Headstart on today's mistakes

Item the first: If your printed object has long, thin towers, don't just grab and yank, assuming it'll come loose:


On a better note, the bridges worked very well on that piece.

Item the second: the towers are weak because I turned off all extra shells. I did that because the shells were not connected but rather had a tiny gap between them. Chatting on IRC reveals that this is a known Skeinforge bug and I should grab the tip off Github to solve.

The break is, in a way, a good thing - it'll get me to fix my SF problems and my printed parts will be stronger.

Today's progress...

was actually progress. Installed all new software (printer control (printrun) and gcode generator (Skeinforge)) - the new Skeinforge is MUCH improved - guys on the IRC channel told me that some of the trouble I was having was due to a bug in SF. This fixes it and it does much less stupid stuff.

I also reduced my com speed to 19200 baud and it made no difference in print speed but should completely eliminate com errors. I had one this morning but it was probably due to me trying to set up a webcam so I could watch the print finish from work. It's also due to the fact that the lab computer is an old piece of junk and it has the crappiest USB host chip in existence.

I printed an X axis idler end, but just as it was finishing I was IRCing with some guys who convinced me to not even bother messing with the printed bushings. Apparently they work OK at first but after just a few hours of printing they get loose and wobbly and prints go to hell. LM8UU linear bearings are absolutely ideal and a full set of them for a printer costs $20 shipped so it's silly not to use them if it eliminates a problem. This new printer will cost hundreds less than my existing one and will be a better printer in almost all respects. When I get it up and running I'll probably see if anyone wants to buy my existing printer. Assembled and calibrated RepRaps go for a few hundred more than I paid for the kit so it could all work out pretty well.

So anyway I won't use the part I just printed, instead I'm printing an X axis motor end with LM8UU linear bearing sockets and a slightly tweaked nut trap size. I'm not going to press my luck by starting a print when I'm sleepy and not really thinking clearly and letting it run unattended.

I do still have to put the power supply back together again, it's running naked, but I should take the opportunity while it's open to nibble out the fan hole and install the fan cover I printed the other day.