A while back I decided for safety reasons not to put laser cutting capability onto my Art Bot. So, the laser and heatsink bundle arrived today. I’m going to be messing with the voltage and current using a dummy load (also bought that, rather than scraping it together) and nothing could possibly go wrong.
I’ve already written the software.
accu.co.uk is a very dangerous website for me. I needed to buy screws and nuts for a client project.
Yep, my own collection got some new sizes. Judge me if you want. My M8x20 says I don’t need to care.
Well, those lead screws didn’t need to be that long anyway.
Wait, they did? Oh. Well, let’s cut about 5mm out of this solid block of aluminium and add in some washers … there…
*lead screw falls out*
And use the pulley grub screws to hold everything together against the backlash. It’ll be fiiiiiiine.
If necessary, I can put the solid block of aluminium on the other side of the metal it’s attached to right now, and that’ll tighten things up, although I’d need to take some more length off the lead screws. That could never go wrong.
I need a small coding project, that will let me practice more than just simple procedural problems, but is not so massively complicated I die of it. I also, let’s be clear here, need sleep, food, and to remember what the inside of a bathtub looks like. Nevertheless, I’m looking for things in about that order.
I don’t know enough to know what an ideal project would be, which is making this hard.
I recently decided I needed to be able to convert raster images to vectors in my software so I could drive Itchy the Artbot. I looked up OpenCV as a way of comprehending computer vision, because it was the only thing I had heard of.
Apparently ‘vector’ and ‘std::vector’ are different things.
This was an error of ignorance that I could only put right once I’d learned enough C++ to feel I could go back to the OpenCV help files. So, I spent a month getting to the point where I could find out I shouldn’t waste time on this. Fucksake.
Sometimes, I find I no longer know the original ‘I need to wash my car’ incident or problem for which I am now shaving yaks. Here, I believe it is ‘I want to improve my drawing machine’.
- So I have to be able to interact with the visuals, programatically; OpenCV looks good. But OpenCV is hard to use, because it’s powerful.
- So, I need to be able to write software.
- And if I go at a project this size I will just bounce off, so I need a smaller project first.
- Project Euler exists, and the maths is relatively easy within the scope of what I can learn. Converting the maths to code is hard, though.
- I need to learn how to handle big numbers, and #include BigInt.h is cheating.
- So here I am, learning how to write bigint.h
- Which means as this pdf isn’t easily readable that I’ll have to print it out
- Hence, I’m trying to make CUPS work so I can print
There is nothing that can be said in favour of this plan, except that it will be a really well shaved yak when I leave it to go back up the list of items. Fortunately, attempting to make my print server actually print is always enough to pull me up and make me reconsider my life.
- So, here I am on the Epson website, buying ink…
I tore apart the head on Itchy today, to start the rebuild. I want to have the plate that’s attached to the gantry, and then the plate that arbitrary heads attach to. Between those I’ll have the linear bearings, and if I stand the bearings off a little from the gantry plate, I’ll have room to put the lead screw down between them. In the course of this, I became a lot better at putting 0.5mm balls back into the linear bearing. I also got really good at cutting the same part several times and forgetting the changes I had mad and failing to update files properly. Then I remembered to eat.
Then I realised it was 9pm and went home. Fuckit, life’s too short for sleep anyhow.
Well, the big servos didn’t work, but nothing melted. The small ones do work. And, even better, I made someone tell me my electronics made god cry. I was using my LED array instead of an oscilloscope – it’s quick, and I only wanted to know where a signal was, not what it was. So hah!
Related: I really need to set up a proper connection to the oscilloscope on my main laptop.
I’ve got a few choices to make with Itchy right now. Do I want my own machine for making art, or do I want to make a set of machines that others can use? If I get this ready for kickstarting, it’s one path, if I make it into a better machine for me, it’s another. I really want to make the better machine, but at some point that will probably run out of money as a project. Things that I have to do for both are relatively limited in crossover, but the big one is a redesign of the head so I can put different tools on there. The pen holder and the lino cutter are two different set-ups, and I’ll want a servo for the brush pens, and so on. For me, I want to make the Y axis and the gantry far more rigid. If I make it for lots of people, I’d want to make the frame in MDF with steel rods. I guess I could convince myself I’m doing that to find out how rigid it is…
So, I’m starting with asking tech companies and art councils about funding. If that comes down to nothing, and I’m not getting enough commissions to let me do my own thing on it, I’ll streamline it for kickstarting in my spare time.
It just occurred to me that if I had an item of known width I could rest my pen on that while adjusting height, and I wouldn’t have to zero it or have scrap paper. I could change pens cleanly in the middle of the work, so I could then do multi-coloured works more easily.