[Oldmanandhistoy]

Well I’ve made a start on my third build. I have plans to build a CNC wood router which will be use to cut Plywood, MDF and Plastics; defiantly NO Aluminium. Cutting area of around 850mm X, 850mm Y and 75mm Z.

This will be used in a light commercial environment. It is not going to be used 8hrs per day and 5day a week; More likely 10 hrs a week.

Basically I’m not going to use commercial available linear ways or ball screws. As I feel these would be an unnecessary extravagance for a wood router working to a tolerance of +/- 0.1mm and this will only be really utilised for plastics.

First question, will it be ok for me to talk in millimetres? For now I will continue and hope this is ok as I am likely to make mistakes with conversions.

I’m looking for cutting speeds of around 1000mm to 2000mm/Min

Rapid speeds to be worked out, any recommendations?

Obviously I could go out and buy a second hand machine but to be absolutely honest I would rather build and also I don’t have a bottomless pocket. Having built two other machines I now have the bug big time and see this as a recreation rather than a chore.

Now you’re all probably thinking Aluminium but no. I am planning being a wood man to build using a hardwood multiply reinforced with steel where needed.

Before you all go to avoid boredom I will add a picture. This is my Y axis test set up. Basically it is two pipes with a torsion box from 12mm ply and skinned with hard board. I will not be using this in my machine it is for test purposes only and will be scrapped after testing. The torsion box was cut using my other small cnc router. You may be able to see my bearing arrangement I will elaborate soon on these.

More to follow.

[Oldmanandhistoy]

Let me explain my self when I say “time to get serious”.

My first build was primarily a MDF construction which I used to learn the basics. This machine was scraped after cutting parts for my second. The second cnc router was made using multiply, pipe, skate bearings and draw slides for my Z axis. I am still using my second. This has a working area of 350mm x 610mm x 50mm.

I designed and built this router with information I got from this excellent forum.
Now I want a more serious machine to work in my cabinet shop. Not being an engineer I would like some help with the design process. So where better than here. I would like to build a machine that is as efficient as possible.
When I said no commercial linear ways I should of add only in the Z axis and these will be 12mm round rail with linear bearings to suit.
The X and Y will probably be drill rod and skate bearings (because I have experience with these and like the set up. This may change as I go along but probably not.

Where I really need help will be with working out cutting forces and the like so I can add reinforcement using steel where necessary.

For instant I would like to turn my Y axis test set up on its side and add weight to represent cutting force and measure the flex and twist adding steel where necessary to counteract the effect..

Now how do I work out the weight to be used? I have worked out that the heaviest cut I will make will be 12mm wide by 6mm deep using a TCT cutter with two flutes in plywood at 2000mm/min hopefully.

Can anyone help me with working this out or knows of a site with information I need?

[Oldmanandhistoy]

I’ve done a Google and been unable to find the information I need where cutting forces are concerned.

So now I’m thinking I may need to set up my test Y axis with an X axis and do some pulley and weight tests.

How I would do the tests.

With a pulley and weight attached to the y axis calculate as near as I can the weight needed to move the Y axis 1 meter in 30 seconds.

Then fix a router with the depth set to my 12mm wide and 6mm deep and drag this through ply at 1 meter / 30 seconds.

Subtract the first weight from the second and that would give me the weight to test my Y axis turned horizontal.

Questions.

Does this sound like a good method to calculate my cutting forces?

How do you other builders calculate how strong or how light you need your build to be?

John

[ger21]

Now how do I work out the weight to be used? I have worked out that the heaviest cut I will make will be 12mm wide by 6mm deep using a TCT cutter with two flutes in plywood at 2000mm/min hopefully.

The force required is dependant on tooling type and sharpness, and spindle speed. It also depends on the material, as there are different types of plywood.

On our commercial router, we use spiral chipbreaker tooling to cut plywood. It cuts faster and quieter than regular spirals, and much faster than straight bits. We do on occasion, though, use carbide insert tooling, but cut much slower than with spirals, about 3000mm/min.

From my experience, and from what I've read in the past, you'll probably be looking at between 10-30lbs force. I'd design for 30-50lbs, personally.


While I'm currently finishing my first homebuilt router, I also have plans for a much faster machine, and also made of wood. Instead of using solid hardwood, though, I plan on gluing up laminated beams, probably with epoxy, to get a bit more stability and strength vs solid hardwood.

One area of concern I noticed on my router, is the connection of the gantry beam to the gantry sides. This can be a very weak point on a wooden router.

[Oldmanandhistoy]

Thanks ger21

Nice to have your input as I know you have a large amount of cnc knowledge.

Just because I like to understand things could I ask you how, where you got the information; did you do your own testing?

I plan to add steel section which will be 16mm x 3mm to the torsion box and epoxy them into place.
Adding as many as necessary to prevent flex and twist.

What are your thoughts on that idea?

I will add a picture tomorrow of my design if it makes things clearer.

John

[spalm]

Hey Old Man, (I always wanted to say that)

Looks good. I don’t have any numbers for you, but I have built a Y set up very close to what I see. (I would like to see a close up of your bearings) The suggestions I have so far are two things.

Gerry mentioned the first: a good connection of the Y beam to the gantry walls. Maybe mortise a rectangle in the gantry to receive the beam or extend the back side of the beam through the gantry walls. Looks like you did a sandwich for the pipe support rather than a torsion box. Seems to me this is sufficient. I don’t think flex here will cause you any problems. I spent a weekend trying to dovetail my beam into the gantry with frustrating results. Anyway, give this some thought.

The second is the probable tilt of the wrap-around Y carriage. I don’t know what your bearing setup is but both Gerry and I saw some of this. Try to keep the Z as close to the beam as possible should help.

I tested my carriage with a 30 pound motor clamped to the front and a metal ruler clamped to the carriage that just skimmed the cutting surface, By just pushing the carriage back and forth I could see the gap and make any design changes.

Steve

[whirlybomber]

could I trouble you for some close-up views of your bearing mountings?

cheers,
Brad

[Oldmanandhistoy]

Hi spalm

Lol @ “hey old man”

So do you guys think if I mortise the beam into the gantry sides it would eliminate the problem?
If necessary I could also add a couple of bolts either side through the gantry wall and into the beam. Would this be over kill?

I am trying to keep the weight of my Y axis as low as possible for all the obviously reasons. I’m trying to think like a model aircraft designer would think. Only add weight if absolutely necessary to increase strength.

Thanks for you tip on testing, but I am more interested in the horizontal forced which would cause the beam to bend and twist. I probably should of said bend and bend as it is the fact that the top of the beam would bend one way and the bottom the opposite.
Having said that, I would be interested to know the forces necessary for plunging.

I also understand the cantilever forces on the Z axis and always try to keep the Z axis as close as possible to the Y. My first machine actually had the Z axis screw behind the Y axis screw. Although this is ideal it did restrict my Z axis movement. But it would be an ideal approach for others building a small machine say for milling PCB’s.

The weight needed to test this I would say would be critical to the whole machine as it will affect every component in all 3 axes.

I know it is impossible to give an accurate figure as it will depend on large number of different factors and my proposed method of testing would only get me so close because of the acceleration factors and the like. I need to give this serious thought before I proceed.

I have added a couple of photo of the bearing arrangement. Hopefully this will show you how I do this. I use this method in the router I am using at present and it works very well for me. The first picture shows the bearing set at 120 degrees and this is how I did my small router. The part you see is for a cnc router I am building for a friend.

The second picture shows the arrangement I might be using on this build.
Please feel free to pick this idea to bits if you see fit.
.
If this does not make things clear or you would like more info let me know and I will elaborate more.

John.

[ger21]

Just because I like to understand things could I ask you how, where you got the information; did you do your own testing?

I read it on one of the Yahoo groups, I believe someone did some testing using a fish scale, or something similar.

[ger21]

So do you guys think if I mortise the beam into the gantry sides it would eliminate the problem?

I don't think the mortise would help that much at all. The beam acts like a large lever, which can put a lot of force from one side to the other. If you're driving the gantry with 2 screws, it won't really be a problem. What I'd recommend is to use some 1/4" aluminum angle, or 1/8" steel, maybe 2"x2", on each side of the beam to attach it to the sides. This will spread the load over a much larger area.

[Oldmanandhistoy]

I don't think the mortise would help that much at all. The beam acts like a large lever, which can put a lot of force from one side to the other. If you're driving the gantry with 2 screws, it won't really be a problem. What I'd recommend is to use some 1/4" aluminum angle, or 1/8" steel, maybe 2"x2", on each side of the beam to attach it to the sides. This will spread the load over a much larger area.

I have been planning on using two screws on the X axis but see your points.

I know I may have got away with one in the centre but I also know I will end up building a larger router in the future so the two screw method will be good practice and I believe a better method even on a small machine like this.

Another thing I would like others to ponder on is; how I will build my X axis rails.
I have been thinking of using two Y axis assemblies turned horizontal. So that would be four rails and six bearings per rail. I believe horizontal forces will be reduced to manageable levels but am concerned about the rails sagging over time.

Your thoughts would be very much appreciated. Maybe I should think about using them vertically instead?

[WhiteTiger]

I'd go vertical, personally. The static vertical weight load is forever but the lateral cutting and inertial forces are relatively minor and only when the machine is cutting. Imo, of course



Tiger

[ger21]

If you can figure out how to mount them, I think the best place for the bearings is vertically, on top (and bottom) and horizontally on the outer sides of the tubing. What I've found with my bearings at 45° is that if you overly tighten them down, they start to get very rough due to the excessive axial load. This is pulling them tight together with threaded rod. If you have 100% radial loads, you can get them very tight and they'll still stay very smooth.

[Oldmanandhistoy]

WhiteTiger

I concur with your opinion which is very welcome. The more I think about it the more sense it makes.

Obviously if I make the sections as I would for the Y axis they would only need to be half as stiff as the same horizontal force will occur as will in the Y axis.

But is this a good method to use for my X axis or should I stick to more conventional methods?

Ger21

I plane to use a configuration as in my second picture (the one with bearings fitted). But have noticed in my test set up that they also stiffen up with higher pressure to the point that when turned on it end so the bearings can travel down the Y axis, they actually stick.

What I have been hoping is that when the bearings form a flat on the pipe which will increase the contact area they will move freely with adequate pressure.

I have noticed also how quickly the bearings wear the steel tubing I have used in my test set up. After maybe 50 times travelling up and down the Y axis they have formed a flat of about 3mm wide. That would work out to be roughly 0.2mm into the pipe, or 0.4mm adding twice that for two bearing set either side.

If instead of using drill rod I used stainless steel pipe and ran the axis up and down its full length until it formed flats on the pipe (could do this with some gcode) then tightened the bearing to a pressure that was necessary to run the machine; would it then wear at a more reasonable rate so as not to need regular readjustment? Also would have problems with wear nearest the home position as this area would get the most use and less at the opposite end of the axis.

Or should I just use square section instead of the round even though it makes things more complicated where construction is concerned?

[whirlybomber]

Oldmanandhistoy,

It looks like your frames are routed out of timber. How/what are you using to support the axles for your bearings in them?

If you were to use a piece of steel rod, would it be able to be clamped tight enough without distorting/damaging the frame?

cheers,
brad

[Oldmanandhistoy]

Hi Whirlybomder

This part of the frame is made from 24mm multiply.

Basically each bearing is mounted on an 8mm diameter axial (steel rod) with a washer either side.
I use the two screws per bearing to adjust the pressure on the pipe; tightening them causes a wedging effect between the axial and the taper on the screw head.

One other point is that the ends of the axils have a 4mm radius which helps with this effect.
I can get more pressure with this set up than I need.

I have noticed no distortion of the frame.

What I would really like to do with this idea is cut them in an aluminium block and improve the adjusting screws. Unfortunately my cnc router is not stiff enough to cut aluminium.

John

[Oldmanandhistoy]

Up date

Having given my Y axis beam some thought I am considering changing the design.
I have added a picture to show possible new design giving an end view of the beam. Red items are skate bearings, blue is steel, yellow is 12mm ply and green is 4mm ply skin.

Dimensions A and B have yet to be decided.

The steel will be replaced with an alternative, yet to be decided upon.

I am planning on giving the ply surfaces a skin of epoxy resin (still researching this idea).

Your comments are appreciated.