[Dan]

So there’s three major parts: hardware, firmware, and software. Hardware could be done like this:
https://www.instructables.com/id/Make-a-CNC-Hot-Wire-Foam-Cutter-from-parts-availab/
With two 2D gantries on opposite vertical faces of a box. A hot wire runs between the two gantry heads.

I would probably use Makelangelo-firmware, add a few commands to heat the wire. one gantry would use X and Y values for position. The other would use U and V. This way the wire could make angled cuts.

The real challenge is the software. Personally I’d start with 2D cuts where U=X and V=Y all the time, because that would be easy to calculate. I fail to imagine how to cut, say, a plane wing. There’s probably a clever way to take a 3d model of a wing and extrapolate the “shadow” on each box face, such that following the shadow edge cuts your wing profile.

[Dan]

I’ve been out of town for ten days. Before I left I placed an order for bearings and motors to go with my latest design. I even printed some parts. Then the motors arrived and I discovered that there were critical differences between my understanding and reality. I thought some holes in the motors could be used to attach to the bones. Those holes are already in use for the gearbox. In at least one case the motor body shape did not match my model, and the design has to be radically changed to compensate. So it’s back to the drawing board!

View this post on Instagram

Was looking great until the parts arrived and I discovered that mounting holes on the data sheet were occupied my screws in the real thing. #redesign #startover #socloseandyet #robots #openhardware

A post shared by Marginally Clever Robots (@imakerobots) on Oct 3, 2017 at 10:49am PDT

View this post on Instagram

Parts have arrived for my new #robot arm. #3dPrinting has resumed.

A post shared by Marginally Clever Robots (@imakerobots) on Oct 1, 2017 at 2:14pm PDT

View this post on Instagram

Robot arm parts ordered, reckless printing begins! #3dprinting #3d #factorio #inserter #prusa #pla #robot #robots

A post shared by Marginally Clever Robots (@imakerobots) on Sep 21, 2017 at 5:13pm PDT

[Dan]

@prenic – interesting. if you can provide me with a mathematical model, I will convert it to code.

[Dan]

If you need it, consider making the change and sending me a pull request. I’ll merge your change (after testing) and then it’ll be there for everyone.

[Dan]

@Desterline – on the Makelangelo 3 and 5 draw a line between the center of both motors. From the middle of that line, go down 217mm (8.54330709 inches). That is the spot where the pen should start.

It used to be in the center of the paper, which was crazy annoying to use on the 3. With the 5 it’s on the edge of an A2 sheet, so well placed.

It’s not friendly for M3s that are very very big – you might not even be able to reach the home position without a ladder. I’ll see about adding a “set your own home position” option for the next release.

[Dan]

BTW:

M101 let’s you adjust the size of the machine by setting the top, bottom, left, and right from the origin. It works the same as the Makelangelo.

M101 TH B-H L-W RW

Where W is width/2 and H is height/2.
Width and height should be measured from the edge of the motor mounts where they touch the edge of the board.

[Dan]

Oh no, the drivers! :O

I’ve retested for all the machines styles and for RAMPs. Should be better now. Sorry for not testing every option more thoroughly. I got hung up on a polargraph issue and when I was done I thought “that’s it! Ship it!” Never make a major commit on a friday afternoon, they warned me…

[Dan]

I set up my zarplotter with a RAMPs board this morning and ran some tests. I can confirm that the current firmware appears to have a bug that makes one motor turn only in one direction. Thank you, @kukomiguel, for the report.

I have updated the makelangelo-firmware to clean up the confusion that allowed D0 to work but not G0. now they are the same so that if one doesn’t work, they are both broken.

I’ve also removed all the firmware inversions. If you want a motor to run the other way, change the physical wiring. This will run faster and use less code that could break.

Please let me know if that helps.

[Dan]

Have you tried to invert a motor in software (M101) and then jogging it (D1)? I want to check if it reverses direction.

[Dan]

Can you send me your configuration files and the firmware config to check all your parameters like board size, etc..? I’m not sure how to define the limits.

Everything I have is checked into github:

https://github.com/MarginallyClever/Makelangelo-firmware/tree/dev
https://github.com/MarginallyClever/Makelangelo-software/tree/dev

[Dan]

Did my previous suggestion about M102 and M101 help? I don’t remember seeing a reply.

[Dan]

it sounds like your motors are inverted. What does “M102” say? It should reply

I1 J-1 U1 V-1

and if it doesn’t, try

M101 I1 J-1 U1 V-1

This is a way to reverse the direction of motors in software. Another method is to physically flip the wire where the motor attaches to the controller, which should only be done with power disconnected.

[Dan]

FYI, 1.27323954 is (20 teeth * 2mm per tooth) / pi = circumference of pulley.

[Dan]

Try

D1 L1.27323954

if the pulley diameter is set to zero, the inverse kinematics fail in interesting ways. I’ve seen on the zarplotter that (from home position) X+anything does nothing, and Y+anything moves only a few of the motors, as you described.

So… probably something with failing to initialize the pulley diameter. I’m strongly in favor of ripping out this pulley diameter changing code, since I only use one size of pulley ANYWAYS.

[Dan]

I have acquired a RAMPs board and I’ve installed it. testing to begin shortly.

[Dan]

The only thing that changes between RAMPS and RUMBA are the pin assignments. If your RAMPS isn’t working right, I’d try changing the driver first, then treat the RAMPS board itself as very suspicious.

[Dan]

I don’t know. Can you? You’ll have to be more specific.

[Dan]

I purchased

11HY3406-9SK842 http://www.robotdigg.com/product/798
Adapter http://www.robotdigg.com/product/799
Nozzle http://www.robotdigg.com/product/800
Connector are both fine KSH04 is for 4mm OD tube and KSH06 is for 6mm OD tube
http://www.robotdigg.com/product/606

and a solenoid valve to release air pressure. Maybe https://www.robotdigg.com/product/1069/TG-series-Solenoid-Valve-TG23-06,-TG22-08 will work?

[Dan]

And the prize pool just went up by one beer! …I don’t have that image yet.

[Dan]

It’s easier to follow your issue if you keep the conversation to a single thread.

Maybe you’re putting power backwards into the shield. get a multimeter and check. also read the documentation for your motor shield and be sure you have the jumper setting correct. Old Makelangelos using the AMS1 shield required 12v2a power. I don’t remember the jumper state.

[Dan]

This conversation has been moved to product support.

If everything worked until you plugged in the power supply, I’d take a hard look at your power supply.

There is a good chance you have no valid serial connection to the board. http://forum.arduino.cc/index.php?topic=83079.0

> avrdude: ser_open(): can’t open device “\\.\COM4”: Access is denied.

Well right here it says you try to connect on com4 and com4 isn’t available.

Code compiling ok can happen with or without hardware. uploading can be affected by hardware. Try disconnecting everything from the board except the USB cable, finding the right com port, and uploading. Then disconnect usb power and put things back on one at a time. Test between each new thing by opening the serial window to talk to the board. You might find things stop working when you plug in a specific part.

[Dan]

View this post on Instagram

Version 2, somewhere between a hot glue gun and an AR-15 receiver. Every part can be printed separately for lower cost customizing and prototyping. #glue #gluegun #solder #solderiron #ar15 #receiver #notagun #3dprinting #functional #functionalprint #tool

A post shared by Marginally Clever Robots (@imakerobots) on Aug 24, 2017 at 6:32pm PDT

[Dan]

Found this, which looks simpler than my ratchet system: http://3dtrek.com/long-reach-hot-melt-glue-gun/

[Dan]

@bulp –

When I try and run the demos I just get a load of error messages such as libEGL warning: GLX/DRI2 is not supported/failed to authenticate etc

The chances are this is because ‘something’ (such as gedit) has installed mesa which added its own versions of libEGL and libGLESv2. If you run:

$ sudo find / -name libEGL*
$ sudo find / -name libGLESv2*

on the Raspberry Pi you should just get /opt/vc/lib/libEGL.so and /opt/vc/lib/libGLESv2.so if other ones turn up i.e. /usr/lib/arm-linux-gnueabihf/libEGL.so.1 you could try creating symbolic links for them all like this:

$ sudo ln -fs /opt/vc/lib/libEGL.so /usr/lib/arm-linux-gnueabihf/libEGL.so
$ sudo ln -fs /opt/vc/lib/libEGL.so /usr/lib/arm-linux-gnueabihf/libEGL.so.1
$ sudo ln -fs /opt/vc/lib/libGLESv2.so /usr/lib/arm-linux-gnueabihf/libGLESv2.so
$ sudo ln -fs /opt/vc/lib/libGLESv2.so /usr/lib/arm-linux-gnueabihf/libGLESv2.so.2

NB on other systems such as Debian, ubuntu etc the “good” files will not be in this location, they will be somewhere like /usr/lib/x86_64-linux-gnu/mesa-egl/ so you have to alter the commands accordingly Use the actual paths as listed by find. After creating new symbolic links you will need to run:

$ sudo ldconfig

This issue is being looked into by the maintainers of Raspbian so, hopefully, it will be fixed in later releases.

— https://pi3d.github.io/html/FAQ.html answer no 7

[Dan]

As for a command line only version… interesting idea. I’ve heard it once before. No idea how I’d approach that. Separating the GUI from the command line might mean I could automate testing for the subsystems…

[Author]

Posts