[mduckett]

I decided to tackle making the hand wheels first. My model drawings show these as single piece wheel/shaft, but it's much easier to make the wheels and shafts separately, especially on the mini-lathe. The scale of this controller constrains how large the hand wheels could be, and I looked at the pendants, and also the radio control car controllers for reference, and decided on 2.5" as a good size. Larger might be better, but this size will allow good gripping of the wheel for precise movements, and the handles will allow for rapids. I had access to some cutoff pieces of 2.5" round 7075 aluminum that served as the stock. I found a .75" long and 2" long piece. On the mini-lathe, the outside jaws on the 3 jaw chuck were used. The .75" piece was faced and turned enough to make it smooth and round, then turned around in the chuck to face the other side to final thickness (0.6"), and then center drilled with #7 drill for 1/4-20 tapping. I waited to tap until after the center was hollowed out to save on tapping depth. Using a boring bar, a pocket of about 1" was cut into the face to a depth of .3" (half thickness of the wheel). With the compound set at 45 degrees, the boring bar was also set to 45 degrees and then the edge of the pocket was worked back until the taper reached the bottom of the pocket. The edges were chamfered, and then the center hole tapped to finish the part. The 2" piece was faced, turned, then turned around to face the other side, and the diameter turned for a distance of about .7". Then I moved to the bandsaw to part off the two pieces about .7" thick. Returned to the lathe to face the cuts to finish thickness (.6") and repeated the center drilling, boring, tapping, chamfering to complete the other two wheels. See attached photos of the finished hand wheels. Each wheel took a little under 2 hours to complete, not fast but the finish turned out nice. I used brazed carbide bits for roughing and boring, and HSS for finish facing/turning. One comment/tip that I've heard about and used here was to keep a set of diamond files handly for touching up the brazed carbide tools. I touched those up after finishing each wheel and it really worked to keep them cutting well. Next up I will get started making the wheel handles and shafts. I have not decided on a handle material yet, chromed steel would be ideal if I could find some pre-made ones at a good price. Delrin might be good also.

[Brass_Machine]

Glad to see you are back at it!. Will be following along.

Eric

[mduckett]

Glad to see you are back at it!. Will be following along.

Eric

Thanks. Did you make the mods to your machine? --md

[Brass_Machine]

Not yet.. Been too busy with other projects, unfortunately. I have been gathering the parts though. I plan to do something close to your build. I have an older HF mill with the tilting column, but have purchased from LMS the X2s base and solid mount column. Trying to find a machine shop that can recut the dovetails to 60* so I can use thicker brass gibs. And I have to finish modding the Lathe so I can cut my ballscrews with more precision.

I like the idea of a stand alone control box so I can use it manually without powering up the shop PC.

Eric

[Eremius]

I really like this idea and can't wait to see more.

To ease the "wall of text" effect that makes your posts really hard to read, would you put some paragraphs in your posts?

[mduckett]

I really like this idea and can't wait to see more.

To ease the "wall of text" effect that makes your posts really hard to read, would you put some paragraphs in your posts?

Yeah, sorry about that. I worry about the posts getting too long, and adding paragraph breaks makes it longer...I'll try to improve it when I can, and to break into smaller posts. Thanks for reading and the feedback. --md

[Eremius]

No worries. I wouldn't sweat your posts being too long if that's what it takes to properly convey the information.

[Brass_Machine]

Yeah, sorry about that. I worry about the posts getting too long, and adding paragraph breaks makes it longer...I'll try to improve it when I can, and to break into smaller posts. Thanks for reading and the feedback. --md

Don't worry about that... keep posting up. I am enjoying the reading.

[mduckett]

Got a little time to get started working on the hand wheel shafts. The final design deviates a little from the model in that there is a shoulder at the end that connects to the hand wheel. This provides the hand wheel side retention for the bearing and is a built-in spacer to reach through the eventual housing/bearing pocket. The hand wheel will have about 1/16" clearance with the control box. This was cut from some œ" aluminum stock. With about 2" of stock protruding, the diameter was turned to approximately 10mm OD for about 1". I was trying for a slip fit for the bearing area and over cut it, so I just extended the turned part another 8mm (bearing width) and was more careful to get the slip fit. The other area was about .002" under sized, but this was where the 10mm x .1 thread belongs, so that actually worked out fine. The threads were cut using a die in my home made die holder. Next was the 6mm hole in the end for the encoder shaft--I don't have metric drills or reamers, so I picked the closest 1/64" size (15/64) and drilled the end (after center drilling). It came out just slightly over sized, but ok as I plan to use opposing set screws so the encoder shaft can be tweaked into the center. Probably overkill, but I don't mind the machining work. The shoulder areas was turned round, and the shaft parted off, then chucked carefully and the shoulder faced and chamfered with a file. Finally, the shoulder was center drilled, drilled with #7 bit, and tapped for 1/4"-20. I cut a short stub of 1/4"-20 all thread to serve as the coupler between the shaft and handle. When assembled, this has a small amount of runout, which I could remove but probably won't because it is not critical. The pictures show how it goes together. There will be a washer on the thread side between the nut and bearing in the final assembly, and a little blue loctite will secure the hand wheel to the shaft stud. One down, two to go. It took about two hours start to finish, hopefully the next ones should go much quicker.

[mduckett]

The remaining hand wheels have been finished, and I got on to making the handles. On a related side note, on Cyber Monday, Enco usually runs their 20% off with free shipping offers, and this year I took the opportunity to stock up on some end mills, and also bought some 4' lengths of Delrin rod. These usually cost a lot in shipping due to the long packaging, but with no minimum order I was able to get several sizes with the discount and free shipping. I planned to use Delrin for the handwheel handles because it is strong, easy to machine, and is slippery. The handles are non-revolving to give good control, but need to slip easily when gripped, so hopefully the Delrin will work out..
The handle design is a simple taper from about 3/8" down to 1/4" over a length of about 1", with about 3/8" of 1/4-28 thread ( the 1/4-28 die was used because my 1/4-20 die in the HF tap/die set was defective, but it turned out to be a good choice to use the finer threads). I cut about 10" piece of 3/8" Delrin and chucked in the lathe. The compound was turned nearly parallel with the bed to about 2 degrees, and with about 1.5" extending from the chuck a sharp HSS tool was used to cut the taper, and then the last 3/8" was turned to a 1/4" cylinder. A file worked good to remove the turning marks, and some 1000 grit emery made them really slick. The 1/4-28 die was used to cut the threads on the end, and then the piece was parted and turned around. The final 1/2" or so of the big end of the handle is not tapered, so that was gripped and the end rounded off with a file, and a little concave cut made into the end to finish. The handles are about 1.5" long + .375" threaded section. It took about 30 minutes for the first one and about 20 minutes for the other two.
With the handles complete, the next step was to drill/tap the hand wheels to accept the handles. The holes were drilled with #3 drill at the midpoint of the outer flat area. I used through-holes to simplify the tapping, which was done with the 1/4-28 tap. The handles fit perfectly, and later will be secured with blue Loctite. This completes the handwheels except for the set screw holes on the shaft, which will be done later.

[mduckett]

With the hand wheels finished, I've been working on making the g-code to cut the housing front panel bearing pockets, switch holes, and drill pattern for the bearing retainer and encoder bracket. The model started as a sketch and didn"t have accurate, final sizes, so separate models for the front panel and top display panels were created, and then the tool paths were made for each part using the model geometry. I use TurboCAD with the CAM plug-in for this. The CAM plug-in has some quirks, but overall it works pretty well. I took a couple screen shots of the front panel with the tool paths.
Since my mill doesn"t have enough X travel to cut all three bearing pocket/switch holes in one setup (the piece is 14" long), I opted to make a single g-code setup centered on the bearing pocket, to be used three times across the front panel. The setup just needs for the center of each bearing pocket to be marked on the panel, then the spindle is zeroed on that location, and run the program.
The material for the front panel is 1/2" birch cabinet plywood, which has about 8 plies. I wanted to test the g-code before committing the actual panel, so a small piece of the ply was mounted on the table on top of a sacrificial piece using strap clamps. The g-code is actually two programs; one for drilling the pilot holes, the other mills the pocket and switch hole. The drill size is 3/32", and a 1/4" 4FL end mill was used for the milling. Using the top spindle speed (2500 RPM), the drills and milling cut pretty cleanly. The milling leaves some stringly edges at the top of the pocket, but these are easily removed with some light sanding with 150 grit sand paper. The sides and bottom of the pocket are very clean and square. In the picture you can see that the bearing fit is pretty good, but could be a little tighter (clearance is about .004"). I also discovered that the switch hole is the wrong diameter. Both features have been changed and new g-code posted. I"ll try to run the front panel tomorrow, and then move onto the top display panel. I also need to make a bearing retaining ring, and also a washer to go between the nut and bearing inner race.

[mduckett]

The front panel was cut from a piece of the 1/2" birch plywood and mounted to the mill table on top of some clean scrap pieces to serve as sacrificial spacers. I used the front edge of the mill table as the alignment standard. A 123 block was held against the edge of the table, and the ply piece against the 123 block while the clamps were tightened. The mini-mill table was barely long enough to put a strap clamp on each end to hold the ply down. The center bearing pocket, switch hole, and screw pilot holes were run without incident. I then put a T-bolt through the center bearing hole and slid down the panel so that the left end could be run, then slid down the other way to do the right side pocket. The strap clamp was used each time on the end to make sure the ply was held down securely. As with the test piece, the mill leaves some stringy remnants around the top of the pockets, but it is easily removed with light sanding.
For the top display panel, I broke the g-code into three segments; left button holes, display cutout, and right button holes. The display panel is made from 3/8" aircraft plywood, Midwest brand. Found a 1' x 2' panel at AC Moore. The reason for using the thinner panel is two-fold; when cut at 45 deg. It will sit on top of the front panel with only about .065 difference in width, and it will allow the panel mount buttons and display to mount with at least smaller effort than if I'd used 1/2" material. The setups and code runs for the left side buttons and panel cutout went without incident. Right after the right side code started, the mill suddenly decided to skip a bunch of steps and lost it's position. Hit the ESTOP and tried to re-run the program but it did it a second time so I called it a night. I couldn't find anything wrong with the system, so I re-mounted the work more toward the center of the table and ran the program again without incident. The muffed first attempt put a few divots around the first hole location, but nothing that a little wood putty won't fix. Once all the holes and display cutout were complete, the piece had to be flipped over and some pockets cut on the sides opposite the switch holes in order to allow the switches to reach through enough for the retaining nuts to grab the mounting threads. I wasn't sure if I needed this step as it is very close, but I needed about 1/16" more thread exposure. I chose to do this free hand with my small router, which was much quicker, if not as neat as using the mill.
Assembled the bearings into the pockets on the front panel, and installed the hand wheels. The bearing pockets were decreased by .002" in diameter based upon the test piece and turned out very nice with a slight press fit for the bearings. I'll use #6 screws (4x) with washers to retain them. The axis enable switches also fit great into their holes. Those have a rubber O-ring under the top lip and seal down tightly.
The switches were installed into the display panel using their plastic nuts for now, but the final assembly will use the rubber screw covers to keep the switches clean. The display pocket needed the corners cut square, which was done with a razor saw and file. The display fits into the cutout very nicely, about .035 below the surface. It needs an outline pocket cut around it to install a sealed, clear window protector. The long shaft on the right side of the panel is the 8-position rotary function switch. It needs a pocket cut to bring it up enough to be fastened properly. So, a few more steps needed to finish the mechanical assembly, but it's getting there. Propped up the display on top of the front panel to show roughly how it is going to look.

[Brass_Machine]

Interesting! I like the wood on the enclosure...

Keep at it. I am still reading.

[mduckett]

Interesting! I like the wood on the enclosure...

Keep at it. I am still reading.

I have a good stock of 1/2" birch cabinet ply, and for making a prototype it seemed like a good idea to use something failrly cheap and easily repaired or replaced. My mill spindle speed is a little low (2500 max) for cutting wood but it worked pretty well. It's finally getting close to working on electronics and software. Thanks for reading. --md

[mduckett]

Got the pocket cut for the rotary switch and mitered the edges of the display panel, and also cut the side panels. Took a few minutes to figure out how to fixture for gluing it up, but it worked iout pretty good using the scraps left from cutting the 45's on the side pieces. First, the front panel was clamped down, then the side pieces were clamped down to that, then the display panel was pulled in using the scraps with clamps. Some minor adjustments and all was good. I used Titebond III glue, which sets very fast. FWIW it's also waterproof. I unclamped after about 2 hours and started filling some of the nicks and dents on the exterior, and just decided to fill all of the exterior surfaces, which meant working wood filler into all of the surfaces (I use Elmer's wood filler) and sanding it when cured using a flat sanding block. You can see in the pictures that there are a number of low spots and dents that filled in. Next step is to prime and paint, probably use epoxy paint, but I'm also thinking about maybe using model airplane iron on covering.
I also did some test fits for the electronics and everything looks good. The back cover will be either a piece of Plexiglas, or 3/16" ply. The bottom will be 3/16" ply and will have connectors for power and cabling to connect to the BOB in the controller box. This will be the last update on the box buildup, it's going too long.
Next update will be for the electronics installation and wiring.

[mduckett]

Here's the box after painting and with some of the parts assembled. The blue metal flake Rustoleum turned out nice and everything fits well. Note the day glow green bezel taped temporarily in place for the display. My friend got a 3D printer for Christmas and has been looking for stuff to print so I made a quick model and he printed it up using the ABS filament he had loaded. The bezel has a pocket for a plexiglas window on the back. It will be painted black using a spray paint made for plastic, and then it will be glued/silicone sealed in place with the plexiglas window to make a watertight seal for the display.
For the wiring, I'm using ribbon cable from SparkFun Electronics that has female header connectors pre-installed to save time and reduce the amount of soldering. The processor board already has headers for all the connection pins, as does the pre-made DB25 connector assembly that will interface with the BOB. The switches will need to be soldered to half of a connectorized wire. The encoder and the rotary function switch will have small PC boards with headers, and will use the longer, pre-made ribbon cables to connect to the processor board. The wiring diagram is in process, and there's still have a bunch of ToDo list items, including making the digital encoder brackets, mounting the processor board, making circuit boards/headers for the encoders and the rotary switch. Need to make the bottom for the box and make the DB25 cutout, install and paint that, and of course the back panel with water seal.
Final progress item is for the software. The code project is setup, and all the I/O from the processor to the various hardware have been mapped and all the initialization code is done. The rudimentary design for the step/direction control and update process is complete, as well as the strategy for controlling motor acceleration/deceleration. Looking forward to working on the display and menus. A lot more to come on that later.

[mduckett]

Almost finished the ToDo: list over the weekend: Fabricated the three metal brackets to hold the encoders over each of the hand wheel shafts. These are made from 1/16" x 3/4" aluminum flat bought at Lowes. I took the drawing measurements for each bent segment and added 1/16" for each bend and used that for the raw material stock size per bracket. The bend lines were measured/marked from the center, and the bends for the center U-shape part of the bracket were made in a vise. I find it hard to bend brackets so they have perfectly matched legs, so I ended up cutting a wooden block to fit inside of the bracket and at the proper height. Clamping this into the U-shaped pieces, the legs were then inserted and tightened into the vise, then bent at that position. This gave fairly repeatable and decent results. The mounting holes were then marked and drilled. The scrap piece I made previously for testing the bearing pockets had the correct hole pattern, so I drilled the holes through and then used that with a transfer punch to mark the holes in the brackets. Finally, I bolted each bracket to the bearing test piece and milled the top flat to make sure the top of the brackets would be perpendicular to the hand wheel axis. I only removed enough material to make them flat, maybe .015" total. A large transfer punch was used to mark the bottom center of the bracket for the encoder mounting hole, which was later drilled from the bottom. I did some test fits with the encoders and it looks like everything is well aligned with the hand wheel axes.
Display cover: A piece of plexiglas was cut and and epoxied into the display cover frame. I used too much glue and it flowed out of the frame to the outside joint and made an uneven bead, but it isn't over the display so I didn't try to trim it. I will seal with clear silicone anyway so it won't show much in the final product. The cover fits nicely, with the plexiglas making a .020" lip that fits down into the cutout. The hole thing will get glued onto the box and sealed with clear silicone.
The bottom piece for the box was cut to fit snugly from 1/4" ply, and the NewFangledSolutions wizard in Mach3 was used to make the cutout for the 25pin D connector. I had to enlarge the holes on each side of the cutout to accommodate the female side connector nuts, and a couple small blocks of pine were glued in to add some support for the breakout board, which will be secured with two small screws into the blocks.
The back panel was cut from a piece of 1/8" aircraft plywood (got that at AC Moore). I also bought some 1/8" x 1/2" basswood strips and used those to stiffen the plywood and make a frame. I messed up when I cut the display panel for the box, so the display sticks out about 1/16" past the back edge. Adding the frame on the back panel allows the back to clear the edge of the display's PC board. I'm not sure how I'll fasten the back, but likely just some small diameter screws.
The wiring has not yet started because I decided to make PCBs for the encoders, the main rotary function switch, and also for the display power regulator and contrast power sources. It only cost about $22 to make the board, and though it will add 2-3 weeks to the schedule it will make the wiring much simpler. I also ordered all of the electronics, which added another $35 (ordered lots of spare components) to the project. I laid out everything on one PC board with outlines and mounting holes for each section (see pic.) that will get cut apart. Other than painting the bottom and back, I only need to drill the hand wheel axles for set screws to hold the encoder shafts. What remains after that is mounting the computer board, connecting the encoders, assembly details for the rotary knob, assembly of the PCB components, and of course the wiring.