with rubber bands holding the end caps on so they act as safety valves...) I think the rule of thumb used to be "an hour per inch of thickness"when "cooking" the frames...

[waynelocke]

How much will your client pay for these? If you make it out of solid wood, you are going to have to glue up at least an octagon and probably a 16 sided or more column because a solid glued up slab 20" in diameter would be a monster to deal with. The pictured piece looks like it may be mahogany, but whatever the species, the cost (with labor) of getting the basic column is going to take a big bite out of the $800 before you ever get to the spiral.

It is a real interesting project for theShopbot, but I don't think you could come close to $800 each to build only two of them when you fiqure the engineering and experimentation it would take.

If they cost $800 and your client doesn't want to pay that, maybe you need to sell him a base he can afford. There is always the nexus between what the client wants and what they are willing to pay. Maybe you can design a good and doable base that they will like.

[drodda]

The client paying the $800.00 is not the issue. He is willing to pay whatever it takes but I feel that the $800.00 is a little excessive for the column. Since I have purchased my shopbot I have always eaten the time it takes in R&D to accomplish a task. So far I look at it as if I am paying for training. Even if I am the instructor.

This is just one of the many ideas that are in the development stage that the indexer would be used for. I am not trying to solely justify the indexer for these two pedestals. I would have bought them already if that was the case.

If you have bought a shopbot I am sure all of you have your list of future projects that you are going to get to "when you have time" This just seems like one of those things.

We sometimes get into the "anything is possible" mode of thinking because we own these increadible machines. Sometimes I get to a wall in my design process and this was one of those times. Sometimes you are so focus on being able to do it yourself that you overlook the cost of just buying it from others.

Wayne I have spent countless hour looking at your table modifications to use your indexer. I am very impressed with how you designed your table.

[artisan]

This spiral pattern is a very common pattern for foam columns. As suggested earlier....if you can make the column from something other than wood, you could make a mould from a foam column and then resin cast the pedestal column in say....resin/pecan or something of that nature....D

[davidallen]

Dave

What I meant by quarter is:
view from top

2606


view from side

2607

view from end

2608

Depending on your profile, you should be able to cut 4 flat then join them together for a full helix


da

[drodda]

David,
Would all four files have to be different to offset the spiral up to align with the one you are joining to? Thanks for the pics it shows it easier than trying to read it from a description.


Dave

[davidallen]

Dave,

You can make 4 copies of the same file and offset each by 1/4 of the distance between turns. The pieces will need to be at least 2 turns longer than the finished length.

You'll trim at least 1 turn off each end to get it square.

If the radius is too small for the bit, the profile too narrow or the height of the arc too tall to fit under the carriage, you may need to cut the base into 6 or 8 pieces.

da

[davidallen]

As suggested in other posts, bending the spiral might be faster and more economical.

Another interesting option would be to bend rattan around a form to make open spirals. You might need a metal rod up the center to provide support but it'd be barely noticeable .

da

[marshawk]

Actually, $800 isn't too bad for columns like that.

IMHO...you are better off buying the columns. There is a pretty big learning curve associated with turning on the ShopBot and chances are you would far exceed anything you would make on the job.

Save up for the indexer and practice with it in your spare time. Save up more and get a steadier motor, especially if you want to carve large items.

Don't get me wrong...we make some really nice stuff with the indexer and it is invaluable to us. But the motor that comes with the kit is not terribly suited to turning big things. Call Gordon and ask him what the options are.

Chip

[gerald_d]

...that's why I would prefer to go with a conventional motor and do my own gearing with something like sprockets and chains.

[Brady Watson]

FYI Chip is referring to the PRT indexer...NOT the Alpha version of the indexer. The PRT indexer had a 7.2:1 gearbox (some had the 3.6), a 300 Oz torque rating at the output shaft and no positional feedback. The new Alphas have a 10:1 gearbox on them, with a 560 Oz Alpha motor (with positional feedback) and it will twist your arm off if you try to hold it while it is turning...there's gobs of torque on the new ones. I believe a 25:1 is also available for even more torque (also can get a PRT motor in this ratio)...like Chip & Gene did.

-B

[gerald_d]

I once tried to tighten a bolt on an Alpha (with the motors switched on) and was surprised to feel that the motors on the gantry "gave way" probably at around 40lbs of force. It was a weird sensation because the whole thing just moved back into position again when I released the pressure. All very smoothly and without fuss, but the system was way out of position with far less force than I would have used on that same bolt on my PRT (that doesn't have position feedback). On that experience, I would be nervous of an Alpha motor's "holding torque" for an indexer.

For turning 20" diameter work, you would theoretically need a gear ratio 20 times bigger than for linear work where you have about a 1" pinion gear into the current racks. So, if a standard Alpha has 3.6:1 gearboxes on the linear axes with a 1" pinion gear, you will need gear ratio of 72:1 in an indexer that will turn 20" diameter work. Hence my leaning towards things like sprockets and chains.

[gene_marshall]

Hi guys,
Sorry I have been very busy with a recent muckafuss at my house... My garage burned down.

Seems I have bad luck in that regard.

I hope you don't mind my humble opinion being interjected here.

1) a diameter that large would be a SERIOUSLY slow cut on the bot indexer.
The torque created by the large diameter would undoubtedly screw up the project and burn out the driver. These motors are not designed for that kind of preasure.
2) the idea of making a repeating section that could be glued up after is interesting. The project could be designed and tested in pink foam. My opinion ...good way to go... But think about the alignment of so many sections that closely and sanding the joinery smooth after. Muck work and oportunity for error. Although the programming would be pretty easy in artcam.
3) stem bending... another great idea.
Cylindrical plywood is available and bending half rounds will get you there quickly. Assuming you have experience stem bending. The steamer would have to be a long one, the peices need to wrap several times. the wet boards resist bending as they cool down so act quickly. Seams in the middle will not work too well.

4) chip( this kills me) is absolutely right (arghh)
$800 bucks is a steal for the supplied part.
We charge by L x DIA. x PI plus material.
This part to turn would cost a bunch more than that. Although making this project would have rewards beyond the dollar value.
If the part was one that had to be different somehow, or even if you could sell 25 of them, the setup will cost beyond what the profit will allow.
If you can get it, get it, if you can't, then make it...I guess is my $.02.
So I guess the
hope I was helpful,
If I can help more I am here for you.
Gene

[gerald_d]

Sorry to hear about the garage Gene


An indexer that produces 560 Ounce.(inches) of torque will create a force of 56 ounces at a 10 inch radius. That is less than 4 pounds for a 20" diam job that Dave is considering. (You would be able to stall the indexer with one finger). After looking at the specs, I really agree with Chip's concerns.

[drodda]

Ok now you guys have done it. Chip is going to be walking around all day with an inflated head and ego.
Patting Chip on the back was not my intentions when I posted this. I had already considered calling Gene and just getting it straight from the source.

I had already come to the conclusion that you are making before posting this question. With my replies I was not trying to disagree with any of your suggestions. I was just trying to stir your imagination and see if this was a solvable problem without an indexer. The making it out of four or eight parts was something that I had not considered but feel as Gene does that this is way too much hand work and this is why I bought the Bot in the first place. It's hard to hold your beer if you have to work with both hands?

With that said I have advised the customer to find another pedestal that would be suitable for him or we can just purchase the ones already done. The one major drawback is that the manufacturer of the spirals needs a minumum order that exceeds what I am willing to sit on till another person wants to pay the price for the spirals.

I have put many hours into trying to figure out this solution and appreciate all the time invested here in the forum to try and help. Like many before me have stated you are the reason that I bought the shopbot over all others. the members of this forum never cease to amaze me with their help.

Sorry to hear about your garage Gene. Hope it all works out in the end.

Dave

[Brady Watson]

Geez Gene...Did you buy fire insurance YET??? Sorry to hear about the mishap!

4#s of force on a 20" diameter piece is not unreasonable...especially when you are going to raster along the long axis anyway. There will be very little force against the piece to knock it out of position...and if for some reason it did get knocked...well the Alpha would self correct. It would also be prudent to do a roughing pass 1st and then go back and machine off the .02" skin left by the roughing tool, reducing the finish pass cutting force quite a bit. The biggest concern that I would have at that diameter would be backlash...Yes...a higher ratio would be better at the expense of speed.

If I were doing this on a PRT, I would be inclined to try to get a treadmill motor with belt reduction & a Gecko working for the indexer. I just so happen to have the components for this setup...but I am holding out for an Alpha upgrade.

-B

[drodda]

Is there someone who can make the file for me to cut the spiral pedestal with a 18" diameter. 36" long. With an indexer and the peice in line with the Y axis. Across the tables short side. the rotation will replace the X axis.

Gerald, you mention using an excell spreadsheet to poduce the file to cut this on an indexer. I have someone who may be able to help with the cutting if I can come up with the cutting file. I don't think I have access to the indexer software that SB provides. Someone also mentioned that this software cut spirals in a column easily. I don't know this first hand though.

We are going to test cut a foam one first to try it out.

Thanks in advance if you are able to help with the file.

Dave

[gerald_d]

If your cutter is the right shape, the file for cutting a simple spiral on an indexer is literally one line of code. You have to figure out how many inches of x-travel it takes to now turn the indexer through one revolution. Then you give a x,y move (M2) on a diagonal and you get a spiral (instead of a diagonal across the table if you still had the x-motors plugged in).

[bill.young]

Dave,

The indexer software is included in the Virtual Tools in the Windows software and will do just what you want to do....give it a shot. It uses the B axis to turn the indexer and can be setup with the blank parallel to either the X or Y axis.

If using the B-axis to turn the indexer is a problem you can download the DOS indexer routines from the documentation section of the Download part of the ShopBot web site. The DOS routines are just ShopBot part files and you can be pretty easily modified if you need to set things up differently.

Bill

[lto]

I noticed the suggestions on using steam bending earlier in this thread. I've done something similar to this in the past. This particular pedestal would require a rather long piece of 2" half round moulding in order to avoid a joint, (Not very practical).

Steam bending;
www.barevillewoodcraft.com (http://www.barevillewoodcraft.com)

[drodda]

I was drawing a spiral in Autocad tonight and decided to check to see the length of a piece for steam bending. The piece would have to be over 42 foot long. I don't know about you but I don't think I have anything that will produce enough steam to keep a 45 foot box warm enough to get the pieces to bend. So I think that rules out the steam bending around a column.

[marshawk]

The garage is insured and we are being fairly treated by the insurance company, so far...

About 75% of my worldly belongings were in the attic and are now ashes. The firemen managed to pull out 3 items of mine before they burned: a statue, a camera and MY WEDDING ALBUM!!! It figures that the thing that I should have burned when I got my divorce was salvaged...

Anyway, I am leaving Marshall's Millwork to re-join the Navy so I can re-buy all of my stuff... The insurance company won't pay for a 'round the world trip for me...

[gene_marshall]

HOORAY

[normand]

Hi Dave Did you read this thread?http://www.talkshopbot.com/forum/show.cgi?tpc=2&post=20400#POST20400

[drodda]

Need help converting a file into a shopbot cutting file. Please see my post under conversions.

http://www.talkshopbot.com/forum/messages/31/11401.html?1138918975

[drodda]

Ask and you shall receive. I have the file that I needed to give this a try. thanks for all the help.

Dave