Additional Axises: Tandem Z Axis v. Independent YZ?

[henrik_o]

Hello all. I'm trying to spec a Shopbot target system for purchase (finally have financing!) and one thing that I haven't quite figured out is the options involving added axises.

In the ShopBot price list (Dec 1, 2006) there are three additional axises listed (besides the indexer);

#14510 -- Additional Tandem Z Axis - PRSalpha @ $ 1,995.00

#14511 -- Additional Independent YZ – PRSalpha @ $ 2,895.00

and

#14525 -- Air Drill Head – for automated drilling @ $ 1,750.00

Now, Shopbot's site, the 'Production Aids' page, has entries for the Air Drill Head and a "Second Z Axis", the latter of which is listed as;

"Second Z Axis
A pre-assembled second z-axis that can be set to run independently. With the second Z set to run independently of the primary Z, the Control Software will allow you to switch between Z's. You can put a different cutter in each Z axis and have a 2-tool, tool changer. Most panel processing operations require no more than 2 cutters and having 2 Z's can significantly improve efficiency for a reasonable cost.

(#14521) Additional Z Axis (PRTalpha; includes motor, driver, wiring and router bracket); PRICE $1,995"

Since this obviously (looking at the price) refers to the Tandem Z, what about the Independent YZ?

What is this, is it new, and why the $900 price gap?

[Brady Watson]

There is a $900 difference because the independent Z also includes an additional Y axis motor to allow it to be parked off of the active machining area.

What is your application? An independent Z can be used for machining with a 2nd tool, creating a poor man's toolchanger. The tandem is for cutting 2 simultaneous & exact parts side by side. The air drill is only for drilling. You will need to ask yourself if you really need a 2nd Z axis.

For those that don't know, ShopBot offered tandem and independent Z axes on PRTs. I believe the option was just on hiatus for a while...

-B

[henrik_o]

Brady, thank you for the prompt reply.

I'm afraid I don't quite understand your explanation. The 'Production Aids' page states that what I assume is in the Price List called #14510 (Tandem Z Axis) --but on that page is listed as #14521;

"[...] can be set to run independently. With the second Z set to run independently of the primary Z, the Control Software will allow you to switch between Z's. You can put a different cutter in each Z axis and have a 2-tool, tool changer."

Are these two different things? The numbering gives no clear guidance, but the price is the same so I assumed they are one and the same.

Your explanation that the Tandem means an axis that works in parallel, and only in parallel, with the main axis, while independent YZ is the 'poor man's toolchanger' certainly makes sense, but the material on Shopbot's site is confusing.

I hope to get to my production application(s) later on, but for now I'd like to just understand what is the difference between these two.

[Brady Watson]

Your best bet is to call ShopBot directly. They can explain the differences in detail & help you to choose the best setup for your application.

-B

[henrik_o]

Yes, I'll contact them directly.

For what it's worth, the main reason I have finally decided on getting a bot is because we'll be launching a line of caskets towards the fall, and our non-cnc machinery won't cut it for this application.

I'm looking for an additional axis because the caskets will be dovetailed, so I figure a second axis with another tool is somewhat of a must.

[ttwark]

Henrik,

We purchased a 2nd Z on our PRT alpha for the purpose of being able to use multiple tooling with out bit changes. We are finding that the control software doesn't support a 2nd Z very well. It is almost like the Engineering dept added as a second thought. The user manual is about worthless with regards to the setup of the second Z. Not being very savy with the software and design parameters I have all but given up on the second Z until we get the basic machine down. I believe that the value ot the second Z will be hard to recoup in my particular case. Eventually I know that I will be happy but the learning curve will be steep.

[henrik_o]

Brady, you were right.

I recieved an explanation from shopbot stating;

"Independent Z’s have a very limited application. They are really for someone who wants to cut two of the same parts at the same time. The Tandem Z is to use a different bit, instead of needing a tool changer. This option is more versatile."

[henrik_o]

Tom,

Hum, that's not very heartening to hear.

I've been interested in a shopbot for years, but so far investments have gone towards other machinery since we do quite well with the standard equipment we have.

However, as previously said, later this year we're going to start producing funeral caskets that will be dovetailed, and our non-nc dovetailing machine wouldn't be able to do the long ends easily, plus there's some other intricate geometry that would make a cnc machine a necessity, more or less. I've calculated a ton on this, and we could do it with our current machine park plus some standard additions, but that would mean we'd have to re-arrange the shop for this, and that's not ideal because we'll still do plenty of other jobs so re-arrangement downtime would be a killer.

What do you other guys with tandem-z's say?

[Brady Watson]

Henrik,
I think what you are after here is a custom setup. Correct me if I am wrong, but what I am hearing is you want one router to do conventional cutting, just like any other CNC router and you want a 2nd Z to be dedicated to doing dovetails. Correct?

If this is true, then you will need to modify the 2nd Z's tool mounting bracket so that the 2nd tool's shaft is pointing horizontally instead of vertically like the conventional tool. The dovetail bit would then be lowered to do the dovetail joint on the right side of the table. This is a challenge to setup, program and implement in a production environment without having the CNC on the 2nd floor with the board being fed up thru the ceiling. Do you follow me? You can put the 2nd tool with it's shaft vertically and do the dovetail joints (if they are to be done on the ends of the boards), but you'll have to feed the boards thru the ceiling/floor. Others have done this...and it doesn't require a 2nd Z. Only a couple flicks of the wrist to change out cutters.

Does that make sense or am I not completely understanding your application?

-B

[henrik_o]

Hello Brady, thank you for your reply and for taking the time to look at my application. Much appreciated!

quote:what I am hearing is you want one router to do conventional cutting, just like any other CNC router and you want a 2nd Z to be dedicated to doing dovetails. Correct?

Correct. There's a catch here however, in that the short ends would not need to be machined on the shopbot.

Here is an image of a prototype casket (http://members.arstechnica.com/x/hiphink/shopbot-casket1.jpg) to give a better idea of what we're dealing with.

Today, we do the joints on a 'sinkmaskin' (dovetailing shaper). Here is an image (http://members.arstechnica.com/x/hiphink/shopbot-casket2.jpg) of the setup in that machine as we do it today.

Currently, the process on this machine is slow and cumbersome, because aligning and handling the short end and long side that are to be joined is a mess -- the machine was not designed to work with lengths like our casket sides but rather for kitchen-style drawers. Actually routing is a snap, once everything is set up.

So, my idea is to have the long sides machined on-sheet by the shopbot. The short ends would be cut on the shopbot, but then taken to the dovetailer where they could be routed very fast since one could work with a fixture, just slap it in and go, no need to align and mess around in general.

Of course, if the shopbot could do this as well, that would be great, but then we'd need to be talking about it doing many of them in one run, otherwise there's plausibly no time saving doing it on the 'bot, and precision is great on the dovetailer too.

I hope this makes sense. Please tell me what you think about my idea, it would be great to get feedback from someone that's proficient with the shopbot.

I should perhaps also add that the shopbot won't be doing this application day in and day out. It will be 20-25% of our shop time, not more than that to begin with. So the bot will do many other things as well. But this application is why I want to get into cnc, so it'd have to work great for it, everything else essentially being a bonus.

[henrik_o]

Brady, I'm trying to think a little about the solutions you discussed that I didn't address (sorry -- and also sorry that I didn't specify halfblind dovetails since I realize that may make a big difference).

If I understand you correctly, with a vertically oriented cutter you need to orient one component of the joint vertically, which is how our current machine works. With a horizontal cutter added to the main vertical cutter axis, both components can be machined horizontally, but the parts routed by the vertical cutter can only be machined in a long but narrow perimeter on a specific side of the table?

If that's what you mean then I follow you since it accords with the geometrical properties of the joint.

[lex]

This reminds me of a material I saw some years ago that is very intersting. It was a composit made from crushed or broken pecan shell. It made a very pretty casket, was extremely light, and quit strong. If I remember the material name I will post it.

[oddcoach]

HELLO Henrik
if the long side is to get the 1/2 blind dovetails
then the shopbot can do everything you want. to do the short ends you could make a fixture on the end of the table to hold them vertically and machine the ends.i would mount some stops on the table top to register the part. they could then be slid out of the way for machining. possibly mount them on destaco clamps then the slide mechanism is done for you.
good luck with your project
john

[henrik_o]

Hello John,

Thank you for the encouraging words.

As I wrote to Brady, the short ends is currently not a problem for me since our dovetailer can handle them with good precision and great speed (if processing only one part per moment).

Doing the short ends on the bot would be great, and something I would like to look into more, but it's not a necessity.

If the 'bot could do the long sides, optimally without a manual tool change (hence the second z and this thread), I'd be a very happy camper.

[rhfurniture]

I think that to do that fully, you would need a horizontal router/spindle and a lot of tricky toolpathing (or a 5 axis machine with software).
It would not need a lot of power, so a small spindle or even a Kress router head would do.
I have a 900watt router bolted to the back of my single Z on my benchtop and it cuts round end tennons and mortices flawlessly:

7353
(See the collet on the right).
For your purposes it would have to be aligned along the long table axis.
The wonderful thing about shopbots is that they can be rebuilt to do exactly what you want (prt anyway).
On a new PRS I would have thought that a second independant ZY axis with a horizontal spindle/router head aligned along the X axis would do it, but talk to shopbot as the spindle/router mounting would need consideration.
Good luck, R.

[henrik_o]

Ralph,

Very interesting modification. Is this supported in software, or would one have to assign the added axis to a free axis in the control and hand program its use?

[rhfurniture]

If you had it as an independant Z on a PRT it would run as a forth axis in the control software, but it is just another use of the X, Y and Z(or A) axes. My guess with a PRS is that you would have 2 additional axes - a new Z and a new Y.
For tennoning and morticing, I have developed some interactive partfiles using INPUT and PRINT commands that make it clear what I am cutting, as I would find it very easy to get confused - for instance JH works as it would for the vertical router with possible unforseen results. You could draw up the whole cycle in a 3D cad program, with a vector representing the path of the middle of the end of the cutter, export as DXF and use the FC command (file convert) on it.
R.

[henrik_o]

Ralph,

Regarding the programming bit, it doesn't click for me now, but I very much hope to get back to you and other proficient 'botters in the future.

Thinking out loud on a couple of issues here, smack me if this is silly or just plain unworkable.

First, if one were to add a horizontal-rotation axis, could one not work around the z travel problem by vertically offsetting the piece(s) to be machined? I don't know how much room there is on the gantry, but adding an axis that would work parallel to a side of the table, from the outside-in, sounds complicated if at all possible. There's also the issue of a very limited machining perimeter.

If one had sufficient z allowance, and my target system if I choose the Shopbot is a 14" z, offsetting the pieces such that the added horizontal axis stayed clear of the spoilboard for machining, and also clear of the total offset for travel (not channeling), you might end up with something like this rough sketch;


7354

Second, and now I'm really out of my depth, if we return for a moment to the unsolvable problem of machining the tails in a dovetail with a horizontal piece and a vertical-rotation cutter. This shouldn't work, ever.

But nor should inside corners. Yet I understand, though not exactly how, that this is in fact possible for some inside corners (those that are at an angle) by sloping the cutter motion out of the cut. In other words, in 2D space it is impossible to solve, but in 3D a fix exists.

I'll readily admit that my spatial intellect is insufficient to sit here and analyse the joint in my head, but intuitively, if we’re talking half-blind dovetails and the tails had a sloped cut down and into the material, wouldn't it be possible to use another cutter to either plunge or back out of the material at an angle, cleaning up the radius? I.e, one would effectively create a "halftail and reverse" dovetail joint -- it would need to be joined backwards by sliding from inside out into the joint rather than from the outside sliding in, but it should still mate. As such, the pin channels would need to be elongated (and widened?) enough to allow for the reverse slide, which is a problem and compromises the structural properties of the joint. Still, from the outside, even though it isn’t technically a ‘proper’ dovetail joint, it would look absolutely right. The extruding pins channels could be backed by an insert, which might actually serve to both strengthen and allow for more flex than a proper dovetail joint. Another benefit of the reverse sliding motion is that a tight joint would cause compression rather than splits in the tails.

Ok, I’m totally way out of my league here. There’s probably a very easy answer as to why this is not possible.

But the problem is interesting. I’m new to thinking of machining in 3D, and I just can’t get enough.