Posting doesn't seem to be too heavy in this forum, so hopefully this post will stir up some of the lurkers. I'm hoping to gain the interest of those "in the know" so they can get this project off of the ground and headed in the correct direction. Posts will be sporadic from me as this project is intended to be made over a number of months... I have waaaay too many things on my plate at the moment, but this little toy would be quite useful to have in my stable of tools, so I'll give it a bit of attention from time to time.

The grand idea (as fantasy-like as some portions may turn out to be):
1) Large cutting area -- 3'x3' is the desired size and should be plenty big enough to handle most projects I could think of while not being so gargantuan I can't fit anything else into the room.
2) Lots of power -- I figure a 100W laser should more than cover any materials I would need to slice and dice. Depending upon the focal length I set, I want to have no issues with getting at least 500 W/mm^2 out of this.
3) Stable bed -- It should have a vacuum bed for material stability. This doesn't need to be fancy, just an array of small holes. I believe I've figured out a way to have a mesh above the bed so the laser doesn't cut the bed itself while still providing a reasonable level of vacuum.
4) Accurate and fast gantry -- It should be reasonably fast so I'm not waiting two hours for a 4" square to finish cutting, but precision should still remain decent. I plan on using preloaded ball screws rather than Acme threads because of this. A thousandth or two of accuracy would keep me happy.
5) Inexpensive/light gantry -- Although welded steel would be structurally solid and less expensive, I think aluminum is the way I'm going to head. With the proper bracket setup, it should be just as stable as steel for these purposes and also allow me to disassemble the system for moves.
6) Quality optics -- I don't plan on ripping open a copying machine. I've found all I need for less than $400, including gold-coated FS mirrors, stackable lens tubes, etc. I can change or adjust any item through the use of Kinematic mounts (much cheaper than I thought they would be!).
7) Z-axis -- I may not include this in my first build, but it's definitely a desire for the future.
8) Lathe mount -- Same as the Z-axis... not in the first build, but desirable for the future.
9) Quality software -- I'll probably go with Mach 3 on this one... it has proven itself to be a fairly versatile piece of code time and again, and at $160 the price is right.
10) Quality motors -- I haven't spent much time finding exactly what I want, but high-resolution (for the money) is obviously high up there on the scale.
11) Motor drivers -- I will probably create my own set of these, though if I get antsy I may break down and purchase something pre-existing.
12) Computer -- I have plenty of Pentium II and similar computers lying around, so I'll just repurpose one. Or, maybe I'll just run a wireless link... something to consider.

Well, that's all I can think of off the top of my head, but please feel free to comment or add your own ideas.

I'm still a newbie here, so I'll have to wait until a mod verifies this post before I can add any more to it. Once it's verified, though, I'll continue adding in my thoughts and what I've come across so far in my search. Optics will be first up...

This is what I've determined my optic needs will be, so far (prices are right out of the catalog... Edmunds, Thor, etc.):
1) Kinematic Lens Mount -- 2 @ $35 each (for holding FS mirrors that route unfocused beam to gantry)
2) 1" square FS Mirror, gold -- 2 @ $30 each
3) 1/2" Right-angle Kinematic Cage Mount, threaded -- $108 (for angling the beam down and into the focusing lens)
4) 1/2" Round FS Mirror, gold -- $17
5) 1/2" Stackable Lens Tube -- $15 (for the focusing optics)
6) 1/2" Retaining Ring -- 2 @ $5 each (holds focusing optic in lens tube)
7) 1/2" Focusing Lens, ZnSe -- $??? (focus optics)

I have left the price tag on the focusing optics open as it will depend heavily upon where I get it from and if I decide to go for 1" rather than 1/2". I've seen 1/2" with a 5" FL for $60 on overstock up to $250-$300 for regular price 1" with 5" FL. This puts me in the general area of $400-$500 total for the optics.

I do not know what the typical beam diameter is coming out of the tubes, but I imagine something on the order of 4-6mm... if that's true, a 1/2" lens is about right. Anything larger and I'll step it up to a 1". My research indicates the lens size should not exceed roughly 3x the beam size to prevent local heating from stressing the optics.

The use of kinematic mounts gives me the ability to rough mount the optics and make fine adjustments whenever I need to. I considered a beam combiner to allow for alignment using a secondary HeNe laser, but I think I'll pass due to the extra cost involved for little gain.

If anyone knows any relatively inexpensive optics suppliers, either for ZnSe lenses or kinematic optical mounts, please feel free to post them up here. If you are an asian supplier and do not accept credit cards, please do not post any "for sale" items in here... I trust those suppliers only as far as my credit card will save me from scam artists.

I actually think the optics portion was the easiest. Next up, I think I'll post what I've found out about ball screws, guide paths, etc., as I've already done some research into that area.

I didn't expect the ball screws to be as pricey as I've found them to be. I will continue to search for less expensive suppliers, but the prices I list below seem to be somewhat average (limiting myself to sites that list prices directly on their websites or offer online ordering capabilities).

The following data includes Major Thread Diameter (MTD) in inches, Threads Per Inch (TPI), Cost per foot of thread, Cost per nut, Cost per flange (one needed per nut), Operating Load (OL) in pounds:


MTD | TPI | $/ft. | $/nut | $/flange | OL
=============================================
3/8" 8 $17 $ 73 $36 135
1/2" 4 $31 $115 $35 750
3/4" 5 $22 $ 36 $30 900
1" 4 $28 $ 83 $41 1,600+
As you can see, unless you're going for long distances, the 3/4" is the most reasonably priced choice due to the low cost of the nuts. However, I really like the extra precision the 3/8" gives me, so I may throw in some extra cash and go that route.

With two rods (one for either side) moving the gantry platform (Y-axis) and a single rod moving the carriage (X-axis), a 3/4" setup would cost just shy of $400, whereas a higher-precision 3/8" setup would be a bit less than $500 (all assuming a 3'x3' machine).

Next up, I'm trying to get some more hard info (i.e., pricing) on guide paths, as well as lock down the type I would like to use.

This screws are fine stuff, but I think you can also use belts, as you will not have contact with the working piece, are cheaper and very exact also.
The budget will get down dramatically...also the work ..

This screws are fine stuff, but I think you can also use belts, as you will not have contact with the working piece, are cheaper and very exact also.
The budget will get down dramatically...also the work ..

I originally considered belts or drive chains but decided against it for several reasons. The main reason is the amount of slop in their positioning, particularly as drive length increases. I can preload the ball nuts and get positioning accuracy significantly higher than with a rubber belt that stretches faster with temp increases than a metal rod. With metal chains, I have to worry about supporting the middle of the heavy chain as length inceases, or tighten the chain to prevent sagging, which leads to higher lateral forces on the drive motors.

By using a track and ball screw, I get the long-distance support from the track with the high precision I desire from the ball screw. Since the screw doesn't need to support any weight, the drive motors can be smaller and less expensive, which also has a similar effect on the drive electronics. I may spend 40% more on the drive mechanism to save 20% on the electronics and drives, but the trade-off for the accuracy seems reasonable to me. Hopefully I will also see a reduction in the amount of mounting hardware "beefiness" required since the lateral forces are not on the drive system.

So there's no confusion as to my intentions, I feel spending twice the money on a custom solution compared to the typical Chinese system is perfectly acceptable as I know mine will be more precise, handle a larger array of materials, and tech support will always be within earshot. At the rate things are shaping up, I imagine I should bring this system in (including laser) for under $3k. Considering the Chinese systems of this general size, such as the 2'x3' system often copied, run about $5k, I think I'm on the plus side of things (even the smallest US made system is 2.5 times my estimated price and has a much smaller 1.5'x1' working area). Should I decide to go smaller/larger, I know what the fixed price items will be: 1) Optics, 2) Laser, 3) Drive/Electronics, and 4) the basic Gantry mechanicals. Once that fixed cost has been dealt with, the price will vary almost in direct proportion to the size of table I choose to use it on due to: 1) Rails/Screws, and 2) Bed/Frame metal.

I've already shown how the optics, laser, and screw system will each be around $500, and I expect the steppers and control electronics to be in the same area. That's $2k, so I'm pretty sure I could bring in the remaining metalwork in under $1k.

I'll discuss in more detail about the carriage and gantry design once I get to those posts, but I have several ideas to help cut down on issues with keeping the optics and drivetrain clean.

Hi Dave,
I don't have much time right now to respond but if you don't want belts, which are perfectly acceptable and really don't stretch that much, then opt for the polished multi start acme screws with polyacetal antibacklash nuts. I use these on one of my routers and don't have an ounce of problems with them and with a little bit of lubricant they work a treat. ball screws need to be sealed from dirt and debris which is a p.i.t.a and the added expense is just a waste. A 1/2 inch acme screw with 4 start 2 tpi or there abouts will give you a great little mover over 3 feet and because of the long lead, you won't have much whip because you won't have to spin it very fast for the rapids. Mine are 7/16 over 2.5 feet with teflon coating and have no whip at all at the limit of my stepper drivers (Geckos)
McMaster Carr have a very good variety and fairly cheap.
http://www.mcmaster.com/ goto page 1053
You don't want any more accuracy than what they will give you as you won't be cutting steel etc with a 100 watt laser.
I have used Ball screws, acme rods and belts and for these types of applications I will go for the belts everytime. There is just NO maintenance on them and they hold enough accuracy for a 0.007" thou beam point which will expand to probably 0.011" after exiting the material anyway. Do the math, I couldn't be bothered right now as I am tired.

Just a thought.
Richard.

You can look at the plasma forum http://www.cnczone.com/forums/forumdisplay.php?f=62 they use gear racks instead of screws or belts, because of the dirt around, also very exact, if you want a heavy duty table...standard steel and Macmaster carr

Dave,

Go with belts and pulley these can be very accurate.
Think of an inkjet printer which employs this method of transport can't get much precise then that.
Due to low friction ie your not trying to cut into any metal such as a mill does the belts drive system can run very fast with low cost steppers or lowcost servo's such as Pittman Lo-cog woth 500 CPR (2000 per rev).

The ballscrew route is way over engineered and overpriced for the requirement.

Just my two pence worh.
Good Luck with the project.

Paragon, in several PMs with MonoNeuron, I think I've decided to take a stab at using belts over the screws. As you said, I'm not pushing/pulling a heavy tool through material, only a lightweight flying head. This is a learning process as much as it is an exercise in design and building, so some of my previous misconceptions are being remedied.

I've also been mulling over purchasing one of the smaller Chinese-built machines from eBay (I've seen them as low as $900 shipped) and retrofitting it onto a larger table. For the price, it would give me a good baseline design to start with, all of the optics I would need, a flying head, and a laser tube. As time goes on, I could change what needs changing, such as possibly higher quality slides, more powerful laser, drive motors and program, etc.

I need to start cutting and engraving 1/4" acrylic soon, and this would get me up and running. I will update this thread as time permits and the project proceeds.

Hey Dave,
Just wondering how the research is going and what you have come up with so far. It's an interesting pastime but can get costly as you always seem to want more speed or power, and in a year or so I will be in the market for a 100 watt laser as the 35 watt just isn't fast enough for anything over 2mm in chipboard or 4-5mm in Acrylic. I have linkmotion software which acts like a print driver but it ties up the computer while cutting so I have to sit there until the job is finished to be able to start another project after. I think I will also look for another controller as some jobs take over 3 hours to cut.(4" high full alphabet with numbers in 1mm chipboard)
Rich.

Hey Rich (btw, it's Dan, not Dave :) ), thanks for all of your help in pointing me in the right direction. I intended on pushing for an 80-100W laser from the get go as I knew I would want to cut some semi-beefy stuff... posterboard is for targets ;)

I've spent quite a bit of my free time working on my business (LED lighting) and cranking out new projects, so little has been left for research. I have a large stack of catalogs on my desk at home, ranging from linear guides to optics. As I mentioned in my last post, I may attempt to get one of the smaller machines from eBay (where I'm reasonably protected through my credit card) and rebuild it to my specs. I can pick up a small machine for under $1k, which gets me the optics and mounts, laser tube, etc. I could get longer rails and drive belts for a few hundred dollars and have a workable machine within a few weekends of custom building and adjusting.

I'm more interested in getting something working, even if it isn't 100% perfect, so I can start working on some projects that have been sitting on the drawing board. Once I'm pushing stuff out of the door, then I can go back and redesign the pieces that I'm not completely happy with. I imagine the entire machine will be rebuilt as time goes on, including drive electronics, motors, gantry, etc. Once I'm at that point, I can document the entire build. I'd rather have complete plans for a system that can be built from scratch rather than having to retrofit someone else's machine.

Still looking for that reputable laser tube reseller, though... not an easy find... impossible so far, actually :(