What would you charge?

I know some people use their MakerBots as small scale rapid prototyping services.  What I don’t know is what or how they charge for providing these services.  Obviously, there’s a cost associated with the plastic, electricity, wear and tear, failed prints, etc. 1 Setting aside the intrinsic value of what you have produced 2 , what’s a reasonable way to set the price of an object printed on a MakerBot?

I can think of several ways to look at this:

  • Resource cost. $0.04 per cubic centimeter, of pure cost in terms of just MakerBot ABS/PLA.  Electricity is probably around $0.02 per hour of print time.  The laptop connected to the MakerBot is capable of multitasking, so there’s no real opportunity cost there.
  • Shapeways. Their gray ABS is $2.50/cc.  They also have a white nylon polymide for $1.50/cc with a $1.50 start up charge.  Their colored prints are made with a plastic powder that is relatively fragile.  At $2.50/cc and assuming it takes 4.5 minutes to print 1cc, this comes to $0.56/minute.3
  • Ponoko. Their UV curable resin is $2.76/cc and their white nylon polymide is $0.80/cc. 4 Their colored prints are made with a plastic powder that is relatively fragile.  At a MakerBot’s printing speed, $2.76/cc comes to $0.61/minute.
  • Metrix:Create. They charge $0.50 per actual minute of build time.  However, Metrix:Create members get 20% discounts and anything printed from Thingiverse gets a 20% discount.  Without any associated discounts and assuming we print at the same speed, this comes to $2.25/cc.

I believe transparency makes for a better customer experience.  That way a customer can see right up front what they may be spending, avoiding “sticker shock.”  It is probably a lot easier for the average customer to estimate the volume of a digital design than it is for them to guesstimate how long my printer would take to print such a thing.

What metric and pricing structure do you use to charge for your printing services?  Do you go by volume, weight, plastic used, machine time, or something else entirely?

  1. It takes about 4.5 minutes to print one cubic centimeter of ABS or PLA.  This is a rough average of several different types of settings, but assumes a roughly 0.36mm layer thickness. []
  2. Pretending 19cc of a puzzle cube is worth exactly 19cc of a toy rocket, window latch, or plastic part that will fix the international space station. []
  3. This is more of an analogy to a MakerBot’s printing speed than an actual assumption of the printing capabilities at Shapeways. []
  4. There are discounts for their “prime” members. []

Block Production

I’m churning out Beco Blocks.  I’ve got 5 sitting in front of me with another 6 printing up right now.

This makes me happy.

Peanut Butter Mousetrap design notes

I treat this blog as part of my lab notebook on printing and designs to help keep me organized.  Having it automatically add tags, adding a time stamp, organizing everything in chronological order, and then making it available for others to learn and comment back are all just a huge bonuses.

Anyhow, here are some of my design notes for this Peanut Butter Mousetrap Insert:

  1. I was going for a minimal design – minimal plastic, printing, and machine time.
  2. I wanted something that would be “ABS warp” resistant.  Even if you end up with a warped flat surface, clamping the insert between the lid and the jar should even it out.
  3. I wanted a resilient design.  Even if the insert is badly warped after clamping down, a flat surface is not critical to its function and it should still work.
  4. I wanted something that could be adapted easily.  I don’t know how to use OpenSCAD, so a parametric design is kinda out the window.  Even so you could print this object at full size and then cut it down to what you need.  1
  1. For the record, I did download OpenSCAD before designing this.  I just wanted to get it out there. []

Making Blocks II

Lego nubs and underbits

Lego nubs and underbits

My lego prints connect slightly better to other printed legos than real legos.  I suspect this is in part to the ridged texture to both parts.  This also makes me think that printing legos is a fool’s errand.

While lego nubs are circular, the parts they fit into touch the circular nubs tangentially.  For instance a nub fitting into a corner piece would be held in place by two straight sides and held against a circular tube.  The picture to the right shows the underside and nubby bits for several lego (lego compatible?) pieces.  This system probably works because of the exacting tolerances of real lego parts.  I’m guessing they tested the nub heights and contact points to find an optimal mix of most of the same criteria I have for an interconnecting building block system.

Without tighter printing tolerances, I don’t think three tangential points are going to be enough to hold printed parts together.  Thus, it may make sense to have taller nubs on printed parts.  Then again, the ridged nature of printed parts may allow them to better fit together without having taller nubs.

One benefit to having circular nubs is that you can make interesting components where pieces pivot.  But, this is more of a lego-hack than an actual useful feature.  I’m not as concerned about this particular usage because I could always design a printable pivoting part.

So, other than that one narrow usage, why circular nubs?  Why not circular nubs with circular holes?  Why not square nubs with square holes?  Or circular nubs with notches in them?

A rainbow of colors!

Printing in red, green, yellow, and blue ABS?  Hells yeah.

Now, I haven’t even made a dent in my black ABS supply, so I can’t really justify dropping $320 + shipping on more plastic.  However, I’m certainly hoping someone asks me to print something up in color so I have an excuse to buy some…

How are you printing with PLA?

I’d like to print in PLA, I really would.  No warping?  No rafts?  Smells like syrup?  Makes kittens and rainbows?  Sign me up!

Rick from MakerGear has the following suggestions for printing in PLA:

  • Have a 150-160 C degree temperature for the entry of the barrel
  • He notes that the nozzle temperature is roughly 190 C degrees
  • However, later he suggests printing at 220-235 was possible
  • He suggests not to worry overmuch about PLA in the threads – its just one of those things

Nick McCoy offered several pieces of wisdom:

  • He found that 10W-30 oil added lightly to the PLA as it entered the extruder, by adding oil to a piece of cotton wrapped around the filament, helped it print without jamming
  • He noticed that pausing the extruder for more than a few seconds would jam the barrel
  • He found he could print at 180-210 degrees (I assume he’s measuring the nozzle temperature as you would with an ABS setup)

Nick’s results suggesting a nozzle temperature of 180-210 seem to match up with Rick’s findings of 190 or so.  I think I’ll give Polly another shot and see if I can extrude at those temperatures.

Skeinforge settings changes

I figured, why not treat this blog as my lab book, right?  It’s a good way to document changes in settings, as well as my rationale for doing so.  Thusly, and without further ceremony, delay, pomp, or ado, I present for your kind consideration and review my latest Skeinforge changes:

  1. Skeinforge->Raft->Activate Raft: checked->unchecked
    • Now that I’m printing in PLA, I won’t need the raft.  As you can probably tell from the pictures below I’ve covered my acrylic build base in blue painter’s tape.  I think I’ll probably just take one of my spare foamcore build surfaces and cover that with blue tape and give that a shot.  I’ve noticed that as soon as my extruder heats up the PLA starts to ooze out.  This isn’t a problem as far as I can tell.  However, it does let some PLA drip – but when I start to run the extruder again it doesn’t extrude at all.  My guess is that the oozing PLA is dripping out of the nozzle, which then has to be filled back up to extrude.  ANYhow, my point is that if I heat up the extruder and don’t immediately start to print it will take a while to fill the nozzle, which means the raft gives the extruder sufficient time to get going before the part is being printed.  Rather than print a raft, waste the plastic, and then have to chip it off later it makes more sense to make sure I advance the extruder a little before I start printing.
  2. Skeinforge->Fill->Infill Solidity (ratio): 0.4->0.3
    • I’ve printed with as little as a 20% fill with no noticeable problems.  I suppose it would really depend on your model.  If the top of the model begins to taper up to a point, then you could probably print with 0% fill.  However, if the top of your build is flat, the topmost layers won’t have much to rest on and will sag just a little.  This doesn’t appear to be a structural problem, but more of an aesthetic one.
  3. Skeinforge->Fill->Infill Begin Rotation (degrees): 45.0->90.0
    • I tried reprinting Zaggo’s whistle in PLA and it didn’t come out – just like it didn’t come out in ABS.  I’m not sure how or why, but my infill rotation got set to 45.  I figure I might have a more solid fill/wall if I had this rotate to 90 degrees.  We’ll find out!
  4. Skeinforge->Fill->Solid Surface Thickness (layers): 3->4
    • This was just a slight tweak to see if I could make solid layers slightly more solid and waterproof/airtight.

The only problem with that system is…

Not really a problem after all.  I’m printing up a (well, another) teapot right now.  Having just run the PLA through rather than trying to actually remove the ABS the teapot has these black streaks throughout.  I kinda like it.

Crystaline teapot

Crystaline teapot

On second thought…

After seeing the dramatic difference printing from the SD card made, I think I’ll try my hand at printing the replacement insulator retainer ring again.  I think I’ll end up with a much better result on the small screw holes by doing this.

Sooo…  if you break your acrylic (or ABS) insulator retainer ring, drop me a line.

Tab A, Slot B?

Okay, I’ve laid out all of the wooden lasercut parts, printing side up.  Since I’m painting the interior a different color than the exterior, I need to make sure which parts will end up being attached where.