[e97]

I plan to use a lot of 6063 / 6061 ~3.175mm thick aluminum and would like to make a machine capable of cutting and welding it, hence the combo CNC + laser!

The minimum working area is x2' x y2' x z1.5' x'360, y'330, if its not too much more expensive and with the same accuracy, I'd like (~3ft) 1m x 1m x 0.5m

My biggest concerns right now are stability / accuracy and then cost.

The plan is typical 3-axis gantry and instead of the CNC on the end of the Z - we have a tube and servo, at the end of that is a coupler and servo to a small tube and/or plate with the interchangeable head. Tubes could be swapped to modify distance for different types of parts.

X Y Z, from reading a few build logs, the most accurate would be an ACME screw and linear glide rails. for x' and y' rotation the idea is to keep them light and stable - x' would have 360 or close to it rotation, y' would be 330' degree ish, to account of the base of the arm. I'm thinking of using gas pistons (kind used to raise a car hood or cabinet) to add rigidity and potentially buffer any vibrations, dont know if this is useful or a waste.

This type of setup would let me weld / CNC inside a structure from above.


The one more thing feature here that makes it magical is a computer vision using a camera + depth sensor to do real time scanning to incorporate feedback into the system to improve accuracy and help optimize fabrication / machining. Ideally one would clamp down the piece and select the component with its corresponding part in CAD and the machine would do its thing without needing to be home'd or further re-calibration.

I think I'll need heavier than 20x20 aluminum extrusions and then CNC milled brackets based on the CAD design. Not sure what size nuts/bolts/washers, ACME threads or motors and motor controllers. I would also like to open source the designs because I was inspired by RepRap and all your projects.

for the CNC router I was thinking of this: https://www.amazon.com/Dewalt-DW618M...dp/B0009E6JHK/

Dewalt 618M - 2-1/4 HP / 12.0 A router Electronic Variable speed (EVS) control for constant 8k-24k rpm - $154.99

The most expensive part will be the laser and components - I'm looking to get a cost effective diode laser of about 400W but they seem pretty expensive. The other options is to build one myself based on this research: https://news.mit.edu/2014/startup-te...-cut-weld-0723 which uses Fresnel lens to focus and combine multiple wavelengths to get a more powerful beam.

Diode laser (typically found in CD-ROMs / DVD / Blu-ray all have different wave lengths) so finding a way to build alot of them on a board, focus then and cool them cheaply would be amazing.

Re focusing and placement - the two options are mirrors and lens or fiber optics and lens. I'm leaning toward mirrors based on this: https://www.youtube.com/watch?v=2d0S6QtHSgQ

Let me know what you think.

[wizard]

Im not sure i follow all of this post but one thing that stood out is that you will need machine frame members much larger than 20 mm T Slots! A lot larger actually, it might also be cheaper to go with steel. You will want to read some of the build logs and define what your expectations are with respect to the machining. In the end your expectations define how you build the machine.

In a nut shell machining is nothing like a Reprap as you have to deal with reaction forces from the cutting tools. That and stiffness or the resistance to vibration dramatically impacts surface finish.

As far as ACME screws go do realize that you also have ball screws as an option. Which is better is open to debate but in either case you will need protection from weld splatter. The same goes for your linear rails.

Third if you dont know what you are doing industrial lasers are extremely dangerous. I would not even bother with a laser facility unless you are willing to fully invest in learning all the issues involved, the standards you must build to and the legal responsibilities in your area. You can end up blind or with severly damaged eyesight in a blink of an eye and maybe more horrifically an innocent bystander.

All of that being said the future is in fiber lasers. I really doubt 400 eatts input power is enough to do much with aluminum. Further it doesnt make sense to me to try to turn a router into a welder. At least not as i understand your post. Now cutting might be a different story but there are dramatically different needs with respect to the table (work holding) and the Z axis tooling.

If at all possble avoid mirrors! They become a maintenance nightmare

Designing interchangable End Of Arm Tooling ( EOAT) is pretty common in industry but the change must make sense with respect to the rest of the machine. You can get around some complexity here by having room on Z place two tools.

[e97]

Im not sure i follow all of this post but one thing that stood out is that you will need machine frame members much larger than 20 mm T Slots! A lot larger actually, it might also be cheaper to go with steel. You will want to read some of the build logs and define what your expectations are with respect to the machining. In the end your expectations define how you build the machine.

In a nut shell machining is nothing like a Reprap as you have to deal with reaction forces from the cutting tools. That and stiffness or the resistance to vibration dramatically impacts surface finish.

Yup, I was wondering what minimum was necessary. 4040? or maybe using steel. I'm searching for some sort of equation or maths to calculate rigidity for vibrational forces, the only thing I can think of is torques across the supports. Advice here would be helpful.

My target benchmarks: for the CNC part alone is based on the best existing product I could find (https://www.cnczone.com/forums/diy-c...e-forum-5.html)

X6-200LUSB is US$2,199.00 - says .0025 mm/step resolution and 0.05mm position accuracy
Shapeoko 3 ~ US$1200 - doesnt mention resolution or position accuracy

As far as ACME screws go do realize that you also have ball screws as an option. Which is better is open to debate but in either case you will need protection from weld splatter. The same goes for your linear rails.

[strike]I'm lacking the terminology here, but isnt the ACME just the threaded screw? I thought the ball component was necessary to convert the rotational motion of the ACME screw into translative motion.[/strike]

This was helpful : https://www.progressiveautomations.c...ew-ball-screw/ . Ball screw maybe smoother, less friction but I'm concerned about braking and vibration movement and noise.

There is no weld splatter when using a properly tuned laser welder with the appropriate settings for the material, it uses micro/nano pulses.

Third if you dont know what you are doing industrial lasers are extremely dangerous. I would not even bother with a laser facility unless you are willing to fully invest in learning all the issues involved, the standards you must build to and the legal responsibilities in your area. You can end up blind or with severly damaged eyesight in a blink of an eye and maybe more horrifically an innocent bystander.

Good safety warning. I worked with lasers in physics labs to experimentally re-discover quantum properties of light and waves so I'm familiar with safety procedures.

All of that being said the future is in fiber lasers. I really doubt 400 eatts input power is enough to do much with aluminum. Further it doesnt make sense to me to try to turn a router into a welder. At least not as i understand your post. Now cutting might be a different story but there are dramatically different needs with respect to the table (work holding) and the Z axis tooling.

Most lasers I've seen quote their average constant power aka 400W, ~ 4kW pulse

If at all possble avoid mirrors! They become a maintenance nightmare

This is good consideration, however I believe a vaccuum sealad tube with auto-aligning mirror could be easier to get the right amount of power at a lower cost then a flexible fiber. See the MIT link and TUMPF laser video.

Designing interchangable End Of Arm Tooling ( EOAT) is pretty common in industry but the change must make sense with respect to the rest of the machine. You can get around some complexity here by having room on Z place two tools.

Yes, this is a possibility but laser are very sensitive and I would not want to damage them parts by exposing them to CNC vibrations and debris.