I love how he has a cnc on his desk that cuts aluminum well enough and still proceeds to make parts for the cnc with the printer while knowing rigidity is the issue.
Impressive! For machining gear-like things, I've found the best process is to first cut out the inside holes, then add more workholding in the center, cut a circle around the outside, and then start working the curvy profile. First use downward motion with the end mill to cut away most of the material inbetween each tooth, and finally do the continuous profile cutout (one rough pass, one finish pass). The circle allows you to use a large and tough bit for full width slotting, and together with the downward motion, the small bit never has to take a heavy side load. You also get much better chip clearing compared to cutting a curvy full width slot. This procedure is particularly necessary when using very small bits in very hard materials (I came up with it when making titanium cycloidal discs with a 1/32" end mill), but it's a great speedup for aluminum too.
Interesting toolpath sequence... That sounds like it could be much faster than continuously going around the tooth profile. Ill have to try it next time! What CAM software do you use?
@@ericinventor C++ :) CAM software is outside my budget. Python would probably be easier, but C++ is what I'm used to. I also use 3D models in Blender to help visualize things. For best results when nibbling with vertical motion, always keep the center of the bit over open space. Involute gear teeth need several bites to get all the way in to the base of the tooth.
@@ericinventor Mine is all written from scratch, but now that you mention it I should clean it up and make it available for others to use :) As is, it would probably be confusing trying to understand how it's meant to be used At its lowest level it's just functions like G01_XY which takes an x,y position, applies backlash compensation, and prints a G01 command to the output file. But I also have a functions to do more complex operations with a single call, like drillHoleCircle(centerX, centerX, numHoles, circleRadius), or cutHole(centerX, centerY, radius) which will cut a helical path downward, with tool offset toward the center of the circle, and then do a finish pass with .001" or whatever finish chip thickness you have set.
Recently I had to look for someone to cnc me some brass parts for one of my projects, and the only one company that actually replied to my email gave me an astronomical price for their service, and now I found your video about buying and upgrading a cheap cnc machine on a affordable budget. With 500 bucks spend on cnc, 80 bucks on materials and a few days of work I can produce same parts much cheaper and also will be left with an amazing machine for God knows how many years. Good Sir, you are my hero! Thank you!
So what I got out of your video is to buy the most expensive tool you can afford... do your research and make sure that $$ is spent for what you are going to do with that tool. Use a hammer for nails, a screwdriver for screws, a drill press for holes, a saw for cutting, a mill for milling and a lathe for turning, etc.... to me, the fun part is thinking about the projects and designing them out on the drawing board. Also enjoy making my own tools, but you can just buy what others have designed for you, as you can always trade-off time for money. It's not the destination, it's the journey. Thanks for the video!
This reminds me of the upgrades I made to my 3D printer kit, where I gradually replaced each new weakest component until I had pretty much built a new, better printer, just using the original kit as a scaffolding. I've got a 3018, too, and I guess it's about time to start giving that the same treatment.
I did the same thing with my Tronxy X5S. Had military money every month and just threw funds at it. The only things left over from the original printer are some of the frame and the motors. Everthing else has been replaced and upgraded. Sad thing is, it still isn't working
A tip for hole "plunge" toolpathing, a far better method is to "helix bore" rather than plunge. 2 flute mills tend to chatter when plunging in the z axis only. If the hole diameter will allow (you will need a smaller tool to interpolate helical within the hole). Try it, I think you will find it mills much smoother. Not sure if your router has the capability to interpolate X,Y and Z , but just a general FYI Good luck !
But it is also a lot slower than plunging too. You need no backlash for it to work properly. He is just using cheap threaded rod, so I don't know if it would produce a better result.
for a cnc mill , rigidity is more important than all other factors.. The last one I made which is still use is box steel filled with concrete in a concrete base. 320 lbs , and the weak point is actually the spindle power which is 1.5 kw.anyway , excellent work finding all the weak spots in the design.
Yea. (600/60)*0.75*3.1 mm = 23.25 mm^3/s. If you put in those numbers in fswizard, it's 0.02kW... that's really just peanuts for any 1.5 kW spindle. I have a nice, granite, rigid cnc here that does 20mm deep cuts with a 6mm endmill... If you punch that in, you still only get 0.53 kW, which means that the spindle will still be fine, that is: as long as the cnc is rigid enough... The weak spot in any cnc3018 is just... well everything... these things are just so flimpsy. Throw a back plate against it, and the little nema's won't have enough power. Upgrade the motors, and the excessive backlash will ensure that vibrations kill the router bit. Change to ball screws, and the side panels are too flimpsy to any real work. Replace the side panels with 15mm steel, and the y axis will be your next problem. Next, the motherboard is crap. The stepper drivers are crap... the list just goes on and on. At the end of the day, not much will be left of your router when you attempt to make it good... Really, the only thing good about these small routers is that they are so much fun to toy around with. I had one and enjoyed it quite a bit.
What would you say the total cost to build youe cnc was in the end? I just got a 3020 Pro Max from Genmitsu but anticipating some upgrades pretty quickly.
@@robsproducts well u already have the main components from the 3020.. Mine cost about 750 total , But i used flat 20 mm linear rails on all 3 axis .. that was the majority of the cost ( 250-300 dollars) the concrete is dirt cheap , I used all the 6040 electronics, got a geared nema 23 for the z ( 45 dollars ) and used the 6040 spindle. The steel for the column was scrap from the garbage I picked up on the side of the road.Its probably more rigid than my 700lb benchtop mill , but these chinese spindles lack the grunt needed for heavy passes i can do manually on the mill. Hope that makes sense.. In my experience , the Z ( spindle mount ) rigidity is more important than the x aand Y, so u can go cheaper on those rails. I kept my 6040 running right to the end , to make brackets for the new machine.. then basically swapped the electronics over in one day.. so very little downtime
I bought a 3018 max CNC and stock it can cut aluminum but slowly and tons of chatter. Pretty accurate and with mist/air setup it's amazing now. I also upgraded all three axis to MGN15H rails, now its a beast.
Does anybody knows what travel length does the Z axis have on this machine? I am asking ,for all of those who want to replace it with a stiffer Z-axis.. also any weight restrictions? :)
For me , linear rails on X and Z solved the "chatter" problem. The rods are just not up to the task. The 10K RPM spindle motor is the next to go - probablyy with the 30K RPM router. Thanks very much for all your input.
That's good feedback. I've got a 6040 router I've had for 10yrs. It cuts a lot of alu & brass, but it's still on chrome rods instead of proper linear rails. I've been wondering whether the flex & chatter or the steppers are the biggest limitation to my cutting ability. Note: upgrading to flood coolant with soluble oil made a huge difference to it's metal cutting ability. Nothing I tried spraying by hand even came close. With flood coolant, you can run a faster rpm, so you can take advantage of a high speed spindle rather than trying to cut low and slow.
Excellent video! I have this exact kit that has been sitting in my 3d printer room unassembled for the last two years. I didn't want to put it together until I had some kind of plan to make it machine aluminum more adequately, Now it can be a really useful machine that can help me to create the long lasting and useful designs that I have been wanting to make out of aluminum. So the main issues are the X rails, Z axis and the motor (I knew the stock motor was garbage from the outset). You have shown a clear path to improve the most salient points while leaving the adequate parts alone. You have earned my subscription sir.
You need swan neck clamps to clam the metal down or the clam has to be level with the piece you are clamping. The lesson i learned from this video is to buy industrial level parts so you could use your cnc to make more parts that you need to complete the rest of the cnc machine. Great video.
I think if you stabilize the base more and get rid of the 3D printed plastic parts and replace them with metal ones, chances of chatter and vibrations are going to be very very slim.
Start with 3018, then let the mother cnc make parts for the next stronger cnc, then make another stronger stronger cnc and so on. Finally you can mill anything u want.😊
bro i am soooooo excited i jsut bought my first cnc machine and i intend to use your upgrade ideas... i know there upgrades parts selling now on the internet so I'm looking forward to the growth and process. Thank you much fo this video!
I just discovered this channel by also turning to google for inspiration and guidance. I too have been "developing" an upgraded mini CNC to cut aluminum. I set mine up inside a Harbor Freight blast cabinet that I picked up cheap from Craigslist and modified for this project. Subscribed Boiii!
Thinking about upgrading my 3018 but wasnt 100% sure where to start, this is really helping out. Cause i started questioning basically everything about it, down to the torque of the steppers. Its very nice to see that the most important stuff for somewhat speedy results are the z axis and the spindle. Cause i realy stopped using it alltogether since i kinda hit a wall already with some hardwood, only managed to go at a feedrate of around 100mm/min woth a doc of 0.25mm took me forever to do anything
I am really impressed - I never thought you can do anything with the stock nema17 steppers. First thing I did was upgrading to TB6600 and NEMA23 - but you got so much further.
To add to the other guys comments about your clamps being level (Level is best, but slightly higher is better than slightly lower). Also the clamping screw is best used as close to the part end of the clamp as you can get it. Great video, I'm thinking of getting one of these machines to play with.
Not sure if you touched on it or not, but one of the first things you should be making is replacement parts for all the support pieces that are still plastic. Especially now that you can machine aluminium easily. Replace the side supports, and also the 3-d printed backing plate. You should be replacing everything you possibly can with metal. I like that you changed out your slides for linear rails as well. Doing all your axes will make an improvement as well.
'at a non-glacial pace' :) Got my 3018 back in 2018 i think it was, always had it in the back of my mind to upgrade the spindle in order to cut aluminium, watching this with intent!
this video is great! I have been in the same limbo with upgrading a 3018. Heres a tip with clamping your materials, Its difficult to describe, Use a packer thats is slightly higher than the material your cutting, so your clamp comes DOWN onto the material. The force exerted on clamping the material will be exponentially higher
Excellent! I'm just starting my upgrade journey with the newer Sainsmart 3020 Pro Max... My first upgrade was to add linear rails to the Y-axis and make it 400mm (now 3040). Next, I'm going to beef up the X-axis in readiness for a new Z-axis and bigger spindle motor 😃
@@BadPractices Yes, I'm now using a Makita RT0700CX4. Had to design and 3D print a new 65mm mount for the router though. I have heard there's a way to start/stop the router using a solid-state relay triggered by the voltage output from the control board, but not prop[erly looked into it as yet.
@@paulblythe5899 Awesome! Is it connected to the control board in any way at the moment. If it is separated, do you just run it at full speed all the time? Thank you for answering my question by the way. I love my 3020 so I’ll be really excited to beef up the motor a bit
@@BadPractices The Makita is independent of the control board for now. It has a variable speed dial with numbers corresponding to different rpm (10K to 30K). In general, I run larger diameter bits slower, and smaller bits faster. Adjusting the speed per job/cutting bit. The 3020 is a great little machine and I'm happy I purchased one.
@@paulblythe5899 For the past few weeks I’ve been fussing about how to integrate a new motor with the stock control board but having the motor be independent is a clever solution. Thanks for the advice. I’ll be buying a Makita spindles today. Thanks for the info, Paul!
I just totally blown away from result! Such a solid journey and amazing result! Thank you for sharing your valuable investigation! And yeah my 3018 is need that chrome >:D
all, an important note is that the default Z-axis stepper is not strong enough to pick up the Makita spindle. Eric probably didn't notice it as he used the Z-axis stepper received with the Z-axis he used in the middle of the video. please note. Eric, excellent work thanks for the help
you are absolutely incredible, though i can never buy it while living it 2.5rd world country, its plesant to watch someone succeding about making aluminum cutting in a budget with high success
So I am a gun Smith and that looks like the perfect tool for what I need . Do u think this is something a new guy can do .I need it for milling out gun parts
Thank you Eric, really interesting and useful. A little engineering tip, when you clamp a piece to the worktable, the heel of the clamp should be slightly higher than the toe and the clamping force should be as close as possible to the work piece, this way you get maximum and safe downward pressure.
As a machinist and cnc programmer, I see a common issue with hobbyist using the CNC 2018 , the clamps are not being used correctly. The clamps should be parallel to the top surface with a packer piece to suit. This will greatly assist in clamping as it won't vibrate loose and break the cutter like it did in you video, with the clamps at an angle, it will come loose and move, breaking the cutter.
Most absurd part - replacing to linear rails and beefing up everything.. than mounting it all on plastic adapters so all that added rigidity could be once again wobbly 🤒🤕
Just bought the 3020 chinese cnc router, I shall absolutely go after your ways. Shall buy a small Makita router to it, as my plans are to cut aluminum just like you, thanks for this video it confirms all my thoughts about the linear rails (roundbars) being to flexible. I been looking at linear rails for a while to modify my 3D printer with 2 lead screws, and 4 linear rails in each corner, I have the creality ender 5 s1 and getting the buildplate adjusted is a real b**** I shall add a 5th point where I can adjust the center too. It is really nice that you can utilize the 3d printer to make parts for the cnc and vice versa :) You gained another subscriber too, have a great day!
WD40 cannot be used as a cutting agent, paraffin is better or best still soluble oil. Lubricants dull the edge of the cutter . If you lubricate a cutter it glides over and does not cut efficiently. Also your clamps need to be parallel to the table. I have been milling all my life and now retired. Just an observation to help you get better results.
Wd-40 is actually a solvent. Performance wise for me, it is a night and day difference for chatter and surface finish on aluminum. I honestly think it prevents a lot of wear especially on hss endmills.
yup... but is not using wd40 for cutting agent better than using wd40 though?... because using any oil is better than using no oil at all. or so ive heard 🤔
Hey Eric, would you be able to make a step by step guide on how you would go about making this desktop CNC? Especially for someone that is just getting into it.
Proof of concept right there. Sainsmart could learn a thing or two. I am planning for a Sainsmart Genmitsu PROVerXL 4030 to 1010 xy extension kit Upgrade with 2.2kw water cooled spindle from China. Your video has taken some of my concerns away. Great video - thank you.
When using a router, it's important to try to cut into the material direction only (as when using it for woodworking), Notice the vibration and chatter when cutting WITH the tool travel direction. That is because it is trying to cut with the shaft/tool direction. The tool will try to climb. The shaft needs to cut into the material.
Bro please please make a FULL detailed tutorial on how to either upgrade a cheap cnc or if you know of any cheap cnc that can cut metal like butter would be awesome!
If you plan on continuing to modify this little beaut of a cnc, you could make a new mount with a vacuum attachment seeing as the current one is already 3d printed.
Glad to see your work holding improved by the end, cause those inclined wingnut clamps with the bolts further away from the work made me wonder how many more times you'd lose the stock during a program.
I've also upgraded the heck out of my 3018. Had a lot of fun doing it, but it would obviously have been much smarter of me to buy a better quality CNC to begin with (or parts to build one from scratch), and much cheaper as well when you add ***everything*** up.
I'm in process of a hybrid upgrade following this video. I'm going with 2 linear rail sliders on each x axis rail but the $80 z axis. The x axis ball screw alignment was a bit challenging due to the tolerance stack but I heated up the backlash nut after everything was mounted and ran the axis back and forth to help it self align in the 3d printed mount as it cooled. The files all had to be scaled but that isn't that difficult. The thing that is less fun is that these aren't solid files and they all need to be meshed into a solid which makes them tougher to modify, but hey he did most of the work so its a small inconvenience. Next is to change over the Y axis to linear rail guides and extruded aluminum because its flimsy after the extension in comparison to the x and z now.. Thanks for the vid!
Hi Jeremy, I have posted STL files strictly for 3D printing, but if you'd like I can also upload a more generic CAD format for editing like .step or .iges. Let me know. And good luck with the build!
@@ericinventor will i face more diffoculties if want to CNC 3cm thick aluminum? The oil pump gears are small but are about 2-3 cm thick. Please let me know if you would recommend a different setup then the one in the video!
@@FirstLast-tx3yj If you take it easy with your depth of cut and stepover you should be able to handle it, just might take longer. But theres only one way to find out! I say send it!
I enjoyed watching this because I went through the same learning curve for the exact same reason ...and I was called ScooterFaster. I watched you learning ... Keep the hold downs level and very tight for example. My mill was 5 hp and when you crash or chipweld a 1/2 inch carbide even at 2000 RPM it is very exciting. One thing I couldn't tell is what specific tooling you were using. Try some premium coated 4 flute solid carbide bits. Money spent here shows up bigtime. P.S. My thing was twin (2) brushed or BLDC motors with 400 amp controllers. Fun.
Thank you for a fantastic video and for giving away your hard earned experience. I have thought about getting one of these for aluminum for years, but it is always a steep learning curve, and too much confusion, what to buy, when I want to start out cheap. I have followed your description and ordered a 3018 Pro and all the other stuff now, allthough I live in Europe, and it was more practicel for me to find the parts closer. I don´t know if the upgrade z-axis will work, it would be far to expensive to get it from Canada, so I have found one that looks kind of ok - we will see. Again thank you, now I can´t wait for the parts to come in 🙂
Using 3d printed parts is going to introduce vibration and the temp change will cause it to expand and contract. You need a good metal and solid frame. Otherwise you will have errors and small chatter.
Well I have wanted to get into CNC machines for a minute. But due to a lack of a decent budget I've avoided them altogether. My goal is similar. To use a CNC machine for a few projects (( mainly motorized bicycle disc brake adapter, a chain tensioner and other parts ))
Bro just straight went to aluminum. He said he doesnt have experience so who needs to try first some foam maybe some acrilic if youre feeling wild. My guy said "NO!! FUNK THAT! I might not have experience but I got a new motor and its going to be LIT!" Just got mine and im not patient but for under 200 dollars even I don't expect it this to be fast
I am a trained CNC machinist. I went to college for manual machining and then CNC machining. I'm also a trained welder, fabricator. I have not been machining since 2009. I was a good machinist and then the economy took a downward turn & I lost my job. I'm on disability now and I am looking to use my education to make a extra income because of the economy being what it is. I bought the 3018pro for under $200 delivered. I'm use to milling metal and not plastic or wood. I'm learning how to turn my skills into a viable income to help me with the increasing costs of living, & I am so bored with life that I could scream! Honestly, I went from high performance go-kart racing parts, Jr dragsters & lawnmower racing when that became a thing, to living a sedititary lifestyle. I was at the end of my rope, figuratively & emotionally, with loosing family members & friends to death. I was in a Very Dark place & I needed a creative outlet for my emotions & my skills of building & creating things. I'm the try that will build something that can be bought already made ready to go. But I prefer to build my own pieces of equipment to guarantee that I will have the best machine for my money. It's a lot of work to reinvent a machine, but the rewards are well worth the effort, time, expense & the knowledge that the machine is built as strong as it possibly can be, given the size of the parts that make up the machine & the components that are used in the machine itself. Like upgraded bearings, bolts, materials and everything else that you can use to make your creation better than the commercial variations that are for sale to the general public. I'm definitely not your average customer & I demand only the best in my equipment selection. I will completely redo the entire machine sometimes to add better bearings, linear rails, high grade bolts & fasteners. When I'm done with the upgrades it's almost a completely different machine. BUT MY MACHINE, WILL ALWAYS BE 2X'S-3X'S BETTER QUALITY THAN THE BEST FACTORY MODEL! I know that I could just save up the money and purchase a New Professional Grade machine, but that takes time and money, both are in short supply at the moment. So I IMPROVISE, and create the best machine that I can using what I have, can afford & what will do the job I'm asking it to do. It's a Frankenstein machine, some parts are modified to make them work with a different companies parts, but it's a TRUE PLEASURE TO SEE YOUR FRANKENSTEIN MACHINE COME TO LIFE, & BE EXACTLY WHAT YOU'RE NEEDING AT THE TIME! NOTHING BEATS THAT FEELING OF A HOME ENGINEERED, PROJECT ONE GOES THROUGH TO BUILD A BETTER WORKING MACHINE AT HOME, WHEN EVEN THE BEST COMERCIAL MACHINES HAVE DIFFICULTIES DOING, THE SAME WORK THAT YOUR HOMEGROWN, SELF-DESIGNED MACHINE IS POSSIBLE OF DOING. The PRIDE you have, knowing that you built that machine and it works, BETTER, that it's Professionally Manufactured Counterpart Machine has trouble recreating the same product. There's more than 1 way to skin a cat! The same with machinery. The Professional Companies have pretty well designed machines. They cut costs where they can so they can stay competitive in the market. I'm not interested in competing with anyone. I'm just trying to make the best machine, with the most robust components available & using them to make my machine work better and last longer because I chose to go with upgraded parts from the beginning of the build. Why wait until it starts to fail to redesign the parts so they can handle the Extra Tourqe, Rpm's & the Flex or movement of the linear rails or support pieces. Take care of the items you feel are substandard to begin with, and save yourself weeks of time and misery trying to figure out how to make this piece not flex or not do something it's not supposed to do. A little preemptive thought and engineering will make the machine work better, faster, will increase overall performance and efficiency in your parts manufacturing and the quality of your parts overall. It takes some time to make the changes to your brand new machine and then there's the cost of the improvements to your machine build. BUT in the end, IT IS ALL WORTH IT. THE COST OF THE IMPROVEMENTS WILL PAY FOR THEMSELVES IN A VERY SHORT TIME.
You are still speeding up the video when cutting. It would be really nice to see the speeds and feeds that you used. One other suggestion would be to get rid of those Y access spaghetti tube linear rails and extend the bed across the edge of the main frame rails and use linear rails on top of those. Those linear rails on the bottom bend and cause all kinds of vibration. Having a solid steel bed adds a ton of mass and really helps the finish. I'm following a lot of what you did with my 3018. At about the same pace.
Hi Timmy, at 10:50, 11:19, and 11:52, there are a few seconds of real time cutting that is not sped up. And yes replacing the Y axis using linear rails like I did with the X axis would be the next logical step for sure. I was happy with the results I was getting for now, but if I do more upgrades in the future, that would be the main next one!
I had a fet tries at my Z axis too. In the end I went with 10mm rods, rod sliders and I vacuumed a carbon fibre clad and epoxied cnced wood plate to hold my water cooled spindle motor. Only issue is my X axis is not ball screw, it's belt so it's a bit more flexible over it's 1.8m travel. Will cut aluminium but not ideal.
*Not the Hero we expected, But the one we needed" Seriously though, this answered all the right questions while still being entertaining. Homie even had the parts linked in the description to top it off. Thanks Mr. Inventor, you're a mf gentleman and a scholar sir🙏🏻🙌🏼
Hi. In your CNC V3 Final Upgrades, you refer the 400mm MGN12H Linear Rails. I have a 3018 pro and the size of linear rails is 360 mm and not 400mm. Is 400mm, the right size?
I've had a 3018 machine for a couple years but after some parts started to fail on it, it's been slowly neglected and I've focused on other options for parts. I recently got my first 3d printer and this video gave me renewed excitement for the 3018 that's collecting dust. What's your 6 month review? Is it worth the hassle?
That's really impressive that you can machine aluminium so successfully in such a cheap machine and are still using 3d printed parts and those tiny 2020 rails.
Here I thought it was the plastic arms holding the x asis. I was going to grab some c beam from open builds for x axis. I have parst from open builds for z axis. Great job, saved me some money
I'm waiting to get mine from Monoprice, interestingly right now they're selling theirs for under $200 for what appears to be basically the same machine. I was planning on using it for plastics, but if I can upgrade it for aluminum, that would open a lot of opportunities in terms of making money to afford a better machine. I'd love to have a CNC plasma cutter.
Great video. I think the problem with the V2 was the linear carriage on the Y axis, and the whole Z axis. I think the 500W motor was quite enough, and there were no problems with it. But I can't say for sure.
i have a 3018 (not pro) and have still the stock 50w (realistic speaking ~15w) motor attached and (with some other ridgidity upgrades) am able to cut aluminium, steel and even V4A without any problems (deflections ~0,1um, which some older industrial machines struggle with...) The motor only gets in trouble if you want to cut way to fast (even to fast for the endmills...) or if you want to push your mills to the absolute end of their capabilities in terms of DOC..
That is the plan! Should be easy for the z-axis adapter. For the makita motor mount I would most likely have to redesign it into two clamps that are .5in thick so they could be cut on this machine.
Thank you for the video, it is cheaper to learn from someone else's mistakes...so, thanks again, I'm already a subscriber, let us know how to save more money by showing the don'ts of cnc
A quick question - would upgrading the router have negated some of the chatter in the rails/axis? This is an inspiring video btw, about the best CNC upgrade journey I've ever seen.
Hi Chitus, yes upgrading the router helps reduce chatter for two reasons: one the higher rpm allows you to remove more material at higher speeds, so the chipload can be lower, and more importantly the lever arm (distance between mounting point and endmill tip) of the makita is shorter than that of the other motors.
I got the same cheap CNC you mentioned at the beginning. It totally can cut into aluminm. All it took was a different HSS drill bit (and it was painfully slow, feed rate of 80). However the surface looks kinda rough and I'm looking for ways to clean it up right now :p
It looks like the machine frame is pretty flimsy along the Y axis (if Y is the one the Z is mounted to). There is no diagonal bracing between the gantry supports and the beams that hold the Y rails, so the gantry will easily deflect from a rectangle to a parallelogram (as viewed looking at the front of the machine). Adding a rectangular aluminium plate at each end of the Y rails to triangulate the rails to the black vertical support elements will make it much stiffer. Stiffness is absolutely key in machine design.
Have you tried using a larger collet and tool with the 500 W spindle? it doesn't seem like the 500w motor is struggling that much, and most of the impact appears when upgrading your tool usage. Maybe I am wrong, but the increase in cutting looks like a mix between cutting tool size and your motor upgrades.
I can't see the plastic adapter on the Z axis not being at least part of the problem. That's the first thing I would have swapped out. It's got to withstand all of the X and Y forces, so I think it should be as solid as possible.
Big people with this machines are rigidity & linear shaft (rods). They are not supported throughout the movement. You had changed x and z to Lm rail and guide, it you can change in Y axis, for better results.
1 flute spiral bit will speed this thing up. I cut aluminum since 2009. Use an ethanol mist in the cutting point and the cut will stay cool. With a 4 mm bit, use RPM around 14000, feed around 1500 mm per minute. You may cut 2mm per pass I live in Brazil, and like the rest of the world, I use metrics. I know nothing about Imperial.
now use it to build a better one. using 3018 instead of 3d plastic. nice job. I got a little 9-speed mill from HF to build something like this from scratch using 80/20 extrusions and Dewalt DWP611.
I love how he has a cnc on his desk that cuts aluminum well enough and still proceeds to make parts for the cnc with the printer while knowing rigidity is the issue.
You could upgrade this a lot more now you have it by replacing all those plastic pieces with aluminum replacements, that'd make it a lot more rigid :)
and the cheap 2020 extrusions with something more rigid
I would definitely love to see a follow up on this idea. A self upgrading cnc is pretty sick
Impressive!
For machining gear-like things, I've found the best process is to first cut out the inside holes, then add more workholding in the center, cut a circle around the outside, and then start working the curvy profile. First use downward motion with the end mill to cut away most of the material inbetween each tooth, and finally do the continuous profile cutout (one rough pass, one finish pass).
The circle allows you to use a large and tough bit for full width slotting, and together with the downward motion, the small bit never has to take a heavy side load. You also get much better chip clearing compared to cutting a curvy full width slot. This procedure is particularly necessary when using very small bits in very hard materials (I came up with it when making titanium cycloidal discs with a 1/32" end mill), but it's a great speedup for aluminum too.
Interesting toolpath sequence... That sounds like it could be much faster than continuously going around the tooth profile. Ill have to try it next time! What CAM software do you use?
@@ericinventor C++ :) CAM software is outside my budget. Python would probably be easier, but C++ is what I'm used to. I also use 3D models in Blender to help visualize things.
For best results when nibbling with vertical motion, always keep the center of the bit over open space. Involute gear teeth need several bites to get all the way in to the base of the tooth.
@@dekutree64 Wow, how does one write Gcode with C++? Is there a package or tool that is made for writing gcode?
@@ericinventor Mine is all written from scratch, but now that you mention it I should clean it up and make it available for others to use :) As is, it would probably be confusing trying to understand how it's meant to be used At its lowest level it's just functions like G01_XY which takes an x,y position, applies backlash compensation, and prints a G01 command to the output file. But I also have a functions to do more complex operations with a single call, like drillHoleCircle(centerX, centerX, numHoles, circleRadius), or cutHole(centerX, centerY, radius) which will cut a helical path downward, with tool offset toward the center of the circle, and then do a finish pass with .001" or whatever finish chip thickness you have set.
Recently I had to look for someone to cnc me some brass parts for one of my projects, and the only one company that actually replied to my email gave me an astronomical price for their service, and now I found your video about buying and upgrading a cheap cnc machine on a affordable budget. With 500 bucks spend on cnc, 80 bucks on materials and a few days of work I can produce same parts much cheaper and also will be left with an amazing machine for God knows how many years. Good Sir, you are my hero! Thank you!
How well did it work out? I'm very interested to hear.
So what I got out of your video is to buy the most expensive tool you can afford... do your research and make sure that $$ is spent for what you are going to do with that tool. Use a hammer for nails, a screwdriver for screws, a drill press for holes, a saw for cutting, a mill for milling and a lathe for turning, etc.... to me, the fun part is thinking about the projects and designing them out on the drawing board. Also enjoy making my own tools, but you can just buy what others have designed for you, as you can always trade-off time for money. It's not the destination, it's the journey. Thanks for the video!
This reminds me of the upgrades I made to my 3D printer kit, where I gradually replaced each new weakest component until I had pretty much built a new, better printer, just using the original kit as a scaffolding. I've got a 3018, too, and I guess it's about time to start giving that the same treatment.
I did the same thing with my Tronxy X5S. Had military money every month and just threw funds at it. The only things left over from the original printer are some of the frame and the motors. Everthing else has been replaced and upgraded. Sad thing is, it still isn't working
A tip for hole "plunge" toolpathing, a far better method is to "helix bore" rather than plunge. 2 flute mills tend to chatter when plunging in the z axis only. If the hole diameter will allow (you will need a smaller tool to interpolate helical within the hole). Try it, I think you will find it mills much smoother. Not sure if your router has the capability to interpolate X,Y and Z , but just a general FYI Good luck !
But it is also a lot slower than plunging too. You need no backlash for it to work properly. He is just using cheap threaded rod, so I don't know if it would produce a better result.
for a cnc mill , rigidity is more important than all other factors.. The last one I made which is still use is box steel filled with concrete in a concrete base. 320 lbs , and the weak point is actually the spindle power which is 1.5 kw.anyway , excellent work finding all the weak spots in the design.
Yea. (600/60)*0.75*3.1 mm = 23.25 mm^3/s. If you put in those numbers in fswizard, it's 0.02kW... that's really just peanuts for any 1.5 kW spindle. I have a nice, granite, rigid cnc here that does 20mm deep cuts with a 6mm endmill... If you punch that in, you still only get 0.53 kW, which means that the spindle will still be fine, that is: as long as the cnc is rigid enough...
The weak spot in any cnc3018 is just... well everything... these things are just so flimpsy. Throw a back plate against it, and the little nema's won't have enough power. Upgrade the motors, and the excessive backlash will ensure that vibrations kill the router bit. Change to ball screws, and the side panels are too flimpsy to any real work. Replace the side panels with 15mm steel, and the y axis will be your next problem. Next, the motherboard is crap. The stepper drivers are crap... the list just goes on and on. At the end of the day, not much will be left of your router when you attempt to make it good...
Really, the only thing good about these small routers is that they are so much fun to toy around with. I had one and enjoyed it quite a bit.
What would you say the total cost to build youe cnc was in the end? I just got a 3020 Pro Max from Genmitsu but anticipating some upgrades pretty quickly.
@@robsproducts well u already have the main components from the 3020.. Mine cost about 750 total , But i used flat 20 mm linear rails on all 3 axis .. that was the majority of the cost ( 250-300 dollars) the concrete is dirt cheap , I used all the 6040 electronics, got a geared nema 23 for the z ( 45 dollars ) and used the 6040 spindle. The steel for the column was scrap from the garbage I picked up on the side of the road.Its probably more rigid than my 700lb benchtop mill , but these chinese spindles lack the grunt needed for heavy passes i can do manually on the mill. Hope that makes sense.. In my experience , the Z ( spindle mount ) rigidity is more important than the x aand Y, so u can go cheaper on those rails. I kept my 6040 running right to the end , to make brackets for the new machine.. then basically swapped the electronics over in one day.. so very little downtime
I wanted to say I loved your video. It was the most successful upgrade I have seen due to all of your hard work and perseverance.
I bought a 3018 max CNC and stock it can cut aluminum but slowly and tons of chatter. Pretty accurate and with mist/air setup it's amazing now. I also upgraded all three axis to MGN15H rails, now its a beast.
Could you send me a link to the parts your talking about? I would like to upgrade mine as well.
Does anybody knows what travel length does the Z axis have on this machine? I am asking ,for all of those who want to replace it with a stiffer Z-axis.. also any weight restrictions? :)
For me , linear rails on X and Z solved the "chatter" problem. The rods are just not up to the task. The 10K RPM spindle motor is the next to go - probablyy with the 30K RPM router. Thanks very much for all your input.
You want slower speeds for cutting aluminium. So you want low RPM but high torque.
That's good feedback. I've got a 6040 router I've had for 10yrs. It cuts a lot of alu & brass, but it's still on chrome rods instead of proper linear rails. I've been wondering whether the flex & chatter or the steppers are the biggest limitation to my cutting ability.
Note: upgrading to flood coolant with soluble oil made a huge difference to it's metal cutting ability. Nothing I tried spraying by hand even came close. With flood coolant, you can run a faster rpm, so you can take advantage of a high speed spindle rather than trying to cut low and slow.
Excellent video! I have this exact kit that has been sitting in my 3d printer room unassembled for the last two years. I didn't want to put it together until I had some kind of plan to make it machine aluminum more adequately, Now it can be a really useful machine that can help me to create the long lasting and useful designs that I have been wanting to make out of aluminum. So the main issues are the X rails, Z axis and the motor (I knew the stock motor was garbage from the outset). You have shown a clear path to improve the most salient points while leaving the adequate parts alone. You have earned my subscription sir.
can't wait till you move up to steel, carbide, stainless, etc.... plenty of materials to keep you busy.
@@MikelNaUsaCom Carbide isn't going to happen. Best bet is grinding carbide with diamond. Unless you can afford CBN inserts or a EDM machine.
@@appatula people now are starting to use 3d printer as edm machine, i saw also some crown found around the internet, it's quite cheap too
You need swan neck clamps to clam the metal down or the clam has to be level with the piece you are clamping. The lesson i learned from this video is to buy industrial level parts so you could use your cnc to make more parts that you need to complete the rest of the cnc machine. Great video.
I think if you stabilize the base more and get rid of the 3D printed plastic parts and replace them with metal ones, chances of chatter and vibrations are going to be very very slim.
"Have you ever seen a nicer CNC?", im a machinist by trade, i work on the real deal every day so my answer is yes xD
Start with 3018, then let the mother cnc make parts for the next stronger cnc, then make another stronger stronger cnc and so on. Finally you can mill anything u want.😊
bro i am soooooo excited i jsut bought my first cnc machine and i intend to use your upgrade ideas... i know there upgrades parts selling now on the internet so I'm looking forward to the growth and process. Thank you much fo this video!
I just discovered this channel by also turning to google for inspiration and guidance. I too have been "developing" an upgraded mini CNC to cut aluminum. I set mine up inside a Harbor Freight blast cabinet that I picked up cheap from Craigslist and modified for this project.
Subscribed Boiii!
setting up inside a sandblasting cabinet is a great idea, does it help with overall noise in addition to safety?
Thinking about upgrading my 3018 but wasnt 100% sure where to start, this is really helping out. Cause i started questioning basically everything about it, down to the torque of the steppers. Its very nice to see that the most important stuff for somewhat speedy results are the z axis and the spindle.
Cause i realy stopped using it alltogether since i kinda hit a wall already with some hardwood, only managed to go at a feedrate of around 100mm/min woth a doc of 0.25mm took me forever to do anything
I am really impressed - I never thought you can do anything with the stock nema17 steppers. First thing I did was upgrading to TB6600 and NEMA23 - but you got so much further.
Thumbs up just coz the intro! Got the heart pumping hard hahaha
To add to the other guys comments about your clamps being level (Level is best, but slightly higher is better than slightly lower). Also the clamping screw is best used as close to the part end of the clamp as you can get it. Great video, I'm thinking of getting one of these machines to play with.
Not sure if you touched on it or not, but one of the first things you should be making is replacement parts for all the support pieces that are still plastic. Especially now that you can machine aluminium easily. Replace the side supports, and also the 3-d printed backing plate. You should be replacing everything you possibly can with metal. I like that you changed out your slides for linear rails as well. Doing all your axes will make an improvement as well.
Totally agree! That's the next step!
'at a non-glacial pace' :) Got my 3018 back in 2018 i think it was, always had it in the back of my mind to upgrade the spindle in order to cut aluminium, watching this with intent!
this video is great! I have been in the same limbo with upgrading a 3018. Heres a tip with clamping your materials, Its difficult to describe, Use a packer thats is slightly higher than the material your cutting, so your clamp comes DOWN onto the material. The force exerted on clamping the material will be exponentially higher
I totaly understand what youre saying, noted! Thanks
Thanks for posting this. Wanted to say the same. Using clamps we see here can cause vibration as well!
Excellent! I'm just starting my upgrade journey with the newer Sainsmart 3020 Pro Max... My first upgrade was to add linear rails to the Y-axis and make it 400mm (now 3040). Next, I'm going to beef up the X-axis in readiness for a new Z-axis and bigger spindle motor 😃
Hey Paul, have you found a good motor to upgrade with? I’m considering upgrading the motor on my 3020 as well but I’m not sure about compatibility.
@@BadPractices Yes, I'm now using a Makita RT0700CX4. Had to design and 3D print a new 65mm mount for the router though. I have heard there's a way to start/stop the router using a solid-state relay triggered by the voltage output from the control board, but not prop[erly looked into it as yet.
@@paulblythe5899 Awesome! Is it connected to the control board in any way at the moment. If it is separated, do you just run it at full speed all the time? Thank you for answering my question by the way. I love my 3020 so I’ll be really excited to beef up the motor a bit
@@BadPractices The Makita is independent of the control board for now. It has a variable speed dial with numbers corresponding to different rpm (10K to 30K). In general, I run larger diameter bits slower, and smaller bits faster. Adjusting the speed per job/cutting bit.
The 3020 is a great little machine and I'm happy I purchased one.
@@paulblythe5899 For the past few weeks I’ve been fussing about how to integrate a new motor with the stock control board but having the motor be independent is a clever solution. Thanks for the advice. I’ll be buying a Makita spindles today. Thanks for the info, Paul!
Wow! This was well made! Stepping up the level!
I just totally blown away from result! Such a solid journey and amazing result! Thank you for sharing your valuable investigation! And yeah my 3018 is need that chrome >:D
all, an important note is that the default Z-axis stepper is not strong enough to pick up the Makita spindle. Eric probably didn't notice it as he used the Z-axis stepper received with the Z-axis he used in the middle of the video. please note.
Eric, excellent work thanks for the help
@Aaron Prat good point. The Z axis stepper is just slightly longer. Its a Nema 17 42-48
there it is 🙂
Going to try this in a year or sooner. I need to make motorcycle parts and this costs just about as much as just one of them. Good shit man
Im more impressed that you got your logo carved into the curved surface of the clamp for the spindle!
Haha thats actually 3D Printed Silver PLA!
I finally got my 3018 to cut PCBs with the stock motor. It was a mess at the start but it's really good once you get the right settings.
We definitely need a follow-up on this.
you are absolutely incredible, though i can never buy it while living it 2.5rd world country, its plesant to watch someone succeding about making aluminum cutting in a budget with high success
Thank you so much for sharing…. Totally awesome upgrades. The Makita was the key….
So I am a gun Smith and that looks like the perfect tool for what I need . Do u think this is something a new guy can do .I need it for milling out gun parts
Thank you Eric, really interesting and useful. A little engineering tip, when you clamp a piece to the worktable, the heel of the clamp should be slightly higher than the toe and the clamping force should be as close as possible to the work piece, this way you get maximum and safe downward pressure.
Seen that so many times. On bigger mills with 23 steppers the part can move if the clamps are not done right.
As a machinist and cnc programmer, I see a common issue with hobbyist using the CNC 2018 , the clamps are not being used correctly. The clamps should be parallel to the top surface with a packer piece to suit. This will greatly assist in clamping as it won't vibrate loose and break the cutter like it did in you video, with the clamps at an angle, it will come loose and move, breaking the cutter.
Noted! Thanks for the tip, a couple other guys commented this also.
Most absurd part - replacing to linear rails and beefing up everything.. than mounting it all on plastic adapters so all that added rigidity could be once again wobbly 🤒🤕
Just bought the 3020 chinese cnc router, I shall absolutely go after your ways. Shall buy a small Makita router to it, as my plans are to cut aluminum just like you, thanks for this video it confirms all my thoughts about the linear rails (roundbars) being to flexible. I been looking at linear rails for a while to modify my 3D printer with 2 lead screws, and 4 linear rails in each corner, I have the creality ender 5 s1 and getting the buildplate adjusted is a real b**** I shall add a 5th point where I can adjust the center too. It is really nice that you can utilize the 3d printer to make parts for the cnc and vice versa :)
You gained another subscriber too, have a great day!
WD40 cannot be used as a cutting agent, paraffin is better or best still soluble oil. Lubricants dull the edge of the cutter . If you lubricate a cutter it glides over and does not cut efficiently. Also your clamps need to be parallel to the table. I have been milling all my life and now retired. Just an observation to help you get better results.
Clearly you haven't. Wd-40 is paraffin. LMFAO. The fuck you talking about.
Wd-40 is actually a solvent. Performance wise for me, it is a night and day difference for chatter and surface finish on aluminum. I honestly think it prevents a lot of wear especially on hss endmills.
Yes ,you are correct, but take please advice from a qualified CNC Miller here in the aerospace industry , WD40 is “not “ designed to aid cutting.
WD40 is mostly great marketing. People use it for everything, thinking the stuff is almost Magic...
yup... but is not using wd40 for cutting agent better than using wd40 though?... because using any oil is better than using no oil at all. or so ive heard 🤔
GJ! I was also planning to upgrade mine. I wish I had seen your video sooner.
We used Relton A9 for non-ferrous metals in the machine shop. I’m an instrument maker and that’s what we used.
Hey Eric, would you be able to make a step by step guide on how you would go about making this desktop CNC? Especially for someone that is just getting into it.
Proof of concept right there. Sainsmart could learn a thing or two. I am planning for a Sainsmart Genmitsu PROVerXL 4030 to 1010 xy extension kit Upgrade with 2.2kw water cooled spindle from China. Your video has taken some of my concerns away. Great video - thank you.
When using a router, it's important to try to cut into the material direction only (as when using it for woodworking), Notice the vibration and chatter when cutting WITH the tool travel direction. That is because it is trying to cut with the shaft/tool direction. The tool will try to climb. The shaft needs to cut into the material.
Bro please please make a FULL detailed tutorial on how to either upgrade a cheap cnc or if you know of any cheap cnc that can cut metal like butter would be awesome!
If you plan on continuing to modify this little beaut of a cnc, you could make a new mount with a vacuum attachment seeing as the current one is already 3d printed.
Glad to see your work holding improved by the end, cause those inclined wingnut clamps with the bolts further away from the work made me wonder how many more times you'd lose the stock during a program.
I've also upgraded the heck out of my 3018. Had a lot of fun doing it, but it would obviously have been much smarter of me to buy a better quality CNC to begin with (or parts to build one from scratch), and much cheaper as well when you add ***everything*** up.
Awesome job. The road is the fun.
I'm going to snag one of these base machines tomorrow for $75, I'll definitely have to see how well I can beef it up with some of these upgrades.
I'm in process of a hybrid upgrade following this video. I'm going with 2 linear rail sliders on each x axis rail but the $80 z axis. The x axis ball screw alignment was a bit challenging due to the tolerance stack but I heated up the backlash nut after everything was mounted and ran the axis back and forth to help it self align in the 3d printed mount as it cooled. The files all had to be scaled but that isn't that difficult. The thing that is less fun is that these aren't solid files and they all need to be meshed into a solid which makes them tougher to modify, but hey he did most of the work so its a small inconvenience. Next is to change over the Y axis to linear rail guides and extruded aluminum because its flimsy after the extension in comparison to the x and z now.. Thanks for the vid!
Hi Jeremy, I have posted STL files strictly for 3D printing, but if you'd like I can also upload a more generic CAD format for editing like .step or .iges. Let me know. And good luck with the build!
@@ericinventor will i face more diffoculties if want to CNC 3cm thick aluminum?
The oil pump gears are small but are about 2-3 cm thick.
Please let me know if you would recommend a different setup then the one in the video!
@@FirstLast-tx3yj If you take it easy with your depth of cut and stepover you should be able to handle it, just might take longer. But theres only one way to find out! I say send it!
I enjoyed watching this because I went through the same learning curve for the exact same reason ...and I was called ScooterFaster.
I watched you learning ... Keep the hold downs level and very tight for example.
My mill was 5 hp and when you crash or chipweld a 1/2 inch carbide even at 2000 RPM it is very exciting.
One thing I couldn't tell is what specific tooling you were using. Try some premium coated 4 flute solid carbide bits. Money spent here shows up bigtime.
P.S. My thing was twin (2) brushed or BLDC motors with 400 amp controllers. Fun.
This is exactly what I wanted to do. Thank you for sharing.
loved your approach man. great mind, congratz, it's a good video
Thank you for a fantastic video and for giving away your hard earned experience.
I have thought about getting one of these for aluminum for years, but it is always a steep learning curve, and too much confusion, what to buy, when I want to start out cheap.
I have followed your description and ordered a 3018 Pro and all the other stuff now, allthough I live in Europe, and it was more practicel for me to find the parts closer.
I don´t know if the upgrade z-axis will work, it would be far to expensive to get it from Canada, so I have found one that looks kind of ok - we will see.
Again thank you, now I can´t wait for the parts to come in 🙂
Impressive patience! Good job!
And it is a hobby. Love it
2 years time well spent
Using 3d printed parts is going to introduce vibration and the temp change will cause it to expand and contract. You need a good metal and solid frame. Otherwise you will have errors and small chatter.
Well I have wanted to get into CNC machines for a minute. But due to a lack of a decent budget I've avoided them altogether. My goal is similar. To use a CNC machine for a few projects (( mainly motorized bicycle disc brake adapter, a chain tensioner and other parts ))
Bro just straight went to aluminum. He said he doesnt have experience so who needs to try first some foam maybe some acrilic if youre feeling wild. My guy said "NO!! FUNK THAT! I might not have experience but I got a new motor and its going to be LIT!" Just got mine and im not patient but for under 200 dollars even I don't expect it this to be fast
Congrats! Well done. 👍This helps me a lot for my own project. 🥰
now you can make even more upgrades, like the 3D printed things you did for the Z-axis
I am a trained CNC machinist. I went to college for manual machining and then CNC machining. I'm also a trained welder, fabricator. I have not been machining since 2009. I was a good machinist and then the economy took a downward turn & I lost my job.
I'm on disability now and I am looking to use my education to make a extra income because of the economy being what it is. I bought the 3018pro for under $200 delivered. I'm use to milling metal and not plastic or wood.
I'm learning how to turn my skills into a viable income to help me with the increasing costs of living, & I am so bored with life that I could scream! Honestly, I went from high performance go-kart racing parts, Jr dragsters & lawnmower racing when that became a thing, to living a sedititary lifestyle. I was at the end of my rope, figuratively & emotionally, with loosing family members & friends to death. I was in a Very Dark place & I needed a creative outlet for my emotions & my skills of building & creating things. I'm the try that will build something that can be bought already made ready to go. But I prefer to build my own pieces of equipment to guarantee that I will have the best machine for my money. It's a lot of work to reinvent a machine, but the rewards are well worth the effort, time, expense & the knowledge that the machine is built as strong as it possibly can be, given the size of the parts that make up the machine & the components that are used in the machine itself. Like upgraded bearings, bolts, materials and everything else that you can use to make your creation better than the commercial variations that are for sale to the general public. I'm definitely not your average customer & I demand only the best in my equipment selection. I will completely redo the entire machine sometimes to add better bearings, linear rails, high grade bolts & fasteners. When I'm done with the upgrades it's almost a completely different machine. BUT MY MACHINE, WILL ALWAYS BE 2X'S-3X'S BETTER QUALITY THAN THE BEST FACTORY MODEL!
I know that I could just save up the money and purchase a New Professional Grade machine, but that takes time and money, both are in short supply at the moment. So I IMPROVISE, and create the best machine that I can using what I have, can afford & what will do the job I'm asking it to do. It's a Frankenstein machine, some parts are modified to make them work with a different companies parts, but it's a TRUE PLEASURE TO SEE YOUR FRANKENSTEIN MACHINE COME TO LIFE, & BE EXACTLY WHAT YOU'RE NEEDING AT THE TIME!
NOTHING BEATS THAT FEELING OF A HOME ENGINEERED, PROJECT ONE GOES THROUGH TO BUILD A BETTER WORKING MACHINE AT HOME, WHEN EVEN THE BEST COMERCIAL MACHINES HAVE DIFFICULTIES DOING, THE SAME WORK THAT YOUR HOMEGROWN, SELF-DESIGNED MACHINE IS POSSIBLE OF DOING.
The PRIDE you have, knowing that you built that machine and it works, BETTER, that it's Professionally Manufactured Counterpart Machine has trouble recreating the same product.
There's more than 1 way to skin a cat! The same with machinery. The Professional Companies have pretty well designed machines. They cut costs where they can so they can stay competitive in the market. I'm not interested in competing with anyone. I'm just trying to make the best machine, with the most robust components available & using them to make my machine work better and last longer because I chose to go with upgraded parts from the beginning of the build. Why wait until it starts to fail to redesign the parts so they can handle the Extra Tourqe, Rpm's & the Flex or movement of the linear rails or support pieces. Take care of the items you feel are substandard to begin with, and save yourself weeks of time and misery trying to figure out how to make this piece not flex or not do something it's not supposed to do. A little preemptive thought and engineering will make the machine work better, faster, will increase overall performance and efficiency in your parts manufacturing and the quality of your parts overall. It takes some time to make the changes to your brand new machine and then there's the cost of the improvements to your machine build. BUT in the end, IT IS ALL WORTH IT. THE COST OF THE IMPROVEMENTS WILL PAY FOR THEMSELVES IN A VERY SHORT TIME.
You are still speeding up the video when cutting. It would be really nice to see the speeds and feeds that you used. One other suggestion would be to get rid of those Y access spaghetti tube linear rails and extend the bed across the edge of the main frame rails and use linear rails on top of those. Those linear rails on the bottom bend and cause all kinds of vibration. Having a solid steel bed adds a ton of mass and really helps the finish. I'm following a lot of what you did with my 3018. At about the same pace.
Hi Timmy, at 10:50, 11:19, and 11:52, there are a few seconds of real time cutting that is not sped up. And yes replacing the Y axis using linear rails like I did with the X axis would be the next logical step for sure. I was happy with the results I was getting for now, but if I do more upgrades in the future, that would be the main next one!
I had a fet tries at my Z axis too. In the end I went with 10mm rods, rod sliders and I vacuumed a carbon fibre clad and epoxied cnced wood plate to hold my water cooled spindle motor. Only issue is my X axis is not ball screw, it's belt so it's a bit more flexible over it's 1.8m travel. Will cut aluminium but not ideal.
*Not the Hero we expected,
But the one we needed"
Seriously though, this answered all the right questions while still being entertaining. Homie even had the parts linked in the description to top it off. Thanks Mr. Inventor, you're a mf gentleman and a scholar sir🙏🏻🙌🏼
Hi.
In your CNC V3 Final Upgrades, you refer the 400mm MGN12H Linear Rails. I have a 3018 pro and the size of linear rails is 360 mm and not 400mm. Is 400mm, the right size?
I've had a 3018 machine for a couple years but after some parts started to fail on it, it's been slowly neglected and I've focused on other options for parts. I recently got my first 3d printer and this video gave me renewed excitement for the 3018 that's collecting dust. What's your 6 month review? Is it worth the hassle?
Great video been trying to cut aluminum for a while need to do this upgrade
And in the end he won me over. Now i have to order that router
That's really impressive that you can machine aluminium so successfully in such a cheap machine and are still using 3d printed parts and those tiny 2020 rails.
amazing build, I have that router and that may have to go on my 3018 also after watching this
Great video and thank you for providing links.
Here I thought it was the plastic arms holding the x asis. I was going to grab some c beam from open builds for x axis. I have parst from open builds for z axis.
Great job, saved me some money
and this whole time while upgrading your kit, you were still using those weak nema 17 motors for x and y :D
Super entertaining. Well done.
I'm waiting to get mine from Monoprice, interestingly right now they're selling theirs for under $200 for what appears to be basically the same machine. I was planning on using it for plastics, but if I can upgrade it for aluminum, that would open a lot of opportunities in terms of making money to afford a better machine. I'd love to have a CNC plasma cutter.
Excellent progress and results. I loved that your S3D was registered to GOD 😂. Thanks for sharing. Cheers, JAYTEE
Love it! Thanks for sharing! I would have gone through the same journey. Now that you've gone on it I will build on it ;) Thanks!
Great video.
I think the problem with the V2 was the linear carriage on the Y axis, and the whole Z axis.
I think the 500W motor was quite enough, and there were no problems with it. But I can't say for sure.
i have a 3018 (not pro) and have still the stock 50w (realistic speaking ~15w) motor attached and (with some other ridgidity upgrades) am able to cut aluminium, steel and even V4A without any problems (deflections ~0,1um, which some older industrial machines struggle with...)
The motor only gets in trouble if you want to cut way to fast (even to fast for the endmills...) or if you want to push your mills to the absolute end of their capabilities in terms of DOC..
Just ordered mine, going to increase the size to 300mm by 1100mm, and use it to rotor foam for lost foam casting.
Great work. Next use the CNC to cut aluminium replacements for the plastic frame parts!
That is the plan! Should be easy for the z-axis adapter. For the makita motor mount I would most likely have to redesign it into two clamps that are .5in thick so they could be cut on this machine.
@@ericinventor btw,pcbway can now make cnc parts!
@@TheUnofficialMaker interesting, ill have to check it out, I didnt know. Are they well priced for CNC parts?
good job at the end, the cutting is smooth. Probably because the Makita has speed control.
Thank you for the video, it is cheaper to learn from someone else's mistakes...so, thanks again, I'm already a subscriber, let us know how to save more money by showing the don'ts of cnc
A quick question - would upgrading the router have negated some of the chatter in the rails/axis? This is an inspiring video btw, about the best CNC upgrade journey I've ever seen.
Hi Chitus, yes upgrading the router helps reduce chatter for two reasons: one the higher rpm allows you to remove more material at higher speeds, so the chipload can be lower, and more importantly the lever arm (distance between mounting point and endmill tip) of the makita is shorter than that of the other motors.
Could you please make a follow up video and more part making and testing such as tollerences and accuracy. Would appretiate it thanks.
You should clamp your workpeaces better, this eliminates vibrations and saves you a lot of endmills! Especially at high RPM´s
I got the same cheap CNC you mentioned at the beginning. It totally can cut into aluminm. All it took was a different HSS drill bit (and it was painfully slow, feed rate of 80). However the surface looks kinda rough and I'm looking for ways to clean it up right now :p
Today is the day I realized I need a cnc. This video is the one that made me realize it has to cut metal.
It looks like the machine frame is pretty flimsy along the Y axis (if Y is the one the Z is mounted to). There is no diagonal bracing between the gantry supports and the beams that hold the Y rails, so the gantry will easily deflect from a rectangle to a parallelogram (as viewed looking at the front of the machine). Adding a rectangular aluminium plate at each end of the Y rails to triangulate the rails to the black vertical support elements will make it much stiffer. Stiffness is absolutely key in machine design.
My 3018 is in peices at the moment, now I know how to rebuild it better. I want to machine brass.
Have you tried using a larger collet and tool with the 500 W spindle? it doesn't seem like the 500w motor is struggling that much, and most of the impact appears when upgrading your tool usage. Maybe I am wrong, but the increase in cutting looks like a mix between cutting tool size and your motor upgrades.
I can't see the plastic adapter on the Z axis not being at least part of the problem. That's the first thing I would have swapped out. It's got to withstand all of the X and Y forces, so I think it should be as solid as possible.
Very true. Its the next part that deserves to be replaced with aluminum. I will mill it on this CNC!
Big people with this machines are rigidity & linear shaft (rods). They are not supported throughout the movement. You had changed x and z to Lm rail and guide, it you can change in Y axis, for better results.
1 flute spiral bit will speed this thing up. I cut aluminum since 2009. Use an ethanol mist in the cutting point and the cut will stay cool. With a 4 mm bit, use RPM around 14000, feed around 1500 mm per minute. You may cut 2mm per pass
I live in Brazil, and like the rest of the world, I use metrics. I know nothing about Imperial.
super video, good luck :D
Nice video, thinking of buying a cnc myself. Already have a makita router
now use it to build a better one. using 3018 instead of 3d plastic. nice job. I got a little 9-speed mill from HF to build something like this from scratch using 80/20 extrusions and Dewalt DWP611.