[machineitright]

Hi everyone,
New to the forum and really like all the experience and know-how on here, would like to have your help on my build.

I have started the build with a torsion table out of some Advantech osb board that I had left over from a remodeling project, a cnc router seemed like a great way to use it up and get a new toy in the process. The osb seems to have a high strength to weight ratio, the table I have built setting on 2 saw horses with nothing between only flexes about .004 in the middle.

Here are some pictures of what I have so far and would like your thoughts.










Used plenty of this!!!

I will be building the rest of the router along the same design as the Black-Toe that I found online with a few deviations because of the materials and electronics that I already have. They were actually purchased to cnc my PM25MV mill but I have decided to try all this out on the router first.

Looking for any suggestions or help ya'll are willing to provide.

[dkirtley]

Looks really nice. The OSB is much stronger than MDF.

I am not as much a fan of the Polyurethane glues, but if you have good fit and clamping, it should be fine.

What kind of drive system and linear motion bearings you planning on?

[machineitright]

Thanks for the reply. So far I have had great luck with the Gorilla glue on stuff like the osb because of the open structure along the edges, when the stuff expands it really gets in the cracks and crevasses, usually use plain wood glue on anything else.

Linear motion will be v-groove bearings running along 3/4 x 1/8 aluminum angle, not original by no means but seemed like a good way to keep the cost down.

I was planning on going with ball screws and drive the X-axis underneath the table which is 6' long to get the 4' cutting length that I want, oh the dims on this router are 2'x4' cutting area and would like to have 6" z. I have looked at r&p, acme screws(multi-lead), and ball screws all seem to be about the same cost when you look at extra motors and drives for driving both sides of the x-axis to eliminate racking. Has anyone tried to add skate bearings running perpendicular to the rails on both sides to control racking and just drive from one side?

The motors and drives that I have are from Keling and were purchased for a metal cutting mill as I said in my first post, there are 2-570 oz in and 1-906 oz in which was to drive the heavy head on the mill. I am limited to 1/2 and 1/8 stepping on the drives for the 570's but more flexible with the 906 drive.

I would like to stay with 5 pitch screws if possible because that was what the drives were optimized for. I will be changing to more appropriate motors and drives when the budget allows for the router.

Sorry about the long post but looking forward to your input!

[dkirtley]

This is from a thread here (I think titled rock solid gantry)

It is a technique that has been used for many years on drafting tables to keep the ruler parallel to the table.

cockrum.net

I totally went away from Gorilla Glue once I got used to working with epoxy. It is cheaper, stronger , doesn't require as much clamping force, has a lot longer open time, and fills gaps much better.

[machineitright]

That is pretty cool cabling, what kind of resolution and accuracy you think?

Looks like he used the same kind of linear rail design that I am going to use. Are there any draw backs to it that I need to be aware of?

Thanks for the heads-up on the epoxy, do you use a particular kind? Would like to use it on the rest of the build if it is that much better.

I routed the edges of my table today to get them trued up and smooth, didn't get any pics but at least got something done. I also cut out my gantry uprights and hope my layout works, I will take some pictures tomorrow and get them posted. I thought I would get everything cut out and assembled on the wood side before I tackle the metal work.

I am still wanting to try ball screws and pretty sure I can build a zero back-lash ball nut assembly which should help with any whip in the 6' x-axis screw. I was thinking a 5/8 screw would be big enough am I wrong? Should I go bigger?

[dkirtley]

For general gluing, I like System 3 GelMagic. It is pre-thickened (a little bit thinner than vaseline) You just mix 2 parts to 1 of the resin and the hardener.

They also sell it in a cool caulking tube that can mix in the tip (or just measure out proper mixture for small batches) The tube is more expensive though. I am cheap and bought a 2 qt or so kit. I just use little medicine measuring cups.

There are some other mixtures for doing things like filleting and filling. Easy Fillet and Easy Fair. You can get the plain resins and mix in your own fillers but that is messy and a pain the rear unless you are doing lots. You also use the plain resin for laminating (fiberglass or carbon fiber)

That cabling just keeps the other end of the gantry square. I have only come across it on drafting tables. It works well there.

[DIYaholic]

That cabling just keeps the other end of the gantry square. I have only come across it on drafting tables. It works well there.

If you search for "Moving Knot" you will find several threads discussing it's implementation on CNC machines. ex: http://www.cnczone.com/forums/diy-cn...ock_solid.html

Randy,

[machineitright]

Thanks for the links, read through them last night and will probably be incorporating that into my build.

Is there a ratio for the gantry hieght to table width that helps? I want to get 6" of usable Z. I cut my gantry supports 24" X 12" with a angle cut from 6" wide at top down to about the 10" mark. The way I have it figured I will need my Y axis to be about 9" wide so when the Z is working on thin material and close to the 6" projection it will be supported.

Am I headed in the right direction?

[dave22]

Thanks for the links, read through them last night and will probably be incorporating that into my build.

Is there a ratio for the gantry height to table width that helps? I want to get 6" of usable Z. I cut my gantry supports 24" X 12" with a angle cut from 6" wide at top down to about the 10" mark. The way I have it figured I will need my Y axis to be about 9" wide so when the Z is working on thin material and close to the 6" projection it will be supported.

Am I headed in the right direction?

Check this link: Do it yourself CNC router: Design Considerations, the Gantry

Excellent diagrams showing the geometry and forces involved with the design of a gantry style router table.

Hope this helps and Good Luck with your build.

[machineitright]

Thanks dave22

I have been reading what they have on that site and was how I arrived at what I have so far, just wanted to see if I read it right.

I have been re-thinking the x-axis drive and would like to know what you think of the chain drive system, pros and cons. Trying to keep the costs down as much as possible on this first build. Any dirt cheap suggestions on drive?

After reading more builds and posts on the forums I think I may back off wanting to cut aluminum with this router and just use it for wood and plastics. This is just going to be a hobby machine and if it really takes off I will probably want to build a better machine.

[dave22]

Thanks dave22

I have been re-thinking the x-axis drive and would like to know what you think of the chain drive system, pros and cons. Trying to keep the costs down as much as possible on this first build. Any dirt cheap suggestions on drive?

Here is a link that describes precisely why I would never use chain and sprocket for a CNC motion drive system, see the animation near the bottom of the page:Report on CNC Linear Drive Mechanics

Ballscrew, Timing Belt, Acme leadscrew,and Rack&Pinion would all be acceptable choices.

Price the required components for each type of motion drive system and see which one best fits your budget. Also consider how this will effect your design, ease of assembly, operation and maintenance.


Hope this Helps and Good Luck with your build. :cheers:

[dkirtley]

Just to be fair, that flash animation is pretty biased and contrived to steer people away by exaggerating the issues. Where would you even find a 6 tooth sprocket to make the effect that bad? As the circumference of the sprocket goes up, that jumpiness goes down. The real problems with chain are the same ones that go along with rack and pinion. They are basically a very coarse tooth rack and pinon with a bit of elongation issues as they wear. The minor fluttering in speed would be a challenge to measure in most cases.

All positioning systems have their drawbacks.

[machineitright]

I will stay with either multi-lead acme or ball screws on all my axis. You think I can get by with 5/8 on the 6' + lead screw on my x-axis? I can get 5/8x.200 ball or hi-lead acme screws for about the same price from Roton but if I go to 3/4 the ball screw more than doubles but the acme goes to .250 pitch and is way more affordable. I am leaning toward the ball screws because of their efficiency and better accuracy.

Thanks for all your help on this project, I can tell this is going to be a fun learning experience.

[dave22]

I have never personally used a screw over 4' in length for fear of potentially violent whipping of the screw.
Beefing up the screw should prevent this, as many industrial class machines are screw driven, but I would think the cost would be very prohibitive. A rigid vibration free design would need screws with a diameter more likely 1'' or greater.
Instead of beefing up the screws, have you considered driving the nut instead. This is a proven method even at the DIY level.
See this link: http://www.cnczone.com/forums/diy-cn..._videos-8.html post #92 Grublua's excellent build. The whole thing makes for a great read. Also he could speak with authority as to it's effectiveness or possible changes to the design, with the benefit of hind sight. Although I think the design is sound.

Hope this Helps and Good Luck with your build. :cheers:

[dkirtley]

Instead of beefing up the screws, have you considered driving the nut instead. This is a proven method even at the DIY level.
See this link: http://www.cnczone.com/forums/diy-cn..._videos-8.html post #92 Grublua's excellent build. The whole thing makes for a great read. Also he could speak with authority as to it's effectiveness or possible changes to the design, with the benefit of hind sight. Although I think the design is sound.

Hope this Helps and Good Luck with your build. :cheers:

Thanks for that link. I hadn't seen that one. That looks like a really nice solution for a longer lead screw.

[machineitright]

Thanks dave22!

All I can say is WOW!!!! that is one nice machine, it amazes me how some people are such artists, even when they build machinery.

May have to try his driven nut design, doesn't seem too complicated, sometimes the less complicated the better they work in the end.

The weather is horrible here today, no going anywhere so going to order in my screws and other things I need for the build and start working on the metal work.

[machineitright]

Ordered my screws, v-groove bearings, anti-backlash nuts and some couplers today. Decided to go with Hi-Lead from Roton on the screws, 1/2-10 2 start for Y & Z and 5/8-10 2 start for the X. I ordered anti-backlash nuts from Dumpster for the 1/2 in screws and am going to build my own for the 5/8 with 2 plastic flanged nuts a spring and an idea I had, also going to try to spin my home made nut since whip seems to be an issue with longer screws.

I also ordered some of Dumpsters threaded couplers for the z & y, anyone using these instead of love joys? what do you think? Looks like a good way to do it but wonder if alignment is an issue, have seen some bad results from misalignment.I will be using opposing thrust bearings on the opposite end of the screws if I use this method to attach my motors.

Drilled my angle for my linear guide rails today with 6" spacing, think this should be plenty to hold it in place and stop any side movement, please correct me if this is not close enough spacing.

A question that keeps coming to mind is what kind of resolution and accuracy is reasonable for these DIY machines? I am going for slower speeds with a .000125 res, what repeatability should I be looking for? Some of the guy's on the metal working side are getting up around 300ipm rapids and 100ipm cutting speeds with this kind of setup and seems like it would work for my needs.

[Markwind]

This is from a thread here (I think titled rock solid gantry)

It is a technique that has been used for many years on drafting tables to keep the ruler parallel to the table.

cockrum.net

I find it amusing how some of these designs get spread around here, (monkey see monkey do) This technique is a total joke, to fix the problem of racking, and should not be used, anyone old enough to have actually used a drafting board with one of these cable systems will tell you THEY SLIP it is based on friction only.

[dkirtley]

I find it amusing how some of these designs get spread around here, (monkey see monkey do) This technique is a total joke, to fix the problem of racking, and should not be used, anyone old enough to have actually used a drafting board with one of these cable systems will tell you THEY SLIP it is based on friction only.

Properly adjusted it would do ok, I am not endorsing it though, was just answering someone's question about it. The ones on drafting tables generally relied on a spring to keep the tension so the ruler could be lifted up. Doing it with a high strength wire line under tension is a different animal. They went out of use on drafting tables mainly because you had to treat them well because they were delicate and the drafting machines were so much nicer to use.

Personally, I don't like the added complexity and all the stuff that can get hung up underneath. They really don't rely on friction, it is the tension of the two cables on the rollers opposing each other. The problems with them are: more in the wear and strain on the bearings and pulleys from the tension, constant strain from the high tension on the cable warping things and making things wear out more quickly, and the cable being prone to work hardening failure by constantly being bent so tightly around the pulleys.

There are other ways to try to avoid the racking problem and they all introduce other problems as well. Unfortunately there is no free lunch. Either add more motors to control, increase bearing surfaces and introduce more friction, or add some mechanical linkage to constrain the non-driven end. Whoever finds the magic solution around this issue will be a very rich person. It has been a problem for a really long time now and not going away any time soon. Once you get beyond toy sized gantries, a single lead point quickly becomes noneffective because the moments on the gantry increase with length.

[Markwind]

I would still have to disagree, this cable system is 100 percent based on friction or having traction between the two cables and the pulley, once you loose traction, from cable to pulley all is lost, exactly like the drafting tables,

And just like the drafting tables the more you used it the more it slowly went "out of square" and from what I remember I was constantly having to release the cable tension and re-adjust the bar square again, I don't see this being any different on a cnc router table?


There are many "absolute" standard and better ways to stop a machine slide from racking, the cable pulley idea is not one of them.