[Bartuss1]

What for 6 axis in this small machine?

[ishi]

What for 6 axis in this small machine?

To learn about 6-axis kinematics with a small and "affordable" machine.
To get 2 pairs of 3 axes fully decoupled from each other.
To be able to use it as an HMC or VMC.

[ishi]

I have added some high-resolution drawings on the public drive:

https://drive.google.com/drive/u/3/f...3dQrx?ogsrc=32

They will be updated on a regular basis, but not necessarily at every design iteration.

[ishi]

It looks like the SYK MBB25L-S motor mount will fit our Siemens SIMOTICS 1FK7060 servo motors perfectly.

Sweet!

[RCaffin]

I am a bit puzzled by your labeling of the axes. Every other CNC I have seen and every text book I have seen has the X & Y axes in the horizontal plane and the Z axis going up and down. But you seem to be calling the second horizontal axis the Z axis, and that is going to clash, very badly, with every text book and every CAD/CAM system. Can I ask why?

Cheers
Roger

[ishi]

I am a bit puzzled by your labeling of the axes. Every other CNC I have seen and every text book I have seen has the X & Y axes in the horizontal plane and the Z axis going up and down. But you seem to be calling the second horizontal axis the Z axis, and that is going to clash, very badly, with every text book and every CAD/CAM system. Can I ask why?

Cheers
Roger

Because this is a horizontal machining center, and my understanding is that the vertical axis in a Horizontal Machining Center is labeled Y. This is due to the fact that (as far as I can tell), Z is reserved for the axis that is parallel to the spindle's axis of rotation. I know it's a bit confusing at first, but the more I thought about it, the more it made sense. Once you realize that your spindle could be oriented in any direction depending on the machine's design or configuration, settling on the convention that Z is always parallel to the spindle's axis clears any possible confusion. Here is an example using this labeling convention:

https://www.okuma.co.jp/english/product/hmc/index5.html

And here is an article confirming this convention:

https://www.linkedin.com/pulse/coord...eron-anderson/

Obviously, X goes with A, Y with B, and Z with C. Which is also why HSD labels the two axes of its spindle head A and C by the way...

And that's why the table on a Horizontal Machining Center is called an X-Z table, unlike the X-Y table of a Vertical Machining Center. Exact same tables, but different labels, because of the different orientations of the spindle. Therefore, the axis of rotation for the rotary table on my machine is B, with A and C used by the spindle's 2-axis head.

Clearly, Horizontal Machining Centers are a bit of an acquired taste, but I think I've fallen in love with the configuration. I used to think that vertical or horizontal was a minor detail, but it's actually a major difference, for very fundamental reasons. There are so many things going for it, mostly driven by the fact that gravity is always pointing down. Having your default spindle axis orthogonal to gravity's direction solves a lot of problems actually, like the clearing of chips within milled cavities. And once you go for the 3+3 design that I have selected with two rotary axes directly attached to the tool, you can make the machine very rigid. I do not know any other configuration of 6 axes that would come even close to that.

[ishi]

Since true 6-axis machines are not that common, I would like to share some general thoughts on the design of such a machine. This is on the theoretical side, but it might help justify some design decisions for the machine that we are building.

The first idea is that 6 axes is as close to the "ideal" number of axes that you can get, because it allows you to go anywhere in space, from any angle. Working in a three dimensional space, three linear axes that are all orthogonal to each other let you reach any point within your envelope. From there, if you add three rotary axes, you can reach any point from any angle. But what's really important here is that all three linear axes must be orthogonal to each other, and all three rotary axes must be orthogonal to each other. If two linear axes are parallel to each other, you do not have a "true" 6-axis machine, and the same goes if two rotary axes are parallel to each other. Adding more axes that are parallel to each other might help you use different tools or do multiple cuts at the same time, but they won't improve the essential kinematics of the machine. In order to get ideal kinematics, you need 6 axis, you need all linear and rotary axes to be orthogonal with each other, and you do not need any additional axis (there is no such thing as a "true" 7-axis machine).

The second idea is that these axes break down into two groups: the ones that are attached to your piece, and the one that are attached to your tool. This is a very simple idea, but it took me a long time to appreciate its essential nature. The idea is that axes don't float in space, they have to be attached to something. Of course, all axes are attached to your machine, but the movement of an axis will either affect your piece or your tool. It cannot affect both at the same time, or neither of them. Otherwise, it's not a machine axis. Granted, your machine is somewhere on Earth and the Earth is rotating around its own axis and around the Sun, but these movements do not really affect neither your piece nor your tool.

From that point on, when designing a machine, you really have to think about these two groups of axes. And if you're designing a 6-axis machine, you will want to have 3 axes within one group and 3 axes within the other. The reason for that is pretty simple: adding axes on top of each other will compound positioning errors, for many different mechanical and electro-mechanical reasons. Therefore, 3+3 is usually better than 4+2, which is better than 5+1, which is better than 6+0. What would be a 6+0 machine? A 6-axis robotic arm with a tool at the end working on a static piece. It's possible to make such a machine, and it's actually a good design for some applications that require a huge envelope and a relatively low accuracy, for example when making parts of a rocket engine out of titanium using laser sintering. But for applications that require accuracy, this is a non-starter.

So, now we know that we want a 3+3 configuration, but we don't know how our 3 linear axes and 3 rotary axes should be distributed across these two groups (piece and tool). In order to find out, there is one major factor to consider: in general, the piece (and its fixturing) is heavier than the tool (22kg spindle in our machine). As a result, you will want to attach the more demanding axes to the tool, not the piece. By demanding, I mean axes that require complex and heavy components. To understand what that means, you just have to compare the size and weight of a 2-axis trunnion table on which you attach your piece, to the size and weight of a 2-axis head on which you attach your tool (spindle). When you do so, you will realize that the former is usually 3 to 5 times heavier than the latter. Therefore, a 2-axis head is preferable to a 2-axis trunnion table if all other things are equal.

Why does the weight of rotary axes matter so much? Because rotary axes are usually mounted onto linear axes, not the other way around. Granted, you could design your machine the other way around, but it would be massive, for a very small envelope. This is something that we take for granted, but taking it into consideration helps you understand why machines are built the way they are.

With that in mind, you will try to attach at least two rotary axes to your tool. Now, should you attach 3, and have all 3 linear axes attached to your piece? You could, but that would make for a very poor design. Having all three rotary axes together is not a problem -- robotic arms do that with 3, 4, 5, or 6 rotary axes very effectively. The problem comes from having all three linear axes built on top of each other: this is a problem because either one axis will have to lift the other two (very heavy), or one or two axes will experience a massive amount of deflection. Therefore, you want 2 rotary axes and 1 linear axis attached to the tool, and 2 linear axes and 1 rotary axis attached to the piece.

From there, the last question you have to answer is this: should the single linear axis attached to the tool be horizontal or vertical? To answer this question, you should actually consider the two linear axes that will be attached to the piece. For these, it's good to know that you have only two ways of combining two linear axes: either horizontally or vertically. That's because in a 3D space affected by gravity, what we call "vertical" is the gravity axis. As a result, in that 3D space, the other 2 axes must be horizontal if you cannot have two linear axes that are parallel to each other (Cf. earlier paragraph). You simply cannot have two vertical axes, but you can have two horizontal axes. Therefore, when you consider any set of two linear axes, at least one will be horizontal. So the question becomes: should you have two horizontal axes or one horizontal and one vertical? Well, in general, building an X-Y table (or X-Z when you're building an Horizontal Machining Center like our machine) is usually easier than building an X-Z gantry. Why is that? because when building an X-Y table, gravity plays the same role on the X and Y axes, unlike what it does on an X-Z gantry, and that makes it a lot easier to get things balanced and rigid. Therefore, if you have to pick between the two options, you should go for table rather than gantry.

Therefore, we have two horizontal axes attached to the piece, which means that we must have a vertical axis attached to the tool, and that is precisely why our machine uses a static vertical column onto which is mounted a 2-axis head (which two axes are rotary of course).

Now, this leaves the single rotary axis that is attached to the piece, and this axis must be orthogonal to the two rotary axes that are attached to the tool. With that in mind, there is only one option: a rotary table mounted flat onto the horizontal table, with its rotary axis orthogonal to the table's linear axes. And that is why our machine is designed the way it is.

Finally, I should mention that one could envision many different configurations for the 3+3 design that was constructed above. But to make it even more balanced, one should try to get the three axes attached to the piece orthogonal with each other, which means that the three axes attached to the tool will also be orthogonal to each other. This is not a requirement, but it will further improve balance and rigidity. And this is what we have achieved with our design: the rotary table's rotary axis is orthogonal to both the X and Z linear axes of the table, and the vertical Y axis of the column is orthogonal to the A and C rotary axes of the spindle's head. This makes our geometric design (not taking into account mechanics, electronics, hydraulics, and pneumatics) as close to ideal as possible, at least on paper...

I call this a fully-balanced 3+3 design.

[RCaffin]

An interesting discussion, thank you.
A question: is your design an HMC or a VMC? In fact, does this question actually make sense?
You see, if I understand your diagrams correctly, you can orient the spindle either horizontally OR vertically (or any angle in between). So it could be either - except that the chips are still going to fall on the table or even into a pocket in the job.
Should you rotate the whole machine 90 degrees over on its side so the chips fall clear?

Cheers
Roger

[ishi]

An interesting discussion, thank you.
A question: is your design an HMC or a VMC? In fact, does this question actually make sense?
You see, if I understand your diagrams correctly, you can orient the spindle either horizontally OR vertically (or any angle in between). So it could be either - except that the chips are still going to fall on the table or even into a pocket in the job.
Should you rotate the whole machine 90 degrees over on its side so the chips fall clear?

Cheers
Roger

Roger,

This is an excellent question. If you remove the 2-axis head, you can keep the rest of the design unchanged, and you have built yourself a nice Horizontal Machining Center (HMC). Therefore, it is fair to call our machine a Horizontal Machining Center. But if you tilt the spindle by 90° around its A axis (which can go ±120°), you got yourself a Vertical Machining Center (VMC), at the push of a button. That is one of the (many) things I *love* about this design. Of course, when you do so, your envelope alongside the Z axis is reduced, from 20" down to 14". But thanks to the rotary table (around the B axis), you can flip your part around and you're back to a 20" cutting range alongside the Z axis...

So, what should we call it? How about Bidirectional Machining Center (BMC)? Or Omnidirectional Machining Center (OMC)?

Now, should you flip the machine around? I really don't think so, for the reasons that were outlined in the previous post. This has to do with gravity: you want your vertical axis to be attached to the lighter axis group: the tool, not the piece. But there is an easier way of achieving what you want: simply mount your piece horizontally instead of vertically, by using a vertical tombstone. And when you do so, you might find out that by using a tall tombstone with 4, 6, or even 8 facets, you can mount many pieces at once, while using your vertical Y axis and your rotary B axis to go from one piece to the next, in a single setup. Of course, you can achieve similar results with a trunnion table mounted on the horizontal table of a VMC, but the HMC configuration makes it much more rigid, because rotary axes are harder to make rigid than linear axes when pieces become large, which is why companies selling rotary tables and trunnion tables can charge so much for them (this is how Hass got started by the way).

[ishi]

Here are some more explanations regarding the 2-axis spindle head that we are using. This is important, because I suspect that some readers might not have realized what this assembly really does, or how it works. If you have, good for you, just skip this post. Otherwise, carry on: this will be helpful.

We are using this 2-axis head:

HST310

This assembly has three motors (Cf. attached picture):

1. Torque motor for the A axis (the red box on the left)
2. Torque motor for the C axis (the grey cylinder on the back)
3. Spindle motor (the blue box on the right)

The A axis has ±120° rotation.
The C axis has endless rotation.

Obviously, if there was no A axis, the C axis would be useless, because it would be colinear with the spindle's axis. But as soon as you start moving around the A axis, the C axis is not colinear with the spindle's axis anymore, therefore becomes useful.

To make it work with precision and accuracy, this relatively-small head uses two direct drive torque motors for the A and C axes, with 80Nm torque for A and 160Nm for C. These are very reasonable torque figures for such a small assembly, but to make it even more rigid when A and C are static, pneumatic brakes are added as well. This gives you 30 arcsec of accuracy around A and C, even through fairly demanding cuts. Of course, you could get a lot better accuracy with bigger torque motors, but that would make the head assembly a lot bigger and heavier. The full catalogue of HSD 2-axis heads is available there:

http://www.hsd.it/bo/allegati/Files/2604_ita_web.pdf

if you are not entirely familiar with direct drive torque motors, please read this brochure, it was really helpful to me:

https://www.etel.ch/fileadmin/PDF/Ca..._Motors_EN.pdf

If you stick to the direct drive options (which is what you really want), going to the HST570 will give you 20% more accuracy (24 arcsec vs. 30 arcsec), but at the cost of a 50% larger shaft (diameter of the enclosure for the C axis motor) and twice as much weight. Therefore, the only rational next step is to go to the HS610, which gives you an amazing 4 arcsec accuracy, but for almost four times the weight... From there, you can double weight again (550kg), and you get 2 arcsec on both axes. Interestingly, you get that with a 310mm shaft as well.

Conclusion, if our initial prototype is successful, we will quite likely want to build a larger model with a 310mm shaft that would give us a much wider range of spindles. If I were to draw that on a napkin, that would make for a machine with a 90" × 90" × 90" footprint and a 45" × 30" × 30" envelope, and a cost of parts in the order of $300k. Again, not cheap, but there is nothing quite like that on the market today, and this is not about to change anytime soon, because there are very few trained 6-axis machinists out there. This is the catch-22 that we are trying to break by developing a machine for the education market: we're essentially starting at the source of talent.

But we might skip this step altogether and build this larger machine with linear motors as a replacement for the servo motors and ball screws currently used for our 3 linear axes:

https://www.industry.usa.siemens.com...ear_Motors.pdf

Torque motors and linear motors are the future.

That, my friends, would be really cool...

[RCaffin]

Interesting, thank you. Yes, direct drive motors are known - but $$.

I find the whole business of clearing chips to be a dead issue anyhow. I use pulsed misting with continuous air blast, and the chips do not hang around. So the distinction may be moot, for me at least.

I have one rotary axis on my machine (A or B), and I have the bits to add a second independent C axis to sit on the table. Just haven't had time to build it yet. That's partly because I have not managed to identify a real need for it as a pure C axis on the table.

I can only point out that your design is a LONG way away from the hobby market. But we can admire from a suitable distance.

Cheers
Roger

[ishi]

Interesting, thank you. Yes, direct drive motors are known - but $$.

I find the whole business of clearing chips to be a dead issue anyhow. I use pulsed misting with continuous air blast, and the chips do not hang around. So the distinction may be moot, for me at least.

I have one rotary axis on my machine (A or B), and I have the bits to add a second independent C axis to sit on the table. Just haven't had time to build it yet. That's partly because I have not managed to identify a real need for it as a pure C axis on the table.

I can only point out that your design is a LONG way away from the hobby market. But we can admire from a suitable distance.

Cheers
Roger

Roger,

Sorry, I did not mean to imply that you did not know what direct drive torque motors are. I just wanted to make sure that we're all talking about the same thing.

And I agree with you that many jobs will only need 3 or 4 axes, and the simplicity that comes with it is priceless.

My theoretical understanding is that kinematic complexity really comes when going from 4 to 5 axis. Going from 3 to 4 is rather incremental, and so is going from 5 to 6. Things become really complex when you jump from 4 to 5. It's like a quantum leap. A bit like when going from 2D design with Illustrator to 3D design with AutoCAD. Or a bit like Newtonian equations of motion: with two objects, it's super easy, but add a third and the equations become differential, and you can't really get a simple solution for them. Well, I think the jump from 4-axis to 5-axis kinematics is the same. In that respect, my 6-axis machine won't be much better than a 5-axis machine from an education standpoint. But where it can really shine is by switching from HMC mode to VMC and back. That is something that very few 5-axis machines can do, and I believe that it would be a very effective way of teaching the pros and cons of HMC and VMC to students.

As far as this build being a hobby, I would say that one person's hobby might be another's job, and vice versa. For me, this is very much a hobby, albeit one that I am dead serious about. It certainly has taken much larger proportions than I had envisioned at first, but it's also taking me to where I want to go: at the edge of the envelope. The goal is not to instill admiration (nor contempt). Instead, I am just trying to encourage active participation, and I'll be happy if the project can become an inspiration to some. In the past 4 weeks, I have learned more than in many past years, and I hope that some readers following this rollercoaster of a ride might learn a thing or two as well. And if we actually end up building a working machine, I like to believe that the few people who have provided valuable feedback will take a sense of pride in this achievement, because as you are implying, it's not everyday that a build like that is being attempted in the open, for anyone to see and contribute to.

[pippin88]

skrubol,

The servo(s) for the Y axis will have a brake, and the Siemens SINAMICS S120 Smart Line module can play the role of braking module and braking resistor. With that in mind, would you still put a counterbalance? Also, if straight pneumatic is enough, why do most vendors adding a counterbalance go for hydraulic? This does not make much sense to me, especially considering the cost of these setups ($2,000 for parts, $4,000 if you buy them directly from vendors like Haas).

I'm going with two screws on X and Z because it provides more rigidity and accuracy. They're more difficult to assemble and they increase mechatronics costs quite significantly, but they're the way to go when spacing between rails goes beyond certain dimensions. If you look at box-in-box designs for many DMG MORI machines, they always use two ball screws on their linear axes. Assembling them horizontally is not too difficult, but it's much more of a pain for a vertical axis. And for Y, the spacing between rails is reduced, therefore this really begs the question of whether two ball screws are really necessary there.

Regarding the spacing of rails, I think I'm starting to understand what pippin88 had in mind, and this makes perfect sense. I always thought that you want to space rails as much as possible in order to reduce torsion, but you actually want to strike some balance between spacing and rigidity of whatever the rails are supposed to support. If they are too far apart, you'll get deflection in the middle of the supported assembly. That makes a lot of sense! Thank you pippin88, and sorry for not having properly understood your original comment. I know you asked the same question twice, but I did not understand what you meant, both times. So, without going into complex mechanical calculations, what should be the rule of thumb there? 1/4 - 1/2 - 1/4? 1/3 - 1/3 - 1/3? If I want to keep two ball screws on each of the X and Z axes, I think 1/4 - 1/2 - 1/4 would be about right. Thoughts?

I could have phrased it better sorry.

The point is to have the minimum span (distance between supporting elements, eg the rails.) I don't know the optimum to be honest, I think looking at established machine designs that it is probably at something like 1/4 - 1/2 -1/4. A compromise between supporting / reducing the length of the span and resisting torsion.

[ishi]

I could have phrased it better sorry.

The point is to have the minimum span (distance between supporting elements, eg the rails.) I don't know the optimum to be honest, I think looking at established machine designs that it is probably at something like 1/4 - 1/2 -1/4. A compromise between supporting / reducing the length of the span and resisting torsion.

pippin88,

That makes perfect sense, thanks again for point this out. I think the latest iterations are getting pretty close to where we need to be. If I shorten the distance between the two rails on X, I might have to go with a single motor (as you can see on the attached picture, which only shows one of the two motors). On Z, I can't make it shorter, otherwise there is not enough room for the rotary table's torque motor. And on Y, I am constrained by the C torque motor of the 2-axis head.

[ishi]

In order to improve the design of my castings, I am looking for a reference book on designing metal castings. The best option I could find is this one, but it isn't cheap ($400):

https://hub.afsinc.org/NC__Product?i...a000000AQPiEAO

Any alternative suggestion?

[RCaffin]

Distance between rails
I do not agree that you should reduce the distance between rails. The bigger the distance, the less wobble there can be.
Mind you, for a machine that size you will need BIG rails, to carry the weight if nothing else. Given that, and given good quality rails, adding a few extra cm to the distance between might not be too significant.

Position of axis motor
I definitely don't like where you have put the X axis motor - over to one side. It might in fact work, but I would very strongly urge that you put all motors midway between their rails. You will probably need something like a 25 mm double-nut ball screw anyhow, at least for the X axis. You will probably need a 20 or 25 mm ball screw for the vertical axis as well: it is carrying a bit of a load.
On the other hand, I do not think you will need dual axis motors for this machine, not on any axis.

Base and column
You specified these as granite I think? Do you have any idea what it is going to cost to get the channels or slots milled out of solid granite? Rather a lot I think! I machine granite and basalt, and I have some idea of what it takes. You might consider getting polished slabs and fabricating both the base and the column, with bolts, rawlplugs (or whatever brand) and epoxy. The end result would be about the same but much less expensive. Any monumental stone mason could set these simple slabs up for you: they would just be custom tombstones. (I had a long and interesting chat to one stone mason years ago.) YOU would have to do the alignment of the slabs of course, but you would have to do the alignment of the vertical relative to the base anyhow.

Axis Motors
Whether you are over-speccing the motors or not - not sure. I have a smaller CNC mill (NOT a router), with 500 W motors on the axes. I also have current meters on the motors, and I very rarely see any deflection on any axis meter. By the time the torque goes through a 3:1 belt reduction and then down a 5 mm pitch lead screw, it does not need much to destroy anything in the way. (Errr - yes.)

Cheers
Roger

[ishi]

Distance between rails
I do not agree that you should reduce the distance between rails. The bigger the distance, the less wobble there can be.
Mind you, for a machine that size you will need BIG rails, to carry the weight if nothing else. Given that, and given good quality rails, adding a few extra cm to the distance between might not be too significant.

Position of axis motor
I definitely don't like where you have put the X axis motor - over to one side. It might in fact work, but I would very strongly urge that you put all motors midway between their rails. You will probably need something like a 25 mm double-nut ball screw anyhow, at least for the X axis. You will probably need a 20 or 25 mm ball screw for the vertical axis as well: it is carrying a bit of a load.
On the other hand, I do not think you will need dual axis motors for this machine, not on any axis.

Base and column
You specified these as granite I think? Do you have any idea what it is going to cost to get the channels or slots milled out of solid granite? Rather a lot I think! I machine granite and basalt, and I have some idea of what it takes. You might consider getting polished slabs and fabricating both the base and the column, with bolts, rawlplugs (or whatever brand) and epoxy. The end result would be about the same but much less expensive. Any monumental stone mason could set these simple slabs up for you: they would just be custom tombstones. (I had a long and interesting chat to one stone mason years ago.) YOU would have to do the alignment of the slabs of course, but you would have to do the alignment of the vertical relative to the base anyhow.

Axis Motors
Whether you are over-speccing the motors or not - not sure. I have a smaller CNC mill (NOT a router), with 500 W motors on the axes. I also have current meters on the motors, and I very rarely see any deflection on any axis meter. By the time the torque goes through a 3:1 belt reduction and then down a 5 mm pitch lead screw, it does not need much to destroy anything in the way. (Errr - yes.)

Cheers
Roger

Roger,

Timewise, I can't afford to do the alignment myself. Therefore, I am outsourcing all this work to this company in the US:

Precision Granite Machine Bases - Pyramid Granite & Metals

They will do all the milling, and I know what it costs. This is already factored in the total cost of the machine:

https://docs.google.com/spreadsheets...NTo/edit#gid=0

There are similar suppliers in China and Europe. Here is the one we will work with for the Asian market:

China Granite Surface Plate, Granite Machine Parts, Granite Measuring Instruments Manufacturers & Suppliers - Factory Price - Fortune Machinery

Our European distribution partner is working with a similar supplier in Western Europe. All three can provide the base and column milled, assembled, and squared, including all epoxy-glued inserts.

This is one thing we don't have to worry about. Of course, you could do all that on your own, but we're trying to put together a process whereby we could assemble 10 to 20 units within two months or so, with customers in Northern America, Western Europe, and South-Eastern Asia. And I don't want to touch any part myself. Not that I don't like it, but I want to streamline the supply chain as much as possible. Therefore, the kit will be shipped in sub-components, directly from our suppliers (at the exception of very small and cheap items like fasteners).

As far as the X-axis motor is concerned, I am putting two of them. In fact, I am putting two motors on every linear axis, because it will improve rigidity. It's possibly overkill, but I don't want to take any chances with that first build. The opportunity and reputation cost are simply too high. As explained in an earlier post (on this thread or the "original" thread), it's much easier to downscale than to upscale, as long as you can afford the cost of the initial upscale prototype. I'm definitely going down that path, for better or for worse.

[ishi]

Here is the drivetrain for the X axis, with a couple of servo motors and ball screws.

Ball screw mounts from NSK and servo bracket from SYK:

https://www.syk.tw/product/201703030006

I still need to select the shaft couplers, but the selection of this component has zero impact on the rest of the design, therefore I'm no rush to settle on any particular one. Recommendations welcome though!

High-res pictures available on the project's drive:

https://drive.google.com/drive/u/3/f...3dQrx?ogsrc=32

[ishi]

Take a look at the attached picture. If you're like me, you'll think that the machine feels packed. It's dense. The 2-axis spindle assembly feels huge. The two motors for the X axis feel redundant and oversized. The carriage for the X axis feels massive.

Why is that? Is it because of bad design? Over-sizing? Over-engineering?

Most probably.

But there is a more fundamental reason for this bulk. The fact is, Horizontal Machining Centers are usually very large in relation to their cutting envelope. The reasons for it are too many to explain on this post, but you immediately get a sense for that once you step in front of one. It feels truly gigantic. And that is why very few small Horizontal Machining Centers are available on the market today (at the exception of the cute little Pocket NC of course).

Our machine is no exception to that rule, and if you try to get a small footprint while preserving accuracy and the ability to cut a wide range of metals (yes, we will cut through steel just fine), you end up with a bulky machine like ours. I don't think there is any escape from that.

Of course, you could try to downscale some components and go for a single motor on some axes (X would work for us, but neither Y nor Z would, for reasons explained on prior posts). But you would not save much doing so. Instead, you're much better off sticking to a balanced design, because if you do, you have a ready-made avenue for scaling the machine up. What we are designing today is what we refer to as the ISHI OMC 30 Mark I. OMC stands for Omnidirectional Machining Center (think both HMC + VMC in one package). 30 stands for 30 inches, which is the travel of the X axis. If we are successful with this machine, we will build its big sister, the ISHI OMC 45 Mark I. How different is she from her little sister?

- 100" × 100" × 100" footprint
- 45" × 30" × 45" envelope
- 310mm head shaft
- HSK A100 50kW spindle
- Linear Motors

So, when you look at the 30 model, just keep in mind that it's a just a stepping stone for the 45 follow-on.

And by the way, the OMC 45 Mark I will be a production machine, unlike the OMC 30 Mark I...

***

High-res pictures on public drive:

https://drive.google.com/drive/u/3/f...3dQrx?ogsrc=32

[RCaffin]

A big machine. We watch with great interest.

Cheers
Roger