Large EG 5 axis cnc : design & build log

[yellow_submarin]

Hi everyone !

I have just registered to the Zone. I have read quite a lot of threads on this forum which were quite full of information and inspiration, so I wanted to take part.

I have a freshly retrofitted Sieg SX3 running on TB6560 with mach 3 and ballscrews from linearmotionbearings. This machine runs quite good but I need something with larger dimensions, more rigidity, spindle speed, and off course, axis

I have well in mind what I would like to build, so I will share with you so you can criticize or agree with my design.

The idea is to build a machine which looks like Huron - Graffenstaden fixed table conventional mills. Have a look there :



I suppose you can guess how it works. The only difference is, I want the table to be fixed and the column to travel in X. I had thought gantry at first for stiffness, but I fell in love with this layout.

Now, the special features :

The whole cnc would be made out of granite epoxy mix, with aluminium machined ribs for accurate positioning in the epoxy castings.

I think of at least 1200mm travel in X, 500 in Y, 800 in Z. But as mentionned, I want a 5 axis. So, the 4th axis would be cast with the table on one tip, being parallel to X. The aim is to use the 4th axis as a lathe as well. 700mm diameter capacity would be great. The 5th axis would be a tilting head at the end of the cantilever arm of Y.



I need quality linear rails to ensure stiffness. These would be 20mm Hgw Hiwin guides & rails. Probably fitted with 20*05 ball screws from the same chinese ebayer, with double loaded ball nuts as on my x3. This compensates the small backlash of medium quality ball screws.

Considering motoring, I thought about servos, but I prefer to save money for good rails and get 1200 oz-in steppers for cheap with Gecko drives though. Direct drive for all linear axis of course, with the 10 microstepping mode of the Gecko drives, I would achieve a 0.002mm resolution, no need for less.
About the rotational axis, same stepper configuration for the tilting head I guess, but with a gearhead off course. I don't know which type yet, probably harmonic drive. Backlash is unacceptable here. And quality worm drives cost at least as much as harmonic drives.
I am not sure about the 4th axis though. An AC servo of at least 1kW seemed to be the obvious solution, but a big, direct drive stepper would fit as well. If you think about it, there is no need for speed AND torque at the same time when turning. Torque is required at low speeds. Anyway, a brake would be probably necessary for accurate position holding.

I want to achieve something special about the spindle as well. I think most of the commercial solutions suck. Chinese spindles with a quality VFD for low speeds could be a good solution, with a belt drive for stepping down max rpm and getting some more torque. But it's not that exciting, and I would like something like 5KW for good torque when spinning big mills or attachments. There are awesome inrunner RC brushless motors which are pretty cheap. Have a look at this one :HobbyKing R/C Hobby Store : T5692 Turnigy Pro Comp Brushless Inrunner Motor 1000kv
This little monster is able of 6KW with ridiculous dimensions. Moreover, the price with a suited ESC is just about $250. There is a major drawback though : it will draw at least 150a at 45v for the intended use. The price of such transformer(s) is a no-go. But I would quite like to find a way to use this beast. A TTS-like atc with R8 spindle and 16mm colled would be just riiiight !

Anyway, I hope I made it clear. Feel free to comment upon my ideas, some of which might be utopic or irrealistic. If you think it's great, don't be ashamed to say it as well

By the way, excuse me for mistakes in the language if any.

Greetings from France

Cyril

[jsheerin]

Is this intended for woodworking? If you want to do metal working, you probably will want a lower speed / higher torque spindle and spindle motor. If you want to do metal working with such a high speed spindle, make sure you design it so it has very high movement speeds in order to get decent chip loads. Additionally, the 20mm ball screws might end up as the weak point from a rigidity standpoint if you're doing EG on a scale as shown in your picture.

[yellow_submarin]

Thanks for your answer. I appreciate.

It is intended for metalworking. But off course, if such a spindle is used, it would be belt driven to reduce speed. 10 krpm of maximum spindle speed sounds good for small diameter end mills. Anyway, I would have to gear the motor in order to place a drawbar for the atc. But high speed linear motion is considered, as a maximum of 400 rpm for direct driven steppers would mean 2m/min.

Concerning dimensions, it would be much smaller than the one shown here. I have not detailed that, but it would have to be a tabletop machine, because I would never be able to get it out of my place when I mouve out with a massive frame. I guess 20*5 ballscrews would be stiff enough for what I have in mind, though swapping for 25*05 ballscrews would not be much more expensive.

Cyril

[wizard]

It is nice to see something in the way of a horizontal mill attempted. I'd seriously consider a straight boring head option for such a mill.

As to epoxy granite I'm not sure that is the route I'd go especially for a bench top machine. Rather for the base I'd look for a slab of steel or cast iron and probably start at around 2 inches thick. I suspect it would be cheaper than an epoxy solution for the vertical I'd go with a filled steel tube.

For the table I'm not sure I would want one completely fixed. Instead I'd allow for a table that could be moved in the Y direction if the need arises. That is a manual ability that might be useful if work pieces have trouble fitting the machine. Yeah that means loosing zero but it isn't something that would happen often. Considering this is a bench top machine even a few inches (centimeters if you like) could make this sort of feature very handy.

The big problem I see with this sort of machine is column stiffness. That might require some pretty heavy duty linear solutions. A good mechanical engineer could comment further but I suspect getting anywhere near the stiffness of a Chinese bench top mill will require some design effort.

[RICHARD ZASTROW]

1200 mm X-axis travel and all that goes with that seems to be a bit more than a bench top machine.LOL

Dick Z

[Zach_G]

Hey neat axis setup. I had been considering doing a full moving gantry for awhile but as I planned out how to fabricate and align everything, a more traditional bridgeport layout seemed far easier to build. I can definitely see benefits of this layout if you're going for maximum work volume with relatively small spindle and cutters in a lightweight machine and small footprint.

Metal Gantry Mill

http://www.cnczone.com/forums/vertic..._cnc_mill.html

Something to advise, DO NOT use that motor for a spindle. I bought one for a project and can definitely say while 6kw may be its most efficient operating point for converting electrical energy to mechanical power, it definitely won't run at that point continuously without burning up. Lets say you're running at steady state, 80C (176F) is usually the max temp before the magnets are permanently damaged. Typical coefficient for convective heat transfer is 10W/(m^2*C). With a surface area of 0.01858m^2 and room temperature of 25C it can dissipate 10*0.01858*(80-25)W = 10.2W. My bench measurements at 24V show it dissipates about 100W just free spinning without any load! And it does get pretty toasty after only a couple minutes running on the bench. So how do they get away with advertising it as a 6kW motor? Well its for the hobby industry which egregiously overspecs components to stay competitive in a market where most don't bother to validate. The applications also only involve 10-15 minutes of flight where the motor's thermal bulk can soak up alot of waste heat and there's a substantial amount of forced air convection from prop wash. You'd need some serious active cooling to get even 1kW continously. That's why I went with a much larger induction motor that can operate at 1.5kW passively plus there are readily available VFD drives that run directly from mains voltage. But hey, that was my path of least resistance. Yours might take you down a completely different road!

[yellow_submarin]

Thank you all for your contribution

Well, the reason why it needs to be a tabletop mill is that when I'll have to mouve out of my appartment, I will have no choice but to get it out by 2 doors. So, there are limited maximum weight & dimensions I can consider. I prefer a moving column, despite being less stiff than a conventional moving table, because a moving table needs twice the travel value of room around the mill. For the sole table length of surrounding room, the moving column provides full travel. I hope I've been clear about what I want there.

About the R/C brushles motor, I have given up about it anyway. It would be inefficient considering the tremendous current flowing through. I don't need 6kw of course, the reason to choose a powerful engine is to benefit of low end torque for large spindle attachments. At full speed, which would be used for small diameter cutters, less than 2kw are necessary ...

[wizard]

Thank you all for your contribution

Well, the reason why it needs to be a tabletop mill is that when I'll have to mouve out of my appartment, I will have no choice but to get it out by 2 doors. So, there are limited maximum weight & dimensions I can consider.

This little bit helps a lot in understanding your needs. As such I still think steel sections and plates are the way to go, even if that means getting a machine shop to help.

I prefer a moving column, despite being less stiff than a conventional moving table, because a moving table needs twice the travel value of room around the mill.

Maybe. You have to remember the moving column has to overtravel the table at both ends if you expect to cover the entire table. Depending upon your saddle and cross arm you may have to overtravel in one direction a bit. On top of that you need a base flange for the column, gusseting and the like. I'd like to suggest cardboard cut outs to get a feel for you proposed layout.

For the sole table length of surrounding room, the moving column provides full travel. I hope I've been clear about what I want there.

Yep but im not convinced. With a moving column you will still be using lots of space some of which does little in the way of controlling machine size. In the end I suspect that you will still be using a lot of space.

About the R/C brushles motor, I have given up about it anyway. It would be inefficient considering the tremendous current flowing through. I don't need 6kw of course, the reason to choose a powerful engine is to benefit of low end torque for large spindle attachments. At full speed, which would be used for small diameter cutters, less than 2kw are necessary ...

I'm not trying to discourage you above so just think of the comments as something to think about.

Like you I have limited access to my shop which means I have to be careful as to how large something is when it comes time to move it into the cellar. Thus I've been flirting with different designs including a gantry machine. The problem isn't the weight of the machine assembled but rather the pieces that make up the whole. You are in a similar situation so I might suggest focusing on weight of the components. That is pick a maximum weight you want to handle and design the parts of the machine to come in under that value. You have the additional issue of total weight.

If you have read through the forums you will find that lots of guys use square tubing for their mill projects bases. I believe this would be problematic for a moving column mill, so I might suggest a steel plate that gives you room for both the table area and the rails for the column. I'd suggest going as heavy as possible here even if that means stretching your limit. At the very minimal the plate should be 1" thick though going even thicker night not hurt. Mind you I'm thinking small benchtop machine here. The biggest problem here is stability, if the plate is machined it would likely be a good idea to do both sides.

Next you need to build a column. The problem is the base plate effectively limits your column size. I suspect you will have lots of frustrations with this part of the machine. You will likely give up table space to allow for proper rail spacing.

[yellow_submarin]

Off course, saying that no greater dimensions than that of the table is required for a moving column is extremely optimistic. There has to be some width allowing linear rail spacing for the x axis, as well as some length to allow the tool to access both ends of the table. Yet, there is no table moving of 2m for full travel.

As you did, I consider the weight of subassemblies. I have experienced moving in my sieg x3 which weighs about 180kgs. Removing the column gave 2 subassemblies of 90kg each, which were the upper limits of weight which can be carried by 2 main out by the staircase.

Yet I consider moving in to a house with a garage box which means full access to the machine thus much less restricted dimensions / element weight.

About the brushless motor, I always considered this option with drawbacks in mind, now it appears they are just too proheminent to make it a good solution. Obviously, a 6000w power supply is not quite cheap, the reasons why I liked it are both low price and dimensions. With the price of the PSUs added, it doesn't sound appealing anymore.

I really like the epoxy granite solution as it doesn't include heavy machining. Vibration elimination is also a good thing. Plus raw steel is quite expensive here in France, and hard to find as well. I could easily machine positioning aluminium brackets with my CNC'd SX3 which could then be casted in epoxy granite mixture.

Thanks for your input

Cyril

[wizard]

This idea might be painful but if you intend to move, why not put off the machine build until then? That way you may not be as limited in size and location.

Off course, saying that no greater dimensions than that of the table is required for a moving column is extremely optimistic. There has to be some width allowing linear rail spacing for the x axis, as well as some length to allow the tool to access both ends of the table. Yet, there is no table moving of 2m for full travel.

yes this is what I'm talking about. The machine will likely be bigger for a given amount of functionality.

As you did, I consider the weight of subassemblies. I have experienced moving in my sieg x3 which weighs about 180kgs. Removing the column gave 2 subassemblies of 90kg each, which were the upper limits of weight which can be carried by 2 main out by the staircase.

Rope and ships can change what you can handle easily or even a two wheel truck.

Yet I consider moving in to a house with a garage box which means full access to the machine thus much less restricted dimensions / element weight.

About the brushless motor, I always considered this option with drawbacks in mind, now it appears they are just too proheminent to make it a good solution. Obviously, a 6000w power supply is not quite cheap, the reasons why I liked it are both low price and dimensions. With the price of the PSUs added, it doesn't sound appealing anymore.

Regular 3phase or druse less DC motors are the way to go. At least on this side of the pond a three phase motor and controller are pretty cheap new and even cheaper used.

I really like the epoxy granite solution as it doesn't include heavy machining. Vibration elimination is also a good thing. Plus raw steel is quite expensive here in France, and hard to find as well.

Epoxy granite has a couple of negatives though. It will be extremely heavy and frankly it is pretty expensive. I'm not sure how much you pay for raw steel on a per pound basis but around here it is often considered the cheap solution. Especially if you can find what you need at a junk dealer.

I don't want to deter you from going the epoxy granite route, it is just that I have a hard time seeing it as the cheap solution.

I could easily machine positioning aluminium brackets with my CNC'd SX3 which could then be casted in epoxy granite mixture.

I'd seriously consider steel inserts. If you go to all the trouble of building a machine you want it to last. More so when maintaining the machine you don't want to have to worry about stripped threads and the like.

Years ago I worked on commercial lathes built with Epoxy granite. Those inserts become extremely important because in reality they hold everything together. The other problem is that they need to be installed perfectly. Oh by the way embed this inserts deeply with features on them to allow solid grip in the epoxy.

Thanks for your input

Cyril

No problem. I can't wait to see your machine. It is always good to see people take a different route.

[yellow_submarin]

Yes, the machine will probably be larger with a moving column, yet the surrounding space will not exceed the machine dimensions, which can't be said about a moving table.

Anyway, I'm considering moving out maybe at the end of the summer. So, maximum mass and dimensions might not be that big of an issue. Nothing sure though.

I am not sure what you mean about the motor, but I would probably go for a 5 kw watercooled chinese spindle with a quality (Fuji Frenic for example) inverter, with belt gearing to get the right maximum spindle speed.

I gotta say you are right about steel for the inserts, I was talking about aluminium because it is easily machined with my mill, though I will probably use steel where threads for removable parts are used. For sure, they wouldn't be smooth cylindrical inserts, I don't want them to slip out of the table

I must say I really like the epoxy granite way. It might not be cheaper, but you can easily make whatever you want out of it. I haven't thoroughly considered the welded steel solution, even though I own a MIG welder, and if it appears it could be more interesting, I would probably build a hollow structure filled with E/G.

I can't say the machine will be built in a few time. I was planning to build it during the summer then sell my SX3, which implies to have a least 3000-3500 euros to buy supplies, which I for now don't have, and likely won't have until the end of the summer as no well-rewarded job is in my sight right now

[asuratman]

yellow_submarin,
Is there any progress on this machine build, or waiting for pocket full of bucks?