04/23/2018 at 08:52 •
I haven't read the fine print, but I see other users who are participating in the 2018 Open Hardware design challenge telling their followers to un-like and re-like their projects. There must be a reason..
So please un-like, and then re-like this project if you want to see this project move forward.
I still have zero time right now as I'm moving in 7 days and been busy packing up a huge house with 2 multi-discipline hackers living in it. At the same time I've had to setup our RV for living/working.
Needed to add some shielding here (the aluminum screen) to keep the inside and outside cellular antennas from feeding back and causing the amplifier to shut down. Eventually I hope to document everything about it, but right now it just has to work.
Moving and RV setup have been mind bogglingly huge tasks and have been going on for months now. The end is drawing nigh though, so now I really need to focus.
It still also contains the source for the C1 as well as I haven't had the time to invest to split it up and document some of it yet, but hey... it's out there now. When a 'Hey, I wonder...' project evolves into something public, things can get messy so don't judge. :)
Let me know if you think that the rendered stl files should accompany it.
04/10/2018 at 17:13 •
Sorry for the lack of updates, but it's been busy around here. Moving and stuff.
The lack of a suitable printer to just slap this onto also means the C1 needs to work first, and then I got distracted with this: https://hackaday.io/project/101776-interdimensional-portal-gun
I'll be back to this soon.
03/16/2018 at 03:30 •
You know.. I have a multi-material mixing print head. Perhaps I should actually use it.
So I did.
No actual active mixing, but I programmed the change from white to black at least. I somehow found I could spare the time required for adding 2 lines of gcode (M163 filament percentage commands for the curious) at a layer change... mostly. Well.. I liked the idea of the M163 command, so I implemented it within the guidelines of the actual standard.
My implementation uses non reserved characters... yes, gcode actually has rules people.... and it is 100x more precise. I use a range of 100.0 to 0 instead of 10 to 0. It doesn't matter much here as the swing is from 100.0 to 0, but your get the point. Yes.. you may have assigned meaning to this Mcode definition, but in the process you used characters (in a very simple system) already assigned to other functions cause they fit your mindset.. And now we are all screwed.
Sorry.. but coming from the 'other side' I'm annoyed at the general RepRap movement not following the rules.
But hey... usage equals reality... Um... No.
Standards exist for a reason. If you want to deviate, use the built in variables which allow for deviation, or define your own system. Don't call it gcode unless systems designed to read gcode can actually do so.
Bastardizing existing standards sucks, and I simply refuse to adopt that kind of M$ mentality. Well... I have to admit M$ has been getting a lot better at this in the last few years. But...
You know how many hours I've spent of my life working around the stupid little changes "companies" have made to established standards, simply because it was easier or they wanted to add feature X and didn't want to do the associated work required? I actually have no idea... But I can tell you that I lived through the early days of the internet and that I still deal with this kinda crap for my "real job" on daily basis. Compared to actual problem solving, working around silly choices like these consume an absolutely huge portion of my time.
So... I have chosen to implement something within the existing standard. I had to change one character....
Now is the time. Resist!
White to black is one of the things that the M2 does exceedingly well. No thought or planning involved. The sheer density of the pigmentation of the black filament takes care of everything.
Going from black to white... that requires actual planning.
So now I have the perfect light ring. You can paint yours. :)
03/15/2018 at 16:23 •
Modeled a new groove mount using a single central bolt, and removed the extra plastic.
The bolt screws through the plastic washer and into a nut glued into a pocket inside the body. This allows me to tighten it down onto the end effector flange very effectively.
Printed. Fitted. Looks great.
The slant of plastic I removed makes it lighter and as a bonus I don't have to route all the wires through the body now.
I'm really happy with it.
The main light ring nylon diffuser is still too brittle and splits along layer lines instead of bending though. I recently rebuilt the print head for the M2 and heating capacity was increased, so this may be just the need to do some further calibration.
03/14/2018 at 20:09 •
Added the removable groove mount, and I don't like it.
I'm going to ditch using the three holes for mounting it and use a single central bolt/nut instead.
On a good note, the snap fit for the main light ring works well.
03/09/2018 at 01:21 •
I've been thinking.. sometimes a dangerous thing.
I think the groove mount should be removable/moveable.
Then people can pick which one of the three mounting positions that makes sense for them, and change out the groove mount for... whatever.
It will likely be a couple grams heavier as I'll need to reinforce each mounting position, but I'm ok with that.
Working on it.
03/07/2018 at 23:10 •
Moved the groove mount to the center of mass to see how it looks.
I like it.
- The center of the nozzle is now much closer to the center of the mount, so there will be a lot less torque generated when the nozzle impacts the bed/boards.
- I'm not as nervous about using the base of the nozzle mount to trigger a drag feeder.
- Centered mass is always a good thing.
I think I'm going to go with that. I'll save the endoscope camera version for others to pursue later.
I'll still be printing it standing up though as it will make the groove mount stronger and the bridges are shorter.
I still have some extra plastic to trim now to lighten things back up, and I want to add some ribs to reinforce the new position.
03/04/2018 at 23:45 •
I did a timelapse of the C light being on for 8 minutes.
It got warm, but alas, no melty goodness.
I'm still reprinting it in nylon.
EDIT: Tweaked the mount so its now a real snap fit and reprinted the bottom in nylon. It was too flimsy, so I went back and added a solid bottom layer. This helped greatly, so the reflector part is done.
The top diffuser portion in nylon was also too flexible which made it prone to separation at the vertical/horizontal wall intersection. I will need to reinforce it. It's only a single nozzle width right now so I'll just step it up to two nozzle widths at the bottom joint, adding to the outside
The diffuser part is nicely translucent though and so I can make it a layer or two thicker. I've also found varying the extrusion width for the lines composing those first few layers gives a nice prismatic effect. :)
03/03/2018 at 20:34 •
I didn't like how the down-facing strip lights were looking, so I modeled and printed a new ring light.
It printed in two pieces. The white part is a PLA box with a .2mm bottom which acts as a diffuser, and there is a black PLA bottom with a foil reflector built in. Each part weighed about 3 grams.
The warped corner here was from soldering it after sticking it in. Oops.
By turning the LED strips on their side again I was able to fit 7 segments in half the space of the 4 I had planned. This required the reflector on the bottom, which was made with some AL foil and spray glue.
I used these LED strips, the same ones I used for the part lighting. They are 5630 super bright LEDs rated at 22 lumens per LED so I should have a total flux of 462 lumens. Not all of that light is getting out, but it's still 3/4 as bright as aiming them straight down, and the diffuser makes the resulting lighting much more even.
Need to make a new mirror still, and then start on configuring OpenPnP.
03/02/2018 at 20:00 •
The new Pi camera showed up. I was able to spin the lens with just my fingernails. So.... yeah.
Everything fits. The pocket for the Pi camera was perfect, cable snap clearance was perfect, cable length was perfect, screw holes were perfect, etc. It sure helps to have detailed specifications.
The bearings got moved to fit in the mirror arm and turned out great. The full length hinge pin passes just under the camera and is a lot easier to remove than the two half pins. I also built two washers into the hinge now which serves to keep the bearings under preload with zero tuning required.
Everything is silky smooth and has zero play. I'm a happy camper.
I need to make a new first mirror as I changed the optics slightly to accommodate the wider Pi camera base. It's a little bigger.
Still have to design/print the strip light baffles. Think really thin, white PLA boxes. They will snap on the black tabs of the body, or so goes the theory.