Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: btboone on September 03, 2012, 09:36:24 PM
-
Hi, I'm a new poster and have a Tormach 1100 that I'm retrofitting to become a fiber laser cutter. I make titanium rings for a living and was looking for a powerful cutting laser. I found that I wasn't using my Tormach much, as I also have a full size machining center and CNC lathe. The cost of a finished cutting laser was well past the mid 6 digits, so I decided to attempt it myself. I bought the fiber laser engine, and I'll be mounting a large laser head to the Z axis and will be doing a full sheet metal enclosure to block stray infrared.
The main challenge as I see it will be to get the capacitive height sensing circuit in the laser head to work. The Tormach uses steppers, and the height sensor will output a DC voltage corresponding to how far away from a grounded sheet, eg. 1v for 1mm away, 2V for 2mm away, etc. It sounds to me that it might be a simple thing to get that to work with a servo, but not necessarily so with a stepper. My question is would it be easier to have my Z axis convert from stepper that's spec'ed with the machine to servo and use the output of the sensor as the driver for Z? I know nothing about what it would take to do that. Maybe there's another clever way to do it that won't take much reworking or reconfiguring the machine. I am new to the electrical workings of the machines, although I'm pretty good on the g-code end. I can certainly use some insight on the best way to go on that. I can get by with programming Z for flat stuff short term, but I also plan on using it for titanium rings on a rotary axis that change thickness as they go around. Is there an add on product or macro that might accomplish this?
Another smaller issue is that my rings are pretty close to 1" diameter, so it makes sense to always draw my rotary stuff as if it were flat and simply scale my artwork until it's exactly 3.6" tall. Is there a simple way to have the rotary axis simply think it is a Y axis so that it is based on 3.6" and no feedrate calculations are necessary? By scaling the artwork itself to exactly 3.6", it will always perfectly line up, as any diameter calculation errors will show up in a big way when the kerf is only .002". I run a tabletop MicroKinetics machine this way with good results. It doesn't even know it has a rotary axis. I would like to set up my Tormach to do that. It has an A axis, whereas my tabletop mill has no powered Y axis, so in the Tormach case, I would actually need to keep a working Y axis as well as an A axis that would look the same in CAD. The difference would be in the g-code only. The Tormach setup doesn't look to allow me flexibility in setting up axes like this, but maybe someone here will have some insight to that as well.
Another thing I would like to tweak is the rotary axis speed. I replaced the large 6" Tormach rotary axis with a small 3" Sherline rotary. It will have more clearance for the laser head, less friction, and will have no cutting forces on it. Right now it takes over a minute to do a full revolution. Ideally I'd like to speed that up considerably and run it as it were a 3.6" linear axis. I don't see any way to do that from within the Tormach software. The laser itself will be capable of cutting about 40"/minute in that thickness.
Thanks for any input.
-
It looks possible to design my parts for the rotary axis in a 3.6" length and simply always add a scale factor of G50 A100. That's not too bad to do if setting up the rotary to behave like Y is not possible. This would eliminate the possibility of using arcs and circles, but I could live with that if necessary.
-
Hi Btboone,
On the Mach3 General Config page you can change your rotary axis setup from angular to linear by un-checking the appropriate box. In addition, by setting Rotation Diameter on the Settings page you can then adjust your Steps Per Unit in Motor Tuning to suit.
Tweakie.
-
I'll look for that, but I don't think the Tormach allows me to see such pages. Is there a way to load only specific pages of Mach without affecting any of the settings or screens that they set up?
-
A bit of lateral thinking here....
How about installing a second hard drive, copying the existing drive to the new drive, then you could boot to the new drive and change screens, settings etc without affecting or changing any settings within your existing setup. ??
Tweakie.
-
That very well might be worth doing. As I understand it (and I probably don't), all the critical stuff is stored only on a few files that could then be transferred back. What are those files, and could I simply make a backup of those and then swap those once I have found the correct parameters for my application?
I love your laser project. Very cool stuff and well documented.
-
The machine settings are stored within the 'Mach3Mill.xml' file (or whatever profile you are using .xml). The screen specific stuff is stored in '1024.set' (or whatever screen you are using) and the Macros folder contains any macro's you use. Also don't forget to save your license file (Mach1Lic.dat). That is the basic stuff but be cautious because you may have more Tormach specific files stored elsewhere so don't delete anything or make any changes you cannot reverse. (This is really the reason I suggested the additional HD you could install the latest version of Mach3, copy over the files mentioned above, then make any changes you choose).
Thanks for the complements relating to the laser project, I am hopeful more will follow.
I believe that one day everybody with a CNC will have a laser attached (probably fibre rather than CO2) but it is a bit slow catching on at the moment. ;D
Tweakie.
-
The laser came today!! It's huge!
-
;D ;D ;D
Tweakie.
-
Here it is. It has measured peak power of 4680 watts. Still waithing on the laser head and sheet metal for the enclosure.
-
The laser head came in and I'm getting close. I made a manifold to be able to deliver high or low pressure nitrogen to the head. Sheet metal enclosure should be complete soon. I have run the red pointer laser and am confirming the various interlocks and stuff. Official setup will be Wednesday.
I realized that I can draw my parts in CAD the 3.6" in Y with a normal X to represent A and X, then output that code and simply insert a G51A100 scale command to make that work to the 360 degrees that the rotary is set to. That gets back to eliminating error with a diameter measurement. Does anyone have an idea if a compound rotary move with A and X will have the same feedrate as an A or X by itself?
-
Btboone-
That looks like a sweet setup! I can't wait to see it cut. What thickness will it cut?
Steve
-
In testing, we went through 14mm of titanium at 3000 watts. I've got 4500 to play with. The trick is the fine kerf of either 50 or 100 micron depending on the lens used. That's still less than a hair. That will be the limiting factor. I need to get through about 7mm of titanium on a ring on the stone holding side. I'd be very happy with a reasonable cut that would take about 1/50th the time of it being milled. Finishing needs to be done either way. It will allow a lot of design freedom that I can't even contemplate making stuff like now.
-
What would be the best way to do the variable timed dwell for the pierce? Ideally, it would automatically happen on every M3 without a bunch of extra code. It would be good to be able to either set it in a table or once in the code, but not every time the laser is turned on and off. Anyone have insight into that?
-
I was in a bit of a panic when I realized that the Tormach automatically adds about an 8 second dwell to get the spindle up to speed and about 3 seconds to stop it. I need to run the "spindle", which has actually been removed, at full speed in order to get the 5 volt output signal that it takes to fire the laser. The laser power settings are adjusted in separate laser software and I just need to fire it with a 5 to 24 volt signal. I found I can get an M3 controlled 5 volts, but the spindle needs to be at full speed otherwise I get a linear fraction of that 5 volts. It takes a real motor a long time to get up to that speed. They allow a full 8 seconds, and I couldn't find a way to remove it. The laser would burn halfway to China with an 8 second dwell. I looked for ways to use the m11p1 command as seen in this forum, but those didn't look to be compatible with the Tormach motion board, which has no extra input capacity as far as I'm aware. The 5 volt signal was instant with an M3 command, but it was about 8 seconds before table movement could commence. I was relieved to find that I could set the control to look at any dwell in milliseconds, so the 8 second dwell became .008 second instead. I can then manually add dwell time for the laser pierce in the code. I am still looking for a way to have the dwell stay the same throughout the code. Maybe I could use a variable defined at the top of the program and that gets repeated throughout the program any time the m3 command is activated? Not too elegant, but it might work.
Everything is set up including the enclosure, so I'm just waiting on a tech to come do the official install.
-
Btboone-
Can you just add a relay to control the laser ? I have three relays on my flame cutter, and I can control the time for each one. Just an idea.
Steve
-
The big thing is that I'm using the Tormach setup with its control board and all its specifics. I certainly don't want to mess anything up that they have working properly. If I was doing it from scratch, I'd have little to lose by trying different things in the hardware end.
I did find out that there's a macro that gets looked at when the m3 command is activated. This must be where the elusive dwell resides. I'll check on that tomorrow. Maybe there's a way to specify a dwell at the program top and that variable could be passed through to the m3 macro. I dont know all the rules on that type of thing in Mach 3. I'm used to an Anilam 1400, which is PC based and more powerful and flexible than the Mach 3 seems to be, although the Mach has a lot of potential for tweaking in.
-
I ordered a USB controlled relay bank that has 4 independant relays. You could just add this to your pc and add a few lines to your macro for m03.
Steve
-
I'm up and running in continuous wave mode, which means only up to 450 watts. In order to get to pulsed mode, it seems that I need to feed the laser a signal with a frequency in the range of 160 to 200 hertz and a pulse width variable from 0.1ms to around 2ms or so. Is that something easily doable with the Mach 3 control?
-
Hi BT
May I ask who's laser and were you bought it, I'm very interested in doing one myself.
Thank you,
Ted
-
The laser itself is IPG. It's not inexpensive though.
-
Making progress with the laser. I'm finding that I do need to specify the A axis as a rotary axis otherwise the discrepency in counts per revolution versus the normal X axis is too much variation. I ran a part and the part was cutting off center. I finally realized that I was missing steps because the feedrate that worked for the rotary was way too fast for the X axis to catch up if the line only had an X move in it. I reset the machine to look at A as a rotary axis and set the proper diameter and the axis feedrated worked normally. I do program my parts to 3.6" in Y then in a word processor, convert the Y's to A's and put a g51 a100 to scale it to full rotary. That all works well now.
Here's a question; can the diameter of a part be given in the G-code of a part instead of input on the settings page? I know for a fact that I will sometimes forget to set a diameter and get a ruined part. Even having an average diameter for rings, like .80" will be close enough to get the feedrates to work well.
-
Create this Macro:
SetOEMDRO(825, Param1 ())
end
Name it, (for example "M666.m1s")
Adding this line:
M666 P0.80
in the G-code will run the macro and enter 0.80 in the A diameter DRO.
Al
-
Exactly what I needed. Thanks Al.
-
:)