Constructing Janus, by Dirk

Post your build logs here

Re: Constructing Janus, by Dirk

Postby BenJackson » Sun Jun 12, 2011 12:45 am

If anyone else wants to go this way I'd recommend picking up a RAMPS board: http://reprap.org/wiki/Arduino_Mega_Pololu_Shield it is essentially the same thing Dirk has constructed, but with a more powerful AVR and in a more compact package. You'd have to handle the laser control breakout yourself, but it seems quite straightforward.

Dirk, I bought 4 of the A4498's so I could spare you one if you need it. I'm still waiting for Bart's next batch of boards.
BenJackson
 
Posts: 522
Joined: Fri Apr 15, 2011 6:13 pm

Re: Constructing Janus, by Dirk

Postby naPS » Sun Jun 12, 2011 2:55 am

I printed a ton of the RAMPS v1.2 boards if you need a blank PCB also.
naPS
 
Posts: 202
Joined: Sat Apr 09, 2011 1:53 am

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sun Jun 12, 2011 6:12 pm

BenJackson wrote:Dirk, I bought 4 of the A4498's so I could spare you one if you need it. I'm still waiting for Bart's next batch of boards.


Wow, thanks. I couldn't find one anywhere. Pololu is completely out until mid july! This will help with getting the project set up. I'll be sure to order extras when they come back online. I'll PM you and we can make arrangements!
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sun Jun 12, 2011 6:28 pm

BenJackson wrote:If anyone else wants to go this way I'd recommend picking up a RAMPS board: http://reprap.org/wiki/Arduino_Mega_Pololu_Shield it is essentially the same thing Dirk has constructed, but with a more powerful AVR and in a more compact package. You'd have to handle the laser control breakout yourself, but it seems quite straightforward.


The Mega isn't really any faster than the UNO, but it does have more code space, ram, and IO. I'm also not familiar with the code that would run this thing, but I know GRBL is geared towards general machining and woudl take very little to get working for the laser. We also don't really need any more than 3 axis at this point. The RAMPS will provide more than we need and is more expensive in the end. However, using grbl and Barts driver board doesnt' allow a lot of room for expansion if desired. I am beginning work on customizing the GRBL code for use as a full featured laser controller. I am having issues with making it home correctly right now and getting all the settings right, but it is coming along as I work. I will publish the code as I make progress.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby BenJackson » Sun Jun 12, 2011 7:15 pm

One reason RAMPS used Mega was because Arduino is not as conservative about code space as straight C based firmware (that's a nice way of saying "bloated"). The other is the increased RAM space for things like buffering commands or talking to SD/MicroSD cards (which require buffers that are big by AVR standards). The extra pins allow for things like front panel LCD displays.
BenJackson
 
Posts: 522
Joined: Fri Apr 15, 2011 6:13 pm

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sun Jun 12, 2011 8:29 pm

BenJackson wrote:One reason RAMPS used Mega was because Arduino is not as conservative about code space as straight C based firmware (that's a nice way of saying "bloated"). The other is the increased RAM space for things like buffering commands or talking to SD/MicroSD cards (which require buffers that are big by AVR standards). The extra pins allow for things like front panel LCD displays.


I hear you... have you looked at the grbl code? Is there any reason why we couldn't run that on a MEGA? I understand that we may have to change the port/pin assignments, but that isn't a big deal. I'd like to add support for up to 5 axis solutions eventually. There isn't enough IO on the UNO to do this. Is there something similar to ArduinoUploader that I can use for the Mega to upload *.hex files? Do I need to compile the code into something other than hex files for the Mega? I'm not that familiar with programing arduino's outside of using sketches.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby BenJackson » Sun Jun 12, 2011 9:11 pm

I'm sure the grbl code will run on any AVR large enough including the Mega1280/2560 that RAMPS uses.

The Arduino code uses avrdude under the hood to program chips. Arduino should already know how to program the 1280 and 2560 if you select the right board. You can drive avrdude by hand to program any AVR with almost any programmer (dozens) ever made. You still use HEX files but they must be built for the specific AVR.

Outside of Arduino the build uses avr-gcc and avr-libc to build an ELF file and then avr-objcopy -O ihex to make the hex file. Then avrdude to program. All of those parts are used by (and installed with) Arduino so you can try using them by hand.
BenJackson
 
Posts: 522
Joined: Fri Apr 15, 2011 6:13 pm

Re: Constructing Janus, by Dirk

Postby dirktheeng » Mon Jun 13, 2011 2:05 am

BenJackson wrote:I'm sure the grbl code will run on any AVR large enough including the Mega1280/2560 that RAMPS uses.

The Arduino code uses avrdude under the hood to program chips. Arduino should already know how to program the 1280 and 2560 if you select the right board. You can drive avrdude by hand to program any AVR with almost any programmer (dozens) ever made. You still use HEX files but they must be built for the specific AVR.

Outside of Arduino the build uses avr-gcc and avr-libc to build an ELF file and then avr-objcopy -O ihex to make the hex file. Then avrdude to program. All of those parts are used by (and installed with) Arduino so you can try using them by hand.


I'm going to hunt around for some tools to do this first. I don't want to waste the time figuing out what others figured out b/4. I have been using WinAVR to compile the hex. I think I just have to change the make file to make it compile the hex for the mega (though I don't know what to change at this point). I will also have to find the right avrdude command string. I'll worry about this later when I can't make the UNO work for my purposes anymore. Right now it seems to be working fine.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Problems with z axis gearing

Postby dirktheeng » Tue Jun 14, 2011 11:23 pm

I am having issues with getting the z axis moving at any appreciable speed at the current gearing ratio. Righ now, the gearing ratio is 2.4:1 with the large toothed gear installed on the threaded rod. This is the way that the kit came.

Lets assume that an acceptable z axis rapid speed is 1000mm/min. The treaded rods are 20 threads per inch or 0.787 threads/mm. The rods are single start, so one revolution will move the z axis 0.787 mm. This means that the rods have to turn at 787.4 rpm. at a 2.4:1 gearing ratio, the motor has to spin at 1890 rpm. At 1/2 stepping (800 stepps/rev), this is 25.2khz step rate. While this is fine for a step rate, the torque that the motor can produce is fairly limited at that kind of speed. The acceleration for that axis has to be very low to keep from slipping. I think I was only able to achieve something like 50mm/s/s, while the other axis can go at 1500-2000mm/s/s. I'm thinking it would be better to switch the gears around. That way the motor would only have to travel at 328 rpm, and the step rate would be about 4.4khz. I know I would be multiplying the torque on the motor by a factor of 5.76, but I am willing to bet that I would be at least that much higher on the torque curve for the motor.

I can't find a torque curve for the motors we have. We have Smart Automation p/n: 42BYGH48950-89. I tried googling for the information, but I couldn't find a matching part number, nor matching specs. I found several in the series 42bygh, but that doesn't help much unless I can tell which one we have... the numbers folowing bygh didn't match. Does anybody have a torque curve for these motors or know which one on the web is right?

Based on past experience, the relatively low torque on the motor to move the z axis, and the curves I found for other 42bygh motors, I would say that the available torque at that rpm is nearly zero. However, most of the similarly sized motors have appreciable torque at 4.4khz half stepped. I am willing to bet the system would run much better by flipping the pulleys around (I know I would have to buy new ones, but I think it is necessary). Even at the higher gear ratio (1:2.4) the resolution in the z axis is still 0.004mm, which is way more than we need. What do you think?
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby cpdude » Wed Jun 15, 2011 2:31 am

Dirk,

To comments...1) I think that 1000mm/min is probably way too fast for the Z axis. Just my 2 cents. 2) I'm not sure which motors you are using but most steppers--including the stock ones offered by Bart--are 200 steps/rev. I think that would make 1/2 micro-stepping equal 400 steps/rev.

Brian
cpdude
 
Posts: 102
Joined: Fri Dec 24, 2010 5:21 pm
Location: Cameron Park, CA

PreviousNext

Return to Build Logs

Who is online

Users browsing this forum: No registered users and 3 guests