3D printing Pickup Bobbins

Hello world. It’s been a while.  I have been busy with various projects, the holidays, etc, BUT here is a doozy: 3D printing guitar/bass pickup bobbins.  The standard bobbins that are available are good, cheap, and easy to find, but CUSTOM shapes and sizes… not so much.  I have hit a wall several times with various designs simply because I could not find a time and cost effective solution:

I have been experimenting with various methods of making custom bobbins: molded plastic, machined wood/plastic, welded stainless, etc, etc… and 3d printing has by far been the simplest, quickest, most flexible, and most successful.  This will be a quick overview of  the printer I am using, the build settings, helpful links, and some tips and tricks to get you in the ballpark if you decide to take the plunge.

And speaking of taking the plunge:  it is not very expensive.  All told, I spent about $300 on printer, filament, and modifications.

Anyway, here is the printer:

printer

The model is the Monoprice MP select mini.  It has a print size of 4.75″ cube, so it is plenty big for bobbins.  I could not find another printer in it’s price range that could print as well, or had as many features.  At $190 (and free shipping), I could not pass up the opportunity.  I honestly expected to get poor results and eventually HAVE to move onto a more expensive model for final products, but I got usable results almost immediately.

Having said that, I DID find a few issues that I addressed with some initial modifications.  I had to print a filament guide to keep the filament from dragging on the printer itself.  The aluminum build plate was not flat, so I added a glass bed.  Also, the fan is a low-flow 30mm pc fan, so I printed out an adapter and shroud to fit a low-noise, high-flow 60mm model, which greatly improved both the noise levels and the prints.  (Future improvements will likely include an improved filament spool arm, a secondary cooling fan with external power, proper clips for the glass bed, a better extruder nozzle and bracket, etc.). I also placed the printer on a maple cutting board, which was atop a thing sheet of cork, which rested on two wooden blocks (to stabilize the printer while printing and damp some vibration).

A few limitations of this printer are that the firmware is closed-source, so you are stuck with factory options (apart from GCode hacks), the bed only heats to about 60c so printing ABS might be tough, it prints fairly slowly (55mm/s or so), the bed-leveling is just a few hex screws, and not particularly precision, while the 4.7″ print size is okay you may need a larger area for very large bobbins, it is a single-head extruder so you can’t print in multiple colors/materials (i.e. no water-soluble support prints).  There is also the aforementioned fan issue, and the power supply is not particularly robust (some users have reported that the supply died quickly).

With all that in mind, let’s move on to milking the most out of this machine:

BOBBIN DESIGN:

When you start cranking out prints, you will soon realize the strengths and weaknesses of 3D printing.  The biggest weaknesses that I have run into as relates to printing bobbins are 1) the finish is not smooth, and 2) the bottom surface of overhangs is not smooth. I tried printing classic bobbins in a single piece, but the overhang of the top lip of the bobbin required support structures, which (along with layer-sag due to gravity as the filament cooled in the milliseconds after extrusion) caused a rough surface which was unsuitable for winding.  I tried several different support structures, and was not happy with any of them.  This might be a result of either the printer, the filament, the settings, or the slicer software, but I tried literally dozens of options and was not happy with the end result of any of them.  Also, the extruder head pushes down each successive payer, so even with perfect support and settings, the initial layer(s) of the top lip would never be as smooth as the top of the bottom lip.

This could be resolved by a dual-extruder printer with a water-soluble support structure with a flat roof, but that is beyond the capability of this printer. You COULD use a snap-off support structure and grind/sand/machine the bottom overhanging surface… but I wanted a more elegant solution, so I re-drew the bobbin in 2 pieces so that I could lay the top and bottom on the print bed for best-looking results, as well as having equal-flatness on the inner flats. The bobbins actually snap together and hold very well, but I will be super-gluing them together for additional rigidity (and to avoid any possible microphonic issues):

freecad

The finish of the top and bottom surfaces will still look 3D-printed, and since it is standard practice to print a thicker bottom/initial layer, it will not look as good.  A third option would be to print in 3 pieces so that the top and bottom faces of the bobbin are ‘outer-shell’ prints, and thus the best looking, but they will still not look perfect.  I will include a section on surface finish further down in this post to examine a few options.

SOFTWARE:

I drew the models in FreeCAD (free for commercial use) and 123D design (free for non-commercial use, but requires a commercial license, and it is fairly limited, so I only use it to knock out experimental ideas quickly), exported to a *.STL file, and sliced them in CURA. Blender works as well, but it is geared more toward organic shapes, so it is a bit clunky to use as a CAD editor.  I will include links at the bottom of this post.  There are also paid options for software, but I will not be exploring those in this post.

Your slicer (Cura in my case) is where you will be setting all the print options.  You can peel back through each print layer to preview the patterns in realtime as you set these, so the final optimizations are easy to visualize.  There is also a print time estimator so you can see the estimated time impact of the various options (thinner layers = longer prints, complex patterns = longer prints, etc).  My builds took between 28-52 minutes depending on options.

cura

FILAMENT:

I used PLA filament.  While the machine will print other compositions, PLA is easy to work with, and provides good results.  It is also bio-degradable and made from readily-available vegetable matter material.  ABS is another option that is a bit less rigid/strong but a bit more flexible and forgiving, AND the print can be smoothed with acetone for a smoother surface finish (more on that later).  ABS is also a non-biodegradable petro-plastic.  ABS also has a slightly lower dielectric constant, and is commonly used in injection moulded pickup bobbins, so it’s sonic characteristics are well-known.  There are also other options which I have not tried, many of which will print on the MP mini (including nylon, various plastics, and wood-pulp filament… which would be interesting to try).

SLICER SETTINGS:

Once you have your model drawn and imported, you will have to set up your print.  I spent a lot of time experimenting with the best options (which I will list at the end of this section), and testing material strength, print speed, etc.  Here is what I have gleaned (you can download test models at thingiverse.com and the like to try this out yourself):

-PLA solid mechanical properties-
Strength (MPa) = 40-70
Elongation at break = 4%-6%
Young Modulus (rigidity, GPa) 2-4

-Prints will be 30% weaker on z-axis than on x/y

-LAYER HEIGHT-
MPa = 29@0.1mm, 33@0.15mm, 35@0.2mm, 37@.25mm, 36@0.3mm
0.2-0.25mm = most rigid and strong
0.15mm and 0.3mm = same (lower than above) rigidity and strength

-SPEED-
10-50mm/s strongest (slower is more consistently stronger, faster is almost as strong, but less consistent. Strength decreased linearly with speed)

-FLOW RATE-
100% = 30% stronger than 90% (so flow as fast as you can melt)

-TEMPERATURE-
220c = ~5x stronger than 190c (over 230c = no change)

-INFILL-
MPa = 20@50%, 30@70%, 40@90%, (46@100%)
(Note that this curve is exponential)

30-50% infill is least efficient in terms of cost/time/strength

90% = most rigid and strong (highest yield strength but lowest elasticity)
80% and 100% = same (lower than above) rigidity and strength

45 degree diagonal linear squares and diagonal diamonds are strongest, followed by hex

-MATERIAL-
PLA = 30% stronger (MPa Tensile strength) than ABS
(ABS = 35% more flexible than ABS)

-CONCLUSION-

Strong fast detailed = 70% infill, 0.2mm layer height
Strong fast = 80-90% infill, 0.3mm layer height
Strong detailed = 80-90% infill, 0.15mm-0.2mm layer height
Best settings overall for balance of strength, speed, time =
-PLA
-3 shells
-0.2mm-0.25mm layer height
-85%-95% infill
-highest temp possible
-highest flow possible
-moderate speed (20-30mm/s critical layers & 40-50mm/s infill, non-critical layers)
-fan off first layer, extra cooling higher layers
-bed heat in glass transition range (60c-65c) early on, then decrease
-45 degree diamond pattern fill
-2-3 solid shell layers on face will look good aesthetically

The settings will vary depending on the printer, filament, etc, and may have to be varied for increased detail, speed, build plate adhesion, etc.  You can only really nail down those setting by running off a few test prints.  Different colors typically require different settings as well.

EXPERIMANTAL PRINT RESULTS:

Here are some pictures of bobbins printed at 3 different settings.  I printed square cutouts to observe the detailing and crispness and small shapes (circles are too easy for a torture-test):

45 min build
1st try: too cold @ 195c, too little extrusion @ 85%
2nd try: too hot @ 220c, too much extrusion @ 100% (might work with additional cooling)
3rd try: good @ 210c, 95% extrusion

lcool_mhot_rmed

The first print was  too cold @ 195c, too little extrusion @ 85%.  The layer fill pattern did not achieve good adhesion with the walls, and there were many voids in the print:

too_cool_voids

The second print was too hot @ 220c, too much extrusion @ 100%.  The material melted past the walls on infill passes and made for rough edges which were unusable.  There were no voids, but the blobs were terrible:

too_hot_bad_edge too_hot_bad_edge_2

The third print was pretty good good @ 210c, 95% extrusion.  There was good wall adhesion, the pattern looked tidy, and there were minimal voids in the pattern:

good

Some tweaking will produce even better results, but this one was usable enough.  Here are more comparisons of the three:

lcool_mhot_rmed_2 lcool_mhot_rmed

The hotter print settings could possibly work — and look best — with another cooling fan and some tweaks to a few other settings (like higher speed and lower extrusion).

Here are the settings that I used:

Printer = Monoprice MP Select Mini
Layer height = 0.21875mm
Initial layer height = 0.2625mm
Pre-heat extruder and platform = yes
Print temp = 210c
Plate temp = 60c
Filament size = 1.75mm
Wall thickness = 0.625mm
Wall count = 2
Alternate extra wall = yes
Top/bott thickness = 0.6mm
Top layers = 2
Both layers = 3
Top/bott layer pattern = lines
Infill = 90%
Infill pattern = grid square
Infill overlap = 75%
Print speed = 40mm/s
Wall print speed = 20mm/s
Initial layer speed = 20mm/s
Travel speed = 150mm/s
Print cooling fan = Yes
Fan speed = 100%
Regular fan speed after initial layer height
Build plate adhesion = skirt
Skirt line count = >/= 3
Build platform = hairspray-coated glass plate clipped to heated aluminum bed
(TIP: Do not try USB or WiFi print.  The printer tends to stop between each layer and re-buffer?  SD card printing worked much better)

FINISH OPTIONS:

You can see that the faces of the bobbin will have the 3D-print-looking cross hatching.  The material will take sanding and smoothing, but unless you print the surfaces very fine and very hot, you will still have voids that would require filling.  I tried fill/paint/polish and had great results in terms of adhesion, smoothness and shine.  Again, printing a 3-piece flange+core bobbin with the outer surfaces being printed as the top shell would produce smoother results, but would increase print time, complexity, and likely have to be printed in separate passes for flanges vs core since the vertical print of the core would require slightly different settings for best results.

Printing in ABS is also an option since ABS will dissolve in acetone.  So you can smooth the surface chemically rather than just mechanically.  (PLA can be smoothed similarly, but the chemicals are rather nasty, so I will not discuss them here.) There are also aftermarket surface-finishing options and epoxy which I have not yet tried.

Another option is to just run with the hatch pattern as your pickup face since it looks kinda cool. A cap for the top face is also an option:  a thin veneer of wood, etc.

If your pickups are covered, this is all a moot point, of course.

Here are some bobbins that I printed in a single-piece to show the difference between bottom/initial layer as face and smoother top layer as face:

img_0627

ALSO: 3D printing opens doors to printing custom shapes and patterns on the face of the pickup.  The tech is not great at the moment for producing the smooth and polished classic bobbin face look without extra work, but it excels in easy customizability and fast one-offs.  A simple option would be not to print a top shell so that the fill pattern is visible.  There are some cool-looking fills out there.  Also, just about any custom/artsy shape could be printed: Etruscan patterns, Celtic knots, Nordic runes, steampunk gears, monograms, lettering and names, etc etc.

Here is an example of simply leaving the fill pattern exposed.  Obviously the fill was not printed to be pretty, and this one was for a prototype so it was printed too hat and too fast, but it gives a general picture of the possibilities:

fill_pattern

PICKUP POTTING ISSUES:

Treat PLA bobbins like Butyrate since PLA has the same glass transition temperature (140f/60c).  I have warped Butyrate at typical mid-140’s potting temperature, so the same care should be taken with PLZ bobbins. Beeswax is likely off the menu since it melts higher (143-147f), whereas paraffin has 3 basic grades: high (150-165f), mid (135-145f), and low (122-135f).  There are many grades in-between if you are willing to hunt for them.  Soy wax melts even lower, and I have seen compositions that melt around 128f, as well as soy/paraffin blends that melt around 133f (I have not tried these blends).  So mid-temp-grade paraffin seems to fit the bill nicely.  If you have an attachment to beeswax, I imagine that blending it with low-melt paraffin or soy could work (again, I have not tried this, but I plan to experiment with blends in the future).  I would not go too low on the wax melt temperature grade since you can run into melting problems if you do something horrible like leave your guitar in the backseat of a car on a hot, sunny day.

SONIC CONSIDERATIONS:

PLA has a dielectric constant around 3.1-3.4+ @60Hz.  This is closer to Butyrate bobbins (3.3-3.6+) than it is to ABS molded bobbins (2.2-2.5).  As a result, the capacitance figures will be closer to vintage Butyrate than modern ABS.  Those dielectric constant figures are for solid plastic though.  Your print will have voids of either air (if the print ends up being sealed and/or the pickup is not potted) or wax, so the constant will vary depending on how much wax/air ends up in the bobbin.  (as a refresher, parafin wax=2.2-2.5, bees wax 3-3.5, epoxy 4-5).

That’s it for now.  Hit me up in the comments section if you have any questions.  I did not include basic 3D printing knowledge because it is covered ad-nauseum online.  Don’t be shy about asking around on print forums since many of these people are very passionate about spreading knowledge on the subject.

Next step for me:  3D printing a better coil winder.

LINKS:

Printer:
https://www.monoprice.com/product?p_id=15365

Cura:
https://ultimaker.com/en/products/cura-software

FreeCAD:
https://www.freecadweb.org

123D:
http://www.123dapp.com

MP mini Wiki:
http://mpselectmini.com

Google Doc of MP mini help:
https://docs.google.com/document/d/1HJaLIcUD4oiIUYu6In7Bxf7WxAOiT3n48RvOe5pvSHk/edit

Reddit MP Mini sub:
https://www.reddit.com/r/MPSelectMiniOwners/

Printing Visual troubleshooting guide:
https://www.simplify3d.com/support/print-quality-troubleshooting

Advertisements

About alexkenis

Guitarist, philosopher, tinkerer

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: