Hello Boris and others,Well ... it is probably me who let you down on the crawler. I have puttered with a number of versions and while I think the concept is cool and plan to keep working at it, the one in the video just had too many issues. Here's a little more reflection on that process ...
GeneralFundamentally, the crawler was about creating different ways to work larger areas with a Handibot. A related aspect of this large-area challenge extends the requirements to: 1) being able to do it with a manual version (presumably inexpensive); and 2) to have a more automated version (more expensive, but the kind of thing we were going for with the crawler). The limiting condition for large-area solutions is one where you have added so much additional mechanism, size, and cost that it would be better to just opt for a larger CNC tool.
I personally think there are probably dozens of good, interesting solutions for doing larger areas with Handibots in practical ways. In fact, no sooner had we started shipping our large-area-jig accessory, than a number of Handibotters reported significant improvements on it (Mark Evans, above, has some great variations). Others (such as Eric Schimelpfenig) have reported creative automated solutions using jigging and Handibot-produced material-re-positioning (and see
video at bottom of this blog). There is some really creative exploration going on. It helps considerably in our thinking about which directions to go in working on accessories. More importantly, it helps others understand Handibot as an open platform that is readily built on by us and by others, and which does not depend solely on "official" accessories. Handibot represents a new and different form-factor in power tools and has enabling new capabilities that have not been defined. It has been our goal from the start to have a tool that allows exploring such capabilities.
More SpecificallyI have come to see the large-area challenge as best dealt with as two challenges (I may be wrong, but I think this makes it more manageable). The first is the challenge of handling long sticks of material such as say, a long 2 X 10 for making stair stringers, or a long 1 x 4 into which we want to 3d carve continuous molding. The second challenge is working over a large piece of sheet material such as plywood, to cut out say, some flat-pack furniture.
I have come to believe the solution to the first challenge will be a fixture or jig under which the material is precisely moved, either manually or automatically while the Handibot remains fixed. The crawler moved the Handibot on the material. As we were practicing it, the crawler did do a reasonable job of the stair-stringer type work, but required a long and awkward fence that included gear rack for motion. And, consistently moving the Handibot, even with a powerful stepper motor, was always a little bit iffy. In the case of trying to do 3d carving, the Handibot, which was physically riding on and holding down the material, kept riding-up over sawdust so that the height of the carving was not consistent from tile to tile. We’ve thought that the improved dust collection of the V2 might address this issue – but the overall awkwardness of the long piece of gear-rack on a fence would still be an issue. I thus think moving the material will work better and provide consistent machining height. There are several blog and forum reports on moving lengths of material under a fixed Handibot, both manual and automated that I think offer good suggestions on how to accomplish it. We hope, of course, to come up with some helper accessories to make it easy for everyone to implement.
Regarding the second challenge, cutting parts out of large sheets, I believe this might be well handled by a Handibot mounted upside down in a table. Sheets within a limited maximum size, say 2’ X 4’, could be manually registered over this table for cutting, or they could be positioned automatically for cutting. We found that when we were using the crawler to move a Handibot on the sheet goods, while the principle worked, again we found sawdust getting under the crawler skids required so much fussing that it was almost easier to just use the jigging system. And because it was necessary to manually move the Handibot between positions within the crawler skid for each new row of work, the overall process was not very automatic. I did manage to make parts for a couple of the Handibot Crates this way with the crawler … but it was not much fun. We have also explored a type of gantry structure for the Handibot with moving table underneath … but we quickly found that we had just build a larger CNC tool.
Other ApproachesOne of the questions raised in your post above was about whether we had used wheels or bearings on the Handibot and if we were lifting up the crawler for moves. The answer is no. We just kept things simple. However, I think what you are suggesting is a direction worth exploring in terms of giving Handibots some sort of motion and self-positioning capabilities (even a manual, raise-and-roll system might be interesting). We have experimented with a Handibot drone (you’ve probably seen the
video), but it had no ability to precisely position itself in the way required to do serious tiling work.
Handibots’ control and electronic system allow adding lots of additional capabilities for exploring various approaches to doing larger areas and other types of work. I’m appreciating from your comments that we can do a better job of documenting and providing resources for others who want to do this type of exploration. My “take” is that it may not be necessary or appropriate to invest a large effort in describing the specifics of an experiment like the crawler which did not work well enough. Rather, that it would be useful to Handibotters if we could provide more general info about things like hooking up extra stepper motors and explaining how to control them and integrate their actions for doing Handibot tasks. We probably do take it too much for granted, since it is easy for us in the environment here at ShopBot to putter with this sort of thing (and Handibots are designed for it). Yet the materials and resources are not always readily at hand for others. We are probably guilty of presuming it is all more obvious than it actually is.
So, I’m thinking that in addition to continuing our development of accessories, we’ll make a point of delivering more general info about how others can do the same sort of puttering. At the same time, we’ll see if we can put together some experimenter’s kits with things like motors, gearing, and other effectors that hook up easily to Handibots and allow others to push the tool’s capabilities. Handibot is really a quite hackable smart power tool and we look forward to everyone’s participation I moving it forward …
Ted Hall, Handibot / ShopBot Tools