Constructing Janus, by Dirk

Post your build logs here

Autofocus fixture

Postby dirktheeng » Sun Oct 02, 2011 2:40 am

I finished the design and preliminary instal of the autofocus fixture. I designed it in inventor and it features micro adjustability to find and set the optimal distance of the work surface from the focus lense. It is removable and hangs on to the corner of the laser cart with a pair of magnets. I was able to make it work with stuff I had around the house, except for the limit switch. I got that at radio shack.

Here's a view of the inventor model:

AutoFocus Assembly.jpg
Inventor Model


The top is made of 2, 1/8" plies with round holes that fit the magnets. The bottom hole is sligthly over sized to allow the magnet to slip on and off. The top magnet is glued into the top ply and the bottom is glued to the laser cart. The botom 3 piecws have hexagons cut out to accept threaded nuts which are glued in. The holes on the bottom are sized to accept the screws that hold the switch.

Hre a re a couple views of the assembled fixture:

DSCN4209.JPG
View 1


DSCN4210.JPG
View 2


The screws are a bit long for this, but they just fit. I could cut them down, but I'm a bit lazy. It works as is. The vertical cap screw is used to micro-adjust the lower part so the exact height can be found. The 2 screws on the sides lock it in place.

Here's the magnet glued in place:

DSCN4211.JPG
Magnet in place


Here's the fixture installed on the cart:

DSCN4212.JPG
Fixture In place on Cart


The good thing about the magnet is that I can stick the thing in a place that is out of the way but easy to get at. Here's a pic of the placement I chose.

DSCN4214.JPG
Put Away


I have yet to attach the wires, but that is the easy part.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Video Demonstration of Laser Auto focus

Postby dirktheeng » Sun Oct 02, 2011 9:19 pm

Ok, the auto focus jig is installed, wired, and functional in Mach3. I made a video to share. It is very easy and quick to use. I hope others will find this information helpful. It is much easier to do than a ramp test, even with the code I set up earlier. Everything tucks back out of the way easily and the wires don't get in the way. It works very well. I hope you enjoy the video.



(sorry for the bad shot here... youtube pics a random frame from the movie as a title frame and it picked this one. I don't know how to change that). The movie doesn't look that bad.

Let me know what you think.

I've attached the inventor files along with the DXF and the *.tap file which is the gcode from Heeks cnc. It has the 8 pieces you need to cut out to make this thing.
Attachments
AutoFocusSwitch.zip
Inventor Model and Cut Files
(1.73 MiB) Downloaded 1207 times
Last edited by dirktheeng on Mon Oct 03, 2011 12:21 am, edited 1 time in total.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Another good reason to have a magnet on the carraige

Postby dirktheeng » Mon Oct 03, 2011 12:19 am

All,

I was messing about with aligning the table height with the focus jig I made ( checking the height of the table corners all around to make sure it is very planar and flat) when I bumped the gantry a bit hard... hard enough that I thought I should check the alignment. I had been using some rough squares I cut out and a glue gun to stick them down. The thought of the mess and time that it takes for the gun to heat up was a turn off, so I thought about something I could use with the magnet that would be quick and easy. I happened to see one of the snap together boxes I made and thought that it would be big enough to be useful. I made these boxes from a dxf file and kept adjusting my tool diameter in Heeks CNC to get the the pieces to just snap together and hold well enough by friction. I set the tool to 0.12mm and it seems to work well enough.

Anyhow, the boxes come apart with a little effort and snap together easy enough, so I threw a magnet in side and set the box in place. Even with a fairly high acceleration, the ring magnets I have were more than strong enough to keep the box from moving about relative to the carraige.

Here's a couple pics:

DSCN4217.JPG
Base Stuck to carraige


DSCN4218.JPG
Box sitting in place


Each face has enough room on it to get about 8 target shots on it and the boxes can be flipped inside out too so you can get a total of 96 shots on one box. Plus the ease of quickly puting a box in place as a target means that I am prone to do alignments more often (as is reccomended) which will help keep my cuts consistant and trouble free.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby dirktheeng » Wed Oct 19, 2011 9:47 pm

Well... it's been a little while since I have had time to work on anything fun. I took a new position at work and it's been a little crazy learning the ropes, but it's good. We also have been traveling almost every weekend for the last month so I haven't had much time. Anyhow, I have a new goal: make some nice xmas gifts with the laser. I decided that I wanted to make some "scroll saw" art with my laser. My dad is a big scroll saw enthusiast and he has made me several very nice scenes. I want to start to do that with the laser for fun. I also want to do marquetry, which is very similar. this involves making very precise cuts on the lines. Even with the M102 command in mach that is supposed to start the laser with the motion of the feed move, it still leaves a little burn hole at the start and end of the mark. What I want to impliment is a "pulses per inch" or PPI style control. To do this well, I have to be able to measure the exact position of the laser head at any given time (or rather the change in position). The DRO from Mach doesn't really gain me anything because it is not the position that the laser head is in. The shure fire way to get that is to use a counter to keep track of the pulses from the smooth stepper (or any other input source) and have a hardware/PLC based system that will turn the laser on and off as needed. Stay tuned, I have some parts picked out and will be making the board layout as soon as I can.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby BenJackson » Wed Oct 19, 2011 10:33 pm

Dirk,

You probably already saw this but I did implement PPI in EMC2 with a small custom HAL component. Come to the dark side. ;-)
BenJackson
 
Posts: 522
Joined: Fri Apr 15, 2011 6:13 pm

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sat Oct 22, 2011 2:48 pm

BenJackson wrote:Dirk,

You probably already saw this but I did implement PPI in EMC2 with a small custom HAL component. Come to the dark side. ;-)



Yea man... I saw what you did and it's pretty sweet! I am modeling some of this after what you did already. However, I'm after a universal solution that will work with any system. Plus, I have a lot invested into Mach3 so I don't really want to switch it up now. Basically, I want to create a little board that will plut into the stepper driver board and "listen" to the signals being passed and talk to the logic via I2C. I have the chips picked out and am making the circuit diagram now. It will be less than $10 worth of IC's and whatever it costs to make a board. I've got a buddy with a small cnc router and he's going to help me do some isolation routing. I will also need help reviewing the design b/4 I make the board. Would you be willing to look at it for me?
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby BenJackson » Sat Oct 22, 2011 6:33 pm

dirktheeng wrote: I will also need help reviewing the design b/4 I make the board. Would you be willing to look at it for me?

Sure. That sounds very similar to what I thought I would do before I actually started working with EMC2. For engraving I was going to add a serial port so that the axis motion could be used to clock out engraving data loaded via serial.
BenJackson
 
Posts: 522
Joined: Fri Apr 15, 2011 6:13 pm

Wiring Diagram for Axes pulse counter

Postby dirktheeng » Sat Oct 22, 2011 8:05 pm

All,

I finally finsished the first draft of the eagle cad wiring diagram. I have never done this before, though it was fairly intuitive from my EE class back 10+ years ago. I am no EE, but this should work. If there are people out there who have experice with this stuff and want to point out problems to fix, I would appreciate that. I have not done the board layout yet as I am fairly sure that this will need revision.

Basically, this board will plug into the pololu driver board and will have a direct pass through for the db25 connector signals. I take taps off of those signals for the x and y direction and pulse signals as well as the limit switches. The pulse signals go to the clock of a bank of 4 bit up/down binary counters connected in such a way that I can read 17 bits for the x direction and 16 in the y. The outputs go to a bank of 16 bit I2C digitial IO expanders. Because the x axis needs 17 bits, I needed 2 for that, which allowed me to get 15 individually addressable digital I/O pins. The I2C chips are on the same 2 wire communication lines and have dip switches that I can set the address and hardware reset state. I have a bank of screw terminals that will allow me to get to the extra IO, power, I2C bus lines, and ground. This board can be used to read and track the instantanious position of the laser head down to ~0.0005". This information can be used by an arduino to turn the laser on and off very quickly. I can control all of this through the modbus address I have already instituted.

LaserPulseSchematic.pdf
Wiring Diagram
(22.43 KiB) Downloaded 1280 times
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby BenJackson » Sat Oct 22, 2011 11:05 pm

This is overkill if you have an Arduino. It can easily handle all of the counting you need without any accessory hardware.
BenJackson
 
Posts: 522
Joined: Fri Apr 15, 2011 6:13 pm

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sun Oct 23, 2011 11:50 am

Ben,

I originally got really excited about maybe using an atmega 328 on the board to do the counting, but the problem is that there isn't a way to reliably count pulses at frequencies above 20kHz and still expect to do other things with the chip. Even ataching interupts to a pin is dicy if you have other things going on. The hardware solution is the way to go, I think, because it will never, ever miss a pulse and it is an instant update. Especially given that i am working on a way to command Mach to generate a roster pattern through the smooth stepper at much higher frequencies (200-300 khz).
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

PreviousNext

Return to Build Logs

Who is online

Users browsing this forum: No registered users and 25 guests

cron