Blog


2015/03/06 - No dust, just geekery / upcoming projects

posted Mar 5, 2015, 10:11 PM by Jeff Balderson

It's been way too cold to do anything meaningful in the shop lately because of the single-digit weather we've been having.  I had a few pieces that a local cub scout pack asked me to make by mid-February.  After that was complete, I've been avoiding the shop due since it's been averaging around 45-50 degrees and it's just too cold to comfortably work within running the electric heaters for hours ($$) beforehand.

In the meantime, I've been playing with my 3D printer a little more, and doing some electronics work for me and with the boys.  GeekInTraining-#1 decided he wanted to get a calculator kit to solder together, which is about halfway done right now.  He's actually getting pretty good at soldering.  I need to start teaching him how circuits work and what the components do soon.  I have a couple of small LED displays waiting for me, one needing to be soldered and one finished, just needing testing.  

One of my recent prints was the parts for an EggBot clone.  The body was an 11 hour job which I was quite happy with when it was complete.  I had turned on the time lapse in Octoprint, but the video didn't get generated correctly.

I posted this video on Youtube of a job I just finished running tonight.

The EggBot is next on my list of projects.  Hopefully, once it's done the weather be warm enough, 40 degrees or warmer outside should do it, and I can get back into the shop.   

As far as Shop projects go, the list is:
  • Redesign my existing hold downs for the CNC machines and cut them out.
  • Finish the dust boot prototype for the OX CNC machine, or at least at a minimum put some plexiglass shields in place to keep chips/dust off the side guide rails.
  • Ron Paulk-inspired torsion platforms for the X90 CNC machine
  • New assembly/work table cart
  • There are a few signs that I want to make in the queue, including one of my logo that I'll probably post as a start (design) to finish (lacquer) Youtube Video.  I'll probably do those somewhere in the middle of all of the above

2015/02/12 - Adventures in upgrading software

posted Feb 11, 2015, 10:06 PM by Jeff Balderson   [ updated Feb 11, 2015, 10:30 PM ]

The other day I sat down at my workbench, fired up my CNC control program to start running the first of two jobs I'm working on.  A prompt that there was a newer version of the software greeted me and I figured, "Sure, why not?"  The upgrade went smoothly, as well as the controller firmware update.   Little did I know that was the last thing to go smoothly that evening.

I started running the job and it was acting a little funny the entire time.  The movement was jerky and it actually missed a few steps while doing the final outer profile cut for the piece, but I was able to shave a little off the sides to square them back up.  Fortunately, no harm done and the finished piece came out very nice.

So, why not?  How about the fact that my second job has to be ready in a week?  

Long story short: Start second job; catastrophic problems with tons of lost steps, much worse than the previous job; check job and hardware, find nothing; eventually roll back to previous version of software to find problems mysteriously gone.  Fortunately, I have enough extra stock it shouldn't set me back anything more than time, but it's too late in the evening to re-run that part.   Next day, research an inordinately long time to find an obscured mention of a checkbox that should be checked when using my controller, reinstall new version, check checkbox and voila, it works perfectly.

So all my problems and frustration turned out to be a simple checkbox that apparently was benign in its previous setting on the older version of the software, but is now troublesome.

Moral:  Upgrade only when you don't have something with a deadline looming.  

It's a nice moral. It's simple and makes sense.  It'll probably fail entirely to cross my mind the next time I'm presented with a similar situation.

2015/02/01 - Putting together a woodworker's tool set

posted Jan 31, 2015, 10:23 PM by Jeff Balderson   [ updated Feb 6, 2015, 1:47 PM ]

I've talked with a few people over the years that are interested in woodworking, but don't know where to start or what tools to get.  How learn, what to lean and what tools to buy are completely dependent on what you want to build.  If you're really interested in building outdoor furniture, your skill and tool needs will vary quite a bit than someone interesting in building Queen Anne dressers.

Getting Started
Its hard to get started in something when you don't have an end goal in sight.  Find some simple projects to tackle until you start developing a repertoire of skills.  Working on shop projects is a good way to develop those skills, but not have to worry as much if you make a mistake.  Remember, the sign of a good craftsman is how well he hides (and never mentions his mistakes).  

Here's a few places I visit for ideas and inspiration:
Learning techniques
I like to feed my brain while I'm doing other boring tasks.  The easiest way to do that is always have an earbud in your ear, listening into podcasts.  Hint: listen at 1.5x or 2x speed to make the information download even faster.  Here's a few podcasts I listen to:
Read Books.  Visit a library (remember those?).

If you prefer to take classes, here are some places to look into :
  • Hacker/Maker Spaces - Some may teach classes and lead workshops on all aspects of woodworking.
  • Community Centers - Some have Woodshops with various levels of commitment and cost.
  • Woodcraft - all woodcraft stores offer classes, as far as I'm aware.
  • Community Colleges
  • Woodworking Schools - here's a list at Fine Woodworking
Last but not least, once you hear about something that you think might be useful, try to practice it.

Initial "required" tools
The list below are the tools I think you'd need to tackle most small househole woodworking projects.  I'm assuming you have a few tools needed for basic home repair (e.g., screwdrivers, pliers, wrenches, basic claw hammer, safety goggles and ear protection) and will focus on the pure woodworking ones.  

Just remember that with tools, you usually really don't get what you don't pay for in most cases.  Low-end tools (e.g., Ryobi and Skill, just to name two) are usually just that. They typically will be lower powered, use more plastic components (e.g., gears and housings) that will wear out much faster,  potentially deform more, be less feature-rich and be less accurate than their higher priced counterparts.  There are exceptions to every rule, however, and it depends on how frequently you'll use the tool and how accurate you need the tool to be.

The couple of links I provide below are for clarification since sometimes the tool terms can be confusing or overly generic.   The links are to the ones I have or have used or newer models if mine are no longer in production, solely as a basis of example.  Don't take my word for it, however, definitely do your own research.
  • Circular saw - crosscut and rip stock; cut down plywood.  Get a rip blade, combination blade, and a high tooth count (e.g., 60) plywood blade.
  • Straight Edge/Track - Used with the saw to cut long, very straight lines
  • Drill and bits - Simple plug-in drill with a decent selection of bits.  I'm still my grandfather's ancient plug-in drill and bits (I resharpen them)
  • Jigsaw - Cut anything that's not in a straight line.  Still using my grandfather's here as well.
  • Small set of chisels - cut notches, clean up cuts, resize pockets, chamfer edges
  • Set of clamps of varying sizes - 4x6", 4x12", 4x18" should get you started.  It really depends on the size of your projects, however.  This is one of those times when cheap is definitely good enough.
  • Router - trace templates, treat edges.  Get a 2-1/4 HP or better, will accept 1/2" shank bits, variable speed in a two-base kit (fixed and plunge).
  • Random Orbit Sander
Later, I'd add the following, more or less in order
  • Cordless Impact Driver
  • Corless drill
  • Router Table - joint edges, cut profiles, mortise and tenons.  You could make one or buy one. 
  • Block plane and Jack Plane (#5)
  • Set of diamond sharpening plates and angle guide for sharpening the chisels and plane irons.
  • Card Scraper and burnisher - leaves a smoother surface that accepts stain and finish better than even find sandpaper.
  • Table Saw
  • Drill Press
  • Bandsaw
  • Air Compressor with brad, pin and finish nailers (in that order)
Conclusion
So, there's my brain dump on getting started.  

Just remember 
  • Inspiration is everywhere.
  • Start small and keep it simple.
  • Don't over-analyze things or you'll never get started on your projects (i.e., Paralysis by Analysis)
  • You will make mistakes.  All woodworkers do.  Fix them as best you can. Make it a feature.  Never say a word to anyone, since they probably won't notice if you don't tell them.
  • Failure is always an option, but it shouldn't be a reason to not start on a project.  You will *ALWAYS* learn something from any project you tackle, even if it fails.
  • You usually get what you pay for in tools, unless you happen to get a good deal on a closeout model.

2014/03/19 - Software toolchain and artwork design

posted Jan 27, 2015, 10:52 AM by Jeff Balderson

An overview of the general terminology and workflow followed by the actual list of software I use at present.

High Level

There are three phases to take a CNC project from an idea to ready for assembly. 

Design -> CAM -> Controller

Below, I'll briefly discuss what is supposed to be accomplished during that phase and list the software I use during that phase.  I'll also provide the links to the relevant software so you can check it out if it interests you.

Do keep in mind as you're reading that everything I'm writing about is based on my experience working with my Probotix X90.  There are lots of other CNC Router designs and companies out there, and lots of other work-alike pieces of software.

Design

CNC machining is a subtractive process.  You start with stock, and after the process it will have less volume than it had when you started, even though the outer dimensions may be identical.  A 3-axis CNC router like mine can make "2.5D" objects.  The objects can be 3D but there may not be any overhangs/undercuts, so it's not full 3D, hence the 2.5D moniker.  This limitation is because the cutter only moves up and down, so all cuts must be accessible from above.

Some of the "no undercut" limitation can be mitigated by cutting out the work in multiple layers and gluing the layers together.  It's also possible to cut one piece from the front face of the stock and other from the back, so when they are glued together, you have overhangs.  

In CNC machines with additional axes, you can rotate your workpiece and/or the spindle, which can effectively eliminate this restriction.

CAD, Drawing and 3D design

The first step to any design is to get the idea out of your head and into electronic form that can be read by the CAM software in the next step.  

Almost all of my work has been primarily flat work with pockets, profile cuts and drilling.  I've also done a good amount of sign work with Vee bits.

My go-to tool for the flat work is a piece of 2D CAD (Computer-aided Design) software called LibreCAD.  It's free and open source and there are lots of tutorials on the internet, particularly some video ones.  It does have a little bit of a learning curve, but once you get use to the interface, it's easy to use.  

I primarily use LibreCAD for cutting profiles, pockets and drilling.  The things that saws and drill presses are usually used for.  With some special bits and a little more care in the design phase, I can also round over or chamfer edges.  

Librecad example

The example above shows a stacking parts box I found on Thingiverse.  It's a good example of a job requiring a complex profile cut out.  I made a few minor modifications to the artwork to clean up a few lines as well as separate the parts around since I was cutting it with a CNC instead of a laser cutter.

When I need to do some more complex shapes, such as bending text around a circle or trace a bitmap image, I'll use another piece of free, open source software named Inkscape.  It's similar in concept to Corel Draw and Adobe Illustrator.  Inkscape works primarily in vectors (start a line on this point, and end it over at that point) and stores the vectors in the file.  This permits you to shrink and enlarge drawings, usually with little, if any, loss of quality.  

Inscape example

Note that the letters are made of up lines and arcs between points and not colored blocks like in a paint program.

Inscape example showing paths

I haven't designed any 3D items yet, so I don't have any favorites there yet.

Both programs above export DXF files, which I then feed into the programs used in the next phase, CAM.

CAM

CAM stands for Computer Aided Manufacturing.  In a nutshell, that's the process of taking the art generated by the CAD and Drawing software and converting it to code that the CNC controller can understand.  My machine and it seems it's far from alone here, accepts G-code.  It's called that because most of the machine operations start with a G.

Take the following for example, assuming the bit is zeroed to the top of the stock:

G0 Z0.125 (Rapidy move the Z axis so it's 0.125in above the table)

G0 X0.000 Y0.000 (Rapidy move the X and Y axis simultaneous to X=0, Y=0

G1 Z-0.125 F05 (lower the bit into the stock to -0.125in at 5in/min)

G1 X0.000 Y1.000 F50 (Move to X=0in,Y=1in at 50in/min)

G1 X1.000 (Move to X=1in, Y=1in at 50in/min)

G1 Y0.000 (Move to X=1in, Y=0in at 50in/min)

G1 X0.000 (Move to X=0in, Y=0in at 50in/min)

G0 Z0.125 (Rapidy move the Z axis so it's 0.125in above the table)

The code above drew a 1 inch square box, and that isn't even taking into account the width of the bit, so it'll actually be a little smaller than 1 inch.  As you can see, writing the code is verbose, very error prone and time consuming at best.  That's a very simple example, too. Knowing how to read/write it is usually sufficient, because most programs run into the thousands and even tens of thousands of lines, depending on the complexity.

The first piece of software is called CamBam, which is a piece of reasonably priced commercial software.  It does have a free trial, so you can work with it for a little while before you buy.

Cambam storage bin example

To begin, you import the DXF file that was exported out of the design phase and you wind up with a window looking not too dissimilar from what you see above.  Specify feed rate, depth of cut per pass, where the holding tabs are located and many other important parameters and then export the G-code.  The good thing is that it that it has styles (templates) for everything: operation, tools, materials, machines and post-processors. So, if you're constantly cutting the same type of stock, create a special MOP (machine operation) style for it to hold your default settings for that type of stock once you get those settings dialed in. CamBam can do profiles (cut outside or inside a line), pockets (cut inside a line and clear everything out between the lines), engrave (follow a line), spiral drill and canned drill (drill with an exact-sized bit).  You can also insert G-code from other sources.

The other piece of software I use for the CAM phase is a F-Engrave.  It's a Python program that's free and open source.  It will engrave inside of lines using a Vee bit, stepping the bit up if the cut is narrower, stepping it back down when the cut area widens out.  It's not the most full-featured tool out there, but it's done everything I've thrown at it so far.

F-Engrave example

You can save your settings by saving the g-code for a job with a specified filename in either the same directory as the program or your home directory.  As I mentioned on my writeup of the Mini Maker Faire, I gave away keychains.  Some of them had the Dusty Geek logo on them.  If I were to have carved them out instead of laser engraving the log, this is the tool I would have used to generate the g-code.  Once everything's set, hit the "Calc V-Carve" button and it'll trace inside the lines, save the g-code and you're ready to do some Vee bit engraving.  It can take a little while to calculate everything, especially on complicated designs. and the design above is surprisingly more complicated than it looks.  These are basically miniature signs, and this is the same process I do when I do my sign work.

For "3D" work, I use a piece of software called MeshCAM which is commercial software.  It's designed to carve shapes out of wood (e.g., a sphere), versus CamBam which works better for manipulating flat stock (e.g., cut a profile, pockets, etc).  Effectively all of the work I do falls into the realm of CamBam or F-Engrave, so I don't have a lot of experience with MeshCAM yet. 

Controller Software

My controller is the Probotix version of the Planet CNC Mk1 controller.  The CNC-USB software reads in the G-code and allows you to do some simple manipulations on it such as resize the entire job, rotate, step and repeat and others.  When you're ready to run, you hit the start button, the software feeds instructions to the controller, the controller sends electricity to the various pins on the X/Y/Z stepper driver boards, and the steppers move appropriately. There's a lot more detail on how the software works at the Planet CNC website, plus it's only applicable if you're actually using a Planet CNC controller, so I won't repeat it here

2014/03/18 - NoVA Mini Maker Faire - wrap up

posted Jan 27, 2015, 10:49 AM by Jeff Balderson

What happens when you get together 80+ groups that want to show off their projects and combine it with 3000 curious attendees? The answer is a Mini Maker Faire. I had a lot of fun showing off my CNC projects, answering questions about my CNC Router and designs, and it led to a discovery of a future project I want to work on.

I had a great time doing the table at the Maker Faire.  Lots of people stopped by and asked a lot of good questions.  There were a few who seemed extremely interested in doing this type of work and a few showed a *lot* of interest in what I was doing.  Hopefully, they learned something and maybe want to start doing some CNC routing on their own. 

The demo

I was making keychains as a demo and give-away.  The X90 was running pretty much all day cutting them out.  While I came with about 75 done and in the bin when the event started, they disappeared faster than I could produce them.  It looks like I might have underestimated the demand slightly.

I did one run of engraving keychains on-site.  It was good to show, but it took entirely too long to finish this job.  The pre-engraved ones had pithy XKCD-style sayings on them and either the Maker Faire Robot logo or Tux.  Unfortunately, I don't have a shot of that available at the moment.Keychains being made at Maker Faire

With the exception of some of my earlier attempts that were router-engraved, all of the ones cut out on-site were were laser engraved ahead of time partially due to the router-engraving taking so long and secondarily because the laser-engraved ones didn't need any stain to pop. 

I had also intended to make some of my acrylic Maker Faire Robot coasters, but didn't even get through all of the pre-engraved keychains I had made.

Here was the view from behind the table.

View from behind the table.

Follow the links to see a couple other pictures that I didn't take.  It'll also lead you indirectly to the event page that's collecting all of the other photos people have taken.

http://www.meetup.com/NOVA-Makers/photos/20673312/343046782/

http://www.meetup.com/NOVA-Makers/photos/20673312/343325272/

A few of the popular questions I remember were:

Do I make decent money selling my items

I haven't been selling any of my work ... yet.  Maybe someday.  For now, I've been treating this as a hobby, learning how to do things, making projects for myself as well as a few Christmas gifts towards the end of 2013. 

Did I make the CNC Router

No.  I purchased it as a kit from Probotix.  I did have to assemble it, square it up, and calibrate it.  I opted for initially spending my time learning how to use the machine instead of building it.  More on that later, though.

Is the CNC Router expensive / How much did it cost

I purchased the Fireball X90 kit from Probotix.  I included every bell and whistle except for the aluminum table, and also included a copy of MeshCAM.  That wound up being around $1700 shipped to my door.  The router itself wasn't included and had to be purchased separately.

Is that just a standard router?

Yes.  The current spindle I'm using is a Dewalt DW611P which is a palm router you can find nearly everywhere.  I previously used a Bosch PR20, but I'm considerably happier with the Dewalt.  For one, it has the built-in LEDs to illuminate the cutting area.  It also seems to magically keep the bits cool.  I could be cutting a job that lasts an hour, pull the bit out and it'll feel cool to the touch.  When I pulled the bits out of the Bosch, they were very hot.  That makes a huge difference when cutting acrylic.

How do I make it work / Is writing the code difficult?

This was a very common question.  I guess people thought it was harder than it actually is.  Basically, the workflow includes three types of software, CAD CAM and Controller. I'll write this up in a separate post so I can cover it in a little more detail.

How long does it take run run a job?

It entirely depends on what you're doing.  That engraving job pictured above alone took at least 45 minutes to run, and then another 15 minutes to cut out.  I could have run everything faster, but I thought that slowing it down so people could see what it was doing might be better.  Some of the signs I displayed took 45-60 minutes or more.

How long do the router bits last?

Again, it entirely depends on what you're doing.  A bit can last for a really long time in soft stock.  If you run something incredibly hard like the Ipe I used for the iPhone speakers, it's going to really shorten the lifespan.  I usually listen to the router and you can generally hear when it's starting to struggle.

Can you only cut wood?

I have successfully:

  • cut HDPE/UMHW
  • cut and engraved acrylic
  • cut and engraved a project box that I was going to use as an external jog/e-stop.  I think it was made from ABS.  
  • cut aluminum, and while it was successful, I'd strongly discourage it because the machine isn't really made for it and is too flexible to do it properly.  It's hard to say how much life I took off the bit I used.

Where do I get my art?

Mostly from browsing websites like Pinterest, focusing mostly on Woodworking and CNC boards.  When I see something I like, I open up LibreCAD or Inkscape and start crafting my own version using the picture as inspiration.  Other art, including CAD files, have come from Thingiverse, GrabCad or GitHub.  One friend found a stool online that was offering free CAD files, so he signed up, downloaded the file for it and I cut one out for him. 

Regrets

My only regret is that I was so busy talking to people and running the demo that I didn't get a chance to walk around myself.  A few of my fellow Nova Labs members managed to escape their tables for a little bit to walk around, and I wish I had a less hands-on table that would have allowed me to do the same.

The next steps

During one of the conversations I had with someone, they mentioned making a CNC with OpenBuilds, which is a modular system based on extruded aluminum for building CNC, 3D printer and other similar equipment..  After I got home, I decided to check out the site again, and I found a design called the OX which is exactly what I've been looking for.

I think I know what I'm going to be building in the near future.  It's a scalable design that can easily be resized for the dimensions I was looking for (24x36) and can use all of my existing electronics to power it.  Here's a picture of it.

OpenBuilds OX CNC Router

http://openbuilds.com/builds/openbuilds-ox-cnc-machine.341/

Part of me wonders if there would be enough interest to put together a CNC Router build group at Nova Labs.

2013/10/18 - Organization

posted Jan 27, 2015, 10:20 AM by Jeff Balderson   [ updated Jan 27, 2015, 10:50 AM ]

Thanks for visiting!

I think I'll start off by talking about shop organization. I'm horrible at it, but I'm getting better and have learned a good bit along the way.

Organizing a small shop space doesn't seem difficult on the surface and a lot of the ideas are really just common sense, but it can take a lot of thought to get everything as efficient as it can be.

When we first moved into our current house, I set up the shop in the basement. It was a very crowded 14'x17' room, and one corner was taken up by our furnace. With such a small shop, it's really important to maximize your storage to take advantage of all of the wall space and even the ceiling within the shop. Some ways to accomplish that are to install wall cabinets and shelves that are easily accessible, wall-mount tools with cabinets above them, use hooks in the ceiling to hang large clamps and store wood above cabinets or hanging from the ceiling. French Cleats can be extraordinarily helpful for mounting the cabinets and tools.

To help keep the amount of open floor space as large a possible, my only full-size floor-standing tools tools are a Ryobi BT3100 table saw, a 30+ year old Shopsmith Mark V (used as a drill press, horizontal boring machine, bandsaw and lathe), and a 2HP dust collector. All three roll and are stored against the wall when not in use. I also have a rolling tool cabinet that fits under my Shopsmith to make use of that wasted space. My workbench is on a mobile base that's usually used for tools. Smaller tabletop tools such as my spindle sander, belt sander, jointer, planer and pancake air compressor are on a rolling 2x2 wire cart that's about 6' tall. They get pulled out as needed and used on top of my workbench. I built a rolling table for my tabletop CNC with storage underneath. It can double as an assembly table, if desired.

Because space is so valuable, every new tool needs to be carefully considered before it's added to the collection and you need to occasionally take some time to think about everything in your shop. Look around your shop to find and remove the duplicate and unused equipment/tools. Discard wood scraps unless you know you have a use for them in the near future. Only buy/collect wood that you know you're going to use in the near future.

Last year, we finished up building our garage. When we were working with the architect, we thought carefully about how the space would be used and decided that we'd move my shop into the garage. To help with that, we added a 4' bump-out to house the majority of the shop equipment, and another 6' x 4' alcove to house the remainder of it. It's been a challenge moving everything that was in the (mostly) four-sided room I previously occupied onto two walls, and I've had to rethink the organization multiple times as I started using things. It's still a work in progress, but it's quite functional. It's hard to say what I might do in the long term, but for now, I'm working again.

Having a small shop requires more forethought than a larger one. Sometimes you can't have more than one tool out and ready to use. Try to group your operations as much as possible to avoid having to set up and tear down your shop as much as possible. Dry fit parts to make sure you won't have to pull a tool back out unnecessarily.

My next step is looking at installing some upper cabinets to replace the open shelving I'm using right now. I'm also probably going to put my air compressor on a shelf in an unused upper corner of the garage. That should keep it accessible when I need to use it instead of having to pull out the rolling cart it's currently stored on.

Lastly, it's extremely important to keep the shop CLEAN. Put tools away when you're done. Don't allow your workbench or other flat surface to become a dumping ground. Get rid of anything you don't need and don't have a place for. Sure you may save yourself a trip to a home center in the future, but it won't be in your way.

1-6 of 6