PDA

View Full Version : BUILD LOG: 5'x3' Ideas and build.



Wal
21-12-2012, 11:27 AM
Hello again,

It's been a while since I first posted and I've had plenty of time to think (and re-think) my design...

I thought originally that it would be cool to have something capable of cutting aluminium, but the reality is that to build a machine than will do that 'properly' is a different project altogether... The primary goal for this project is to be able to cut hard/soft woods and be accurate/repeatable in doing so.

Specs so far:

Cutting area maximum 5'x3'.

X Axis driven by 2x 20mm ball-screws with a lead of 5mm driven by a NEMA34 motor connected by cam-belt/pulleys.

Y Axis driven by a 16mm ball-screw with a lead of 5mm driven directly by a NEMA23 motor mounted to gantry side-plate.

Z Axis driven by a 12mm ball-screw with a lead of 2mm driven directly by a NEMA23 motor.

All linear bearings will ride on supported rails.

Main gantry to be constructed from 15mm 6082 aluminium plate.

So, here are my ideas so far:

This clip shows how I propose to build the main gantry:


http://youtu.be/4WskSFgqOgA

In order of appearance:

1. The first plate will act as a mount for the linear bearings riding on 20mm supported rails (TBR20 from Zapp) mounted to 80mm aluminium extrusion and will also act as a brace for the...

2. ...vertical cross piece, which also has 20mm supported rails mounted to its front to accommodate the z-carriage. The hole in the arm that drops to either side will accommodate the ball-nuts. There will be a ball-screw on either side of the bed driven by one motor via a belt/pulleys to move the gantry on it's X. The ball-screw for Y motion will run in between the two supported rails and be directly driven by the motor which will be mounted to the...

3. ...side-plate which, when bolted to the other 2 pieces, will hopefully add to the overall rigidity of the structure.

Some questions so far.

I'm using 15mm 6082 aluminium plate for this build. I'm hoping to use M8 35mm hex socket head cap screws to bolt this all together, which means the screws will bite 20mm of the thread I'll tap into the plate edges. Does this depth sound robust enough? Is M8 too big for the edge of 15mm plate?

To allow for a bit of adjustment I'll be making the holes in the face of the plates 10mm in diameter so there will be a bit of slop to play with before torquing everything up once it's all aligned. Does this sound okay?

Where there are slots cut for plate to pass through (the arm which passes through the side-plate, for example) - rather than cut this slot to be exactly 15mm wide, should I make it slightly wider (say 15.25mm) to ensure that the plate passes through okay? .25mm too much?

Obviously there won't be any internal right-angles because the end mill will have a radius. Where there's a square edge that fouls a radius, I'll manually chamfer off the corner to make it fit (see pic below).

7725

Now on to the bed that the gantry sits on:


http://youtu.be/o0c3H3vtQsY

In order of appearance:

1. C channel which connects the 2 X rails.

2. Spacers to fill the gap between the C channel and the spoil-board.

3. L section which adds rigidity along the Y-axis and connects to the machine frame.

4. 25mm x 25.4mm aluminium spacers. Hole-sawing them at the mo - Pff, long and dirty job...

7726

5. 25mm thick plywood spoil-board.

6. 6mm plate. Ball-screw bearing blocks will mount directly to this.

The T-bolts that secure the aluminium extrusions to the top plate will pass through the top-plate, spoil-board spacers, L section c-channel before being fastened with a nut/grip-washer to the underside of the c-channel.

A general and straight-forward question here. Is this viable?

I have some ideas about how this is all going to connect to the frame, but I'll leave that for now and get a diagram together in the near future... Same goes for the Z-carriage. Meanwhile here are some pics of the adjustable feet I made... First time welding and also had a go a bluing the mild steel plate, didn't come out too bad...

772777287729

Is there anything here that represents an alarmingly bad design decision? Or could this actually work?

(Bear with me on the wooden legs, all will be explained...)

Cheers!

D.C.
21-12-2012, 11:46 AM
I'm not 100% certain but I believe the purpose of a spoil board is to get spoiled by chopping into it either accidentally or on purpose. When your spoil board is spoiled past being able to use any more how are you going to change it for a new one without dismantling the whole machine?

Wal
21-12-2012, 12:00 PM
Hi D.C.

Yeah, I hear you on that one - it's not ideal, but I'm going to be clamping a thinner bit of ply between the work and the spoil-board and cut into that rather than the actual table surface.

Cheers.

w./

Jonathan
21-12-2012, 12:18 PM
X Axis driven by 2x 20mm ball-screws with a lead of 5mm driven by a NEMA34 motor connected by cam-belt/pulleys.

Y Axis driven by a 16mm ball-screw with a lead of 5mm driven directly by a NEMA23 motor mounted to gantry side-plate.

Z Axis driven by a 12mm ball-screw with a lead of 2mm driven directly by a NEMA23 motor.

1. The first plate will act as a mount for the linear bearings riding on 20mm supported rails (TBR20 from Zapp) mounted to 80mm aluminium extrusion and will also act as a brace

From your selection of ballscrews and bearings it appears that you are planning on ordering them from England? If you want to save money on them then linearmotionbearings2008 (http://stores.ebay.co.uk/linearmotionbearings?_trksid=p2047675.l2563) on eBay offers any size you need if you send him a message, along with fast service. I'm not sure if he has TBR bearings still, so you may have to use SBR. TBR are stronger than SBR, but you could probably compensate for that by just using one size bigger, i.e. SBR-25mm.

Using 20mm ballscrews is only beneficial on the X-axis, of a router, if they are 10mm pitch. You can get 10mm pitch, 20mm diameter, ballscrews from China, but not on eBay. If you used RM1610 the critical speed will limit your rapid feedrate to about 6-7m/min, but that's still respectable for this size machine. To get any more is expensive. See this post (http://www.mycncuk.com/forums/router-build-logs/4572-new-cnc-mill-horizon.html#post39527).

Most people use RM1605 on Z, but if you want to use something smaller (e.g. RM1204) that would work just fine but there isn't really much point unless you're very short for space.

20mm is plenty of thread engagement for M8 in aluminium.

njhussey
21-12-2012, 12:27 PM
Hi Wal, the single 15mm Y axis (or I think you call it the X axis) plate could do with stiffening as I don't think it will resist twisting very well, especially if the Z axis is at max travel.

Wouldn't an easier solution to your 25mm dia Ali bits you're painfully cutting out if plate have been to buy 25mm round Ali, cut to length, mark the centre and drill?

Wal
21-12-2012, 12:39 PM
>Wouldn't an easier solution to your 25mm dia Ali bits you're painfully cutting out if plate have been to buy 25mm round Ali, cut to length, mark the centre and drill?

Maybe, but I don't have a metal saw - at least this way I can be sure that each spacer is an identical height - buying stock cut to length is always a bit hit and miss... The guide bit drills the centre hole too! Bonus.

Yep, it's a pain - but on the next batch I'm going to try drilling a guide hole straight through, cut from one side, flip, re-align and from the other side. It's the swarf not rising cleanly out of the cut after 15mm or so of depth that's making this job a pain - flipping the piece should solve the problem. I can only hope!

Cheers.

w./

njhussey
21-12-2012, 12:45 PM
I've got a B&Q hand mitre saw which cuts Ali ok and set a stop so the lengths are identical. I'm ok as I'd use the works lathe if i was doing it but cutting my Ali for my cheap router gantry was done in this way.

Cutting from both sides.....ooooooo good luck with that, I can never get cuts to meet in the middle doing that lol

Wal
21-12-2012, 12:49 PM
>the single 15mm Y axis (or I think you call it the X axis) plate could do with stiffening as I don't think it will resist twisting very well, especially if the Z axis is at max travel.

Do you not think that the plate that it's bolted to (which sits horizontally behind it) would be enough? When the Z is at max this is the part that's going to be taking the brunt of the forces - hmm... Will have a think.

Cheers.

w./

Wal
21-12-2012, 12:50 PM
>Cutting from both sides.....ooooooo good luck with that, I can never get cuts to meet in the middle doing that lol

Hehe. Yep, I know what you're saying. Luckily these spacers will be hidden from view... ;)

w./

njhussey
21-12-2012, 01:28 PM
Not sure how you're doing your Z axis but if you put some bracing across the top it will stiffen it right up, something like this....

7724

Probably need to put a brave in between the top and bottom in the middle to stop vibrations too...

Wal
21-12-2012, 06:55 PM
Ah yeah, I see what you're getting at. I'll probably mount mine a little bit lower so that it can accommodate the e-chain. Nice. Two birds, one stone.

Cheers.

w./

JAZZCNC
23-12-2012, 12:02 PM
The primary goal for this project is to be able to cut hard/soft woods and be accurate/repeatable in doing so.

Specs so far:

Cutting area maximum 5'x3'.

X Axis driven by 2x 20mm ball-screws with a lead of 5mm driven by a NEMA34 motor connected by cam-belt/pulleys.

Y Axis driven by a 16mm ball-screw with a lead of 5mm driven directly by a NEMA23 motor mounted to gantry side-plate.

Z Axis driven by a 12mm ball-screw with a lead of 2mm driven directly by a NEMA23 motor.

I see why your thinking 20mm screws to avoid whip so you can have 5ft cutting but in all honesty they are really going to compromise this machine just for wood use.?
Why not drop the cutting area to 4ft which is standard sheet width or better still make the machine 4x4 to accept half sheet.?
This 300mm less screw length puts you within the 16mm screws range, yes it's near the limit but still usable. This will then allow you to use 10mm pitch screws on both X & Y axis which are really what's required to give speeds needed for good wood machine.
Nothing wrong with joining screws(my preferred way) with belts or using 34 motor so long has realise you'll get slight less speed than you will from the single 23 driven Y axis. Thou that said thats the same if you drive X with 2 x 23motors due to having to keep a good safety margin top avoid stalling. Also the speed imbalance can be regained with a bit of gearing.
Whether you use 2 x motors slaved or larger single motor is debate argued many times before so won't go over again.!

The way you have the machine spec now you'll be lucky to get 5mtr min rapids and be in the 3-4mtr cutting limit. This means the motors will be working near flat out all the time while cutting wood and at times not fast enough. This opens up lots of potential for dropped or missed steps when working hard due to being near the motors corner speed all the time. 10mm pitch doubles the speeds which means you'll be cutting with far more torque available at the same 3-4mtr/min making the machine far more accurate and reliable. It also means that you can cut faster if needed which you haven't got an option with 5mm pitch.

Yes the 5mm pitch can be geared 1:2 to give same speeds has 10mm and I know it works because I use my machine like this. But at the time I had no realistic cheap alternatives so had to take that route but thats not the case these days and if building again I'd use 10mm pitch.




Some questions so far.

I'm using 15mm 6082 aluminium plate for this build. I'm hoping to use M8 35mm hex socket head cap screws to bolt this all together, which means the screws will bite 20mm of the thread I'll tap into the plate edges. Does this depth sound robust enough? Is M8 too big for the edge of 15mm plate?

To allow for a bit of adjustment I'll be making the holes in the face of the plates 10mm in diameter so there will be a bit of slop to play with before torquing everything up once it's all aligned. Does this sound okay?

Where there are slots cut for plate to pass through (the arm which passes through the side-plate, for example) - rather than cut this slot to be exactly 15mm wide, should I make it slightly wider (say 15.25mm) to ensure that the plate passes through okay? .25mm too much?

Obviously there won't be any internal right-angles because the end mill will have a radius. Where there's a square edge that fouls a radius, I'll manually chamfer off the corner to make it fit (see pic below).

M6 or M8 will be fine and 20mm thread is plenty.

With the gantry and slot then I wouldn't make it like that at all and your creating lots of work and expenses. just use flat plates bolted to each other far easier and cheaper than wide piece of Ali plate.


now on to the bed that the gantry sits on:

A general and straight-forward question here. Is this viable?

Is there anything here that represents an alarmingly bad design decision? Or could this actually work?

YES this bed design is a VERY BAD IDEA and the base board going under the rails will be a HUGE MISTAKE.!!
First it will absorb moisture and swell turning the machine into a shape shifter with temperature or humidity changes. Even if sealed it's not good.
Also if you mess up and cut thru the bed which is easily done or you spill drink or water anything that will cause damage then it's a major job to change.

DON'T DO IT:!!

If you can weld then weld the frame together and drop the spacers and just bolt base board to frame. Much quicker and easier to build has well.!!

Wal
23-12-2012, 10:43 PM
Hiya Jazz,

Thanks for your post. I'll take your advice on the 16mm/10 pitch - 5'x4' size was the absolute maximum I was considering, but as well as dropping the need for this machine to be able to cut aluminium, I'm also happy to drop my size specs to something that will do the job I need it to do - think guitar making - I'd actually be very happy with a usable cutting area of just over 3' x 2'...

Okay, I'm going to change the way that the spoil-board works. Yep, silly idea to have everything bolted through it.

Right, those wooden legs you saw in the OP are going to be held together by a steel angle iron cross braced sub frame - bolted together. I can't weld this. If I welded it then it's not coming out of the cellar ever again. It's just not going to happen. The upside to this is that realistically I'll probably use this machine once every couple of months, so it's not like it's going to be getting a good shaking 24 hours a day, 7 days a week...

Jazz, when you say that you wouldn't build the gantry in the way that I'm planning on doing it - do you mean that it's a shite design that won't work, or are you coming at it from a purely expense/ball-ache factor?

I've been 'thinking' about this build for the best part of a year and if it takes me another 6 months to chamfer out the tabs once the plate is cut, then hey - I'm cool with that! I already have my 15mm 6082 stock - 2 huge pieces that are more than enough to cut the gantry I have planned so that's not something that's going to trouble me. So long as whoever cuts my plate (and that might be you Jazz, if you have the time and fancy taking on the job...) gets it right, then the 4 piece design just seems like a really neat and simple solution to me... Looks good too, although from a pure engineering point of view that may be low on the list of priorities... What I'm saying, though, is that I don't mind this taking me a while to build and I'm happy to spend a bit on it.

w./

JAZZCNC
23-12-2012, 11:54 PM
Jazz, when you say that you wouldn't build the gantry in the way that I'm planning on doing it - do you mean that it's a shite design that won't work, or are you coming at it from a purely expense/ball-ache factor?

No not shite and with some modification will work fine I'm actually building something bit similar.

7742774177437744

I was thinking more about the cost and the work involved for someone without another machine to help. Which from reading your post's I didn't think you had.!
Thou still I wouldn't do it like that to be honest has it's too inflexible regards adjust for things like ball-screw alignment etc.
Keep the shaped bottom if you like but I'd drop the extended/slotted front plate idea and go with adjustable stronger brackets for ball nut mounting.

Like someone pointed out you absolutely need piece across top for support. I'd also put in some braces between upper/lower and front across the width to help reduce resonance and all though these pics don't show it this gantry has two. They also provide support/fastening for a rear plate enclosing screw and stepper.

Wal
25-12-2012, 11:40 PM
Cheers Jazz.

Nice looking gantry there. I entertained the idea of vee-bearings across the top and bottom edges with the drive passing through a slot like how you have it. Turns out getting hold of vee-bearings/rails etc isn't as easy and certainly not massively cheap enough to abandon supported rails.

Anyway. Yeah, you're right again. The slots are probably over-egging it. It works on paper, but I'm assuming that the plates are gonna be dead on 15mm, which is never the case. I'm going for something like this instead:

7754

That's oversize, but there's the idea... When you talk about building in a bit of adjustment to things like brackets etc. - what do you mean exactly? Loose tolerances on clearance holes? That kinda thing? I'm a total noob, man - you'll have to forgive the stupid questions, but any pointers/links to illustrations always helps..!

In an earlier post, you recommended that I go for 16mm/10pitch ball-screws which, at 4 feet, would be running close to their limit - any reason why I shouldn't go for 20mm/10pitch screws, like these (http://www.zappautomation.co.uk/en/14-20mm-diameter-ballscrews/673-r2010-ballscrew.html)?

Incidentally - when I'm getting my plates cut, is it a good idea to get them skimmed to 15mm if they're a little bit over in thickness? You know, to maybe help get any surface 'bumps' out. Or does it really not matter that much?

Hearing you on the extra brace at the back/top of the gantry. Haven't got a clue where I'd fit bracing across the front of the gantry... If I'm bolting through the front of the gantry into the bracing at the rear, then surely that's going to increase rigidity in both directions (albeit in one of those directions it'll be the bolts doing the work...) I imagine that for cutting wood at moderate feed-rates with fairly shallow plunges the stiffness here should suffice without popping bolts...?

Thanks again for your help.

JAZZCNC
26-12-2012, 02:15 PM
That's oversize, but there's the idea... When you talk about building in a bit of adjustment to things like brackets etc. - what do you mean exactly? Loose tolerances on clearance holes? That kinda thing? I'm a total noob, man - you'll have to forgive the stupid questions, but any pointers/links to illustrations always helps..!

Little clearance but not too much.! Really I mean adjustable brackets that can be removed and shimmed or with slots etc into alignment making it easier to fine tune.
Building the machine to within 95% of alignment or accuracy is relatively easy the last 5% or less can be a proper bitch and it's this 5% that makes the difference between OK and Great.
In ideal world you'd have surface grinders,milling machine etc and precision measuring and surfaces to work from making building and alignment lot easier.!!
In the real world we tend to have Hack saw, pound shop pillar drill and Big persuader with corn flake box for shims so anything you can do to ease the pain the better.


In an earlier post, you recommended that I go for 16mm/10pitch ball-screws which, at 4 feet, would be running close to their limit - any reason why I shouldn't go for 20mm/10pitch screws, like these (http://www.zappautomation.co.uk/en/14-20mm-diameter-ballscrews/673-r2010-ballscrew.html)?

No if you have the funds then they are perfect size and pitch. Only reason why recommended 16mm was that most folks buy from Chai at linearmotionbearings has he's very very cheap but he doesn't sell them. They are available from other Chinese vendors has Jonathan pointed out earlier but I haven't used them.! (yet)


Incidentally - when I'm getting my plates cut, is it a good idea to get them skimmed to 15mm if they're a little bit over in thickness? You know, to maybe help get any surface 'bumps' out. Or does it really not matter that much?

Yes it helps immensely but to far less degree if the rest of machines important surfaces aren't done to same standard. The important surfaces being, Rail mountings, bearing mounts, ballnut and screwbearing.
Like I said above all these matter greatly for that last 5% and the better the surfaces are prepared the easier it will be.


Hearing you on the extra brace at the back/top of the gantry. Haven't got a clue where I'd fit bracing across the front of the gantry... If I'm bolting through the front of the gantry into the bracing at the rear, then surely that's going to increase rigidity in both directions (albeit in one of those directions it'll be the bolts doing the work...) I imagine that for cutting wood at moderate feed-rates with fairly shallow plunges the stiffness here should suffice without popping bolts...?


Not so much to do with mechanical strength but dealing with or lessoning resonances the machine makes. Any resonance or vibrations will ultimately end up at the cutter giving a poorer finish. If bad it can even affect performance of the machine by throwing the motors into a Cissy fit.!!
Bracing here(See pic blue bit) on gantry and else where on the machines frame help dampen them out.
7756

Wal
27-12-2012, 03:49 PM
Right,

Had another look at the gantry and done away with the interlocking plate idea. Made it a bit longer too, so the spindle isn't hanging out the front on its own... You should notice the back all braced up, too.

I think I can get it looking nicer - at the moment it kinda looks like it could plough a field, but I should be able to sleeken it up a bit...


http://youtu.be/vt4WcYf0vaQ

Wal
29-12-2012, 02:03 AM
Looking a bit sleeker. 2 options for the side-plates. Angles, or rounded...

77797780