Until Friday, when my frustration with a certain class of software got the perspective and skill of my being a professional engineer. Instead of complaining, instead of getting frustrated with how someone else managers their project, instead of not being able to pull rank and make them do it my way, I can just go write my own and hand it out.

It’s a nice feeling. I don’t think I’d have gotten here without writing it-cannot-fail code for security and privacy projects as a day job. That sort of rigor is like daily exercise for the brain, like daily exercise or workouts only for the brain.

However, nobody on the MakerBot list really had a reason or a fix I liked, so I took it apart to fix on my own.

The first thing I discovered was that the pressure mechanism for pushing the plastic against the stepstruder was not pushing hard enough, due in part to scratches on the head’s surface.

The second — and more important — thing was that the stepstruder gear was filled with shreds of plastic. It came off easily, so like the Mk6, just pushing the plastic harder would clear the junk out of the motor.

But how to do that?

Like this.

It’s really that simple. Grab a spare M4 bolt from your parts kit, buy a wingnut at the local hardware store and glue/Loctite the wingnut to the end of the bolt.

I set it up, dialed in a temp of 240C (see earlier posts about temperature), and it forced plastic through with no problem at all. Even a 5 minute run ran without any problems at all.

Now to print some test models….

Like the previous set of tests, for temperature reads I used:

• built-in sensor that ReplicatorG software reads
• Fluke 88V temp sensor with temperature probe. Not a cheap device but the readout is 0.1C and it’s fast
• $20 BBQ grill cooking thermometer. Cheap, easy to use, but I think their goal is to make sure I don’t kill someone with undercooked poultry.

As with the Mk.6, the temperatures are lower on Replicator than on the other two sensors, so I need to run with much hotter temps than are required by the instructions. At this point I’m considering adding an offset/calibration value in the software so that the read temp is adjusted properly.

(The images are hosted on flickr for easier commentary)

Mk.7 with software 0033

After some decent experiments with the Mk.7 (more on that later), I decided to try the 0.3mm head and 1.8mm plastic on the Mk.6 after the success with the 0.4mm head and 3mm ABS.

It works, but not well enough to actually print. The drop from 3mm to 1.8mm plastic, almost %50, should have worked well with the drop from 0.4 to 0.3 but the extruder doesn’t want to extrude. There was some talk on the makerbot list that black 1.8mm plastic required higher temps, those were red-lined by the software, so I did some measuring on my own.

Back in the early days, getting your print temperature correct was one step from ritual magic. That seemed likely, so I decided to repeat some early tests and wire up second and third sensors to run in parallel with the default sensors.

With these, I collected some data and made some graphs. The sensor tips were put as close to the MakerBot sensors as possible and I waited as long as seemed reasonable for the temperature adjustments to stabilize after a change.

• built-in sensor that ReplicatorG software reads
• Fluke 88V temp sensor with temperature probe. Not a cheap device but the readout is 0.1C and it’s fast
• $20 BBQ grill cooking thermometer. Cheap, easy to use, but I think their goal is to make sure I don’t kill someone with undercooked poultry.

The short answer is this — the replicator is reading at least 10C lower than anything else, so I need to run with much hotter temps than are required by the instructions.

(The images are hosted on flickr for easier commentary)

Mk.6 tests:

Mk.6 with software 0033 (recent tree build)

Mk.6 with software 0029 (distributed)

HBP tests:

HBP with software 0033

HBP with software 0029

My thinking is doing fine, I’m passing all my medical tests, and I can write code and talk on the phone and drive around town. I was wanting to go back to work a month before they let me, but I think I needed the time off for the physical work.

Physical stuff is a little slower — concussion recovery is nothing to joke about. I’m not allowed to do all sorts of “dangerous” exercises like riding my bicycle, running, or even running on a treadmill or standing on a chair to change a lightbulb. I’m actually proud that I can go for a mile or two walk, do a dozen situps, lift weights, and go back to music lessons.

More importantly for you, I suspect, I am back to soldering-and-coding. I’m finishing up some existing projects like finishing the Cupcake Upgrade docs, working on Lasersaur #4, and starting some new software for digifab. I’ll update things about hacking here and talk design stuff on allartburns.

The trick to getting it all working was replacing the Z stepper that shipped with Cupcake #235 and replacing it with a more powerful one from Lin Engineering (which required a new metric pulley from SDPSI). The Z-table is now quite zippy and does not stall like the old one did.

Photos and parts lists to be posted everything’s a scattered mess in the aftermath of Pittsburgh MiniMaker Faire, the lasersaur build, fixing my mom’s laptop, and a thousand other papercuts.

The STL and images are posted over on thingiverse. The short of it is:

1. make a hollow object
2. determine which way will be up/down when you print it
3. add columns along this axis that are close enough together and to the edge that the spanning shells will hold
4. print!

Here’s an example: The top piece is a completed print, the bottom is a partial print showing the base, rotated 90 degrees. You can see the starts of the three columns that support the rest of the print and the inlet (the big hole on the top print) and the outlet on the other side of the print:

It’s really just that simple.

handwriting ->blog might kick ass.

Today, for example, I got a request to do a tech review in trade for a copy of the book. Here’s my response:

Sorry, but I’m not interested in doing tech reviews for a copy of the book on a subject where I’m already an expert. Back in the 90s I got paid decent bank per-section to do tech reviews, I see no reason to work for less than minimum wage at this point in my career.