Uncategorized

Production Update: One step forward, two steps back (do see do and swing your partner)

thanks to a suggestion from IRC freenode#reprap, I turned down the X axis polulu and it started to work better. This infill test was going perfect, until I noticed the Reprap Prusa was no longer pushing ABS out the heated nozzle. I checked that the plastic filament wasn’t caught on anything. I tried to extrude 1mm. Curious, the stepper makes noise but the gears on the extruder don’t move. I tried again from another angle – AHA!

Now that I look at it, this PLA extruder sprocket isn’t completely filled in. That made it weak und flabbey ya. Today’s goal is now to get down to the VHS and print TWO better, strong, faster ones on the Tantillus and then get back here and kick out the jams. Why two? Backup.

Uncategorized

How not to design an interface

Interfaces are a long time fascination with me. Interacting with people is hard for humans and even more so for inanimate objects. Take a look at this menu I saw recently.


Can you show me an example of a menu with a worse layout? I had to ask for help to be sure I understood which item went with each picture, even though I love japanese food!


The bottom half isn’t any better. Do you see the hiddeous menu background? How about the way the dishes aren’t photoshopped to remove the non-dish elements? I’d at least try to photograph the dishes straight down so every piece looks as big as possible.

Maybe I’m spoiled from online shopping experiences. I expect one picture, one description, and one price for each item. Each should be laid out in a logical manner. Heck, this isn’t even a GridBag layout.

My microwave, dishwasher, and TV remotes are all equally stupid. Someday soon I’ll come back and post about how broken their interfaces are.

Now on the other hand, a nice interface I’ve just discovered is the one minute key duplication machine at my local Home Despot.

Asides from the obvious security concerns, this is a neat idea in an elegant package with a dead simple interface. Stick your key in the hole. If the machine approves it grabs your key and won’t let go until you complete the on-screen instructions. It even gives you a choice of keys at different prices. Best part is the clear window into the guts so you can watch your key being copied.

You’re only there a minute but they still find a way to keep you entertained. Also it’s a great way to debug the system. I wonder how they clean the swarf away automatically…

Do you have an example of an interface you love/hate? Post it online and link it here.

Uncategorized

Production update: first prusa print results


Here’s the first layer height calibration print. The goal is to see that layers are properly sticking together, the nozzle doesn’t drag through the deposited material, and that XY movement on each layer is the same. All in all not bad. The challenge is that the bed sags in the middle. The nozzle is too far from the bed at the start of the print, which is why the bottom is wierd and the outside line (to clear the nozzle?) is wavy. I’m getting a piece of glass cut today that will fit over the entire bed and not sag in the middle. Not sure yet how to make the heater adhere to the glass for good thermal transfer.


This is the start of an infill calibration test. I stopped it when it started to go wonky. It looks like it might not be outputting enough plastic to fill the cube 100%. I’m going to recheck my Slic3r settings first, I may have had it set to 40% infill.

So in the meantime I’m not sure what else I can check. Any ideas? I’m thinking I’ll post on a few subjects close to my heart.

Update 2012-09-20 13:30 – Tried the infill cube again and the extruder made a nasty clicking sound. The screws that hold the extruder stepper in place had come loose (after almost no printing!) and some filament retractions were making the gears skip teeth. Ouch!

Update 2012-09-20 13:45 – Cancelled another infill test. What is causing this result?

Update 2012-09-20 14:40 – More layering tests. I’m going to stick to these until they’re perfect, then go back to infills. I changed the Slic3r filament extrusion multiplier to 0.8 and tweaked the extruder retract parameters.


Update 2012-09-20 16:30 – layering tests look great! Slic3r had a 0.3 nozzle size when it’s actually 0.35.

The infill test started out great, bu then something went wrong with X movement causing layers to become misaligned.

Update 2012-09-20 17:53 – Video! Skip to -40 from the end to see where it goes wrong.

Uncategorized

Production update: 3D printer is now heating, moving, and extruding!

The gentlemen and ladies at the Vancouver Hack Space are truly worth their weight in gold. Certainly more than the $50/month membership! I have been hanging out in IRC freenode #reprap fighting with this Makerfarm Prusa for a week. VHS found the issue(s) in a single evening. Go visit your local hackspace! They are probably even more incredible.

Before I go on, I want to be clear: Makerfarm make a great product and have excellent support. Truly! I had to get a couple of parts that I knew weren’t in the kit and, naturally, documentation couldn’t cover them. The mistakes I made were entirely mine.

The biggest thing was with my power supply. I have one of those ugly ATX boxes from a computer. (A ThermalTake TR2 600W.) It’s the part you plug the power cable into, and on the other sides it has a giant fan and 50 smaller colored cables of different types coming out. As a safety measure the box won’t turn on unless it’s connected to a PC. The way to trick it is to close the gap between two of the wires with a “bridge”. I bridged the wrong two so it was constantly turning on and off very very quickly. Whoops.

Then we found the polulu carriers – the chips that make the steppers move – had to be adjusted. There’s a little knob, called a pot, that needs to be turned very gently with the tiniest of screwdrivers to tweak the amount of juice going to each stepper.

The next thing was the greg’s extruder – it was moving backwards! Thankfully this is also an easy fix: The plug goes in two ways, and I guessed wrong when I put it together. Then I tried extruding some plastic from the hot end and found that there was some slip in the toothed gear that grabs the ABS filament. A drop of gorilla glue later and the gear was moving great.

So happy days! I got my first night of real sleep in about two weeks and woke up feeling fantastic. I felt so good I cleaned the tub while I was in the shower. The relief is palable. (go on, touch it. ha!) The very next thing I do after I hit post is starting making test prints to calibrate the machine. More updates on the way!

PS: Can you think of any tech blogs that are looking for news? I’d love to get my indiegogo campaign mentioned on them. It would mean far more coming from you than from me. Think Endgadget, Slashdot, io9, Hack-a-day, IEEE spectrum, Reddit, etc. If you submit one, please comment below. I’d love to build a list of related blogs for future media blagging.

Update 15:40 – I think I have it fully calibrated for XYZ movement. I’m trying to check extrusion now but Slic3r (the program that makes my model into gcode) is being stupid and doesn’t want to talk to Repetier (the program that runs the printer).

Uncategorized

Do you have a Drawbot? Is there a Maker Fair happening near you?

You know, World Maker Faire day is coming up in less than two weeks. I’d love to see Drawbots at local Maker Faires, drawing while their parents tell the world what they think. I’ve got the setup/teardown routine well practiced now and could provide tips, and even a list of fairly popular pictures that you can draw as you show off your creation. Just planting a seed.