A 5-axis CNC machine including a rotary axis

[synthetiklone]

Hello all,
an introduction..

I have decided to post a build log on CNCzone so that others may see what I am typing about in the forums.
There is a gallery under my name, synthetiklone, with some 10 pictures of the major equipment and parts I have so far. It looks like I can post another 10, & thats the limit. I will post progress reports with pics here if I can, and probably keep the gallery for major updates.

I am an electronics engineer from Wellington, New Zealand - land of the long white cloud, at the bottom of the planet, where all things CNC seem to be rare as hens teeth, or just as costly! eg, local SKF quote was NZ$3600 (US$6000+-) for 2x 200mm long rods with 4 pillow blocks and a 200mm long ball-screw!

This CNC project began with a 3D foam cutting machine, but as I have acquired parts and materials for building a more advanced machine, and after extensive information gathering on the net, especially here on the CNCzone, the CNC bug has bit me! The foam cutter is still going to happen, but most efforts are going towards the 5-axis beast.

The 5-axis machine is to be as mechanically robust as I can build. I happened to come across some good parts from a 2-ton large laser image scanning machine. It provided 40mm thick (!) solid gantry rails, and 40mm thick leadscrews. I couldn't even dream of obtaining buying/importing these for a good price here in NZ (see pictures). I would have liked to cover a 1200x2400 sheet, but again, here in NZ parts are price prohibitive for the home constructor.

So the design is a 1500mm X 1200mm X 700mm table/frame with 860mm X 700mm X 200mm toolspace/workspace.
The x-axis leadscrew will end up going across the top of the whole assembly, rather than the conventional underneath the table approach. I will try it on one side first, but may have problems with "torque-twist" on the y-axis gantry.
There are two reasons the main x-axis leadscrew is placed above the gantry; I need the working space..
This machine incorporates a rotary axis as well - running along the longer length of the x-axis, and the rotary servo becoming the y-axis servo.
The rotary servo pulls in on the flywheel with a solenoid, disengaging the higher speed motor which can be used for conventional lathe CNC work. The job placed here has to be rotated 360 degrees. So if you can imagine it to be possible to carve any details all around an 860mm long section of a 400mm diameter round piece of wood all in one go! It would be possible to carve sections of a totem pole for example, but whether it ever is used for this, who knows, but I can think of numerous applications becoming possible. I'm sure you can think of some too. The rotary end bearings I have can easily accomodate this. See picture.

The other reason for the space required under the z-axis is that I have a friend who has developed a 4th & 5th (a&b) axis servo drive unit for the router. This will be the last addition to the machine, as at the moment, I will be very pleased just to have it run as a 3-axis!

Another thing you will notice (seen in gallery pictures) is that the y-axis rails are horizontal to each other rather than vertical. I see this as an advantage where stability is concerned, however, it limits me on the uppermost rest position of the router, which may have to hang underneath the gantry frame. It wont fit between the y-axis rails! I also have four 20-25mm solid rails about 440mm long from an old scrapped IBM printer, I was thinking of using an all four corner configuration for the z-axis, again for stability, but haven't decided yet, as have also been offered some 20mm, 200mm long THK linear bearings, which I may use (depending on price! ) So Z-axis still to be decided, worse case scenario for me is to flip the gantry up, redesign the bearing mounts, and go for the conventional vertical gantry rails, if I dont have the room between them with the current design.

I am at the stage of building the framework to hold it all together, it is to be made from 40mm square section pre-primed steel, 4mm thickness (I was going to use 3mm, but I'd rather over-engineer now, than be caught out later with flexing problems. The design of the frame is a dilemma in itself, as I have to forward think (and draft drawings) of different possibilities. I don't own a welder, or mill, so am paying someone to do all the metalwork, and bearing milling, and engineering. I don't want to have to pay to do the same job more than once if I can design it properly the first time. Although I probably will

I was going to design my own servo controller w/ quadrature PID feedback, but won't have the time for this, and the Gecko 320's are my first choice, with the possibility of utilising the G-rex in future, although the G-rex option looks like a complicated procedure at first glance through the documents provided, I'm sure it is a simple setup (?) with lots of benefits.
Which brings me to the software side.. err.. I haven't even looked into this as yet, and hopefully will pick it up when I have a working machine to play with.
I know virtually nothing about the software, but am impressed by whats available (Mastercam for US$15,000!??) I have a friend who has laser cutting/engraving machines, and it is hoped that some things will be similar design/software wise, and will work well together, as there is some things a laser beam just cannot do

pictures are here: http://www.cnczone.com/gallery/showg...0/ppuser/36476

Any advice or information on anything regarding thy project is warmly welcomed.
Well, it's getting late here...
Thank you kindly
and CNCleep well!

Tony

[epineh]

Nice score on the parts, I know what you mean on the buying of parts, they aren't much cheaper on this side of the pool either, look forward to seeing what you make of them.

Russell.

[synthetiklone]

Things are pretty slow the last month, my metalworking engineer isnt doing any machining until mid January. I've used this time to redesign the framework and looking at different configuration scenarios trying to make sure I cater for every possible job I might be up against in the future. I am still really unsure if there is enough strength in this to allow 3D work, ie having a 300mm Z-axis drop. It seems a nice thing to have, but will I need it?, will it be stable, and accurate. Especially as I have only one X-axis leadscrew pulling the gantry from the side, or top if the side dosent work.
Also thinking maybe the Y-axis guide rails should be the standard configuration - above each other, to allow more room in the Z-axis. Still thinking.. have a week or so before the welding/machining starts

[epineh]

Nice to see you haven't given up, if you can get the leadscrew in the middle of the rails/bearings it will be better, Z travel is entirely up to you, but I guess you should think about what you will be doing with the machine. If you find you outgrow it you can always build another... not that I have that in mind at all lol

Russell.

[dan dimock]

I would like to hit a home run on some parts. I would like to build one to cut gun handles but have not found and plans for one, so maybe you will give me the know-how as I keep looking at yours.

Dan

[synthetiklone]

Just a quick update, have the main steel box section frame built, and after much research, have decided to have two Y-axis drive threads, rather than the one down the middle. This will provide better torque when the machine is working off to the side. Luckily have sourced a 3 meter drive screw for this, and cut it in half, as was considering having a linear cog manufactured, and this would have been very expensive.
The Y-axis frame is at the engineers having 4 bearing modules made and attached to the ends of the frame. Each bearing unit has three bearings 120 degress apart, and the top one is made finely height adjustable.

Software wise, have tried downloading (on dialup) EMC2, which is the free linux based system, three times now, each time a failure. Some glitch somewhere in my 4 night download session (using "getright" download manager now) seems to muck it up, and wont install, or run preoperly.

Also have a dedicated SUN Ultrasparc machine waiting and ready to run it, just got to find a Sun keyboard/mouse for it. They don't run windows x86 based software at all, just UNIX, and hopefully linux. But if this fails, have a PC ready to go, and one higher end PC for any design work located on a network in a more comfy/warm part of the house!

[synthetiklone]

Blinking thing is still sitting on the floor at the engineers workshop. I asked him if I could pay some more on his hourly rate so that it might get started sometime soon, and he said that even if I paid him with it's weight in gold, It wouldn't make him to get it started any quicker. He is busy with more "important" jobs I guess. So still waiting.. waiting.. waiting as usual on other people taking their own jolly old time with things that aren't important to them. Typical.

[dengvall]

Since you are an engineer; spend some time with FEA before you build. It might be time well spent.
Take a look at the emc project for cnc software as is will handle steppers or servos. http://linuxcnc.org should get you there.

Also once you get a machine running then CAM becomes a real issue for 5 axis. Synergy is not too expensive and should do your job. Some pretty big companies use it for extrusion screws which are not simple to draw or CAM. A google for synergy CAD will get you the link. Best of luck with your project

Dave

[synthetiklone]

Another quick update:finally got the table back of my engineer. I got tired of "popping" in for a progress report, and I was beginning to feel like I was hassling the guy. After 3 months, I am back to where I was, he'd done nothing, as he was just too busy with other jobs already. Funny thing though, while I was there, a guy called in to see him, and wanted a new tow bar madeup for his vehicle, and welded into place. He booked this guys job in there and then, in front of my eyes! So I simply got my trailer hooked up and picked up my machine and left without another word! If this guy would rather weld towbars, then so be it! Pity, as he made a nice job on the adjustable rail holders for me.
So. Three months, back to where I was, but I've redesigned the bearing holders now so the gantry will run on four pairs of bearings at 45 degrees. The weight of the gantry should be enough to stop any lifting (it is heavy!) Having twice the bearings spreads the load on the circular rail, and is much simpler for another engineer to "knock" something up, without having to hassle them for 3 months to do the job.
Well, this machine might take the rest of my life to build, but I'm going to get it done!

[harryn]

I had a scan through your picture gallery and thread - nice concept - PITA to get going, isn't it. That is the main reason I am trying to get going on my own project, it will not be cheaper than hiring out the work, but at least I can only blame myself then.

As far as getting some work done, sometimes local high schools have shop classes for their students. Sometimes, the teacher can be enticed to have one of their better students do some work, especially if it comes with a donation to the school. Maybe call a few school principles / guy in charge before the new school year starts and see what you can get going.

It is probably just me, but it seems a real shame to cut a large screw like that in 1/2. I can just imagine you saying "man, I wish I had that big old screw to build a really big machine". In any event, good lluck.

[synthetiklone]

Wow, It looks like I've had a busy few years. Only part time rarely working on the 5 axis. The 5 axis is very slowly getting completed. I'll post some photos as soon as I can, just to have a visual update here. As a quick rundown, everything is almost finished. X moving table with 900mm travel completed, y gantry with 800mm travel completed, z axis with 600mm travel completed. This machine sits 2 meters tall! Aluminium z support plates completed, which holds a new Doughty Drive b-c axis rotary head. I designed my own dual (bc) rotary head, but as my project has taken this many years to complete so far, I bit the bullet and bought an excellent ready made unit - the Doughty Drive. Currently fitting the wiring harnesses, connecting up electronics. All the drive electronics are housed in a box which sits on the back of the gantry. This shortens all the cable runs, and keeps everything up and out of the way. Recently purchased a 2 hp ER20 spindle motor and controller. I've mounted this on my 3 axis table, and it purrs like a kitten, very fast smooth cutting. One big issue I have found now is with the table, and framing. It was originally designed as a small 3 axis machine, and used 40x40mm box section, including the gantry support. Well, any design engineer can tell you that this frame will flex I guess, and it does - quite a bit, especially with all the weight of the gantry. (which is why the backbone of a good mill is very solid). I've also just bought 100x100mm box section at 5mm thick wall, to re-make a new frame, and gantry support. Also another note, because the x & y leadscrews have 4mm per rev, and the z has 12mm, Ive decided to include a 1:3 belt drive to match up the speeds. It would be nice to have the resolution, but will sacrifice too much in rapid travels. So, I'll update some photos as soon as I can for anyone interested, and just as a personal update too. T

[slowlearner]

subscribed