Abbott Analytical Products

Fishing Lure Mobile:

Brief Summary of Project:

This project did not involve any special CAD/CAM/CNC knowledge. The design concept was borrowed from a similar fishing lure mobile that was seen at a cafe/garden shop. The cafe version was selling for over $80. Hanging on display it was about 36 to 48 inches long. It had about eight/ten large lures with their hooks. Separating each lure was a piece of wood. The wood varied in quality from ink pen blank to "rough-hewn". The cable used to suspend the various lures and wooden pieces was large with correspongingly large metal cable clamps. Both the cable and clamps exhibited a light rust pinta. The lures appeared to be squashed between the piece of wood. The retailer did not back-off the price.

Retail Version:

The "borrowed" design concept has been executed twice now as gifts. This proved to be a good activity for involving the wife: wood, colors, beads, lure arrangement, etc. This version of the borrowed design concept has a shorter suspension length. The hole for threading the cable through the lures were located individually at the "balance spot" of each lure. As such the lures are positioned on the cable without yawl or twisting about their dorsal axis. The center holes for wooden pieces were located at the geometric center of the top face of each piece. Between each lure and piece of wood there is a bead that improves articulation of the assembly. A heavy duty salt water swivel was added to allow additional articulation at the top loop of the cable. Finally at the bottom loop of cable, an appropriate spinner lure was added for additional color and "bling".

Fishing Lure Mobile: Balance, drilling, layout, assembled.

Parts List:

Discussion of Process Flow
This project was designed, kitted, milled, and assembled on-the-fly. The equipment and tools listed below could have been replaced by any array of power and/or hand tools. The process steps listed further below are a good rendering of the actions taken to create the project.

With the exception of purchasing slightly exotic woods, this project was executed "on the cheap".

The real challenges were:

• Vertical Mill Vise setup for drilling the lures.
• Learning to create and to work with the "threading point" of the cable.

Equipment/Tools
Shop Equipment:

• Sherline Next Generation 8580 Vertical Mill
• Sherline Mill Vise #3551
• Vise for Electronic Prototyping with Soft Jaws
• Durcraft SP-30 Drill Press, 3 in Throat
• Straight-Cut Tin Snip
• Ice Pick
• Sharpie Pin Fine Point
• Straight Edge
• Pencil #2/#3
• Ruler or
• Speed Square or
• Combo Square

• T28 P28 Z2.25 D0.0781 ; 5/64 drill DeWalt #1305
• T29 P29 Z3.10 D0.0781 ; 5/64 drill Snappy #49105
• T30 P30 Z0.0 D0.09375 ; 3/32 drill Dremel #628 Set

Process Steps:

The project was executed in the following stepwise manner

Lessons Learned

• It may be wiser to "not cut" the cable to length. Instead start inserting with the bottom ferrel and the individual pieces according to the design. Once all the components are loded on the cable, set teh loops and then cut.
• Re-tap holes as needed to insert the cable with a minimum of effort. Try a 3/32 drill bit using a Dremel or other drilling equipment.
• If the cable becomes difficult to cut and form a point, try oiling and sharpening the snips.
• The assembly may not need the Locking Beads if the cable ends can be "stuffed" into one of the beads used in the assembly at the top and bottom.

Wooden Puzzle Box:

Brief Summary of Project:
The intent was to create suitable wooden puzzle box as a specific gift. The project involve creating matching puzzle box components with interlocking dovetail slots/tabs. This variant of prior wooden boxes required additional CAD skill and on-the-fly design changes at milling. The major project break came from rotating the design center line 45 degrees. The design manufacturibility feasibility was verifeid simulation runs on a standalone LinuxCNC PC and then using scrap material on a Sherline 8580 NexGen controlled by LinuxCNC 2.8.

Discussion of Process Flow

The development of the puzzle box involved a couple of false starts, going down various rabbit holes, using spikes as needed, employing maturing iterations during the design process, and more than a few scrap runs at the CNC bench. The PuzzleBox desgin was rendered using FreeCAD. The desgin was executed in builder's grade yellow pine using Sherline 8580 NexGen running under LinuxCNC/Ubtunu. A short period of sanding, surface preparation, and staining yielded a Rosewood stained Puzzlebox. The presentation of the Puzzlebox as a gift was well received.

The first false start involved FreeCAD 19. This version of FreeCad proved to be unusable as installed. The FreeCAD 19 installation required a complete restoratiion of the system backup image (Clonezilla). Special thanks go to Mark Ganson for his efforts to introduce me to FreeCAD 19 and the Dovetail Design desing feature it offers. Time constraints and my uncertainity of using the feature forced abandoment of this effort.

Upon restoring FreeCAD 18.4 the focus was on creating the bottom of the puzzlebox object first. Early in the design process it became obvious the desgin simplification that could be derived from translating the design (bottom and lid) 45 degrees. This allowed for changing one dimension at a time as one moved along the centerline or above/below.

Once the bottom design of the puzzlebox was achieved the matching puzzlebox lid was created mirroring the surface features of the bottom. At this point there was no attempt to generate dovetail cuts for the slots or tab. Instead the effort centered on identifying cut regions and no cut regions for the bottom and the lid. The concept was to specify first the major milling cuts on the bottom and create the associated toolpath gcode. Then the dovetail cuts for the the bottom and lid would be sepcified. This portion of the FreeCAD project desgin data was captured in the corresponding spreadsheet. As the design evolved the decision was made to put the locking mechnaism (locking pin, strong magnetic material, the locking mechnism hole) on the Lid at one end of the Lid's center slot along the Lid's centerline.

Two "raw" Job tasks outputs were derived from the FreeCAD Phase of the project. These two bodies of gcode were then split into two sets of gcode programs. Each gcode program had a suitable header added, various variables defiend, o-code inserted, substituted variables, adjusted for tooling difference between FreeCAD and the LinuxCNC tooltables, and added LinuxCNC specifc syntax.

FreeCAD 18.04 does not exactly provide features/tasks for creating the long slots and tabs that the puzzlebox required. It does provide the capability to generate gcode outputs that support creation of outlines/boundaries and the creation of channel cleanouts (pockets) that will accommodate the slot/tab features of the puzzlebox.

Both the slot and tab features of the puzzlebox are composed of a long dovetail cut and a channel cleanout. These sequence of development and production followed gcode for Object Outline with Cut/No-Cut Regions, gcode for four Dovetail Cuts, and for Channel Cleanout. FreeCAD design rendered outputs yielded a collection of gcode for the Object Outlines and for the Channel Cleanouts. The FreeCAD spreadsheet for the project also provided a library of data for the manual development of the eight Dovetail Cuts (four each for Bottom and Lid).

LinuxCNC provides three sets of syntax that supported the gcode generation. The first set of syntax was the standard gcode . The next set of syntax provided was o-code. The final set of syntax is composed of user defined programs referenced as m-code (100-199).

1. Set project specific variables.
2. Set general dovetail cut specifc variables
3. Identifiy bevel shape specific variables for usage in dovetail cutting loop subroutines.
4. Set bevel location variables
5. As deemed needed, execute a first test cut for each of the four dovetails/bevels.
6. Execute a dovetail cut loop subroutine for each of the four bevels:
   • Execute next dovetail cut subroutine
   • Set variables values for specific loop
   Execute loop's specific dovetail toolpath
   • Begin from bevel's safe start position
   • Start dovetail cut at loop's speciifc bevely postion
   • Cut along x-axis to stop point
   • Kickback into "channel"
   • Return to bevel's safe start positon
   • Increment as neccessary until loop count obtained.
7. Start next dovetail cut loop for bevel's subroutine
8. Move to the next bevel
9. Repeat Steps 6,7,8 until done

The suggested sequence/order of milling to use in the CAM phase of the project is given in the following two sets of gcode.

gcode File	Tool	Purpose	Comments
Gcode For Puzzlebox Bottom			Use Home and Check Holes to verify before starting
s001_02_104.01c_02_179.983_038_outline_boxBottom.ngc	Dremel Scribe	Displays cut and no cut regions for bottom and slots.	Abort at anytime confidence is reached.
s002_03_104.03_26_038_dovetail_slot_bevel_single.ngc	Dovetail Cutter	Forms dovetail groove of slot	Creates four bevels
s003_01_104.01_04_179.983_038_boxBottom .ngc	1/4 inch longshaft tool	Executes the mill cutting tasks for slots and cylinder.	Provide "rough" mate for lid's "pin-key"
s004_04_104.01_03_179.984_038_contours.ngc	1/4 inch longshaft tool	Cuts tags and contours for bottom.	Same toolpath as used on Lid. If issues arise use a bandsaw with high tpi.
Gcode For Puzzlebox Lid			Use Home and Check holes to verify before starting
s005_03_104.02c_02_179_985_outline_lid.ngc	Dremel Scribe	Marks the cut/no-cut regions of the lid and outline	Abort at anytime confidence is reached.
s006_06_104.04_19_038_dovetail_tab_bevel.ngc	Dovetail Cutter	Forms dovetail shape of the tab	Creates four bevels.
s007_07_104_038_179_982_lid.ngc	1/4 inch longshaft tool	Executes the mill cutting tasks for tabs, cylinder, and lock.	Creates "pin-key" hole
s008_01_104.01_03_179.984_038_contours.ngc	1/4 inch longshaft tool	Cuts tags and contours for lid.	Same toolpath as used on bottom. If issues arise use a bandsaw with high tpi.

Check Holes:

For a level of confidence that "all is well", properly Home-d, and running without issues "Check Holes" were created and readily used to test the validity of CNC control. This is not a requirement, but a best practice in AAP's shop. Test holes were created immediately after the first, original setting and verification of "HOME". Suggested Check Holes' dimensions are provided further below. Drop back tp testing the currently active tool as often as deemed neccessary. One of the best "decision points" is the "Abort or Continue" dialogue. That dialogue window provides enough time to think, decide, and abort to test.

Dremel Scribe Engraver

MDI Tabsheet Commands

One of various possible jogging devices/approaches.

1. FreeCAD Spreadsheet utilization in the design process
2. Rudimentary Dovetail Cut Degsign: 047 FreeCAD dovetails
   • The project was unable to use FreeCAD 19 due to system impatibiliies with Ubuntu 16.0 and 18.04.
   • FreeCAD 18 proved equal to the task once a bit of elbow grease was applied to the design and simulation runs.
3. Ballpark Dovetail Cut Calculator
4. Real World Dovetail Cut Milling: Matching Slot and Tab Dovetails video provided a recommended the slot and then work the fit of the tab.
   • Recommendation: Set the Slot feature first. Then fit the Tab feature to the slot.
   • By being "fixated/focused" on the puzzlebox's bottom slots as the baseline, only the tabs on the lid will need modification/sanding.
   • As originally viewed, the bottom of the puzzlebox was seen as having two slots extending at 45 degrees from the sides.
   • The project was fully designed using this approach. But it was only until after fully testing the actual puzzle box that a better approach "may have" surfaced.
   • Approach To Be Explored Next Time: View the design with the bottom having the "centered" main slot and two corner slots. Then define the puzzlebox lid having a "centered" main tab. The Locking mechanism would be moved to the main tab. The cylinder cut and lock hole would be moved to the main slot. This may aid in the "closing" of the lid over the bottom of the box by improving closure alignemnt and reducing puzzlebox contents from "jaming" between the bottom and lid components during closure.
     As currently desgined
     
     Alternate design path to pursue at a future time:
     

Puzzle Box Design Supporting Artifacts

• The supporting FreeCAD design project files for the puzzle can be obtained at the GIT.com project repository.
• The NGC gCode files for Milling the puzzle can be acquired from GIT.com project repository.

Tool Table Entries Tool and Pocket identifiers are those defined by the toolable used by the LinuxCNC based CNC (simulator and Sherline 8580 NexGen). Corresponding gcode and tooltable modifications are required to conform to each unique CAD/CAM setup.

• T18 P18 Z4.459 D0.233 ;1/4 router bit longshank
• T21 P21 Z1.273 D0.025 ;dremel scribe engraver
• T27 P27 Z1.0625 D0.500 ; dovetailcutter

Useful Images:

General Knowledge	Design	Simulation	Project
45 Degree Triangle 30-60 Degree Trianlge Layout for Dovetail Cuts LinuxCNC Setup Add Lines to *.ini Add Files to Folders Dialogue Messages Verify Location Verify Readiness to "GO" Shop Support	FreeCAD 18.04 Spreadsheet Bottom Lid Jobs to gCode	Puzzle Bottom s001_*.ngc Outline Start-- Stop s002_*.ngc Bevel Cuts ---- First Cut More First Cut Testing Bevel Cuts For Record Cleanout of Channel s003_*.ngc Cylinder + Drill Cuts s004_*.ngc Contour + Tags Puzzle Lid s005_*.ngc Outline s006_*.ngc Bevel Dovetails Cuts s007_*.ngc Cleanout + Drill s008_*.ngc Contour and Tags	Finished Puzzlebox Display Puzzlebox Test Bottom - Lid Test Runs for Lid Almost Good Lid Check Holes Move To Test Hole Not Used

---

---

Please send comments or questions using e-mail, voice telephone at 919-846-7705. (Last update: June 4, 2021 tar)