Very inspiring! I want to entertain designing one similar to your design to see if I could build it. I like your “fly by wire” description in the next video too. Maybe one dial could control forward and backwards and instead of multiple dials for each axis. The operator hits a button on a raspberry pie that tells steppers on each axis to work together at different speed combos to move the machine in the desired path. I know nothing about electronics or computing but I bet something like that could be done somehow to move the front assembly. Maybe I’m over thinking it. Again thanks for the inspiration!
Sure would be good to see you proceed with a concept like this. You certainly do understand all the different aspects of building it and using it. Thanks for this video. Very interesting.
U could use an old table from a bridgeport to mount the headstock and tailstock to. Then u would also have a mill table in the middle. It sure looks like a pretty space/cost effective rigid machine
If I am looking at this correctly, you might have some difficulties gyroscopic forces affecting the xy travel. The forces will be a function of mass and RPM which may be difficult to calibrate out of the servo control. Functionally something like this would work, but it might be difficult to make it precise parts. Enjoy watching your channel. Thanks for sharing
It’s just an idea. The size this machine would be the servo system would have no trouble dealing with the inercial forces. This is no different than a vertical mills table with a rotary table A axis mounted on it. I think you may be referring to the X and Z axis. The vertical axis in this design is the Y axis with the B rotary axis. The C axis is the turning spindle axis. Only the rotary axis would have gyroscopic forces on them. But they rotate slowly. Now the turning spindle does rotate at high speed. But you only generate gyroscopic forces if you tip the axis of a gyroscope. This machine movement parallel and perpendicular to the rotating spindle. Swiss type screw machines move their turning spindles in a similar way to this. And they have no problem holding close tolerances. But as I said it just a idea I was thinking about. Thanks I appreciate your comment.
Awesome to see someone else contemplating actual machine tool building! I’ve been conceptualizing something with more of a vertical turning/milling lathe but I like the looks of this as well!
This got me thinking about a possible product for hobbyists. An 'integrated headstock' (spindle, chuck adapter, motor) that could fit on a Bridgeport table.
Very interesting design Peter. I might build something along these lines and if I do I'll be in touch! Main thing that I see that isnt commonly available is the spindle. Theres tons of decent spindle options for routers etc but something that can both mill and take & index fixed turning tools like your Mazak is a challenge.
Yeah, they tend to be of custom construction and they are always quite bulky. What would be more achievable is just a nice solid and repeatable spindle lock, and if you use toolholders that index repeatably too then you can use turning tools in a pre-set orientation, it's just not adjustable.
Very cool project we have been trying to figure out something similar for years. We think we will start with a small DIY horizontal mill for doing production of small parts. We have a vise we sell that is a competitor to mitee bite and need something in the production area. But man these types of machine are unbelievably expensive and there isn't a scaled down version out there like the Haas mini mill. So we are going to have a go at it. I would love to see you continue with a project like this but we know how much work they are. We have built 5 of our own DIY cnc Machines. They are a lot of work.
Peter has turned mill/turning into an artform. if the post gets traction and he is encouraged to work further on the design it will be interesting to see it the design heads. I am watching with interest.
I would love to help design and build a hobby mill like that! I just bought a hobby 3 axis mill to entertain me while my haas mills are running and I love that little machine but it makes me want a little hobby cnc lathe too, but this machine would be even better!
I like the idea of a small gantry milling machine with capto tooling. With the option of adding a lathe headstock and counterpoint. Maybe a lathe chuck with capto-backplate?
I have been working with Mitsubishi and their PLC stuff. They have been quite supportive and a lot of the CNC software is prebuilt. Gets kinda expensive though. Might be something to look into.
Nice layout! I like how easy it would be to make the B axis, it could even just be driven by a large gear or pulley that extends under the machine, plenty of room to avoid needing a precision drive. The C drive is bit annoying, I'd given it some thought a while back and I think just a large radius worm drive with the worm being adjusted further into the gear as it wears seems the most appropriate. I'm pretty sure that if materials are chosen so that only the worm will wear then you would maintain linearity and precision quite well over time, and it can easily pivot away to let the axis spin. There are types of roller cam drive that allow the cam to pivot away so the axis can be spun, and generating those cams is only a little more complex. Hard design but they are simpler to machine and grind once you are done. The roller alignment that you need is insane though.
My Mazak has a worm driven C axis. It engages by first orienting the tuning spindle, while bringing the C axis worm to it's zero position. Then the worm and it's servo drive are mounted on a pivoting arm that is tipped into the worm gear to a precise stop to control the backlash by a hydraulic cylinder. This work fine but it takes time to do it, and its rapid rate is slow. I think on a small machine like this it may be possible to just belt drive a precision servo drive with a high resolution encoder to do the C axis. Then use some kind of curvic coupling to engage and disengage it, for turning. The B axis on my Mazak is driven by a roller cam drive. The B axis only rotates about 240 degrees total so this could be done in several different simpler ways. A belt drive may also work. Or actual apposing steel bands wrapped around cylinders or driven by a linear slide and ball screw. Worm drives are slow and wear. The Mazak has a special lube system just for the C axis worm drive.
@@EdgePrecision I think belts have kinda poor properties for directly driving the C axis, even the steel core ones are a bit elastic and wear unevenly under load compared to other transmissions, only really good for transmission from a motor to some other transmission if you want precision and stiffness. Might be good enough for the B axis though if the wheel radius is very large. If it's good enough for the Mazak then I think you could get a worm drive to work well on the C axis, maybe with a higher gearing, pneumatic cylinder if you don't want hydraulics, and using the spindle drive to rough position the shaft with pulses, then a little pneumatic cylinder that pushes a bar against a flat on the shaft to quickly bring it into proper alignment, maybe 5sec max for the process? If you have the worm swing up from below then the lubrication could just be a sump that the worm sits in without the gear touching the surface when it spins on it's own, easy separate lube system.
I think definitely needs a quill. I know you loose some rigidity but having a quill certainly makes a machine much nicer to use particularly manually ibecause it give you feel for what is going on which often averts what can be disasters otherwise.
There is a slide on the top of the Y axis that will act like a quill. It could be configured to operate with a lever/handle like a Bridgeport quill. Depending on how the spindle is driven would depend on if the motor was on the top or maybe underneath. I haven't really figured all that out yet.
I really hope you build this machine! I am currently building a B axis head mill that I'd like to be able to do mill turn on. You should look into how you will set tool lengths and what controller you will use to get a B axis head to work. I have found that getting a typical hobby controller to work with a B axis head is challenging because the tool length affects the machines geometry
This sounds like a rotary (lathe axis 😁) stuck on the table of a Bridgeport mill. Just laying down. I like the more vertical design to give operators better access and visibility to the work area
Interesting axis layout. Millturns are fore sure the most versatile type of machining center. I work as a machinist but its also my hobby and I understand your desire to build your own machine because I have it too. Having a lot of weight on the x/z axis might be a problem because the motors needed to be agile and powerful enough to do 5ax milling are pretty expensive. Tool changing might be problematic as well and the type of spindle to do milling and turning also presents a challenge. I have looked for them myself and didnt find any within a reasonable budget, yours might be different than mine but its going to be expensive to build this machine. But I hope you'll find the time because it looks like an amazing and interesting project. Greetings and respect for the work and videos from Germany
It's more the machines control that needs the ability to position the tool correctly. There are quite a few Cam software's that can process the code, given the machines configuration. My Esprit software can. I thing even Fusion could do it as well. Its just a normal 5 axis configuration. Just because the whole lathe bed is moving doesn't really change anything as far as the tools movement is concerned. It may be more confusing to us, but for the control it makes no difference.
Interesting idea and not a axis arrangement I have seen before! Quick idea, if you are going with a fixed rotary tilt axis a very solid rotary axis reduction you could consider is ballscrew tilt system that DMG Mori DMU50 uses (it is also relatively cheap, especially when dealing with large tilt axis). Mind I would be very careful in the position of the tilt for the spindle as the rotary joint is far weaker compared to linear systems, so it should not be too low in the axis stack, as it then starts to influence the linear axis as well. Not to mention at the bottom of the stack the lever working on the rotary axis increases which is adding to the problem as well magnifying the angular error as the offset from center increases. Also I am guessing automation and Automatic tool changes are not a priority on this design since idea is to make it more diy? Personally one of my favorite designs for flexibility is the vertical traveling column design with the rotating head (B tilt axis) and the A rotary/lathe axis (Example Willlemin 301S2 or the DMG DMF with tilting B head, rotary C axis inbuilt in the table as well as fixed A axis lathe spindle on the side).
I think this design lets you get around a lot of the problems you normally have with a rotary axis, since you just have so much room to have the bearings far apart and have a huge worm drive gear under the table if you want. I like the ballscrew tilt idea though, it would limit you to only a bit past 90deg of accurate movement but it is rather rare to want the spindle angled that way anyhow.
I like the design, very unusual, but the rotating cutter holder at the front is certainly innovative. You could sort of hide the motor for that holder underneath the holder body, keep it out of the way. Might need bespoke tooling in some cases for the turning side of things. One interesting thing is that swarf/chips would fall more readily down underneath the lathe rather than get jammed inside the casting ribs. Was this supposed to be CNC or small/hobbyist? Would be interesting if there was a conventional version for hobbyists, or even a kit version to keep costs down. You could even put a slideway on the rotating holder so you could move that in x.
Toolchanger would be challenging.. and like toolroom lathes with the turret in front of the part, vision is partially obscured. Otherwise a very nice setup. Z axis servo would be pretty hefty to move the "lathe" for threading at higher rpm etc.
In this case I figured on manual tool changes. As far as the spindle obscuring the operators vision. You are correct, but it’s a compromise for the sake of simplicity and a more rigid configuration. And yes the X and Z servos would have to move some weight. But really no more then a Bridgeport table with vises on it.
I always wanted a scaled-up version of the PocketNC for some strange reason... I really like your idea, it would be interesting to see a proof of concept. This reminds me of the sort of thing NYCNC would do years ago. Now, the big question is - can I make most of it out of epoxy granite and use air bearings, etc? 😁 At least this is something else to add to my @dgelbart Lathe and @PiotrFoxWysocki envy...and fingers crossed for @cylosgarage to finish that diamond turning lathe...
Hey Peter, that is and interesting idea. I wonder if something more like a engine lathe with an old school versa mill head that could lock the spindle to hold turning tools and on a rotary indexer would be better/simpler? Should beat the limited x negative travel the bigger mill turns have.
That style of b axis where the tool tip is close to the center of rotation are inherently more precise. The opposite way like your mazak, where the longer the tool, the larger the minimum movement at the tip.
Great idea, I like the footprint but have a couple questions/comments. Would you be able to mill on a belt driven spindle? Or are you thinking of a brake as well? Unless you’re planning on making things totally manual I would ditch the quill. It’s one of those things that seems great but really ends up being a negative when it can be accomplished with a synthetic axis.
Put some t slots on the vertical face and you could bolt a vise to it even. I wonder about all that weight carried by the x/y though. On the one hand, weight reduces vibration. On the other, bigger linear rails. (And bigger servos.)
I don’t think it would weigh any more than a Bridgeport’s table with some vises on it. So motors that could do that would probably work for the X and Z axis. I think they use around 400-500 ounce in toque motors on a Bridgeport conversion on the X and Y axis. Again that depends on the speeds you want to move at.
Any idea on machine weight ... 4000 lb ish to 7000 lb ish ? Also wonder about spindle headstock + clearance (behind) , maybe integrated spindle idea like an integrex ? Cool idea.
The only concern I see is you’ll end up with simultaneous axis movement for any kind of drilling outside of 0 and 90 degrees to the rotation axis. I’d make the quill a requirement, it be easier than what would be require to get two axis’ to move that precisely. And for transparency that is me searching for any possible issue. It’s a really interesting arrangement with a lot of potential.
All mill turn and lathes with milling do it that way. Any holes off axis have to sink two or more axi to drill parallel to the tool. This kind of thing isn’t really a problem. My Mazak can even manually jog like that. There are switch settings on the manual jog wheel to let you do this if the B axis is tipped at some angle. If you think about it. What if you stop the machine drilling a hole at a angle. In order to jog the tool out of the hole you have to be able to manually jog straight with the tool. So this isn’t really a problem on 5 axis machines.
Very true! I was just thinking as far as a hobbyists’ machine where quality could be spared. If you start building this machine I’d love to help out with cad design and prints if you ever need it!
@@EdgePrecision Saw the second video. Love the 2 positions on the switches that allow jogging along a different work plane. Would be very cool to implement in home made machines.
My Mazak has the gear hobbing option. But to do that I would need a gear hobb for the particular gear tooth I would be cutting. Or It could be done in the normal way with gear cutters like on any milling machine. I don't really do any gear cutting on that machine. So I don't have any tooling. But the machine can do it.
its pretty close to the layout of an od grinder or a cincinati #2 tool grinder you coul also do a rack tool changer on top of the z axis or gang tooling on the y axis similar to how swiss machines get layed out
A machine with this axis configuration could very well grind cutting tools such as endmills. In fact this is very much like my CNC tool and cutter grinder flipped upside down. The only thing you would need is the software to program the tool path.
What is really not easy to implement are the required backlash-free and absolutely rigid axes during milling. What would really interest me as a hobby machine builder is how this was implemented on your Mazak? Is the A-axis of your Mazak belt-driven? On my homebuilt machine this only works with very small end mills (3mm or so) otherwise the axis starts to swing up because a belt solution is apparently not suitable for this? With a harmonic drive, on the other hand, you do not get the required speeds when turning.
There is no A axis on the Mazak. The turning Spindles Rotary axis is a C axis. This is driven by a precision Worm gear drive. The B axis on the milling spindle is driven by a roller cam drive. Both these rotary axis can resolve to a .0001 degree accuracy. There are no belts on any of this machines axis.
@@EdgePrecision Thank you for your reply and generally for the useful content of your videos. I have already learned a lot through it. If you have the time, I'd be interested in seeing a video of how Mazak has implemented the worm gear with C-axis and motor in your machine 🙂
@@9voltprojects I can describe it for you. The worm gear drive on the C axis works like this. The worm and its servo motor are mounted on an arm with its end pivot in such a location that it can engage and disengage the worm from the gear. So when the machine switches between turning and milling it first orients/zeros the turning spindle with the turning drive. At the same time it orients/zeros the C axis. Now with everything oriented is swings the the worm into the gear. The arm has hard stops to control the lash between the worm and gear. With that done it shifts the spindle's drive into neutral. I believe on this machine the arm is controlled by a hydraulic cylinder.
@@EdgePrecision Ahhh, very cleverly done 🙂 And I have so often thought about how that works, that the axis / spindle can be so extremely stiff on the one hand, but can also rotate fast for turing. Thank you very much for the description.. Have a nice weekend, Chris from Berlin
You can't really use it as a mill like you said, simply because there is no vertical axis. Everything is left/right and forward/back- but there's no up and down.
Looks like a nice machine, but what would be the target envelope like a 8inx15in part? Also what kind of easily obtainable cnc controller supports simultaneous 5 axis?
The max part size as far as just turning would be 8” dia x 24” long. But if you want to mill on the end I would say half that length. Depending on the length of the tool needed. Everything is a compromise. As to controllers I don’t know yet. I have to do some research into that.
@@EdgePrecision seems like a pretty good envelope, but i didnt see if the milling spindle can go up an down like yours can go in and out for pocketing on the circumference, did a quick search and it seems Linux cnc suports some kind of 5 axis kinematics. And i like the idea of a tool room 5 axis mill turn
My idea (I didn't mention this in the video) with this, is sort of a hybrid manual/CNC type of situation. Or maybe more accurately put servo operated axis with manual wheels/controls. Fly by wire so to speak. If this can be done. So when you have the B axis tilted at an angle you can face or drill hole patterns at that angle. The control will sink the axis necessary to make the tool move in a plane perpendicular to the tools axis in X and Y and the Z parallel to the tool as normal. This could also be used for turning. My Mazak can jog like this in the manual mode, or in the normal X,Y,Z.B,C machine coordinates. This would be very useful to do for manual machine work. Or the machine could be put into full auto mode. To better visualize what the manual mode could do. Consider a Bridgeport type of mill. When you tilt or nod the head. You can drill holes with the quill but you really cant face surfaces with mutable passes easily or drill hole patterns. This would require careful calculation to trig out the locations and moving the Knee up and down to do it. But what if the each axis of the machine was actually driven by a servo drive and you just turned pulse generator wheels and the computer figured out where to drive the movements to make it happen. In the new work plane relative too the tools axis/B axis. You can easily turn chamfers or tapers face big surfaces, mill pockets, drill patterns manually. The only thing this really lacks is the tactile feed back you get on a manual machine. But after using my Mazak this way. I think this is a better trade off. Kind of a long comment but I think you get the idea.
@@EdgePrecision If you have access to the servo controller it might be possible to provide force feedback on the control wheels. I've done some experimentation with this type of feedback, you essentially use a very small gimbal motor with an encoder as your MPG wheel and tell it how much torque the servo is putting out and you then use that data to feed a PI loop to have the small gimbal motor push back on the user's input. you can search "DIY haptic input knob" for an implementation of the video I made about it. my video is called "haptic textures and virtual detents" it is kind hard to describe, but the feedback surely works. I think the hard part is feeding with this type of knob with a constant speed. I've not used a Mazak but I'd bet they have some very good algos for ensuring that input ends up as smooth output on the machine.
Hi, Bit late to the discussion however I would like to make a suggestion to you and for anyone to have a look at a Masso CNC controller, 5 axis touch screen, developed in Australia and I believe there is a dealer in the US, Mine cost me a bit over $2000 Australian so cheaper for the US blokes, I can say the system is quality in manufacturing tho admiditly I haven't got my machine up and running yet, tho people using them seem to reckon there good hope this is useful info, Cheers.
I think they did a Kick-starter campaign. But any more I don't know, if they made any machines and soled them. Someone in another comment said they reached their goal. But have their supporters got anything?
I am working on a video on some very small parts on the Mazak. I just haven't had the right jobs for videos. They are either repeat jobs I have already done a video on or a job I cant do a video on. But I should have this video out this week. Thanks!
Once a detailed design is done, you could crowd fund to pay for a couple of staff to start manufacturing. Perhaps using your shop on a night shift basis
@@EdgePrecision Pledge US$ 2,599 or more SwissMak - 3 Axis VMC Kit This gets you the simplified 3 axis vertical mill variant of the SwissMak. Heavy boxways, 30 taper spindle and ODrive servos for all axes. ESTIMATED DELIVERY Oct 2018
Once you get your first real machines, you usually don't want to do anything with chinesium machinery ever again, except woodworking or small botchjobs, where precision doesn't matter and so that you don't get wood dust on your good iron. In a couple of years, the early 2000s 5axis machines hit the used market for cheap, 5-10k, IMO so much easier to refurbish/LinuxCNC those than to build a machine from scratch or modify chinesium to a somewhat usable tool.
Yes but even used mill turn machines are expensive and large machine's not suitable for someone's garage. But it would be nice to have similar capabilities as one, for small projects. Maybe like making small engines or things like that. But I agree with you on the Chinese machines you buy a new one and turn around and rebuild it, to do anything.
This design has huge problems, the lathe is too small and so the machine is not universal. And it has structural issues. Tsugami TMU1 has a design like that but the lathe part is fixed and facing down like it should be. Grob makes some upside down millturns, Index has sliding bed design. But all these machines are cnc chuckers. A universal lathe must be longer even though it can accodate many milling operations. Similarly, deckel FPU type mills (euromills) can also turn parts, but lose their versatility for machining gears and worms if you changed the design. Also way more handy than any of the points you mentioned is the ability to hang huge parts on the table (and the other side by a crane) to mill only a small feature - like the boring of the articulation of an excavator arm. Ther'es a guy on youtube who built a 4 foot press brake on a deckel like the one you showed. On your design this would not be possible. Btw his channel is excellent, his press brake is for making precision angles. If this is too hard to understand, consider this --- you shipped one of these machines to third world india, would the people have use for them to fix their broken down tractors and trucks? Have you seen a third worlder turn a 1 meter part on a 500mm lathe? They put the headstock on a block to accomodate their part. They can extend the bed easily to accomodate longer parts. This is universality. The gap in the gap bed lathe.
The whole purpose of this machine is not to do what you are talking about in this comment. Get a bigger machine. It is not intended for that. You don't turn a tractor part on a watch makers lathe.
@@EdgePrecision The whole purpose of a machine is to use the smallest machine to get the job done. Floor space and iron cost money. Its all about the money.
So what are you optimizing for? Floor space? Money? Who is it targetted towards? And how is it going to turn straight parts without weighing a ton? Because i can see FEA issues from the way things are layed out.
Very inspiring! I want to entertain designing one similar to your design to see if I could build it. I like your “fly by wire” description in the next video too. Maybe one dial could control forward and backwards and instead of multiple dials for each axis. The operator hits a button on a raspberry pie that tells steppers on each axis to work together at different speed combos to move the machine in the desired path. I know nothing about electronics or computing but I bet something like that could be done somehow to move the front assembly. Maybe I’m over thinking it. Again thanks for the inspiration!
seems pretty similar to the setup on a universal tool and cutter grinder, like a cross between a cincinatti #2 and Monoset, definitely a cool design
Sure would be good to see you proceed with a concept like this. You certainly do understand all the different aspects of building it and using it. Thanks for this video. Very interesting.
Talk to the folks at Taig Tools in Arizona. They make small machines in the
This is a really interesting idea!
U could use an old table from a bridgeport to mount the headstock and tailstock to. Then u would also have a mill table in the middle. It sure looks like a pretty space/cost effective rigid machine
If I am looking at this correctly, you might have some difficulties gyroscopic forces affecting the xy travel. The forces will be a function of mass and RPM which may be difficult to calibrate out of the servo control.
Functionally something like this would work, but it might be difficult to make it precise parts.
Enjoy watching your channel. Thanks for sharing
It’s just an idea. The size this machine would be the servo system would have no trouble dealing with the inercial forces. This is no different than a vertical mills table with a rotary table A axis mounted on it. I think you may be referring to the X and Z axis. The vertical axis in this design is the Y axis with the B rotary axis. The C axis is the turning spindle axis. Only the rotary axis would have gyroscopic forces on them. But they rotate slowly. Now the turning spindle does rotate at high speed. But you only generate gyroscopic forces if you tip the axis of a gyroscope. This machine movement parallel and perpendicular to the rotating spindle. Swiss type screw machines move their turning spindles in a similar way to this. And they have no problem holding close tolerances. But as I said it just a idea I was thinking about. Thanks I appreciate your comment.
Awesome to see someone else contemplating actual machine tool building! I’ve been conceptualizing something with more of a vertical turning/milling lathe but I like the looks of this as well!
What a great idea to share your experience working with all these types of machines and parts.
Thanks a lot !
Interesting idea
Looks like a great idea Peter. Have you considered polymer concrete for the machine frame? Relatively cheap and not to hard to do.
Mark
Seems like a good project for when you move to Mexico
This got me thinking about a possible product for hobbyists. An 'integrated headstock' (spindle, chuck adapter, motor) that could fit on a Bridgeport table.
I think that’s a very interesting idea. I am heavily constrained for space an I would love a machine like that.
Very interesting design Peter. I might build something along these lines and if I do I'll be in touch! Main thing that I see that isnt commonly available is the spindle. Theres tons of decent spindle options for routers etc but something that can both mill and take & index fixed turning tools like your Mazak is a challenge.
Yeah, they tend to be of custom construction and they are always quite bulky. What would be more achievable is just a nice solid and repeatable spindle lock, and if you use toolholders that index repeatably too then you can use turning tools in a pre-set orientation, it's just not adjustable.
Very cool project we have been trying to figure out something similar for years. We think we will start with a small DIY horizontal mill for doing production of small parts. We have a vise we sell that is a competitor to mitee bite and need something in the production area. But man these types of machine are unbelievably expensive and there isn't a scaled down version out there like the Haas mini mill. So we are going to have a go at it. I would love to see you continue with a project like this but we know how much work they are. We have built 5 of our own DIY cnc Machines. They are a lot of work.
good video peter..thanks for your time
I want one! Very nice design.
Nice.. You could do a lot of custom attachments for that... I'd love a tool-grinding attachment.
Yes it could be configured as a tool and cutter grinder.
Peter has turned mill/turning into an artform. if the post gets traction and he is encouraged to work further on the design it will be interesting to see it the design heads. I am watching with interest.
I would love to help design and build a hobby mill like that! I just bought a hobby 3 axis mill to entertain me while my haas mills are running and I love that little machine but it makes me want a little hobby cnc lathe too, but this machine would be even better!
Cool idea! I like it because is compact but you can use it in so many ways. Also as you mentioned manual or servo driven CNC like or maybe hybrid
I like the idea of a small gantry milling machine with capto tooling. With the option of adding a lathe headstock and counterpoint. Maybe a lathe chuck with capto-backplate?
I have been working with Mitsubishi and their PLC stuff. They have been quite supportive and a lot of the CNC software is prebuilt. Gets kinda expensive though. Might be something to look into.
If you want help with some of the design cad let me know. Your videos have been very helpful so for this I will work for free.
Nice layout! I like how easy it would be to make the B axis, it could even just be driven by a large gear or pulley that extends under the machine, plenty of room to avoid needing a precision drive.
The C drive is bit annoying, I'd given it some thought a while back and I think just a large radius worm drive with the worm being adjusted further into the gear as it wears seems the most appropriate. I'm pretty sure that if materials are chosen so that only the worm will wear then you would maintain linearity and precision quite well over time, and it can easily pivot away to let the axis spin.
There are types of roller cam drive that allow the cam to pivot away so the axis can be spun, and generating those cams is only a little more complex. Hard design but they are simpler to machine and grind once you are done. The roller alignment that you need is insane though.
My Mazak has a worm driven C axis. It engages by first orienting the tuning spindle, while bringing the C axis worm to it's zero position. Then the worm and it's servo drive are mounted on a pivoting arm that is tipped into the worm gear to a precise stop to control the backlash by a hydraulic cylinder. This work fine but it takes time to do it, and its rapid rate is slow. I think on a small machine like this it may be possible to just belt drive a precision servo drive with a high resolution encoder to do the C axis. Then use some kind of curvic coupling to engage and disengage it, for turning. The B axis on my Mazak is driven by a roller cam drive. The B axis only rotates about 240 degrees total so this could be done in several different simpler ways. A belt drive may also work. Or actual apposing steel bands wrapped around cylinders or driven by a linear slide and ball screw. Worm drives are slow and wear. The Mazak has a special lube system just for the C axis worm drive.
@@EdgePrecision I think belts have kinda poor properties for directly driving the C axis, even the steel core ones are a bit elastic and wear unevenly under load compared to other transmissions, only really good for transmission from a motor to some other transmission if you want precision and stiffness. Might be good enough for the B axis though if the wheel radius is very large.
If it's good enough for the Mazak then I think you could get a worm drive to work well on the C axis, maybe with a higher gearing, pneumatic cylinder if you don't want hydraulics, and using the spindle drive to rough position the shaft with pulses, then a little pneumatic cylinder that pushes a bar against a flat on the shaft to quickly bring it into proper alignment, maybe 5sec max for the process? If you have the worm swing up from below then the lubrication could just be a sump that the worm sits in without the gear touching the surface when it spins on it's own, easy separate lube system.
Having the mill head turn 135 degrees would be nice.. -45 to the spindle, and parallel to the spindle for face drilling of bolt patterns
I think definitely needs a quill. I know you loose some rigidity but having a quill certainly makes a machine much nicer to use particularly manually ibecause it give you feel for what is going on which often averts what can be disasters otherwise.
There is a slide on the top of the Y axis that will act like a quill. It could be configured to operate with a lever/handle like a Bridgeport quill. Depending on how the spindle is driven would depend on if the motor was on the top or maybe underneath. I haven't really figured all that out yet.
I really hope you build this machine! I am currently building a B axis head mill that I'd like to be able to do mill turn on. You should look into how you will set tool lengths and what controller you will use to get a B axis head to work. I have found that getting a typical hobby controller to work with a B axis head is challenging because the tool length affects the machines geometry
This sounds like a rotary (lathe axis 😁) stuck on the table of a Bridgeport mill. Just laying down. I like the more vertical design to give operators better access and visibility to the work area
Could be an interesting project Peter, maybe some kind of kit or plans?
interresting, might be worth exploring the idea further.
Damn I would love to see you build this🎉
I hope someone partners with you and brings this to market.
Interesting axis layout. Millturns are fore sure the most versatile type of machining center. I work as a machinist but its also my hobby and I understand your desire to build your own machine because I have it too. Having a lot of weight on the x/z axis might be a problem because the motors needed to be agile and powerful enough to do 5ax milling are pretty expensive. Tool changing might be problematic as well and the type of spindle to do milling and turning also presents a challenge. I have looked for them myself and didnt find any within a reasonable budget, yours might be different than mine but its going to be expensive to build this machine.
But I hope you'll find the time because it looks like an amazing and interesting project.
Greetings and respect for the work and videos from Germany
Cool idea, the only downside is programming it would be a challenge as there would be no software that could calculate the tool center point.
It's more the machines control that needs the ability to position the tool correctly. There are quite a few Cam software's that can process the code, given the machines configuration. My Esprit software can. I thing even Fusion could do it as well. Its just a normal 5 axis configuration. Just because the whole lathe bed is moving doesn't really change anything as far as the tools movement is concerned. It may be more confusing to us, but for the control it makes no difference.
Very interesting!
Interesting idea and not a axis arrangement I have seen before!
Quick idea, if you are going with a fixed rotary tilt axis a very solid rotary axis reduction you could consider is ballscrew tilt system that DMG Mori DMU50 uses (it is also relatively cheap, especially when dealing with large tilt axis).
Mind I would be very careful in the position of the tilt for the spindle as the rotary joint is far weaker compared to linear systems, so it should not be too low in the axis stack, as it then starts to influence the linear axis as well.
Not to mention at the bottom of the stack the lever working on the rotary axis increases which is adding to the problem as well magnifying the angular error as the offset from center increases.
Also I am guessing automation and Automatic tool changes are not a priority on this design since idea is to make it more diy?
Personally one of my favorite designs for flexibility is the vertical traveling column design with the rotating head (B tilt axis) and the A rotary/lathe axis (Example Willlemin 301S2 or the DMG DMF with tilting B head, rotary C axis inbuilt in the table as well as fixed A axis lathe spindle on the side).
I think this design lets you get around a lot of the problems you normally have with a rotary axis, since you just have so much room to have the bearings far apart and have a huge worm drive gear under the table if you want. I like the ballscrew tilt idea though, it would limit you to only a bit past 90deg of accurate movement but it is rather rare to want the spindle angled that way anyhow.
I like the design, very unusual, but the rotating cutter holder at the front is certainly innovative. You could sort of hide the motor for that holder underneath the holder body, keep it out of the way. Might need bespoke tooling in some cases for the turning side of things. One interesting thing is that swarf/chips would fall more readily down underneath the lathe rather than get jammed inside the casting ribs.
Was this supposed to be CNC or small/hobbyist?
Would be interesting if there was a conventional version for hobbyists, or even a kit version to keep costs down. You could even put a slideway on the rotating holder so you could move that in x.
I dig it!
Toolchanger would be challenging.. and like toolroom lathes with the turret in front of the part, vision is partially obscured. Otherwise a very nice setup. Z axis servo would be pretty hefty to move the "lathe" for threading at higher rpm etc.
In this case I figured on manual tool changes. As far as the spindle obscuring the operators vision. You are correct, but it’s a compromise for the sake of simplicity and a more rigid configuration. And yes the X and Z servos would have to move some weight. But really no more then a Bridgeport table with vises on it.
@@EdgePrecision Would love to see somebody take a stab at it. Go with a CAT or BT spindle at the very least. R-8 has no business here.
@@ilikec don't forget 5c.. But that setup is cool because from the lathe chuck to the mill head could use every tool type of holder around.
@@Jeralddoerr for the "lathe" spindle, 5C or 16C would be nice. Maybe a A2-5 nose. If the tailstock was live, even better.
I always wanted a scaled-up version of the PocketNC for some strange reason... I really like your idea, it would be interesting to see a proof of concept. This reminds me of the sort of thing NYCNC would do years ago. Now, the big question is - can I make most of it out of epoxy granite and use air bearings, etc? 😁 At least this is something else to add to my @dgelbart Lathe and @PiotrFoxWysocki envy...and fingers crossed for @cylosgarage to finish that diamond turning lathe...
I want one.
Hey Peter, that is and interesting idea. I wonder if something more like a engine lathe with an old school versa mill head that could lock the spindle to hold turning tools and on a rotary indexer would be better/simpler? Should beat the limited x negative travel the bigger mill turns have.
Would be easy to build I too
Have come up with similar ideas
That style of b axis where the tool tip is close to the center of rotation are inherently more precise. The opposite way like your mazak, where the longer the tool, the larger the minimum movement at the tip.
A Bridgeport series two head would make a great candidate for a mill spindle assy...
Flimsy spindle and no ATC capability
@@MF175mp mine are cat 40 spindles.. nor sure how flimsy that is..
Great idea, I like the footprint but have a couple questions/comments. Would you be able to mill on a belt driven spindle? Or are you thinking of a brake as well? Unless you’re planning on making things totally manual I would ditch the quill. It’s one of those things that seems great but really ends up being a negative when it can be accomplished with a synthetic axis.
Put some t slots on the vertical face and you could bolt a vise to it even.
I wonder about all that weight carried by the x/y though. On the one hand, weight reduces vibration. On the other, bigger linear rails. (And bigger servos.)
I don’t think it would weigh any more than a Bridgeport’s table with some vises on it. So motors that could do that would probably work for the X and Z axis. I think they use around 400-500 ounce in toque motors on a Bridgeport conversion on the X and Y axis. Again that depends on the speeds you want to move at.
@@EdgePrecision True! Scraped ways though, which are cheaper for the load.
Anyway -- I'd love to see this get built!
Any idea on machine weight ... 4000 lb ish to 7000 lb ish ? Also wonder about spindle headstock + clearance (behind) , maybe integrated spindle idea like an integrex ? Cool idea.
The only concern I see is you’ll end up with simultaneous axis movement for any kind of drilling outside of 0 and 90 degrees to the rotation axis.
I’d make the quill a requirement, it be easier than what would be require to get two axis’ to move that precisely.
And for transparency that is me searching for any possible issue. It’s a really interesting arrangement with a lot of potential.
All mill turn and lathes with milling do it that way. Any holes off axis have to sink two or more axi to drill parallel to the tool. This kind of thing isn’t really a problem. My Mazak can even manually jog like that. There are switch settings on the manual jog wheel to let you do this if the B axis is tipped at some angle. If you think about it. What if you stop the machine drilling a hole at a angle. In order to jog the tool out of the hole you have to be able to manually jog straight with the tool. So this isn’t really a problem on 5 axis machines.
Very true! I was just thinking as far as a hobbyists’ machine where quality could be spared.
If you start building this machine I’d love to help out with cad design and prints if you ever need it!
@@EdgePrecision Saw the second video. Love the 2 positions on the switches that allow jogging along a different work plane. Would be very cool to implement in home made machines.
It would also Interesting for you to experiment on Gear Cutting Gcodes on your Mazak
My Mazak has the gear hobbing option. But to do that I would need a gear hobb for the particular gear tooth I would be cutting. Or It could be done in the normal way with gear cutters like on any milling machine. I don't really do any gear cutting on that machine. So I don't have any tooling. But the machine can do it.
Hey Peter, you taking preorders? :)
its pretty close to the layout of an od grinder or a cincinati #2 tool grinder
you coul also do a rack tool changer on top of the z axis or gang tooling on the y axis similar to how swiss machines get layed out
A machine with this axis configuration could very well grind cutting tools such as endmills. In fact this is very much like my CNC tool and cutter grinder flipped upside down. The only thing you would need is the software to program the tool path.
What is really not easy to implement are the required backlash-free and absolutely rigid axes during milling. What would really interest me as a hobby machine builder is how this was implemented on your Mazak? Is the A-axis of your Mazak belt-driven? On my homebuilt machine this only works with very small end mills (3mm or so) otherwise the axis starts to swing up because a belt solution is apparently not suitable for this? With a harmonic drive, on the other hand, you do not get the required speeds when turning.
There is no A axis on the Mazak. The turning Spindles Rotary axis is a C axis. This is driven by a precision Worm gear drive. The B axis on the milling spindle is driven by a roller cam drive. Both these rotary axis can resolve to a .0001 degree accuracy. There are no belts on any of this machines axis.
@@EdgePrecision Thank you for your reply and generally for the useful content of your videos. I have already learned a lot through it.
If you have the time, I'd be interested in seeing a video of how Mazak has implemented the worm gear with C-axis and motor in your machine 🙂
@@9voltprojects I can describe it for you. The worm gear drive on the C axis works like this. The worm and its servo motor are mounted on an arm with its end pivot in such a location that it can engage and disengage the worm from the gear. So when the machine switches between turning and milling it first orients/zeros the turning spindle with the turning drive. At the same time it orients/zeros the C axis. Now with everything oriented is swings the the worm into the gear. The arm has hard stops to control the lash between the worm and gear. With that done it shifts the spindle's drive into neutral. I believe on this machine the arm is controlled by a hydraulic cylinder.
@@EdgePrecision Ahhh, very cleverly done 🙂 And I have so often thought about how that works, that the axis / spindle can be so extremely stiff on the one hand, but can also rotate fast for turing.
Thank you very much for the description..
Have a nice weekend,
Chris from Berlin
You can't really use it as a mill like you said, simply because there is no vertical axis. Everything is left/right and forward/back- but there's no up and down.
Yes there is. The B tilting axis has a vertical slide way.
@@EdgePrecision Oh, I see it now.
very original I've never seen before, and then what about software....i wonder if Mach could work
Linux cnc can do this, I think.
Looks like a nice machine, but what would be the target envelope like a 8inx15in part? Also what kind of easily obtainable cnc controller supports simultaneous 5 axis?
The max part size as far as just turning would be 8” dia x 24” long. But if you want to mill on the end I would say half that length. Depending on the length of the tool needed. Everything is a compromise. As to controllers I don’t know yet. I have to do some research into that.
@@EdgePrecision seems like a pretty good envelope, but i didnt see if the milling spindle can go up an down like yours can go in and out for pocketing on the circumference, did a quick search and it seems Linux cnc suports some kind of 5 axis kinematics. And i like the idea of a tool room 5 axis mill turn
Levil Technology has a very small mill-turn machine that may give you some ideas.
I was unaware of this machine. I went to their UA-cam channel. Interesting small machines.
Hey Peter, are you taking preorders?
the big problem is availability of 4+ axis CAM for the hobbyist
My idea (I didn't mention this in the video) with this, is sort of a hybrid manual/CNC type of situation. Or maybe more accurately put servo operated axis with manual wheels/controls. Fly by wire so to speak. If this can be done. So when you have the B axis tilted at an angle you can face or drill hole patterns at that angle. The control will sink the axis necessary to make the tool move in a plane perpendicular to the tools axis in X and Y and the Z parallel to the tool as normal. This could also be used for turning. My Mazak can jog like this in the manual mode, or in the normal X,Y,Z.B,C machine coordinates. This would be very useful to do for manual machine work. Or the machine could be put into full auto mode. To better visualize what the manual mode could do. Consider a Bridgeport type of mill. When you tilt or nod the head. You can drill holes with the quill but you really cant face surfaces with mutable passes easily or drill hole patterns. This would require careful calculation to trig out the locations and moving the Knee up and down to do it. But what if the each axis of the machine was actually driven by a servo drive and you just turned pulse generator wheels and the computer figured out where to drive the movements to make it happen. In the new work plane relative too the tools axis/B axis. You can easily turn chamfers or tapers face big surfaces, mill pockets, drill patterns manually. The only thing this really lacks is the tactile feed back you get on a manual machine. But after using my Mazak this way. I think this is a better trade off. Kind of a long comment but I think you get the idea.
@@EdgePrecision If you have access to the servo controller it might be possible to provide force feedback on the control wheels. I've done some experimentation with this type of feedback, you essentially use a very small gimbal motor with an encoder as your MPG wheel and tell it how much torque the servo is putting out and you then use that data to feed a PI loop to have the small gimbal motor push back on the user's input. you can search "DIY haptic input knob" for an implementation of the video I made about it. my video is called "haptic textures and virtual detents" it is kind hard to describe, but the feedback surely works. I think the hard part is feeding with this type of knob with a constant speed. I've not used a Mazak but I'd bet they have some very good algos for ensuring that input ends up as smooth output on the machine.
Sorry, but is this video also available in a higher resolution than 240p?
Sorry It may not be. I think I didn't pay attention to the settings when uploading it.
Hi, Bit late to the discussion however I would like to make a suggestion to you and for anyone to have a look at a Masso CNC controller, 5 axis touch screen, developed in Australia and I believe there is a dealer in the US, Mine cost me a bit over $2000 Australian so cheaper for the US blokes, I can say the system is quality in manufacturing tho admiditly I haven't got my machine up and running yet, tho people using them seem to reckon there good hope this is useful info, Cheers.
I took a look at their web site. You are correct the touch control here for 5 axis cost is around $1,350.00.
Looks kinda like a liechti turbomill for Turbine Blades
was Swiss-Mac are real machine?, i dont think the inverntor produced any Swiss=Mac machines for sale, just a prototype.
I think they did a Kick-starter campaign. But any more I don't know, if they made any machines and soled them. Someone in another comment said they reached their goal. But have their supporters got anything?
Is everything ok mr. Peter ?, i've not seen any activity on the channel lately .
I am working on a video on some very small parts on the Mazak. I just haven't had the right jobs for videos. They are either repeat jobs I have already done a video on or a job I cant do a video on. But I should have this video out this week. Thanks!
@@EdgePrecision It's good to hear that you're ok :) .
Once a detailed design is done, you could crowd fund to pay for a couple of staff to start manufacturing. Perhaps using your shop on a night shift basis
I hope you get rich with your idea, maybe you can put swissmak or haas out of business. Or even sell a handful of tabletop machines.
Has Swissmak even sold any machines?
@@EdgePrecision They met their desired pledge on kickstarter of 120,000USD at a value of 187,000USD
@@EdgePrecision Pledge US$ 2,599 or more
SwissMak - 3 Axis VMC Kit
This gets you the simplified 3 axis vertical mill variant of the SwissMak. Heavy boxways, 30 taper spindle and ODrive servos for all axes.
ESTIMATED DELIVERY
Oct 2018
If you need CAD models for your prototype please hit me up I can assist you...for as long as you going to build the machine
Be a manifector of small hobby lathe. Make good money.
Once you get your first real machines, you usually don't want to do anything with chinesium machinery ever again, except woodworking or small botchjobs, where precision doesn't matter and so that you don't get wood dust on your good iron.
In a couple of years, the early 2000s 5axis machines hit the used market for cheap, 5-10k, IMO so much easier to refurbish/LinuxCNC those than to build a machine from scratch or modify chinesium to a somewhat usable tool.
Yes but even used mill turn machines are expensive and large machine's not suitable for someone's garage. But it would be nice to have similar capabilities as one, for small projects. Maybe like making small engines or things like that. But I agree with you on the Chinese machines you buy a new one and turn around and rebuild it, to do anything.
This design has huge problems, the lathe is too small and so the machine is not universal. And it has structural issues. Tsugami TMU1 has a design like that but the lathe part is fixed and facing down like it should be. Grob makes some upside down millturns, Index has sliding bed design. But all these machines are cnc chuckers. A universal lathe must be longer even though it can accodate many milling operations. Similarly, deckel FPU type mills (euromills) can also turn parts, but lose their versatility for machining gears and worms if you changed the design. Also way more handy than any of the points you mentioned is the ability to hang huge parts on the table (and the other side by a crane) to mill only a small feature - like the boring of the articulation of an excavator arm. Ther'es a guy on youtube who built a 4 foot press brake on a deckel like the one you showed. On your design this would not be possible. Btw his channel is excellent, his press brake is for making precision angles. If this is too hard to understand, consider this --- you shipped one of these machines to third world india, would the people have use for them to fix their broken down tractors and trucks? Have you seen a third worlder turn a 1 meter part on a 500mm lathe? They put the headstock on a block to accomodate their part. They can extend the bed easily to accomodate longer parts. This is universality. The gap in the gap bed lathe.
The whole purpose of this machine is not to do what you are talking about in this comment. Get a bigger machine. It is not intended for that. You don't turn a tractor part on a watch makers lathe.
@@EdgePrecision The whole purpose of a machine is to use the smallest machine to get the job done. Floor space and iron cost money. Its all about the money.
So what are you optimizing for? Floor space? Money? Who is it targetted towards? And how is it going to turn straight parts without weighing a ton? Because i can see FEA issues from the way things are layed out.