DIY Fiber Laser Cutter V2 | Part 1: Better Laser for Only $10k?
ฝัง
- เผยแพร่เมื่อ 28 พ.ค. 2024
- If you want to build your own fiber laser cutter, join our Patreon group: / diyfiberlaser
After using my DIY fiber laser cutter for a year, I couldn't stop thinking about improvements that could be made...so I've design a new machine! Today, I go over the details of the plan and new design. Stay tuned...in the next video we'll start the build!
-------------------------
My apologies to Matt Davey, for not crediting his work. I showed a brief clip of his excellent 3D model of the Ultimate Bee CNC mechanical kit, that he has shared with the community on GrabCad. Sorry I forgot to reference the source in the video Matt, I will be more diligent next time. You can check out his 3D model here: grabcad.com/library/ultimate-...
--------------------------
Buy fiber laser components, complete machines, fiber laser repair and retrofitting services at
Sky Fire
www.alleriastore.com
For a 5% discount on laser components use the code: “FROMTRAVIS”
(this discount applies to all products except for the laser machines)
--------------------------
My website: www.id3ntitycrisis.com
My design portfolio: www.id3ntitycrisis.com/portfolio
I wish more people were interested in content like this. We need a 1000 of you.
Good modeling so far, we're very excited for this. See if Skyfire would offer a 'Travis' bundle of your components with a discount. Good luck, look forward to your upcoming video.
I second this! I will build this pending documentation and video instructions
I placed a hanging mesh of expanded metal just below my grates on my plasma table; keeps everything from dropping to the ground.
Dude im excited for this build! Stoked for this content for sure
Looks great! Looking forward to seeing more info!
one potential problem could be the carrier drag chainon the left, limiting the travel of the y axis when the gantry is at the front. from my experience it is worth the extra time animating the chain to know the runout and the exact number of chain segments.
Thanks for pointing that out! I'm not particularly fast at modeling, so I'll probably just mount it and test the travel before I do any wiring, and add or remove links as needed (I have some extra links if needed)
I agree with this comment, at one end is no problem, but at the other end you wont have enough room for a "loop" because it will hit the vertical post. If you don't have a loop in the carrier then the cables will have to make a sharp bend. Are you familiar with the lumenPNP from opulo? Look how they mount the carrier, I wonder if that would work..?
@@davedave4335 The X gantry never moves all the way to the front, only as far as it needs to go to get the cutting head to the front. To me it looks like there should be just enough room to make the loop in front of the vertical post. If it turns out to be too tight, that lumenPNP solution looks interesting...maybe we'll try it. Thanks for your feedback!
@@diyfiberlaser I see what you're saying, looks like it'll be close, but may very well work as is. Keep up the good work!
Looking forward to the full build, looks cool so far.
Muy bien, sera genial acompañar el proyecto!
Thank you 🙏 from INDIA
Looking forward to the full build, a suggestion is to add an access door/s at the bottom to access the parts
Thanks, several people have mentioned this now...it's great idea, I'll certainly add this in
It’s awesome 😀can’t wait for BOM to be released. I built a co2 laser table designed by someone on TH-cam 2x4 feet area . He post the videos on TH-cam for all the steps that he did while was building it . He sold the plans bom and pdf how to put it together . It will help a lot of people .
If there is enough interest, I could certainly try put some kind plans and instructions pdf together when I'm finished.
I’ll buy it from you !
@@tiberiud1955 Thanks, I'll probably try to do some kind of poll near the end of the project and ask viewers what kind of assistance or documentation would be the most helpful to everyone and see if I can get some consensus about how to best help.
Looking forward to this series. Any way you could make a drawer below the bed so you can pull it out after everything is cut and just pick out your pieces that way instead of reaching in?
Thanks, several people have mentioned this now...it's great idea, I'll certainly add this in
Really excited to see how it goes with the cheaper laser source and controller. I built my CNC Plasma table with the intention of eventually converting it to being a fiber laser machine. Looking at prices of the components mentioned, you could get the fiber laser components for close to the same price as a higher end plasma cutter. Tariffs add to it, but you are trading that for a higher quality cut in my opinion.
Maybe you've thought of this: if you have a bottom hinged door running the width of the machine across the front middle-third section where the parts drop into would make part retrieval much easier [without having to remove spars from the material support bed] and would make cleaning that section with a brush/vacuum a simple process.
Where fume extraction is concerned such a door could be easily sealed using off the shelf commonly available window draft sealer strips or butile strips on the back of the door in order to stop fumes escaping out of the front.
Thanks! Yes, several people have mentioned this...I somehow overlooked this simple addition...I'm definitely going to add in the access door!
Hi Colleague!
I would make the base of the machine welded if you want to increase the cutting speed. The head of a fiber laser is very heavy and will cause inertia at the point of sudden stop and acceleration. this inertia must be dampened by the frame, otherwise vibration will not give you a high quality cut. now you have very short stiffeners due to the small corners for the aluminum profile. if it is not possible to weld the frame, then cut out flat triangular stiffeners from a sheet with a thickness of at least 3-4 mm and fix the frame parts with them to a bolted connection.
also make sure that the sparks will not fly on the belt, otherwise it will sooner or later become unusable. 5m is a good belt, but it can also deteriorate from sparks.
Thanks for your feedback. I want to use aluminum extrusion to make it accessible to non-welders, but I like the idea of triangle braces...we could use the machine to cut them out. Sparks on the belt is a concern of mine also, we'll have to see...they may be high enough above the cut to avoid the sparks.
I have another idea for you. You can cut a square pipe on the first machine. Why don't you use this advantage when assembling the frame? An aluminum profile is needed when you need straightness. Where there are mechanical movements. At the same time, you use a fairly small aluminum profile. 20*20 and even 40*40 mm is a rather thin design. At the same time, the aluminum profile is decent. Plus corners and stuff. You can assemble the main frame from a square or rectangular metal pipe with a cross section of at least 40*40mm or 40*60 mm. You just cut the pipe to size on a laser, make mounting holes (if you are afraid that you will not get into the holes with your teeth, then you can make them oval) and then assemble the entire structure. Bolts can be taken at least 6mm, preferably 8mm. This is much better than the 5mm bolts from the corners of the aluminum profile.
This way you can add diagonal joints that will increase the strength of your structure.
I have a workshop and I have a laser fiber machine with a pipe cutter. He does very cool things.
Another thought is, of course, the question of the size of the working field of the machine. I understand that you have a small workshop and just don't have enough space to place a large machine. Trust my experience, it is very convenient to put a whole sheet of metal at once, which does not need to be sawed beforehand. And on this sheet you will get larger parts and you will save on metal waste. Because the cutting will be more economical. To do this, it is convenient to use the nesting program.
It's just that the main part of the cost of your project is the laser source itself, the head, the chiller, the controller, the motors, the drivers, etc. The mechanical part here does not take up most of the cost. That is, if you had a place where to assemble a large machine on a sheet of metal (we have a local standard of 1.25*2.5m sheets), then the cost of your machine would not be much more expensive.
@@diyfiberlaserI don't know if you use oxygen or air as a gas. We have a 16 bar screw compressor. But it's expensive. It costs about 70% of the cost of your project. I can tell you how a local DIY craftsman solved this issue on his fiber machine. He bought an 8 bar screw compressor, put a booster on it that compresses the air by an additional 2 times (consumption also increases by 2 times, so a productive compressor is needed). For air treatment, he assembled a large cleaning filter. I just welded a filter out of a thick pipe (about 1.5 m long and 120-150 mm in diameter) and filled it with something like cat litter and something else. This is necessary for oil and moisture removal. He generally made a chiller out of a container with water and an air conditioner of 9000 btu (it seems to me that this is not enough, but you have a 1 kW source - it should be enough). It turned out to be cheap.
Although I wouldn't do that to myself. But if there is no money, but you really want a machine, then this may be justified for him.
If there is a desire to communicate, I can use WhatsApp to show by video how our work is organized in the workshop. We are constantly in the process of modernizing production.
@@Pyroprinter My current machine doesn't have the size capacity and the controller doesn't support square tube cutting. I'm considering building a dedicated tube cutting machine in the future so I can do exactly this.
Nothing wrong with info heavy videos. Knowing why is very valuable. Assuming the prices you're showing are the "not sponsored" price list - does that include shipping?
I paid full price for everything except the cutting head that Sky Fire sponsored for the project, but you can get 5% off by using the code "FROMTRAVIS". No that does not include shipping. Do some shopping around and get other prices...you'll see why I'm shopping here : )
@@diyfiberlaser No worries, would do so. Just helps give a ballpark because things like laser sources can be pretty heavy.
@@maficstudios The laser source is definitely the most expensive component of the build. I don't know if you saw my last FAQ video, but I was previously buying laser components through the cheapest source I could find on AliExpress. I could have saved over $300 buying my laser source through Sky Fire instead...you live, you learn. Also keep in mind for your budget, that whomever you purchase though, you'll probably end up paying around 25% tariffs if you're importing into the US.
I am glad to see a v2 build in the planning as really enjoyed the first versions videos. But… I had to double take when I saw you mention the Ultimate Bee 3D model as that is my labour of love, created and shared with the community over on GrabCAD a little while back now! 😊
I hope it was of use to you despite going another route this time with the framework (for good reasons). With your permission I would like to share the model link to the Ultimate Bee in the comments section here in case it is useful to others. Thanks, Matt
My apologies Matt, you're right...I totally forgot to reference the source of your model. I've updated the video description to add a permanent link and give you credit. If anyone needs an excellent model of the Ultimate Bee kit, check out Matt's work: grabcad.com/library/ultimate-bee-cnc-milling-machine-1000x1500-mechanical-kit-1
@@diyfiberlaser that’s very kind and no apology required!… I should have made it more clear that I feel privileged to have some work feature in one of your videos. Great that others will be able to more easily benefit from it now too if needed. Much appreciated & looking forward to the your next upload 👍
SDid you look at nut spinning in the ball screws to avoid the whip.
yes thanks, I didn't mention it because I don't know how to fit it within the tight space of the C-beam like I was trying to achieve
Would you consider using a Masso controller for the fiber laser if they programmed it to work properly. Over the chinese one.
Thanks from Ross.
I did consider it…I even have the Masso controller here that I purchased for it. Though I decided against it, since it doesn’t yet have the controller or software support and I can buy a full featured fiber laser controller for cheaper…I may attempt it in the future
@@diyfiberlaser Thanks for the reply.
Well I know whats my next project gonna be ;)
Option D is static ball screws and rotating nuts...
Yep! ...I didn't mention it, because I'm not sure how to make that work within the C-beam like I'm trying to use in my design
How long do the consumables last? Also why not slant the floor of the machine to slide parts out to a retrieval bin on one side?
It's hard to say as I'm only using my machine sparingly for prototyping and on-off projects, not daily use...after a year's use, I've not had to replace a nozzle yet, just recently had to place my first protective lens. Thanks, slanting the floor would certainly be a viable option.
Any updates - waiting, before I order parts :)
I have the frame done, working on editing the footage so I can get the next video out
@@diyfiberlaser Great ... waiting for controller and fiber source installation :).
@@kbelcar I'm exited to get there too, but it's gonna take some time...there's still much to be done before we can get to that stage
Did you import BM110 head from skyfire? And if the servo inside the bm110 head Ethercat based or simple pulse based? Because the raycus controller supports Ethercat aswell
Yes the cutting head was sponsored by Sky Fire. I got the pulse version of the controller because I already had the pulse servo drives. Had the timing worked out differently, I would have got the Ethercat version of everything.
Are you going core XY for the movement system or independent X an Y belts ?
Controller doesn't support coreXY
When you ordered how did you handle import/tariff tax or customes etc. The control combo you show does it come with opperater software like ezcad or will it communicate with lightburn?
FedEx takes care of the customs process, they just send you a bill for the taxes. Software is Raycus 2DCutAhead, no Lightburn
@@diyfiberlaser I saw on the sky fire site it was 25% does that sound about right?
@@cnckitsplus1 Yes, I believe that's what it is supposed to be...though when I got the bill from FedEx, it calculated to be closer to 19%
@@diyfiberlaser Thanks.
Hey wouldnt it be possible also to use closed loop stepper? Little cheaper and also dont lose steps?
For me the silent operation of AC servos is worth every penny. Ever since the first time I used them on a machine, I can never go back to steppers.
Since speed is one of the main goals, any reason you're not going for a CoreXY motion system? Even with a heavy head, it still moves the X-servo from the gantry to the frame, and servos aren't that light either. While it slightly complicates belt routing and needs longer belts in total, unless absurd accelerations are needed (above 10k mm/s² or so), the additional length shouldn't be a limiting factor (assuming quality belts like Gates or equivalent).
That would be interesting to make, but I don't think the controller supports coreXY...I don't see any mention of it in the user manual.
@@diyfiberlaser oh yes, whoops. Coming from comparatively simple 3d printers, where this is essentially the norm for cartesian movement now, and every firmware has been able to do it for 5+ years, I kinda forgot that your controller needs to support it. I mean mathematically it's trivial, maybe they'll add it in a future patch or revision to the controllers? Still means it's not possible for your build for now though.
1000mm per second seems a lot for me. Which acceleration in x and y direction do you want to achieve?
31 mm diameter pulleys
(100 mm circumference)
Servos max 3000 rpm
With 5:1 reducers = max 600 rpm
600 x 100 / 60 sec = 1000 mm/sec
That’s how I got that number. I don’t know the math needed to figure out what acceleration will be achieveable….I did some searching online, but couldn’t find any resource that explains it simply for a non-engineer. I would love to learn, if you have any reading to suggest?
The motors are 400W 1.27Nm…one for X and dual for Y. I can figure out a close estimate of the weight of the components.
@@diyfiberlaser Theoretically acceleration without friction a = power /(mass x velocity).
With friction you have accelation a= ((power /velocity) - Frictionforce) * (1 / mass)
So these aren`t both useful equations for sure on one hand, because all influences of inertia and friction aren`t calculable in detail.
On the other hand the controller needs a useful value for the acceleration to avoid motion problems.
Usually servo motor has nearly constant torque, so
power P increases linear with increasing RPM.
In this case: 400W at 3000rpm and 200W at 1500rpm
Example: mass m=10kg, P=200W (1500rpm), v=0.5 m/s, no friction gives:
a=200/(10 *0.5)=40 m/s^2= 4G.
Most important is the tuning of all servo motors under real conditions to achieve good mixture
between speed and accuracy.
In my opinion two of the best videos on youtube due to tuning are
from JBWorxStudio (nice German guy):
Part 1: th-cam.com/video/_9Q-VFesnA0/w-d-xo.html
Part 2: th-cam.com/video/jvbw_PMOs-s/w-d-xo.html
Maybe your cutter will be as fast as with acceleration 6G = 60 m/s²:
th-cam.com/video/6h2eE2FOnSs/w-d-xo.html
Wish you all the best
@@stefankrimbacher7917 Thanks for all the info! I'll check out your video recommendations and see if I can learn a thing or two before I start the tuning process
@@stefankrimbacher7917 OMG! I just watched the 6G acceleration video...that is incredible!
Slide through bed for large pieces?
Thanks, that's a good idea! Fo my use case, I've never had a situation in the past year where I would have needed this feature, but it would be easy to implement if someone wanted to put hinges on a couple panels to allow access for longer material.
If it were me, I would get rid of the planetary reducer and just do a 5:1 belt reduction. Much cheaper and you don't have to worry about backlash. I would also go with a bigger belt than 15mm.
That's certainly another possible solution....I would have to do more shopping around because I was having trouble finding the appropriate gear sizes needed, for cheaper than I got the reducers. Do you have any recommendations for where to buy cheap pulleys? I just need the belt to fit within the 40mm space, so I can go larger if it turns out we need it.
@@diyfiberlaser Or just spend that money and upsize the servos and direct drive everything. As for the belt I don't know where you are, but look at mcmaster carr, sdp-si, if you can only get than maybe double it up. You just don't want it springy.
@@moki123g Thanks for your input, that's a good point...for the cost of the reducers, I could have also upsized the motors instead. This is an area that I would love to get some formal education in, as the deeper I get into all the math involved in properly sizing motors, it gets to be more complicated than I can comprehend. The 15mm GT5 belt has a working tensile strength of 444N...I'm not even sure how to start calculating at which point it will start to get springy...I need to find an online course I can take to learn this stuff!
@@diyfiberlaser the limiting factor with the belts will be the stiffness not the breaking strength and that is harder to decifer with a catalog because your whole system will contribute. Think about using as wide of a belt as you can. Also upsizing the motors is not trivial. 5 times more torque will be a lot of mass. I have used a few different gearboxes for simracing projects from alibaba and lichuan had better quality then others. Consider helical planetary for reduced noise and backlash.
Would be nice if you released all the files for free but then get the money by cutting and selling parts for the machine.
But at the same time it's all 20 series profiles so people should be able to replicate the design fairly easily based on the video and the spec.
Also to join profiles I would suggest using the button screw approach like they use on the voron 3D printers instead of the 90* brackets, this way you will save some money on the hardware, it will be easier to assemble and most importantly will be much more rigid.
I thought about that, but I don't really want to do production work...I have to much creative energy trying to get out : ) Also, there will already be a way to easily order all the custom aluminum parts...I'll be mentioning it next video. Thanks for the suggestion... I've seen those connectors, but never tried them. I've already purchased the hardware for this project, but maybe I'll try using them on the next one.
@@diyfiberlaser There's no connectors it's just tapped hole, drilled hole, and a button head screw.
I can send pics if you want just let me know where.
@@Arek_R. Oh I see! I found it this time...last time google showed me something similar, but there was an insert also. You could use this method for some connections. A lot of the extrusions butt up to a 40mm one, so you'd need longer screws. There are roughly 150 corner brackets used in the design...I don't know that I'd want to drill and tap that many holes if I could avoid it : )
@@diyfiberlaser Trust me, dealing with 1 bracket, two screws and two t-nuts is even more work, it's very fiddly.
If you make some drill setup it will be quick and easy to drill, same with tapping if you get a spiral tap and just quickly drive it in (don't need to do the whole in and out dance like with a generic tap).
I see on the front you have 2020 profile butt up against 2040 profiles, not sure but I think you could still use this attachment type with it, first you screw 2020 and then there should be enough space around either side of 2020 to reach 2040 screws.
Now that you told me it's 150 of them, isn't that like $300-500 of hardware?
@@Arek_R. The corner brackets are fiddly! You're right, if I made some jigs the process might not be too bad. I actually have an electric tapping arm on the way, but it's probably on a container ship somewhere and I don't think it will arrive in time for this project. It is about $150 of hardware.
Have you ever been told that you bear a striking resemblance to Sebastian Vettel? Or do you have a secret career as his doppelgänger?"😅
Haha no...I'm too busy for a secret career on the side!