Go Back to handibot.com
sign up or log-in

Advanced

Handibot 3 - is a Lowcost DIY International machine possible?

Posted by AndrewG 
A discussion started here, that touched on a number of issues in the world of Handibot.

Brian shared a lot of the history of Handibots evolution and why rather than the $1k machine originally envisaged, it is now an almost $5k machine.

My personal agenda as an engineer turned furniture maker who uses CNC services but want to get a first machine into my workshop, is that Handibot is the closest thing to the solution I want but is:

A.Too expensive - my target would be $2k - $2.5k (yes the Kickstarter/intro price...but I'm not an early adopter!)
B. Not International - I'm UK based, so don't want to be reliant on
C. An evolutionary dead end - I accused this forum of being dead, but a few good people have cleared that up!)
so

As Handibot is Open Source, I wanted to see if I could build my own.
- The evidence is inconclusive that anyone has actually achieved this ( a guy was selling plates on UK Ebay after an attempt ,see end of thread here
- A Swedish guy has some Youtube videos up suggesting he is trying here

All the part models and BOM are available here and build guidance here
BUT you quickly run into problems, for example:

The Stepper/leadscrew are proprietary and not available. 3D printing market has made Nema17 steppes with leadscrew available but not Nema23 with quality bearing and 0.5" pitch leadscrew, so a redesign would need to address this

So why persevere? Well I believe the original brief for the Handibot was spot on:

- A reliable workshop machine needs to use proper profile linear rails to give rigidity and low maintenance. Open Builds have done some great things and currently the Workbee represents probably the best entry level CNC router at $1.5-3k. But all those skate wheels need careful adjustment and maintenance. Now linear rails have dropped in price a lot and increasingly designs on Open Builds mount profiled Rails on their VSlot extrusions..the future is linear rails!

- Small is beautiful,Most of us in Europe and Asia just don't have the space and large workshops you Americans seem to have. Whilst larger format CNC router have their advantages that is a considerable commitment of space. Where Handibot pioneered, Shaper Origin follows and with great software and a European launch last month, it is tempting but misses a key advantage of CNC - it cannot work autonomously. I need that productivity advantage.

- The handibot has evolved as a design into a lovely package of CNC capability and the applications and additional jigs and fixtures have demonstarted how it can be used. Sadly small does not also mean cheap - the Handibot still has the same number of motors/rails/controllers as a much larger machine. Most hobby woodworking CNC routers come down to :
Frame= aluminium extrusion and plates
Linear motion = profile rails (or V slot in the past) plus ballscrews/leadscrews (or belts) plus stepper motors
Controller = Arduino and gbrl (or similar) plus PC or SBC like Rasberry P

Where the Handibot differs from the rest is the frame, where HDPE and Aluminium plates replace extrusions. Whilst in theory the Handibot design leverages CNC routing to make the plates, the complexity makes it an expensive process to get a one off set cut (I think, I am still exploring that)

Meanwhile, new designs throw up new concepts. Unsupported rails as seen on the cheap 3040 routers from ebay are terrible as they flex excessively. A new market entry Onefinity have revived the idea by supper sizing them. I like the low part count of the approach but fear the addition of a 'stiffy' accessory suggest they still have some rigidity issues). Another to watch.

So trying to share and summarize - but the question is simple. Could I build a version of Handibot DIY to get the benfits at a reasonable price?
The payoff for the community would be more European and Asian makers taking the format seriously and contributing to the community.
Ok got some time over the weekend to consider the options, and made the following decisions.

1. Stick with 6x8x3.5 work envelope (or it wouldn't be a Handibot!)
2, Exoframe - prototype will be plywood and alumiunium - objective to be able to machine with manual woodworking tools. Possibly cut panels eventually on larger CNC and upgrade to HDPE.
3. Linear motion - copy use of HGR15 Hiwin style linear rails and 7 bearing carriage design, though probably source cheap chinese vesions
4. Stepper/leadscrew - Significant redesign needed as Haydon Kerk motor/screw packages not available. Means separate motor, coupler, bearings, leadscrew assy that needs to be fitted into limited space. Stepper seems to be Nema23 2Nm (289OZin) and leadscrew a 1/2" pitch. Current thinking is to adopt OpenBuilds standards and us their T8*8 leadscrew, couplers , bearings and antibacklash nuts. as on the Workbee
5. Electronics - No firm decisions. Not wedded to Fabmo /G2 as gbrl seems more common in open source community and increasing number of Gcode senders available with web interfaces to run on a RPi like Octoprint. In UK Workbee by Oozenest(the original designers) is now supplied with Duet2 controller that get very good reviews, lot so support as a result and has 2.8A stepper drivers build in.

Also Duet 2 uses TMC2660 drivers that support sensorless homing - it would be nice to avoid adding homing switches to design.Though wonder if Handibot homing is achieved just through lost steps?
_____

SO priority seems to be to look at impact of swapping Workbee style stepper/bearings/leadscrew into Handibot design and seeing how much that compromises neat layout of Handibot.
Andrew,

Thanks for starting this new thread!! I'm hoping to have something interesting to share soon. Saturdays are the day I usually get to tinker a little bit with new designs and I had planned to put something up last saturday -- I'd spent the week modeling and toolpathing and came in Saturday to cut out the parts and assemble -- turns out the rails I had picked out are about 0.25" longer IRL than they were in my model ... which meant changes to almost every part because they wouldn't fit! Fail! So I'm gonna try again this saturday...

Haven't solved the issue with the motors yet...but I'm always looking around for alternate vendors for motorized leadscrews ... and I think they're becoming a lot more of a commodity these days.
Brian,
Good to know you are on the case and await with interest your reveal! Know all about IRL snafu derailing plans, thank goodness for parametric modelling.....yet one never seems to have accommodated quite that particular change!

I think to me the key objective is retain the key 'handibot' characteristics that will allow us to leverage the innovation in accessories and work practices, whilst generating a design variant that allows DIY home construction to reduce the cost if getting started. I hope that will grow and internationalize the community whilst still supporting Shopbot's original vision .

Motors - whilst 3D printing seems to have made Nema17 steppers with integral lead screws more widely available, Nema23 versions don't seems to be available. Having to accommodate a motor/bearing/leadscrew assembly badly impact the tight design layout and simplicity. Still pondering the best approaxht...

Andrew
The handibot has evolved as a design into a lovely package of CNC capability and the applications and additional jigs and fixtures have demonstarted how it can be used. Sadly small does not also mean cheap - the Handibot still has the same number of motors/rails/controllers as a much larger machine
Hi everyone! I wanted to provide a long overdue update to this thread.

Pretty much the week that I really started to get excited about all the input from everyone and dig in on some new design ideas -- a couple of changes happened throughout ShopBot (as was the case for most of the world around that time!). A couple of people moved on to new jobs at other companies. People that we'd been contracting with either could no longer travel to Durham to work with us or became overwhelmed with other responsibilities -- etc, etc...point being that it is probably the same story you're all living through at your jobs right now (unless you're lucky enough to be retired!). We all regrouped and reorganized to tackle the new challenges.

As a result; my role at ShopBot has shifted a bit and become much more focused on inventory, manufacturing and what we're calling "customer experience" which includes everything from quality, to customer service response time. For at least the next 3 months, these things are going to consume as much energy as I can provide (and more!).

So, however much I want to engage in more of this back-and-forth with all of you about handibot, I've got to be honest that I won't be able to jump right on things as quickly as I want to -- that it'll be more like waiting for a lull in the "storm" to hop on and do a little bit of design work and share what I can. In the meantime, I'm getting my creative kick writing database queries and building spreadsheets with google apps script; working on new documentation and policies; sourcing parts that have been cut off by supply line disruptions. All of that is surprisingly a lot of fun!

This past week, I was able to finally dig my way out of the weekly pile of little problems that needed solving by about 2:30 on Friday afternoon -- a new best! I'm hoping to eventually be free by 10am on Friday most weeks and be able to devote a couple hours to the things discussed in this thread. Hope to talk to everyone again soon!

Brian
Hi Brian,

Not at all surprised if you, like many others, are having to work hard to just help your business survive and prosper in these challenging times. Good luck with your expanded duties.

As and when you have time to share your thoughts on the evolution of the Handibot concept, I am sure we will all be very interested. For me, having looked at the Handibot design in more detail - it is not well suited as a first DIY project, so I am looking more to the OpenBuilds ecosystem which has the design evolution and community to support a DIY builds.

Always interesting to see how thing evolve
Best wishes
Andrew
Sorry, only registered users may post in this forum.