HandiBot accuracy in X, Y, and Z?

Dear HandiBot users,
I am considering getting the HandiBot V2.0, trying to get used to the idea of tiling to do larger pieces, instead of getting a larger CNC.
For those of you who have done tiling, how onerous is it? Has anyone made a really large project, like a house or even a garden shed, with their HandiBot?

And for carving small things, what kind of accuracy can I really expect? The specs say 6.3 micron steps, but I imagine actual accuracy is substantially worse, and may vary in X, Y, vs. Z. What is your experience with the accuracy/repeatability/ minimum feature size of a carving, with things well held and careful use of feeds and speeds?

Thanks,
Steve

Steve,

Like any tool, you have to mask its weaknesses and play to its strengths.

Tiling is not really onerous once you get the concept down and learn how it is handled in Vcarve.
Tiling is a no go for precise carvings as you will never re-position the unit with enough precision to not have the tiles show.
For something long with gaps it works well, say a valiance, you can make the tiles any size you want and have them fall so that that 1/32" or so error just doesn't show.
If you can work out re-positioning to the accuracy of the handibot you will be a Handibot community hero!

On the other hand, for large non art work it shines. Several people have worked out ways to hold 2xX lumber and re-position it. The optional laser works great for doing cutouts on sheet goods. The jig they sell is fast and reasonably accurate.

As to accuracy, I've cut aluminum and then miked it out to plus or minus .005 , it is not a milling machine but for a lot of uses it is "good enough".

I have cut a 39" diameter clock with the indexing jigs on tongue and groove pine flooring scrap. It worked well enough but there seems to be induced "slop" - user error in my opinion. I have been battling the looseness of the indexing jigs the farther I go from 0, 0. My friend and I are going to make a set of 48" long aluminum indexing jigs. Our thoughts are to eliminate the wiggle room in the jigs, and reduce the amount of jig repositioning. Our goal is to only have the squaring issues of the bit itself.

For the 1/16" errors on the farther out tiles, most have been "hidden" with the use of a Dremel tool.

Hopefully the clock image loads for your review.

This forum hates attachments for some reason. You can link to dropbox and it will come through okay.

Steve K Wrote:
-------------------------------------------------------
> I have cut a 39" diameter clock with the indexing
> jigs on tongue and groove pine flooring scrap. It
> worked well enough but there seems to be induced
> "slop" - user error in my opinion. I have been
> battling the looseness of the indexing jigs the
> farther I go from 0, 0. My friend and I are going
> to make a set of 48" long aluminum indexing jigs.
> Our thoughts are to eliminate the wiggle room in
> the jigs, and reduce the amount of jig
> repositioning. Our goal is to only have the
> squaring issues of the bit itself.
>
> For the 1/16" errors on the farther out tiles,
> most have been "hidden" with the use of a Dremel
> tool.
>
> Hopefully the clock image loads for your review.

Uploaded the image to Dropbox.

[www.dropbox.com]

[www.dropbox.com]

[www.dropbox.com]

Maybe it'll work now.

I drew the 48" jigs in SketchUp (version 7) and exported them to a .dxf file. Sent them to our local machine shop. 1/4" aluminum was quoted at $100 for each template. I ordered one base and two 8" x 48" jigs. I couldn't afford 1/2", so that's unfortunate. I should have them in by Wednesday. I'll take some pictures and post them. I think I will be able to eliminate a lot of slop from the plywood index jigs. I'll let you all know. I hope the image of the clock comes through this time. Very proud of how it turned out, especially as a novice.



Edited 1 time(s). Last edit at 06/18/2016 10:40PM by Steve K.

Steve,
the dropbox links are all working and that is nicer than anything I have built to date. I may be the self-proclaimed biggest fan of the handibot but I definitely am not the most talented. Any jogles as I call them, don't show in the photos but I am with you that my biggest complaint is nat's butt accurate indexing is hard to impossible and I really hope there is a secret program going on in the skunkworks for this.

I would love to see an instructable on how you worked through this.

It has a few joggles on there, but I smoothed most out to almost unnoticeable. The problem with building anything is the builder always sees the errors, even when there aren't any. The clock is pretty large at almost a meter across. I did cut the 12 border pieces out with the HandiBot, too. I used VCarve Pro to draw the entire layout, then drew the border. Once done, I made overlapping 69 degree arcs into a single file. It still required three tiles, but it cut them perfectly. After that, it just turned into a puzzle.

I'm a novice woodworker, just making things for friends and the honey-do list. I prefer country style because errors and imperfections are expected. I found the HandiBot makes me look like I know what I'm doing.

The CNC is fascinating to me. I went with the HandiBot for the price, the versatility, and the community. Everything gets better with usage and understanding. i think the best use of the HandiBot is going to be with user developed / created jigs. It's small footprint and ease of use lets you focus on the jigs and how to improve workflow. I'm looking forward to messing things up on the scrap wood. That's where all the learning happens.

Beautiful work Steve! I'll definitely be sending this around to everyone at ShopBot--they'll love it!

As for your squareness issues--I spent last night thinking about how to write a document about squaring the new tool. Of course, when we build them we don't have the Y motor in the way and squaring is pretty easy (done in the same way the old doc shows). Once the tool is assembled you have to remove the Y motor to get to the bolts that you'd need to adjust. I left a slot where you can slide the Y motor out of the way for this purpose but its still kind of a pain in the butt.

One alternative way to square the tool would be to put a shim between the HDPE mid-wall and the aluminum plate that the X and Z axes are built onto. The whole X-Z assembly is built and squared outside the tool and then installed once the frame is complete--it is very rigid on its own. I had meant for people to be able to adjust for any depth change from X=0 to X=6 by loosening this plate and tipping it up or down. But now that I think about it, you could also square the X axis to the Y axis by loosening the bolts on one side of the plate and pushing a 1/32 or 1/64 shim into the gap and then re-tightening. You might also be able to get a little bit of adjustment just by tightening the bolts on one side and pulling the plate inwards that way...

I'll give it a try on my tool and see what works best. Will report back with a doc when I figure it out. Let me know if you try something similar.

Brian

Ok, I tried both methods and I'm not quite happy with either. Tightening the screws doesn't have any noticeable effect of alignment. Shimming the X Rail Backing Plate does--but it just doesn't look right to me...I'm going to give it some more thought before documenting a procedure. But feel free to try shimming your handibot to make fine adjustments to alignment of Y and X