Jill of all trades, but mistress of few !

Hi,

Where DIY is concerned, I'm a bit of a Jack (or should that be Jill?) of all trades, but mistress of few (if any!). I have been establishing a woodworking workshop which, for a hobbyist, is reasonably well equipped (table saw / spindle moulder combo, planer / thicknesser, band saw etc.). For some time I have thought that a cnc router would be useful to add. In truth, I probably can't afford it or justify it, but hey, since when did such details impede the female of the species when in need of a little retail therapy? However, it does mean that I should seriously consider the DIY build route.

My interest is centred on a router for woodworking. Ideally, I'd like to leap in the deep end and go for a full 4' by 8' working area. In order to be able to undertake reasonably heavy work in either soft- or hard-wood, I imagine I should be thinking in terms of a router spindle of around 2kW. What workpiece hold down system is appropriate? Is a vacuum system at all practical for a home builder? What are the best sources of equipment in the UK? (I've come across Marchant Dice so far.) I've started doing a bit of research online, but still have a lot to learn.

As I'm not really into metalwork, I'm thinking of construction primarily from wood / MDF. My existing wookshop set up should enable me to undertake such construction with sufficient accuracy. I've found various links to websites offering plans for specific designs and / or showing videos of DIY build projects. However, I'm still trying to find a good basic "idiots' guide" on the subject that would help me to identify a suitable construction and the best / most suitable components to specify.

Any pointers are most welcome.

Julie

Quote:

---

Originally Posted by Julie

I'd like to leap in the deep end and go for a full 4' by 8' working area. In order to be able to undertake reasonably heavy work in either soft- or hard-wood, I imagine I should be thinking in terms of a router spindle of around 2kW.

---

Excellent, another Sussexite and one who thinks big :beer:
The cheapest way to track it will be round, supported rail. Driving it will be fun, as will keeping the 4' gantry running square. Stepper motors are cheap but there is always a trade off between accuracy and speed when you go large size. Servo motors could be the way to go. You don't want to wait 5 minutes while it goes end to end. You don't want to rely on fine microstepping to do fine detail.

Maybe allow for a vacuum bed in the design but start by using a crummy adhesive such as the humble Pritt stick while you get a feel for the problem.

Do you have a specific item for cutting in mind? Suggest you post preliminary design sketches here and see if you get a big thumbs up or howls of anguish :whistling:

Hi Julie,

Welcome to the forum.

Only advice I can give is:

Know what you want to make beforehand and size your machine accordingly. It's pointless having an 8X4 router and then having it only making 18" long house signs.

CNC is really, REALLY good at multi-run jobs, but not for one-offs unless you can price it right. It ISN'T just another tool like a drill or a planer, it is a manufacturing system that can do the jobs of all your machinery in you workshop but I bet 90% of the time you will use the drill or the planer or whatever simply because it is easier and quicker to do than writing the program or generating the CAD/CAM. WHen I first built mine and I wanted to say cut a 20mm groove across a sheet for a shelf to fit into, I would clamp it down and zero the machine, then type in a move to the appropriate distance up the sheet, lower the Z to the required depth and then simply zip the cnc across the sheet, job done. I don't do that anymore, I just clamp a straight edge across the sheet and run my triton router across it by hand like I used to, quicker and easier.

If you want to get into 8x4 size machines and cutting the likes of oak etc. think of getting into welding classes and build it from steel with supported rail on all axes, rack and pinion on the 8' axis and ballscrew on the other two. IMHO an 8x4 from MDF or baltic birch can't hold a candle to a steel or aluminium framed machine for rigidity, Irving will be able to fill you in on the deflection of wood vs steel. If you are thinking of doing this professionally, i.e. to earn a living from, then you really need to be building in longevity and accuracy IMO only a steel framed design will give you that on the scale of machine you are looking at.

You are looking at splashing £2,500+ to get a half decent machine up and running, 8x4 really is getting into commercial territory size-wise and it might be cheaper or more sensible to look at a second hand commercial machine. If you want to build one for the experience, knock yourself out, that's why I built mine and I can honestly say it's the best thing I've ever made and the most fascinating hobby I've ever had.

I don't want you to think I'm trying to put you off but do yourself a favour and decide in the first place what you want to actually DO with the machine and stick to it!

All the above my own humble opinion as always but good luck whatever you do and don't forget if you start to build one, we want to see pictures!

Jeff

Hi Robin,

Thanks for the quick reply. I guess my continuing to set ambitions above abilities is an on-going victory for hope over experience!

I'm interested in what you say about servo motors, rather than steppers. If you servos at all, I guess it's best to use them on all axes to have common control logic?

I appreciate your point about a 4ft gantry. I had seen the Blacktoe 4 x 8 project on buildyourcnc.com, which uses MDF for construction. However, I did note that the gantry structure was made pretty beefy. As for keeping it running square, at that size I'm guessing you need to provide a motor on each side to drive the x-axis?

Still a lot to learn.

Julie

Hi Jeff,

Thanks for the reply. I'm not put off by your words of warning and do appreciate all advice and points of view.

I do have some clear ideas on what I want to do. My most immediate requirement is to do engraving on a large number of wooden panels. They will not be very wide, but will be of varying length, with many of them being in the 2 to 2.5 m range. My current solution (which I have prototyped) has been to make up a wooden jig then use a bearing guided cutter in my hand router. It works, but I would expect to able to get better accuracy and finish from a cnc tool. Beyond that, I have other ideas, but nothing so specific yet. However, I think I'm sufficiently clear on my needs and ambitions to decide that if I do this at all, I should go for the size I've suggested. Equally, other functionality that might be nice to have, such as additional axes, are not high on my priority list at present. In summary, yes I do think I''ve got a suuficient handle on what my needs are.

I certainly wouldn't dispute your point that a cnc router would never be the best option for all tasks. I have many other machines and hand tools in my workshop that will continue to earn their keep.

I think welding is a skill too far for me at this stage. However, since my original post yesterday, I've noted one or two websites advertising ranges of extruded aluminium profiles. This has started me thinking. I should be able to handle that, provided I can lock everything together by mechanical means (i.e. brackets or other fixings).

I'm interested in your suggestion to go rack and pinion on the long axis and ballscrew on the others. Is this because rack and pinion is better over such lengths or is on the basis of cost for a ballscrew system with sufficient rigidity at that length? (I'm not doubting your advice, just keen to understand the rationale behind it.)

I was wondering whether anyone was going to come back on me with regard to my idea that the router spindle should be of the order of 2 kW. I note that commercial machines are at this rating or, indeed, much higher. The only DIY offerings I've seen so far though are the Kress models that only go up to about 1kW. I've noticed that large commercial machines often talk about water cooling. I wouldn't want to go to that complexity. I could go 3-phase (off an RPC that I've already installed to drive my saw / spindle moulder) but would probably rather stick with single phase.

I have had a long term search set up on ebay for cnc routers, and while they certainly come up regularly, anything that might fill my wishes (adequate size, not too ancient) still goes for more money than I could contemplate. For example, there's currently a Pacer Cadet on offer. It has only a 1220 x 700 mm bed, albeit with computer and vacuum pump thrown in, but the seller is seeking around £6k.

As you can see, I really would benefit from a good, comprehensive idiots guide if one exists.

Thanks

Julie

The problem with ballscrew is that the longer you need, the thicker it has to be, or the slower you need to drive it (otherwise it whips).

Rack and pinion doesn't have problems with length (and if you need it longer, you can "just" bolt on another length).

Welcome to the forum Julie, promise I won't swear (too much)!

I have been considering designing a larger machine also and have been lurking around youtube and various 'other' DIY CNC sites for idea's before I put mouse to screen.

Found this site on my travels - http://www.automatedwoodworks.com/

I think this guy has done a half decent job, take a look at his gallery and his youtube video's. He is using steppers which are a much cheaper option than servo's and he is getting decent rapid speeds.

Like other people have said, with steppers its a trade off between speed and accuracy, on my current machine I am using 16 x 5 ballscrews which are directly driven off of a 23 size 3.1Nm stepper, with a 5A driver I can run comfortably at 5m/min and I have had it up to 7m/min. I am running at 1/4 step so this give 800 steps per 5mm, this = 0.0063mm which is fairly accurate!

I am thinking of using the same steppers and drivers for a larger machine but gearing them 2:1, this should double the rapid speeds and half the accuracy but still give me 0.0126mm per micro step. Pretty sure that the 23 steppers will be fine but may consider going up a size to be on the safe side.

Will probably belt drive it and just use one stepper for the long X axis with a shaft to link the drive belts on either side to avoid any crabbing.

Just my two penneth worth!

Quote:

---

Originally Posted by Julie

I'm guessing you need to provide a motor on each side to drive the x-axis?

---

Hi Julie

That's to be avoided like the plague :nope:

First you have to decide what kind of accuracy you want to achieve for the cut. Think resolution. The motor moves the table by one increment. A small increment improves the accuracy of the cut. A large increment improves your top speed which is handy when you want to move 8' ASAP.

When you have an accuracy figure in mind you can decide how to drive it. Screws, rack and pinion, belting? The lower the accuracy the cheaper it becomes. It's all trade offs, what are you trying to achieve? Are we talking mass production where speed is king? Are we talking bespoke and it doesn't really matter if it takes 2 minutes to move end to end at a lowly 2 cm/second?

Robin

I think she means the Y axis.
Moving a large gantry is better to drive from both sides or you can get crabbing even with good quality profile rails.
This is more promenent if you are cutting on one side.

Quote:

---

Originally Posted by Robin Hewitt

Hi Julie

That's to be avoided like the plague :nope:

First you have to decide what kind of accuracy you want to achieve for the cut. Think resolution. The motor moves the table by one increment. A small increment improves the accuracy of the cut. A large increment improves your top speed which is handy when you want to move 8' ASAP.

When you have an accuracy figure in mind you can decide how to drive it. Screws, rack and pinion, belting? The lower the accuracy the cheaper it becomes. It's all trade offs, what are you trying to achieve? Are we talking mass production where speed is king? Are we talking bespoke and it doesn't really matter if it takes 2 minutes to move end to end at a lowly 2 cm/second?

Robin

---

Hi all
Just an idea regarding the x axis 2 motor thing, if i were doing it i would put a lead screw/ballscrew on each side conected by a timing belt and pulleys, with a double pulley on one side then drive the second pulley from the motor, also this gives you some flexibility with pulley sizes, small motor pulley for accuraccy large motor pulley for speed. Shoud also work with rack and pinion systems with a long drive shaft.
Thats the way i would go if i were building it from scratch, that way its just the one motor driving both sides of the axis, and a swap out pulley set up with the progam,
Thanks for listening
Still a newbie so dont take me too seriously..

Hi Julie,

RE: driving the long axis with 2 motors. You don't have to with R&P!

All I did was to make a shaft with a pinion on both ends and a drive pulley inboard of one of them. Then simply mount the motor in line with the pulley and you are driving both sides of the gantry with 1 motor, no crabbing, simples.

If you want to get fancy you can use a zero backlash pinion on one end but I didn't think it needed on a router.

Jeff.

The principle is the same if using screw or rack.
drive it from both sides and you won’t have a problem.

Quote:

---

Originally Posted by Gary

The principle is the same if using screw or rack.
drive it from both sides and you won’t have a problem.

---

Of course it is. However, once you get into long travel lengths, unless you need the super duper accuracy of ballscrew then IMO R&P wins hands down. It is also much easier/mechanically simpler to drive both sides of a rack with a single motor and a single motor is what I would propose. Maybe Techserve could give us some insight on how they do things on long axes.

End of the day this is a router and R&P will give Julie all the accuracy she needs whether she is carving long panels or cutting parts for furniture. All I would suggest is she drives both sides with a SINGLE motor (bought from Zzap).

Jeff.

Totally agree R&P is far better for long axis, and cheaper.
Techserve use R&P, infact most plasma machine manufactures do.
The accuracy that is needed with a router, will be fine with R&P, however, normally will requre some gearing, but simple belt and pulley will do.

Quote:

---

Originally Posted by Smiler

Of course it is. However, once you get into long travel lengths, unless you need the super duper accuracy of ballscrew then IMO R&P wins hands down. It is also much easier/mechanically simpler to drive both sides of a rack with a single motor and a single motor is what I would propose. Maybe Techserve could give us some insight on how they do things on long axes.

End of the day this is a router and R&P will give Julie all the accuracy she needs whether she is carving long panels or cutting parts for furniture. All I would suggest is she drives both sides with a SINGLE motor (bought from Zzap).

Jeff.

---

Hi,

Thanks for all the replies so far. I haven't disappeared as fast as I came, just been doing a lot of reading and research.

This immediateley throws up a load more questions. I hope others will excuse a few really basic ones. For starters:

1. I originally suggested wood / MDF construction, but having seen the construction methods others have used and investigated various suppliers of extruded aluminium profiles, that seems a much better way to go. I have so far come across a number of such suppliers, having downloaded the Valuframe catalogue and checked out the Rhonmac (same system as Valuframe?) and Marchant Dice websites. I've also seen references to Bosch. Is there anything significant to choose between these, or other, systems / suppliers?

2. I've read through all the posts concerning HiltonSteve's very impressive build. If I've understood it correctly, Steve, you've used just one guide rail each side on the x-axis (but 2 for y- and z-axes). Your gantry appears to have additional rigidity through the Al plate that connects each side beneath the bed. At the sort of size table I'm considering (around 8' x 4') I'm expecting to need to use at least 3 legs per side (I envisage it being free-standing, not bench mounted) and, very possibly, at least one under the centre. These additional (non-corner) legs would preclude an under-bed connection between the gantry sides. I'm presuming I would need 2 guide rails per side for the x-axis and 2 for each of y- and z-.

3. Again with reference to Steve's build, the entire gantry / y-axis structure seems very solid with 15 / 20 mm Al plate throughout. Does it really need this for sufficient rigidity? I imagine I will have a greater need for care here with a 4ft + y-axis, but was hoping that the use of extruded profile with, perhaps, rather thinner Al sheet skinning would suffice. Am I kidding myself?

I will shortly be exiled from the internet for a couple of weeks, so won't be replying again for a while, but this will give me some quiet time for further planning.

Julie

Hi Julie

The cheapest way to rigid is steel box section, but after making the frame you want to make your fittings adjustable because it won't be quite as straight as you require.

If you want long lengths of aluminium bar or box section, I suggest you stray outside the hobby market and go directly to a stock holder such as http://www.parkersteel.co.uk

Robin

Robin,

Thanks for the reply. I understand the reasoning for using steel, but my problem would be fixing it all together. I'm not into welding at all. Also, if I do go ahead with this, then I'm seriously contemplating installing it in the loft above my workshop and garage. The garage itself would be the only alternative as the workshop is already full! The garage would not be ideal as the floor is rough concrete with a gentle drainage slope towards the door. Besides, my supposed double garage is already almost full with wood store, racking for off-cuts / work in progress, workshop overspill etc. (Well, you know the way it goes. Nothing wrong with my workshop that doubling its size wouldn't put right!)

Assuming it goes in the loft, it will have to be assembled in situ due to access and, should I ever want to move it, it will have to be taken apart.

I accept what you say about leaving scope for adjustments.

Julie

Quote:

---

Originally Posted by Julie

1. I originally suggested wood / MDF construction, but having seen the construction methods others have used and investigated various suppliers of extruded aluminium profiles, that seems a much better way to go. I have so far come across a number of such suppliers, having downloaded the Valuframe catalogue and checked out the Rhonmac (same system as Valuframe?) and Marchant Dice websites. I've also seen references to Bosch. Is there anything significant to choose between these, or other, systems / suppliers?

---

I have not had any experience with many aluminium extrusion suppliers, I used Hepco for my 80x40 and when I ordered it I told them that the 3 lengths needed to be within 0.5mm and cut 90 degrees, when they arrived they were spot on, just tapped the holes and bolted them in.

Quote:

---

Originally Posted by Julie

2. I've read through all the posts concerning HiltonSteve's very impressive build. If I've understood it correctly, Steve, you've used just one guide rail each side on the x-axis (but 2 for y- and z-axes). Your gantry appears to have additional rigidity through the Al plate that connects each side beneath the bed. At the sort of size table I'm considering (around 8' x 4') I'm expecting to need to use at least 3 legs per side (I envisage it being free-standing, not bench mounted) and, very possibly, at least one under the centre. These additional (non-corner) legs would preclude an under-bed connection between the gantry sides. I'm presuming I would need 2 guide rails per side for the x-axis and 2 for each of y- and z-.

---

You are correct, I used 1 20mm rail each side for the X axis. For the Y axis I used a rail top and bottom of the gantry crossmember which is why I made the crossmember out of 20mm (same size as the rail), made it easier to mount and align. The Z axis has 2 rails running down the front face. So each axis has 2 rails and 4 bearings which make it very solid, no play anywhere at all.

Quote:

---

Originally Posted by Julie

3. Again with reference to Steve's build, the entire gantry / y-axis structure seems very solid with 15 / 20 mm Al plate throughout. Does it really need this for sufficient rigidity? I imagine I will have a greater need for care here with a 4ft + y-axis, but was hoping that the use of extruded profile with, perhaps, rather thinner Al sheet skinning would suffice. Am I kidding myself?

---

I used 20mm plate for the gantry sides and the crossmember, the rest is 15mm. I went for these sizes as that is what suited it best when doing the design, bolting 20mm rail to the edge of 20mm thick plate is quite easy! Then thought if the crossmember is 20mm then the side plates need to be so it looks right, then used 15mm for the rest because....well, I did!

Yes it may be over the top and I may have got away with 10 and 15mm plate but it would have made my design more difficult to put together. The cost difference of using thicker material outweighed any potential build problems I may have encountered and also guaranteed a good solid machine. Also I thought about how I could manufacture the parts and what machinery I had available to use, John S did the side plate profiling for me on his beaver CNC then I machined all the rectangular plates one Saturday morning on my mate's bridgeport, then drilled and tapped everything in the afternoon at home, so most of the main structure was made in a day!

The way I did it made sense to me because it was quick and easy to build because of what machinery I had available to make the parts, not everyone has these resources available so you have to do things differently. If I had a CNC mill to play with for a day to make my bits then I would have done things a lot differently!

Steel bolts together okay, but how are you equipped for drilling holes?

If space is at a premium, a low profile machine could have one long axis hinged to a wall so you can fold it up out of the way when not required.

A bit caravanesque perhaps :heehee:

Steve,

Thanks for the extra detail. That all makes sense.

Robin,

Have a heart!! I'm not into metalwork. My interest is very definitely woodworking. Hence, my desire for a large woodworking sized machine. In just one week I've gone from thinking in terms of wood / MDF for this build to conceeding that aluminium would be better. Now you want me to jump straight to steel. You'll want blood next.:sad:

A fold up solution wouldn't help me. My workshop is around 5.5m square, which isn't bad for a hobby workshop, but it's absolutely full. All my main machines, including 800kg of table saw / spindle moulder, are on various forms of wheels or castors so that I can pull out the particular one I need at any one time. There's just not sufficient wall or floor space available without discarding things that are already there.

You ask how I would drill holes. Therein might lie another problem! Before I got quite so seriously into developing a workshop, I bought a Shopsmith combination woodworking machine. I've subsequently replaced most of its functions with much more capable separate machines, but still use the Shopsmith as a vertical drill. In that regard, it's seemed adequate for my purposes to date without going to the expense of a dedicated pillar drill. It has a reasonable sized table to which I can attach a fence, and depth setting ability. Other than hand-held drills, that's my lot for drilling at present. Although I say I'm not into metalworking (beyond drilling and tapping the occasional hole), I can see that some limited capability here could have uses from time-to-time. To that end I've started giving a bit of thought to getting a milling machine. Don't know much about them, so guidance would be welcome. So far I haven't got beyond the Axminster catalogue. In there, the possibilities that have caught my eye include the SIEG U2 and Axminster RF40. Anyone have any experience of these machines by any chance? (Perhaps I shouldn't admit on a CNC forum that I'm contemplating a manual solution here! Perhaps a retrofit one day in the future might be on.)

My table saw can take an aluminium cutting blade claimed to be able to cut up to 30mm, which should be more than adequate for basic straight line cutting to size. (If it can cut 30mm solid, I guess it would be OK on 40mm extruded profile sections. With a 300mmblade, depth of cut isn't the issue and with a 5.5hp motor, I guess it would have adequate power.) That would still leave detail shaping and drilling, of course. Obviously, trying to cut steel would be another matter altogether and not something I'm equipped to deal with at all at present (beyond a hacksaw!).

All-in-all, while I hear you, I still think aluminium, whether extruded profile sections or not-too-thick solid, is better suited to my needs and capabilities.

Julie

Hi Julie,
i have been making cabinets for the home in solid Oak (not wood effect or verniered mdf) as i abore "rubbish" furniture. My workshop is approx 25sqMeters and i have a Fox 250mm table saw and Dewald radial arm saw, Dewald planner/thicknesser and a Ryobi Router table with a 1800W Dewald router.

Most of my joints that i do on my router are of the types Finger, 45deg Finger, and dado. I also do my raised panels on my router using a 1/2" 75mm Raised Panel bit and a Rail/Stile bit.

If i had the money I would have gone for a Scheppach Spindle Router.

If you use MDF please make sure you have excellent ventelation and dust extraction. MDF is a lung killer.

Regards
Roy

8X4 ... that's www.mechmate.com territory It may not be your cup of tea but it's worth a look and there is a new member on this forum who has built one ....

Just a silly suggestion which I haven't really thought through:

By analogy to the circular saws on the wall they have at B&Qs, if you are short of space could you mount the bed of the machine vertically? - essentially screw a normal machine to the wall. It would mean you would have to think about how you clamped things to it slightly more carefully but once they were clamped I don't see why it wouldn't work as normal. And if you were mostly working with sheet materials they are often easier to handle vertically than horizontal.

I guess you might need some form of brake on the Y axis as it might otherwise fall down if the power went out for some reason, and you may need a beefier motor to pull all that mass uphill.

Anyway I am sure there is something I am missing as I haven't seen any photos of this configuration.