Tag: CNC

Pink plushy on CNC Router

Designing a CNC Program for Everyone

Smiling people at table with takeout and documents.
CNC Program planning in progress with Sharps(left) and Rachel (right).

We love building kick-ass programs at Ace Makerspace, and this past year we had a special opportunity to apply our innovative, collaborative approach to program development to a much-needed community resource—our new Full Bed CNC Router. Like many small businesses, we took out an SBA Loan to survive the pandemic but also wanted to put some of the money towards mission-driven growth that would benefit the broader Oakland/East Bay community around us.

Did you know the Ace CNC router was the only public access router in the east bay not on a college campus? The community needs a more robust resource for people to use, run their small businesses, and gain job skills. It needs to be accessible for beginners and easier to maintain.

–Rachel Sadd, ED

We’d been thinking about investing in a new Full Bed CNC Router for our program since 2019, and after a few false starts over the years, we finally gained enough traction to bring our dream to life. Providing an approachable point of entry for folks of all skill levels to learn modern fabrication skills in the East Bay drove our subsequent program development process.

Green full bed CNC Router
Our new full bed CNC router, X axis mural, and workstation.

The new program we imagined would be:

  • Beginner-friendly
  • Affordable
  • Self-guided
  • Highly accessible

And would serve:

  • Folks interested in learning job skills
  • Small businesses
  • Folks making repairs
  • Hobbyists
  • Current and graduated students looking to continue their education and practice their skills
Smiling person with protective googles and carved wooden bat.
Student showing off their first CNC Project after class.

The Early Adopters

In order to build our dream CNC Program we put out a call for volunteers to join the Program Development team as Early Adopters.

We are seeking women, non-binary and BIPOC folks to join the CNC Early Adopter team. Ace firmly believes that a variety of lived experiences are needed to grow truly inclusive programs. And we are at a magic moment with the CNC Router Program as we re-design the program for the new router. You do NOT need to be an expert CNC Router Operator. The perspective of new users is also very valuable to program development.”

–Early Adopter Recruitment Post

For a truly accessible program we needed a group with diverse perspectives, skill levels, and lived experiences. We ended up recruiting:

  • Sylvia- A novice CNC operator with a technical background, who played an instrumental role in curriculum and knowledge check development
  • Bob- Someone with extensive CNC experience at Ace both as a user and an instructor
  • Frank- An advanced tool user with no previous CNC experience and an old school approach to documentation
  • Dave- Someone with CNC experience in both Ace and outside contexts
  • Sharps- A former tool manager for a shared shop with extensive CNC experience outside of Ace
  • Liz- Ops manager in charge of the executing the environmental design for the physical space and limited CNC experience
  • Rob- A technical contractor who helped build out the space and design the X and Y axis murals
  • Rachel- Executive director of Ace Makerspace
CNC Router in action
Taking the new CNC Router for a test drive shortly after delivery. Part of developing a new program involves testing the machine in a variety of ways and troubleshooting issues.
Checking the X and Y axes for precision after troubleshooting an early issue with router bed alignment.

Building a Beginner Friendly Program

Building an approachable CNC program requires bridging the gap between the advanced user approach and a more equitable, beginner-friendly approach.

X, Y Axis Murals
X and Y Axis Murals for easy reference make our newly designed space approachable for novice CNC Operators.

Sylvia, a novice CNC Operator with extensive technical writing experience, played a major role in developing the curriculum and writing the knowledge check. Ace Communications Associate, Jacky, sat down with Sylvia to learn a little bit about her perspective as a novice CNC operator on the team and what beginners can expect from the new program.

1. What prompted you to join the Early Adopters Team?

I joined because the call was for a BIPOC beginner and that was me.

2. What was your experience as a beginner CNC Operator working on curriculum development?

My first experience with CNC Routing was with the Forest Scientific trainer who came to Ace to work with the Early Adopters. That training session, which lacked visual aids and handouts, wasn’t helpful to me as a beginner, although it did seem to resonate with the more experienced users and did take a hands-on approach.

Writing the curriculum was a highly iterative process, and as a beginner I was still able to contribute at our first team meeting by simply asking questions. Rachel provided background on Ace culture with respect to collaborative class development. Sharps wrote an initial draft outline covering CNC Router concepts. Bob contributed his expertise from years of teaching beginner CNC users. Frank [another New CNC Operator] documented the step-by-step process of operating the CNC router. During team revisions Frank and I identified things that were unclear to us as beginners. For example, the Router has 3 axes: X and Y and Z axis, and one of the training slide examples mentioned ‘2.5 axis.’ A beginner wouldn’t be able to decode that.

3. What is one (or more) thing you really thought about/kept in mind as you were contributing to the writing of the CNC knowledge check?

The more brains that attack this the better. I thought the team effort was key, which included leveraging the Laser Team’s tried and tested Laser Knowledge Check as a starting point. Sharps and I generated an initial draft of questions. Then there was a technical review by Bob and Sharps, an inclusivity review by Rachel, a re-write by Rob to transition questions from a true/false format to a multiple choice format, and a review by Jesse M who provided valuable feedback from a test taker perspective.

Color-coded Knowledge Check revisions.
Revising the new CNC Knowledge Check as a team.

4. What did you learn through the process?

How important it is to address different kinds of learning styles. Some folks are visual, some folks like a checklist, and some folks need things said aloud, while other folks need to touch the machine for the info to take hold.

5. What can beginners expect from the new program?

“An evolving process that they can contribute to in order to make it even better!”

Pilot Classes Begin

CNC carved wooden bat coaster
Wooden bat coaster practice project for New CNC Router Basics Workshop.

Ace is looking for 4 beginners for the next CNC Pilot Class! It will likely be mid-month and definitely on a weekend. Reply to this thread if you are interested. Update: Please if you have previous CNC experience do not put yourself on the list.

-Pilot Class Announcement

Finally, after installing the New Router, learning how to use the new machine and experimenting with it, designing the curriculum, handouts, knowledge check, and other user documentation on the Ace Wiki, installing a new desk, painting X and Y Axis murals for easy reference, and designing a beginner friendly “data pack”, it was time to pilot the new CNC Basics and Certification class starting with….beginners!

What’s next for the CNC Router Program?

Our program is evolving and we plan to continue iterating our collaborative approach to design as we fine tune our existing CNC Router offerings, grow our program, and develop more great beginner-friendly learning opportunities for our East Bay community.

Here’s a taste of what to look forward to as we continue developing new learning opportunities for CNC Routing at Ace:

  • Software Design Classes for CNC Routing
    • V-Carve Design for CNC Routing (less powerful, but less complicated)
    • Fusion 360º Design for CNC Routing (more powerful, but more complicated)
  • CNC Router Advanced Operation and Certification Workshop focused on 3D Routing and handling specialty materials (like plastic) safely
  • Open Labs with focused topics like “planing contours” and “cutting plastics”
  • Practice Projects like dice trays, bee-houses, and clocks

In order to create robust and equitable resources, we invite folks interested in being a part of this exciting process to bring their unique perspectives, skills, and talents to the table. We also plan to continue working side by side with our community as we evolve other current and future programs.

Guitar body on router bed
Guitar in-progress on new CNC Router Bed.

Electronics Tool Rack

Having the right tools when you are puttering around making things is invaluable and having those tools near at hand is a wonderful thing. When revamping the Electronics Area and turning it into a place you are not afraid to work from the pit of darkness and despair it used to be, I noticed we didn’t have the necessary tools needed to do the job. Also we had no place to put said tools even if they where available, so I decided to m
odify the standard tool rack I developed for my workbench at home. It is a standard rack with basically vertical piece of wood between two frames, like an I beam. On each side of that vertical piece you put a collection of PVC tubing to hold whatever you want to hold. The trick is to have an idea of the tools you want to have on it and lay those out before hand, so you know what kind of tubing and how much you need. Some #6 screws, a drill press, bandsaw and CNC up the wood bits and away you go. Rubber feet on the bottom make sure it doesn’t slide away from you as you reach for that next screwdriver. The tools selected for the rack are based on my own experience in doing electronics and detail work. Precision screwdrivers are a must but so is a regular #2 Phillips and flathead. The #2 and #1 Phillips are probably used more than anything. From there it all makes sense to have enough tools but not too many. You can go crazy on the tool rack but it starts to get too large and would take over the tabletop if you let it. Keep it basic and know what you use.

3D CNC Milling a new part for an old lathe – part 2

Continuing the saga of the bed bracket for the Monarch 10EE lathe taper attachement, we left off with a nicely squared up rectangular block of metal, and a Fusion 360 model of the part. The next step is to figure out how to hold the work and machine it to shape. I’ve never done much in the way of 3D milling, so this would be a learning adventure for me.

It’s important to think through the sequence of operations, and how for each operation you are going to 1) hold onto the part, and 2) find the origin for your WCS (work coordinate system). For the first setup it’s usually pretty straight forward as you are starting out with a nice rectangular block with flat sides and square corners, just clamp it in the vise, make the top back left corner the origin, and carve away. But then you have to flip the piece over and do the other side and you’ve just cut off all those flat sides and corners, and the vise might not have anything to grab onto. Having the 3D printed version of the part in hand was definitely helpful in visualizing how to do things and in what order.

For this particular part (see http://a360.co/2iXVoIu and part-1), I could choose to start with milling the top or the bottom. There flat areas on the top, one end, and 2 sides of the part which could be used for references. All the critical dimensions are on the bottom, so my first thought was to mill the bottom first and if I screwed it up I wouldn’t waste time milling the top. But thinking about the work holding once the bottom is milled there isn’t much to grab onto any more for milling the top. Only a very small area of the sides are available for clamping. So, looks like I should do the top first, and hopefully there will be enough of the sides left flat for clamping.

In Fusion 360 created my first setup, Since I had already machined the width of my material to size, picked my stock size to be the “model box”. I made the origin the top back left corner of the rectangular block of stock. It’s good to use one of the back corners for this because the rear jaw of a milling vise is fixed that location should stay the same if you take the part out of the vise, flip it around, etc. That back side of the part, and the face of the fixed jaw of the vise are a solid reference point for all the setups. All of your part will be at negative values of y axis. I like to put my origin on the left size, and then all my x axis positions are in the positive direction. If you do this consistently you will be less likely to make mistakes. If you look at the Mach3 toolpath display and things are going into positive Y or negative X, it will look funny, and you should check what is going on.

Drilling the hole in the top seems like a no brainer, so decided to start with that. Long drill bits are wiggly and can wander when starting. So to get an accurately located hole we can use a spotting drill or center drill. They are short and stiff, and will make the hole wherever you put the point to the metal. At AMT we have a bunch of center drills. The problem with these is that if you make a mistake the tiny little tip can break off in your hole and it’s a bear to get it out again. I had just recently experience this, and decided to buy some spotting drills and see how they work. Turns out they work well, and you would really have to screw up bigtime to break one off in the hole. So first operation is a 0.1″ deep spotting drill hole.

Next we need to drill a clearance hole for a 3/8″ bolt. Looking on the chart a close fit clearance hole for a 3/8″ bolt is size W. Oops. No letter size drills at AMT, or in my tool box (add that to the wish list for next year). So, looking at a drill size chart I see that next size up from W is 25/64″ and I do have one of those, so I will use that. Only turns out that when you put the block in vise, and the drill chuck in the mill there isn’t a lot of vertical space left between the bottom of the chuck and the top of the part. The drill is too long and won’t work. Fortunately, the shank of this particular drill bit is turned down to 3/8″, and it fit securely into the R8 3/8″ collect, which doesn’t take up nearly as much vertical space as the drill chuck. The drill fit in the available height, but just barely.

When drilling a deep hole (more than about 4 times the drill diameter) you need to worry about getting the chips out of the hole. The chips can clog up the flutes of the drill bit, jamming it in the hole and causing the bit to break. So when setting up the CAM for this operation, make sure to set the cycle type to “Deep Drilling”. This will cause the drill to go down a bit, then retract out of the hole to clear the chips, then dive back in, drill a little deeper, and repeat until the correct depth is reached.

Unfortunately, in the process of screwing around with the drills, chuck, collets, etc, I somehow (and I’m not really sure how this happened) screwed up my x-axis location. I fired up the g-code and it started drilling the hole in the wrong place. Yikes! Hit E-stop, but it was already in about 1/4″. So much for my beautiful part. Oh well. It’s right next to the real hole, and a nice washer under the bolt head will just cover it from view. Anyway, I re-homed the machine,  used the edge finder to locate the left and back edge of the block, and successfully drilled the hole in the right place.

Next comes carving out the shape of the top of the part. In the CAM software I picked 3D adaptive clearing since I had heard that this was a good way to do this. It uses the whole length of the side of the end mill and side mills off a constant amount of material on each pass. This supposedly removes a lot of material quickly, and improves the life of your cutter since it uses the entire length and not just the tip. And since I didn’t know how well this was going to work I decided to  leave 0.1″ of extra material to clear out on later finishing passes.

One of the parameters for adaptive clearing is the “engagement” of the cutter. This is how much material it will attempt to side mill off on each pass. I was using a 3/4″ end mill, and so I thought I could take pretty big bites, and picked 0.050″. That’s a pretty small fraction of the 0.75 cutter diameter, so why not. Turns out that a 1.7″ depth of cut with that much engagement and that size cutter take a LOT of horsepower to turn, and it puts a lot of force on the machine. When it started to cut the motor immediately stalled out, and the stepper motors started skipping. Argggh. Back to the CAM program, cut the engagement in half to 0.025″. Re-home the machine, re-find the edges, and try again. Boom! Same thing again. Do it all again, but this time try 0.010″. This time it starts to work, but it makes a terrible noise. On no! E-stop again! The bottom of the end mill is cutting the top of the vise jaw! I forgot to check that. Take my part out of the vise, put a parallel underneath, re-find the edges, and try again.

This time it seems to actually be doing the right thing! It’s taking a bit of material off the back of the part going from left to right, but when it gets to the corner and starts to come around it stalls out again. E-stop again! What’s going on? After some thinking it occurred to me that I had used the bounding box of my part for the stock size in my setup, but when I actually cut the metal, I had not accurately cut it to length. I had left a little extra to mill off later. Well the CAM software doesn’t know that metal is there and happily drives the cutter into what it thinks is 0.010″ of metal, but what was really there was 0.1″ of metal. Back to the CAM software and add the extra metal to my stock size, re-home, re-edge find, and try it one more time.

This time it’s working! The big end mill is actually carving out the shape of the top of my part! Time for a celebratory beer? No wait. Taking off only 0.010″ of material on each pass it is going to take about 90 minutes to finish. And long needles of aluminum are flying all over, and so I watch, vacuuming up the mess as it goes until it is finished.

Remember that 0.1″ of material I left in the setup of the adaptive clearing pass? Well now it’s time to take that off. I look at my options in the 3D milling CAM operations and see that “Contour” is a finishing strategy that is good for steep sided things like my shape, so I give that a try.

Surprisingly this works with no problems. How can that be? Something that does what I expect the first time? I decide to quit while I am winning and save the rest for another day. Here’s a picture of where I stopped.

Next steps are to go over the curvy areas of the top with a 3/4″ bull nose end mill, then flip it over, and mill the bottom. That will have to wait until I get back to it a week from now. Stay tuned for part 3.


3D CNC Milling a new part for an old lathe – part 1

As some of you know, besides my Tool Steward duties in the ATM Metal Shop, I also have a similar role with the Eastbay Astronomical Society. I help take care of their metal shop located up at Chabot Space & Science Center. One of the key tools up there is a 1964 Monarch 10EE Toolroom Lathe, which was donated by the USDA a few years back.

This is an amazingly accurate and powerful lathe that has all kinds of useful features and functions, one of which is a taper attachment. A taper attachment is an accessory for a lathe that, as the name implies allows one to cut very accurate tapered parts. For example, the tailstock of AMT’s Jet 9×20 lathe has an #2 Morse taper socket into which can be plugged all sorts of centers, drill chucks, drill bits, etc. Unfortunately, one pice of the taper attachment for the Monarch lathe was missing when we got it, and the that has prevented using it. My project is to create a replacement for this part, which is a bracket that securely attaches the sliding part of the taper attachment to the bed of the lathe.

A bit of research on the web found this drawing of what the original looked like

It looks like kind of a complicated part to make, and the material would be expensive, so I decided to try a 3D printed prototype to make sure it will fit. First I created a model in Fusion 360 based on the measurements in the drawing. I haven’t tried this before, but I think you can view the model at http://a360.co/2iXVoIu.

Then I 3D printed it on AMT’s Makerbot. Here’s what I got

I then took this plastic version of the bracket up to the lathe to try it on for size. Good thing I did because it didn’t fit. The arm of the bracket was too long for my lathe and I would have wasted a big chunk of aluminum of I had just dived right in. Perhaps Monarch made multiple versions of the lathe over the years and the plan I found was for a different one? Anyway, I corrected the model, printed out another version and this time it fit perfectly. It’s time to commit to metal!

Drove down to Gorilla Metals in Hayward and bought enough 2″ x 2.5″ Al to make 2 since I figured I would screw up somehow. Also found a bunch of nice materials in their cutoff pile, paid for it all, put the scrap in my car, and drove home, leaving the actual bar I had driven all the way there for on the counter. Next day drove back, and picked it up with some mild ribbing from the friendly folks at the counter.

First thing to do was to cut a 5.5″ piece of the 12″ bar I had brought home. Turns out the 10 TPI blade for the horizontal bandsaw is terribly dull, and after 5 minutes had barely made a dent. Time to order some new blades for that thing. So then moved to the vertical bandsaw to see if that was better. Definitely! Was done in about 5 minutes.

Just like wood from the lumber yard, extruded aluminum bars are not flat and square when you get them, so clamped my raw stock into the milling machine vise, loaded up a fly cutter, and proceeded to flatten and true up all 6 faces making a close to a perfect rectangular block as I  could. The fly cutter left some marks, but not too bad, but it may be time to take a closer look at the tram of our mill. the column seems tilted ever so slightly towards the front. A project for the new year.

Next post will discuss a bit about CAM and workholding for turning this blank into the actual part.