I've finally got some parts assembled for my gantry router. Basically have complete y-axis and z-axis without the motors and leadscrews. Z-axis uses 3/4" shafts with linear bearings, Y-axis uses rollerblade bearings on 2" EMT conduit mounted to a baltic birch torsion box gantry rail, which is extremely rigid. X-axis will be similar to Y. Overall size will be about 40" Y and 60" X. Working area should be about 48" x 30", with a 5" Z clearance and about 7"+ of Z-axis travel. Router to be Porter Cable variable speed 690. Here is an overall view.


Gerry

View of Rollerblade bearings, mounted to 1/4" thick, 1-1/4" x 1-1/4" aluminum angle, spacers cut from 1/4" brass pipe

Close up of mounted bearings.

I glued a 1/4" aluminum plate to 1/2" plywood to stiffen the top and bottom plates, also to draw heat from the motor. Here's a view showing the motor mounting location, as well as the linear bearings in the Z-axis.

Gerry

:banana: Nice job on the frame work. It must of took a long time to cut all those wood pieces out by hand.

Actually, I cut them all on our cnc at work. I'm going to take it all apart and drill a lot of holes tolighten it up, as it's a bit heavier than I'd like. But it is very solid.

Gerry

looks great! I like the 2" emt.

Eric

Where is the lead srew

It'll be between the z-axis shafts.

Gerry

Could you post a pictur of your your X axis rails and bearings. Interested in seeing how you mounted them.
Thanks
Hager

I Haven't assembled it yet, but the parts are all cut and waiting. I had to put everything on hold for a few months. Basically the X-axis is the same as the y, just larger. I'll have threaded rod through the Y axis tubing to keep everything tight to the X-axis tubing, as well as threaded rod under the table to keep everything tight there, too. The table will be suspended from the 2 ends. It's a torsion box, with the inside framework from 3/4" baltic birch plywood with 1/2" baltic birch skins. I'm hoping this is stiff enough, as I havent assemble it yet. The Y-axis shown in the pictures will also have skins attached to both sides, to stiffen it up. For information on torsion box construction, look here:

http://www.diynetwork.com/diy/shows_wwk/episode/0,2046,DIY_14350_26946,00.html

Gerry

Gerry,
I looked at the layout of the torsion box table and the first thing that popped in my mind is that it would make a great "zoned" vacuum table because of all the individual boxed compartments.
hmmmm, not bad
Thanks for the link. I hadn't seen that site before

Gerry,
Integrating the features of Torsion design will make your CNC a rigid and light machine. And CNC cutting the grid to fit the tubes then bolting everything together completes the torsion sides, good planning and design. Wish I had access to a CNC to cut my material out. I'm guessing that the tubes were mortised into the gantry sides also.
One of my hobbies is woodworking and I am building out of 3/4" MDF. Going for rigidity, but I’m taking the brute force method. The gantry sides are two 3/4" layers and have mortised the gantry cross piece which is also two 3/4" layers into the inside layer of the gantry sides. Everything was glued and clamped together. Because of the mortising the gantry is quite rigid. However at the sacrifice of weight. It weighs a ton.
My objective with this first machine was to make a machine that would accurately cut out my second machine, which before reading your posting was going to be aluminum. But now your design has stirred my interest in "Torsion" construction methods.
Keep us posted on your progress, interested to see it complete and running.
Hager

Hager, with a few jigs, my parts could have been easily cut with a table saw and drill press. As for light weight, I way overbuilt it and it's actually quite heavy. Before I put it back together I'm going to be drilling a lot of holes to lighten it up. But the weight shouldn't be a problem, as I'm driving the gantry with 2 250oz-in steppers.

Gerry

Gerry,
2, 250 oz-in steppers should do the job.

Excuse my I'm a newbie and gotta ask. I suppose you will be driving on each rail. What is the likelyhood of loosing steps of one motor, and causing the gantry to bind?

Does the one driver module drive both steppers on that axis?

Hager

From everything I've read, you only lose steps if you try to go faster than the system is capable of going. If you set up your software correctly, and stay within the machine's limits, you shouldn't lose steps. Ive got a Xylotex 3-axis board + a single axis board. I'll probably use the single axis board for my z-axis, and the 3-axis for the 2 x motors and the y-axis. All the motors are 250 oz-in. Z-axis leadscrew is 1/2-10 acme, X and Y will be 2 start, 1/2-8 acme (4 turns per inch). Hopefully I'll be able to get pretty good speed out of this setup. I'm hoping to be able to cut at at least 50 in/min.

Gerry

gerry


its been.. 5 months since this post - updates? hows the build? im curious how the torsion box worked out, as i imagined something like that a long time ago... it reminds me of the bucks ive built to make shaped plywood furniture...

pictures?

Originally posted by vacpress
gerry


its been.. 5 months since this post - updates?
pictures?

I feel the same way...have been waiting more info!

Hopefully I'll be back to work on it by June, maybe sooner. Just have to finish the kitchen.

....upload pictures of your kitchen!

Forget the kitchen! Get out in that garage and start building.

ger21 -
At least the weather kept you outta the yard!
I jackhammered and redid the driveway and walkway - cobblestone pavers - looks great but ohhhh my aching back!
cheers - Jim

Hopefully you didn't do what I did. I built all the kitchen cabinettes from scratch, and I didn't even have a cnc machine to help!

I'm looking forward to seeing more pictures of your machine, especially since I have to design a new one for myself!!!

Mike...

P.S. Otherwise I'll just have to drive over to your house and check it out in person! :D

Originally posted by ger21
Hopefully I'll be back to work on it by June, maybe sooner. Just have to finish the kitchen.

Hopefully you didn't do what I did. I built all the kitchen cabinettes from scratch, and I didn't even have a cnc machine to help

That's exactly what I did. The boxes have been done and installed for a while now. Except for 3, + an island. Mainly, I just have to cut and seam a bunch of veneer, press it in onto some mdf doors, dye and spray them. Oh, and build some pull out drawers, and numerous other small but terribly time consuming items.

Mike, what happened to your router? Did all that flex make you decide to stop, or did you finish it?

Ger,

I finished the router, and will use it for light cutting until I get a second one built. Do you have any pics of your kitchen. I'd love to see em. Since this is off topic, how about sending em here.
mykes@comcast.net

Mike....

Originally posted by ger21

Mike, what happened to your router? Did all that flex make you decide to stop, or did you finish it?

I would like to see them also. The General Wood Working forum might be a good place for pictures like that.
Bill.

Hey Gerry, what happened to the thread, did you stop building your machine?

I'm very slowly building a base right now, while I try to finish a bunch of other stuff I have going on. Should be able to get going again in about 2 months. I'll start a new thread when I get going again.

After a long wait, construction has resumed! Everything in the previous pics was broken down and packed away for about a year. About a month ago I started reassembly. Here are a few pics of the gantry beam being assembled.

Here is the main table framing. All the baltic birch parts I cut 2 years ago were still perfectly straight (they were wrapped with stretch wrap). I drilled a ton of holes to lighten up the framing. After a few I figured out that it was faster to leave about 1/16" from going all the way through, so I wouldn't have to keep cleaning the circles from the forstner bit. After sliding everything together, I layed it on my table saw and it was perfectly flat. This was important as the table holds the rails that the gantry rides along. Each side was then skinned with 1/2" mdf, stapled and polyurethane glued. (for more working time). Once the glue dried, I set the table on two sawhorses, and checked for flatness with a 4 ft level. Perfectly flat in every direction. And I have to put well over 100lbs of pressure on it to see any deflection.Table size is 59" x 32". Y axis travle will be the full 32", but that will be right up to the gantry uprights. X axis travel will be around 48".

Welcome back Gerry.

Good looking torsion box. It will be super strong too, as well as light.

It does my old wood working heart good to see lots of wood in CNC machines. Not everything nas to be alum.

Can't wait to see more pictures.

Hager

I wanted to build a sturdy base, without spending a lot of money. I bought a few 2x4's, and resawed them into ~3/8" x 3 1/2" strips. I then glued these up, jointed and planed them into roughly 2" x 4" legs. I drilled a hole in the bottom of each, and inserted a threaded coupler. I then made feet using this tip from Industrial Hobbies. http://www.industrialhobbies.com/howto/tips/hockeypuck.htm

For about $6 for all 4, they work great. 1/2" mdf panels were mortised in to the legs to tie them together and stiffen everything up. ( the table does most of the stiffening, though). And on each end, a 3/4" baltic birch panel is mortised into the legs, and supports the table and tubing. The X axis steppers and leadscrews will also be mounted to these plates.

I clamped everything toghether and took a few pics, with the gantry beam sitting on the table. This thing is pretty big. This weekend I'll get a coat of paint on it, and start on the gantry uprights. Hopefully everything will be rolling in a few weeks.

Welcome back Gerry.

Good looking torsion box. It will be super strong too, as well as light.

Can't wait to see more pictures.

Hager

It's not that light due to the mdf. Next time, I'll use 3/8 or 1/2" baltic birch for the framing, and probably 3/8 baltic birch for the skins. It would probably cut the weight in half.

I put all my weight on it (200+) with a level laying on the table, and it only sagged about 1/64". The entire gantry should be under 50lb, so It should stay perfectly flat. I was a little concerned with the 59"span, but it looks like it'll be OK.

That's right, I missed the MDF skins.. They are heavyweights. Did you do some of the cutting on a CNC?
Hager

Did you do some of the cutting on a CNC?
Hager

All the Baltic Birch pieces were CNC cut. I'm going to CNC a template for the gantry sides and use a pattern cutting bit to make them, hopefully this weekend.

Why are you pattern cutting? Is that because it is faster than to CNC them?

Wer'e really busy at work right now, and I'll barely have time to CNC the template.

That's understandable. Being busy at work is a good thing.

Thanks
Hager

Any updates on this? I'm particularly interested in how you are driving the X and Y axes.

Thanks
Jon

No updates yet, although I've got the gantry uprights mostly assembled. X axis will have dual 1/2-8 2 start acme, same screw on Y axis. 253 oz-in PacSci steppers, 3 axis + single axis Xylotex (2-X, 1-Y, 1-Z)

I doubt it will be as fast as I'd like, but that's all the budget allowed. I may upgrade to Geckos and a higher voltage power supply down the road.

That is an amazing effort Gerry.
It looks like you are taking no shortcuts.
It looks more like a wing structure than a cnc router. Will it be dual purpose?
I've seen a flying lawn mower b4, but never a flying router.

So it looks like your x-axis screws will be situated below and to the side of your x-axis tubes?

Will the Y-axis have any connection to the rest of the machine besides these screws? Is there any concern for the Y-Axis "tipping" since it's essentially just riding on top of the X-Axis tubes?

I'm concidering a similar design and I am just trolling for some pointers.


Thanks
Jon

The Gantry will attach to the rails the same way as the Z-axis. 1/4" 1-1/4" x 1-1/4" aluminum angle, with rollerblade bearings. The gantry beam shown in the pictures will be between the uprights, and threaded rods will run under the table pulling the bearings tightly against the tubing. The bearings will be about 12" apart, at 45° angles. It won't be able to tip at all. The threaded rod underneath was the reason for the stiff torsion box table. The entire table is suspended between the 2 end panels, and spans 59". And it appears it will have no flex at all.

The leadscrews will be under the tubing, actually more under the edges of the table. The sides of the gantry extend below the tubing and will have blocks that go under the table a few inches to mount the nuts to. The threaded rod will go through those blocks as well.

That is an amazing effort Gerry.
It looks like you are taking no shortcuts.


There are plenty of shortcuts due to time and budget constraints. Like most of the people here, I already have plans for a bigger, faster machine. Still made out of wood. But much lighter, and even stronger. And of course, much more expensive. :) Already have some THK an IKO rails for it.

I love the looks of that thang! I just love torsion boxes. I even came up with a way of building them. Don't know if anyone else does it that way (not sure they should :D). I like the choice of routers too. I recently chnged my mind from laminate trimmer to the 690. I have a 7518 in my router table but it's just too heavy. I think the suggestion for using the deadspace for suction was a great idea. You could even have little blast gates for zoning it off. Great idea about leaving the Forstner holes unfinished. Not only does away with prying the disc out but eliminates the tear-out too. Just a lot of good thinking here. Love it. Oh yeah, are you using dual motors for torque or symmetry?

The 690 is out, I have a brand new variable speed 892 sitting in its case. Dual motors are for more speed (torque).

ger21]The 690 is out, I have a brand new variable speed 892 sitting in its case. Dual motors are for more speed (torque).

Yeah, like I say... you probably won't like that little 690 :D

I do have a 7518, just in case. ;)

Gerry,

Are you planning on fastening your conduit to your torsion box with JB Weld or socket head cap screws, or are the two end plates your only fastening points?

What's going to keep the conduit from wiggling around in the torsion box? Let me guess, it's all press fit? :stickpoke :)

Mike...

Just the end plates. It's a tight fit to get them through the end plates, and there is no bowing at all. Also, there will be threaded rod under the table pulling the gantry sides together, and holding it very tight. Also, the bearings on the gantry sides will be about 12" apart, so it will be pushing tightly over a large area.

Also, I think one end of one tube will be adjustable, in case I need to get any twist out of the table. Although it seems perfectly flat right now.

I should also mention that the tubes are not press fit into the sides of the table. They fit perfectly, but not tightly.

ger21,
I used threaded rod to pull my machine sides together and now the it is almost impossible for the gantry to push the y axis rods outwards.
Thanks for the tip.
Hope to see your machine completed and cutting soon.

Jason

Hope to see your machine completed and cutting soon.

You're not the only one. :)

Ger 21
If you don't mine what bearing did you use? I'm using what I have on hand but they are too small. I have 2" SS 16 gage with 2 x 2 x 1/4 angle that I'm trying to use.
I started Jero's design but found out I don't like mdf plus I have most of the wood on hand to make one like yours. Thanks I like yours alot

Just regular rollerblade bearings.

I started Jero's design but found out I don't like mdf plus I have most of the wood on hand to make one like yours.
I had mine mostly assembled last week and discovered some flaws with the Z-axis. The 4 threaded rods do not keep it from moving from side to side and front to back. I'll be adding side panels, and front and back panels to keep everything square. Basically boxing in the whole thing.

Gerry, do you have any more pictures available? Your work looks top notch. I'm just hoping I can come up with something half as good.

Here's the template I used to route the gantry sides, a partially finished side, and a pic Mike took the other day when I had it partially assembled. I should be able to start finishing it up in 2-3 months. Maybe :)

Here is the main table framing. All the baltic birch parts I cut 2 years ago were still perfectly straight (they were wrapped with stretch wrap). I drilled a ton of holes to lighten up the framing. After a few I figured out that it was faster to leave about 1/16" from going all the way through, so I wouldn't have to keep cleaning the circles from the forstner bit. After sliding everything together, I layed it on my table saw and it was perfectly flat. This was important as the table holds the rails that the gantry rides along. Each side was then skinned with 1/2" mdf, stapled and polyurethane glued. (for more working time). Once the glue dried, I set the table on two sawhorses, and checked for flatness with a 4 ft level. Perfectly flat in every direction. And I have to put well over 100lbs of pressure on it to see any deflection.Table size is 59" x 32". Y axis travle will be the full 32", but that will be right up to the gantry uprights. X axis travel will be around 48".


Gerry,... That is one super table!! If I didn't have the steel to use, I would have copied YOUR table!

Gerry,... That is one super table!! If I didn't have the steel to use, I would have copied YOUR table!


Thanks John. I'm pretty happy with the way it turned out. Now if I could only find time to finish it.... :)

Gerry, how thick are the sides of your gantry? I noticed you're going with the xylotex 3 axis board, are you going with the complete kit in the box with the 30v supply or the non-boxed kit with the 24v supply? and how whill this effect performance?

I'll have to check, but I believe the sides are about 1-5/8". Two layers of 1/2" MDF with a 5/8" hardwood frame in between.

I have the old style Xylotex, a 3-axis + a single axis, with a 24V power supply. I'm not sure if the difference between 24V and 30V would be that much, but possibly up to 20% higher speed, best case. The Xylotex purchase was based mostly on cost. The HobbyCNC chopper wasn't available at the time, and Geckos were too much money for this project. :)

Gerry, may gantry is set up just like yours torsion box design, but I was trying to figure out how to attach my y torsion box to the gantry sides. I use 2" square tubing for my rail rather than pipe. I just don't know whether I should bolt & glue to the side or just bolt it in case I need to make changes to the Z axis, how are you connecting the Y torsion sides to the gantry side? I was going to follow your recommendation on using 1 piece panels with a hole in the middle to close up the sides of the Z. Would just using 4 bolt on each side keep the bridge from flexing?

Ed.

My gantry sides will have pockets to accept the Y-axis tubes. The box itself will be shimmed as needed to keep the sides parallel, and then screwed through the sides. I don't plan on having any adjustment, because My template for the sides was CNC routed, so the pockets will be perfectly perpendicular to the X-axis tubes. I wouldn't use glue, that's asking for trouble. :)

I don't exactly follow what you mean by using 4 bolts, but I'm pretty sure it needs to be completely boxed in to eliminate flex.

I got a new camera yesterday, with video. :banana:

Here's a few pics of what I've been doing. The gantry sides are ready for paint (tomorrow). They were lightweight MDF, and very soft, so they've been coated with West System Epoxy to both seal them, and make them a little tougher. Where the bearing angle will be mounted, the 2 slots at the bottom, I epoxied some aluminum strips into some routed slots so when I tighten the angle down, it won't crush the MDF. The Baltic Birch ply piece at the bottom will be drilled for the threaded rod to go through, tying the 2 sides together, and the leadscrew nuts will mount to the bottom of it.

I also started making some leadscrew mounts, using 1/2" phenolic plate and rollerblade bearings. I cut the plate in 3"x3" squares, and, using a 22mm forstner bit (made by CMT if you need to find one), I drilled about 3/16" into the center, followed by a 1/2" hole all the way through. I then epoxied 2 back to back, using the bearings and an 8mm bolt to pull them together and make sure they spun freely before the epoxy set. Once cured, I trimmed them square on my table saw, and now just need to drill the mounting holes. These should be plenty stong, and a little easier to work with than aluminum. I took a video of the block in motion (had to play with the new camera ;) ) The bolt is just sitting on the table, not attached to anything. I held it still with my finger, and gave it a spin. Very free with no play.

http://home.comcast.net/~cncwoodworker/CNC/ScrewMount.avi


I also weighed some parts, and it looks like I've substantially overbuilt. :)
The gantry sides are about 8 1/2 lbs each, and the gantry torsion box is about 22 lbs. That's 40lbs, + 8 lbs for the router, and probably another 20-25lbs for the Z-axis, screws and steppers. Somewhere around 75lbs probably. I think I was originally thinking somewhere around 50, but it adds up fast.
But, as I've mentioned before, The table deflects very little with over 200lbs on it, so I don't see any problems, except maybe for the load the bearings put on the pipe. I've got an idea to possibly add more bearings on top to help carry the load, but I'll wait to see how it works out.

So, if all goes as planned, and I get everything painted tomorrow, I plan on assembling the gantry on to the table in the next week or two. I'll be making the X axis bearings and mounts today and tomorrow, so when the paint dries (hammer paint can be tacky for a few days) I'll be ready to go.

I was wondering if you could share how you are going to attach your leadscrew nuts to your gantry walls. Some kind of bracket or block, I ‘spose.

Steve

You mean the driven nuts? Oh, at the bottom of the sides, right? I was going thinking of buying some 2x2 (whatever size will work) aluminum angle and mount the nuts to that, and mount the angle to that plywood "wing" on the bottom of each side.

I did the final assembly of the table to the base last night, and provided the bearings I bought on Ebay arrive today or tomorrow, I'll get the gantry mounted and rolling this weekend. No leadscrews yet, though. Then I'll start making the necessary changes to the Z-axis.

Finally got something that actually moves, although not by itself yet. :) Still have to install the threaded rods under the table to pull the bearings tight to the tubes.

Here's some better pics of the bearing mounts. 1/4" thick 1-1/4" angle. The spacers were cut from 1/4" brass pipe on the table saw :eek: (careful if you try this) and the length adjusted on the disk sander. Somehow I cut most of them within .005 of each other. The notch in the top of the angle was done on a router table. I cut a 45° dado in a piece of plywood, to hold the angle with the point down. Then just slid it along the fence on the router table. Go very slow, as the bit will want to pull the aluminum into it.

You have a very nice looking machine there. Glad to see your progress.

Hager

Congrats on the progress Gerry, Its looking great.

Nice machine Gerry!

How wide is the base of the gantry uprights and how did you decide on this?

How do you folks keep the wood dust/chips off the Skate bearings? will you fit covers? do you need wipers? or can you 'get away with' some dust n chips?

Very Impressive looking machine, congrats Gerry. Do we see plans for sale? :D

Jason

What do you use to adjust and tension your rails?
Thanks. :)

How wide is the base of the gantry uprights and how did you decide on this?

12" wide. I just guessed at 12" as a compromise between travel and stability.

The wider the better, but at the expense of travel. The rails are about 59 1/2" long, and leaving a little clearance on each end will give me about 45" of travel.

How do you folks keep the wood dust/chips off the Skate bearings? will you fit covers? do you need wipers? or can you 'get away with' some dust n chips?

This is a big concern for me. When the thing is finally running, I'm going to see if I can make some delrin covers that hold a felt wiper over the bearings. We have a large saw at work that uses a system like this. The felt is saturated with oil, which gives you an oiled surface for the bearings to ride on, while wiping the dust out of the way.

Very Impressive looking machine, congrats Gerry. Do we see plans for sale? :D

Jason

No plans for this one. I think I'm pushing the envelope for size of a wooden machine using basic construction techniques. Building the torsion boxes required precisely cut parts to keep the rails perfectly straight, and you'd need a 5 ft router to cut them. This is an untested design, and I'm finding a few issues as I go. The Z-axis needs additional support, and if it wasn't 90% complete, I'd have completely redesigned it. And while the round pipe seems adequate, I'm not completely happy with it. I may add 2 more bearings to the gantry sides to help carry all the weight. I think having the bearings running on steel angle is a better way to go, but I'm not sure if anyone pulled it off yet. See the open source hardware store thread for more on that. Also, my drawings only covered the gantry and table, from here on out I'm making it up as I go. I think when complete it will work at least as good as a JGRO machine, most likely quite a bit better, but to sell plans for it would require a complete redesign, and it would cost a lot more to build then. At some point in the future the machine will probably be for sale, but probably 2-3 years from now. I already have a lot of parts for it's replacement. :)

I am (slowly) working on some plans for a smaller machine (24x30), which I'd like to sell plans for, but also would like to sell precut kits, and some prefabricated parts like the torsion boxes. It may be a while though, as I can't seem to get all the other things done that I need to do to give me the time foor that. :)

What do you use to adjust and tension your rails?
Thanks. :)

There is no adjustment. Only on one end of one rail, because I needed a litttle bit of room to be able to get the rails through the end plates, as it's a very tight fit. The gantry and base torsion box ribs hold the pipes in place, and each is perfectly flat. (As flat as my 4 ft level, at least). There are threaded rods under the table holding the gantry sides tight to the rails.

This is a big concern for me. When the thing is finally running, I'm going to see if I can make some delrin covers that hold a felt wiper over the bearings. We have a large saw at work that uses a system like this. The felt is saturated with oil, which gives you an oiled surface for the bearings to ride on, while wiping the dust out of the way.

Thanks, I see, bit of a difficult kinda shape to cover ! hope that goes Ok

I'm working on my Z-axis, which need some additional panels attached to keep if from flexing (design flaw ;) ). The problem is, there is very little extra room to mount the panels, but I think I have it figured out. I only have room for 1/4" plywood, which is kinda flimsy, so I'm laminating some aluminum skins (.025) on to the 1/4" with epoxy. I'm going to throw them into my vacuum press shortly, so they'll be ready to cut and mount tomorrow.

In the meantime, I made a jig to turn the ends of my acme down with a router table. I bought some 1/2" ID bearings from Mcmaster Carr, as well as some 1/4" shank grinding stones. I set my router to 10,000 rpm, clamped the jig to the router fence, and fed it into the stone while turning it by hand. I'm using rollerblade bearings, so I need an 8mm shaft. I soon realised that grinding the 1/2" down to 8mm would take a loooong time, and if I got to aggressive, the screw got very hot. So, I pulled out the screw, and using my 12" disk sander, sanded all the threads off in about 1 minute. Then back to the grinding jig for basically a finishing pass.

Now, this is not an extremely precise setup, as fine tuning the fence turned out to be a bit tricky. Because of the lack of high precision, I had to end up a little on the small side, but I was able to get a very good fit. The ground area is within .001 it's entire length, .312-.313. 8mm is .3149, so I'm within .002-.003. And the tranition into the threads is radiused, so the bearing self centers. I then used an 8mm die to thread the end, and mount a nylon lock nut. The bearing will mount to the bottom of the screw, with the outer bearing in a 22mm counterbore under a 1/4" aluminum plate, and the top bearing resting on top of the plate. Here's a few pics.

Very inovative, you did an excelent job with what you had, now that something i would do :)

Joe

Ger
Very innovative solution to turning down the end of the acme threaded shaft. Thanks for the explanation & great photos.
I've been wrestling with a similar problem recently & this certainly helps.
Perhaps you could post a photo of the end of the shaft assembly as you explained.
Thanks, WT

Sorry if these are a little blurry.

Ger
Your bearing mounting technique should do the job quite well!
Thanks for the fast response.
WT

The X and Y axis will use the blocks I've shown earlier in this thread, Just the Z- is like that. All the screws will be done the same way.

Your router is very impressive. Some great ideas here, very innovative indeed.
Have you thought about buying a mini-lathe? They are reasonably priced, and are perfect for things like turning down shafts of leadscrews. I would not have been able to do half of my projects without it, including parts for my own router.

My wife would kill me if I bought a lathe. I don't like to buy cheap tools, so I'd probably end up spending a lot of money on one. I'll probably get one someday, but not in the immediate future.

Yea these little lathes do have their limitations. I would have saved up for a large one, but I have access to a 13x48 Clausing everyday. So for anything large, i just use that.
FOr hte price though, it cant be beat!

Got a little more done.

That's one stylish looking machine. Great workmanship !

Dang Gerry, you do good work!

Did those side plates on the Y carraige help? I'm guessing: yes.

What kind of motor couplers are you going with?

(BTW, my wife has killed me several times, really doesn't hurt that much after a while :) )

Steve

Nice Z axis!

Very nice work indeed, keep the pictures comming :)

Thanks, Joe

I liked how you wrapped the z carriage around.

Have you noticed any skewing side to side on the gantry, I could see on some of the wider machines having problems if the bearings ride on only one side of the pipe so you only have a total of 8 contact points in 3D. It's something that I've been fighting in my design process. Best of luck the machine looks great.

Niiiice !! going to follow with interest - I've just m/c'd a Beam/plate for my m/c that will use two 20mm silver steel dia bars top n bottom, I'll have channel section on the back of a 10mm 1.3m long ally plate, still undecided about bearings tho.

Like it. keep posting

Pat

Have you noticed any skewing side to side on the gantry, I could see on some of the wider machines having problems if the bearings ride on only one side of the pipe so you only have a total of 8 contact points in 3D. It's something that I've been fighting in my design process. Best of luck the machine looks great.

use two drive screws instead of one (Ok I'm talking Acme ones) - end of prob, at a price!

Finally got something that actually moves, although not by itself yet. :) Still have to install the threaded rods under the table to pull the bearings tight to the tubes.

http://www.cnczone.com/forums/attachment.php?attachmentid=12693
That's a very solid looking machine! nice work!

Dang Gerry, you do good work!

Did those side plates on the Y carraige help? I'm guessing: yes.

What kind of motor couplers are you going with?

(BTW, my wife has killed me several times, really doesn't hurt that much after a while :) )

Steve


The side plates do help, but I'm going to make new ones. There is also a back plate that you can't see (That I accidentally cut too short). The problem is that because I made all the panels 1/4" thick, I had to use aluminum angle to bolt them together in the back. But I could barely reach inside due to the 2 1/2" clearance back there, so it took me about 2 hours to assemble it. The panel behind the router had no access at all behind it, so I had to epoxy some screws in place to bolt it in. Hopefully that works, as that's the only panel I'll be using from this batch.

OK, so what happened is that it's pretty difficult to get aligned, and the too short back panel allows a little flex. So I'm going to make new side panels, only use 1/2" baltic birch. That will allow me to slide a 3/4" panel between them at the back and screw it in place from the sides once I get it aligned. I originally used 1/4" because I didn't want to lose any travel, but I have to sacrifice another 1/2".

So, bottom line is that the panels will eliminate all the flex, but I just don't like the way these came out. They look cool, but are a little lacking in functionality.

I'll cut the panels tonight and get the first coat of paint on, so I can install them in a few days. I can now figure out the screw length, and cut it and turn the end for the Oldham couplings I have. I wanted the helical's , but didn't want to spend the money.

Have you noticed any skewing side to side on the gantry, I could see on some of the wider machines having problems if the bearings ride on only one side of the pipe so you only have a total of 8 contact points in 3D. It's something that I've been fighting in my design process. Best of luck the machine looks great.


If you mean one side moving along the rails without the other, yes, it does, and far more than I'd have thought. But, as was mentioned, the dual lead screws will take care of that. If you mean from side to side, with the bearings actually sliding around the pipes, then no. The 12" tall gantry beam keeps that from happening. With a lot(at least 50-100lbs) of pressure on one of the gantry sides, it will probably deflect a very small amount, but it would be barely noticaeable, and should't cause any problems. You'd need very stiff sides to eliminate this completely.

Change of plans. Because most of the side panels have to be cut away to clear the gantry box, It semed like my laminated 1/4" panels would have been stiffer than the 1/2" baltic birch I was replacing them with. So the new panels will have plastic laminate on both sides. The panels are in the vacuum bag now, I'll cut them tomorrow night. This will give me stronger panels, add a little color, and I won't have to paint them (Just seal the edges).

and cut it and turn the end for the Oldham couplings I have.

My mistake. I have the LoveJoy couplings.

I got the new side panels done. I'll try to get them on tomorrow.

I got the new Z-axis side panels on. Much better. Also got the Z-axis screw mounted. Tomorrow I'll try to get the nut finished. It's already tapped. Then the entire Z-axis will be finished.

Gerry,

I noticed you machined off quite a bit of your Z axis lead screw in order to fit the coupler. Surplus center has the larger diameter L-050 series halves at $2.19 each so you would not need to machine so much material for the other axes.

http://www.surpluscenter.com/sort.asp?UID=2006010805551560&catname=&byKeyword=yes&search=lovejoy

Careful on that list as they have the larger L-095 series next to the L-050

Jason

They're only $2.10 at McMaster-Carr, where I got mine. I don't know what I was thinking when I ordered them all with 1/4" bores. It seemed really small when I had to turn it down to fit. I'm going to order some with a 1/2" bore and see if I can mount them right on the screws without having to do anything to the end. I'll probably get some 3/8" bore as well which would need minimal grinding down.

Btw, McMaster-Carr doesn't list them as LoveJoy, and doesn't use the L-050 #'s either. Search for spider couplings, and the L-50's are 1-5/64" diameter. Or just search for 6408K11.

There should be no problem mounting directly onto the 1/2" screws, that 1/4-20 set screw really locks down when tightened. You could also flatten the end of the lead screw to allow the set screw face to sit on a level surface.

When I tightened one half to the 1/4" motor shaft and the other to the
5/16"-18 stainless allthread the set screws dug into both surfaces (I have a problem with overtightening :D ), definitely no slipping is going to occur here.

I will need to order more halves but 6mm bore size as some steppers I liberated from old toshiba copiers have this shaft size.

My only problem with these couplers is that the rubber spiders allow rotational play between the outer hubs. I followed Joe2000che and cut platic spiders which eliminated the play but they fit some of the hubs tighter than others.

Jason

They're only $2.10 at McMaster-Carr, where I got mine. I don't know what I was thinking when I ordered them all with 1/4" bores. It seemed really small when I had to turn it down to fit. I'm going to order some with a 1/2" bore and see if I can mount them right on the screws without having to do anything to the end. I'll probably get some 3/8" bore as well which would need minimal grinding down.

Btw, McMaster-Carr doesn't list them as LoveJoy, and doesn't use the L-050 #'s either. Search for spider couplings, and the L-50's are 1-5/64" diameter. Or just search for 6408K11.

I ordered mine with 1/4" on one side and 1/2" on other just slip it on leadscrew and tighten down works good, very nice fit.

Yea i dont know why i ordered 1/4" either, got em from McMaster. It just seemed like the thing to do until after i got the package and i wonder what i am thinking. But I managed to get ahold of two helical beam couplings with a 1/4" bore, so im using that anyways

The helical beam type couplers are good to use as they are zero backlash, but getting different bore sizes on either side is a problem. I found the sdp-si catalog 790-1 Handbook of INCH Drive Components that has bore sizes on either side.
Problem is though the largest bore looks like just under 1/4".

The other option is the Double Disk Flexible coupler shown in pic 2.

Jason

I am pretty sure mcmaster sells helical beam couplers with different bores. Even some with metric to inch.

I like the couplings more than the lovejoy though, they are indeed zero backlash and can take a decent amount of misalignment. They dont mar the finish of whatever shaft they are on, and they look cool.
BUt your also looking at 20+ a piece vs 2 for a lovejoy

Thats true phantomcow2 you just cannot beat that price. :D

Jason

My only problem with these couplers is that the rubber spiders allow rotational play between the outer hubs.

Jason

How about using silicon caulk (100%) to "glue" the spider in? Once everything is set up, apply silicone and assemble. If you need to disassemble it, you may need a new spider, but it should get rid of the backlash. Most of it, at least.

I never thought of that but I guess its worth a try.


Jason

I got the Z-axis nut done. I drilled and tapped a round piece of Delrin, I think it was 1-1/2" D. Using my wood lathe, with a 4 jaw chuck, I mounted it on a piece of acme, and made a bracket with a bearing to support the other end. I then carefully turned it down to the shape shown. I found a parting tool was best (safest?) as most woodturning tools seemed to like to grab the plastic. :eek: Once turned down, I used my disk sander to grind a flat so it would fit in the aluminum angle bracket. I also cut a slot in the end, and put an O-ring over the end in a groove I had turned in it to eliminate backlash. But the screw doesn't really seem to have any right now, though. The nut is threaded for about 2". The router mount was aligned and clamped to the drill press, and I used a hole saw to "drill" the hole in the mounting bracket. The nut was turned to just fit into this hole, so everything lined up perfectly. Good thing, because there's really no adjustment, although I could probably shim here and there if I needed to. I was a little worried about alignment and binding, but the screw seems to turn very freely, even with the router mounted. It's probably lifting about 12-13lbs, and I can turn it with my fingers pretty easily.

Here's the competed Z-axis, with the router mounted.

It's looking great gerry.

How's the rigidity of the whole thing seeming now?

It seems to be very solid. And the gantry moves really smooth, although it's quite heavy. The only thing is that the 2 sides can move quite a bit independantly, but the 2 screws will take care of that. And I'm not sure you can build a 40" wide gantry out of wood that wouldn't do that.

I'm thinking about building a 24"x36" cutting area version of similar construction... My current machine has been serving me well, but it's shortcomings are starting to limit what I can do with it. So a rigid machine with a full size router is high on my list to build. Just can't afford to go to an aluminum frame machine right now.

I'm thinking about building a 24"x36" cutting area version of similar construction...

For the most part, I'm pretty happy with it. The torsion box table is perfectly flat, and can support a LOT of weight. There's no flex or movement at all with the ~75lb gantry roilling it.

What would I change?

One issue is the EMT conduit. The galvanized coating wore off very quickly, and flat spots developed very quickly. Only time will tell how well it holds up, but I have a feeling that after the initial wear in, it will be fine.

If you followed this entire thread, the Z-Axis you see hear was basically a patch on my original design, which needed some help. The original design is in there, but needed to be boxed in. I'd recommend something like either Lionclaw's or Joe's Z-axis, with the Y-axis screw in the middle, as opposed to on top like mine.

One other thing would be to figure out how to make really, really stiff gantry sides, and a really soilid way to attach them to the gantry beam. That's the only place I have any flex. But if you can't eliminate it, dual screws will do it for you.

The only other issues were caused by building it without finishing the design. Overall, it's pretty much what I had planned, only a lot heavier.

Gerry, did you notice a problem with the Y screw drive on top instead of in the middle back? This does not seem like a big difference to me. Did your side plates on the Y carriage work as planned?

The two problems I found with my mini version of your machine were the twist of the gantry (as you said it should be solved by dual X axis leadscrews) and “lift flex” caused by I think poorly designed (my fault) linear bearing holders on the Z that were not strong enough to hold the Z plate steady. Have you tried driving your Z (without the router turned on) into the table to see if it flexes? Maybe this is a illegal test, but it looked pretty bad on my machine. I was never able to determine if it was the carriage tilting back or the Z axis flexing.

Steve

Did the sideplates work? Yes. If I really push on it, I can get it to flex a little, but very little. This is because it's screwed together, rather than dadoed and glued. It's acceptable for being a fix.

As for the screw on top, I haven't made it yet, so I don't know. I'm sure it would have been better in the middle back, but I didn't think it would be a problem when I designed it.

I haven't hooked up the drives to see how the Z-axis works, but there is no play or flex at all. I'll check again tonight, but I'm pretty sure it's rock solid. It's 3/4" SS shafts with Thompson bearings (2 per shaft). By wrapping the bearings with a layer of tape, I can actually add a little preload to the bearings, but I can already feel a very slight binding at the bottom of the travel, so it's not needed right now. They're very tight. I thought the shafts were perfectly aligned, but apparently they're not :( . But they're as close as I could get with a drill press.

I don't plan on running the motors until it's done, unless I can borrow a laptop for a quick test. I don't want to drag everything in the garage and then have to move it back in the house. I'll have a new PC in about 2 months, and the current one will be running the machine. I'm going to build a box to house the motherboard, power supply and Xylotex.

Have you thought of using a torsion box for the side panels? I am building them on my router from 1/4" plywood for the ribs and skins, all glued and nailed. From memory they are 25" long x 10" wide , fairly light and I can't detect any twisting or flexing.

If you mean for the gantry sides, that's not really the problem. What happens is that the gantry beam itself acts like a 40" long lever. It doesn't take much force to move one side with the other side held in place. With dual drive screws, there will be absolutely no flex at all. With a single screw, I'm not sure you could avoid it.

Have you tried driving your Z (without the router turned on) into the table to see if it flexes? Maybe this is a illegal test, but it looked pretty bad on my machine. I was never able to determine if it was the carriage tilting back or the Z axis flexing.

Steve

According to Nook, it takes .625 oz to lift 1 lb with 1/2-10 acme (My Z leadscrew). So, in theory, my 250oz stepper should be able to drive the router into the table with 400 lbs of force. I'm sure that would cause some flexing. ;) I went with 1/2-10 because I wasn't sure I'd have enough torque to lift my heavy Z. But I can turn the screw so easily with my fingers, I'll probably be switching to 1/2-8 2 start to double my speed if I find it's too slow.

I just went out and pushed on it pretty good. No movement. The only place I can flex it is moving the Z-axis side to side along the Y axis. Just a little bit. If I have to, I'll make a new panel to fit behind the leadscrew to eliminate this. The one I made got rid of almost all of it, but I can still move it a little bit. I'll wait until I run it for a bit, because I think I'll need to remove the threaded rods on the front side to get enough room.

So how is your skate bearing setup working? Ridgid enough?
I was going to go a very similar route as you, with the mounted pipe. BUt I am also considering adding an extra pipe so i have 2 per side for the extra strength

So how is your skate bearing setup working? Ridgid enough?


It's plenty ridgid, with 1 minor issue, which I'm sure affects all similar designs. As you tighten the bearings (preload?) to the pipes, there's a fine line as you go from a nice smooth tight rolling setup, to a slightly cogging type motion. This is when you put too much axial load on the bearings. I think that the heavier the rolling components, the more apparent it becomes. It doesn't appear to really be a problem, though.

I am pretty new here but I really like your design do you have a link to it? Also are you done with this machine?

thanks

Sorry, no plans, I never finished them, and it keeps changing. And no, it's not finished, I've been too busy. But I plan on finishing it up this winter.

Just following this thread. Interesting build.

Gerry:) where is ur machine?

I actually have my Xylotex and power supply in front of me, and with a lot of luck, I'll be building the control box in a few weeks. Maybe. ;)

How is it coming along, Gerry? Inquiring minds want to know. You've done so much for the community with your builds and detailed notes, we feel like we're part of it! :cheers:

Trying to find the time to build an electronics box. Seeing the Z axis moving should motivate me enough to get it running. Hopefully in the next few weeks, If i don't start too many other projects first. :) Lately I've been doing some research for another new design. :D

Looking at the dates of these posts, it looks like I haven't done anything for 3-1/2 years?? :eek:

A month ago, I pulled out the acme screws I bought 6 years ago and "turned" down the ends.
http://www.cnczone.com/forums/showthread.php?t=86180

Spent a while figuring out how to mount the motor, but got the Y axis done today.

Made the 2 X axis nut mounts today, and worked on the two motor mounts a little. I might have it mechanically done in a week or two. Maybe. :)

After 6 years, the purpose of this machine has changed a bit. It's main function now will be to build the new one, which will not take 6 years. At least I hope not. I do have some other things I need to make with it, some custom crown moulding, cabinet doors, probably some signs and stuff.

I still have a few things around the house I need to get done before it gets cold again in a few months, but I've been trying to do a little bit a few days a week, so progress is now steady, although not extremely fast.

Looks good so far Ger.

Sounds like I work just he opposite of you. I work extremely fast in short spurts, then rest for months at a time. :D

CarveOne

At last he returns to finish this machine, I was beginning to lose hope.:)
I hope you are able to run through to the finish.

Jason

I'm mounting the two X screws today, and hope to hook up some temporary wiring and get some movement tomorrow.

After 6+ years, I had my Z axis moving last night. :banana: Unfortunately, I may have killed the extremely temperamental PC I was using. Messed with it for about 4 hours to get 5 minutes of movement, then it crashed and windows freezes while booting. I'm gonna try to steal one on Ebay today.

I only installed one of the X axis screws, but have everything ready for the other one, so it should only take an hour to get it done. Then I just need to build an electronics box, and install cable carrier. And probably a bunch of other stuff, but the end is near. :wee:

Congratulations Gerry! :)

CarveOne

Great news Gerry.

Alan

Thanks guys.

I bought a new motherboard on Ebay, and 10 minutes later discovered my memory was bad. But the new motherboard comes with memory, so the $20 doesn't all go to waste. And I'll have a spare. Still can't get into Windows, but I think a reinstall will fix that. The PC's been sitting in the garage for 3 years untouched. I also bought a CD-Rom, as I didn't have a spare lying around, and need it for the re-install. Hopefully I'll get it this week, but it probably won't be for another week.

Anyway, if anyone needs a cheap PC that runs Mach3 pretty good, look at old Dell GX150's on Ebay. With a PIII 1Ghz, it runs Mach3 with a rock solid 25Khz kernel speed, zero fluctuations. Motherboards with RAM and CPU's go for ~$20. You'll also need a special Dell power supply for another $10-15. Or, you can buy a complete PC for ~$50.

Put the new memory in and installed XP, and am back up and running. A few issues, though. One, my Aqua screenset really needs a faster PC than this. There's a 1 second delay when changing screens, and I need to turn off the toolpath display. However, Brian posted a test version of Mach3 the other day that greatly reduces the toolpaths CPU usage, and allows you to move and rotate it while code is running. I'll try to test it today or tomorrow and see if it helps. Not a big deal, though, as my original plan for the Aqua screen was to not have a toolpath screen.

Second issue, is that it appears that I have a serious resonance issue on the Y axis. I got it up to about 110 ipm, but it runs smoothest at around 70-80ipm. As soon as I get all the axis working, first thing I'll do is make some dampers. Unfortunately, my Z axis has a single shaft stepper, so I may need to replace it. I also need to swap the 1/2-10 Z screw with a 1/2-8 2 start. The Z runs pretty smooth at 45-50ipm, but also has some resonance, mostly near the bottom of it's travel. (The cutting area :( )

I got the other X screw installed a few nights ago, so I just need to wire up all the motors. And of course build a box to hold the PC and electronics. But I'm getting there.

I'll try to take some video later, too.

... it appears that I have a serious resonance issue on the Y axis. I got it up to about 110 ipm, but it runs smoothest at around 70-80ipm. As soon as I get all the axis working, first thing I'll do is make some dampers. Unfortunately, my Z axis has a single shaft stepper....

I'm confused (ok, it's my constant state)... By 'Dampers' are you talking about putting a flywheel sorta thing on the stepper or using methods to fight deflection in the screw?

Mil

Flywheel sorta thing, like here.
http://www.cnczone.com/forums/showthread.php?p=256639

YouTube - First Movement

I'm confused (ok, it's my constant state)... By 'Dampers' are you talking about putting a flywheel sorta thing on the stepper or using methods to fight deflection in the screw?

Mil


There are lots of home brew damper designs. A double ended stepper motor is needed for mounting the damper. This Solsylva (http://www.solsylva.com/cnc/dampers.shtml) site page has a few designs and a description of how they work. I use the first one shown that is made from a caster wheel with my Xylotex 425 oz-in motors. They definitely help tame the resonances when the spring tension is adjusted properly. It uses a controlled slippage method that is adjusted by varying the spring tension for best effect.

There are commercial ferromagnetic fluid and viscous fluid designs available also.

CarveOne

Gerry,

First motion is always exciting. Great job.

Alan

Wired up the two X motors, slaved them in Mach, set a conservative speed, and started jogging. Jogged to the end of travel to square the gantry, loaded up some g code, and did a dry run.
YouTube - cnc test1

Then ran a piece from scrap ply to be sure it was right (it was) , then loaded some Baltic Birch and cut my first part.

No measuring, no testing was done. Just got it running, and started cutting. Now, if the dampers work, I'll be extremely happy. I have some 5/8 steel bar that I'm going to drill on my wood lathe for the shaft, then use the drill press to drill and tap for a set screw, and epoxy it into the damper. I bought some 1/2" x 10" bolts that I'll cut up for the slugs, then I can test it out.

Finished the damper, and was able to change the Y motor tuning from 80ipm with accel of 7, to 180ipm with accel of 10. It'll run at 210 and 12, but doesn't have much power. Now I just need to make 2 more, and this thing will be pretty quick. Far better than I'd hoped when I started building it. :banana:

That's a lot more improvement that I saw with the caster wheel and spring design from the Solsylva site. I saved the information I found for an aluminum damper with steel slugs in nine holes. I was going to use wood instead of buying aluminum.

My new build will probably need something like this. I have the Xylotex 425 oz-in steppers but am not sure yet if I'll get new Keling 495 oz-in steppers instead of scavenging motors off of my first machine.

Keep us updated with the results of your dampers please.

CarveOne

I was going to use wood instead of buying aluminum.


The Baltic Birch damper works great. It's 3" diameter. The 9 holes for the slugs are on a 2" diameter circle, and are .55 diameter and ±.625 deep. The slugs were cut from a 1/2" bolt with an angle grinder, and squared and cleaned up on a 12" disc sander to a length of about .59 to .60.
The hub was cut from a piece of 5/8" round bar I had lying around. I chucked it in my Nova chuck on my wood lathe and drilled the 1/4" bore, then drilled and tapped the 8-32 set screw on the drill press. I glued the hub in with cyanoacrylate glue. The cover is 1/4" birch plywood with 3 screws holding it on.

I've already routed two more bodies for the X axis.

The results are absolutely amazing.

Before it ran smoothest at 80-85ipm, and was very rough at 50ipm and unusable over 90ipm. Now it's perfectly smooth at all rpm up to 234ipm (the max at 25Khz kernel speed), although it doesn't have much force at that speed, due to being wired bipolar series.

The X axis has a lot of resonance right now, but I'm confident it'll run as smooth and fast as the Y axis.

Thanks much for the details Gerry. I was hoping you would share it, but was hesitant to ask before you were convinced that it was final. From your comments it sound like you won't be needing to make any changes.

Since neither of my CNC machines are able to cut it right now (dead computer on the first one) I can use my metal lathe to turn the diameter and use the DRO on my mill to program the bolt circle drilling with a .500" end mill. Finished hole size will probably be a little over .500" in wood anyway. (Or use the keyless chuck I have with the most run-out. :)) Maybe I'll get the first machine back in operation before I need to make these.

Just out of curiosity, can you weigh them? The weight info may be useful for comparisons with other designs.

CarveOne

Just out of curiosity, can you weigh them? The weight info may be useful for comparisons with other designs.

CarveOne

I don't have anything to weigh them with. The BB body only weighs a few ounces. I'm guessing it weighs pretty close to the same as the 1/2" x 10" bolt I used for the slugs. Maybe a pound?? Or a little more?

Ok,

I can do it when I make 3 for the Solsylva 25x37 and 4 for the new machine. Mine may be a little heavier than the baltic birch. I think just for the fun of it I'll use some stable tropical woods I have on hand and use clear Lexan for the covers. :)

I drew up the top view in TurboCAD this morning.

CarveOne

Nice work Gerry - Good kick to get my own project going.

Got the two X axis dampers on tonight. Smoothed it right out. But it's not quite as fast as the Y. And when one motor stalls on a slaved gantry, it's not a pleasant sight (or sound) :eek: Now I forgot what I set the velocity at?? I think it's at 150 with an accel of 9. The main problem with the X is the 60" length. I think the 60" long suspended torsion box bed contributed to the resonance. I also get some screw whip at over 150ipm if the gantry is close to one end. It's OK in the middle, but at 180ipm at the far end, it's whipping pretty good. Part of that problem is that they have very slight bends in them. Too little to even think about trying to straighten them, but enough to facilitate the whipping.

Yesterday I thought I could set my Y axis rapids at 200ipm, but it stalled while shooting this video. So I backed it off to 185ipm, with an accel of 10. I won't try cutting over 150ipm, and if I do it'll probably be light passes.

Now that everything is running good, there are a few must do items before I can start making parts.

1. The PC and drives are lying on a bench. I'll start laying out a box design tomorrow.

2. Need to install Igus track on the gantry. I bought it about 5 years ago for $1/ft. Didn't think I had enough, but just went out and looked and I think I do. May be able to get that installed this weekend.

3. Need a dust shoe. First project will be some MDF crown molding, so need to keep the dust in check.

Here's a quick video showing the improvement that the dampers make.

YouTube - Stepper Motor Damper Test.

I have the same issue with slightly bent 1/2-10 5S screws. With the ends supported, I have tried to push in the middle against the bend just enough to get them straighter but they just act like springs and come back to where they were. I didn't get to a pressure level that would make them take a set.

I keep telling myself that my machine should be fairly easy to convert to R&P when I win the lottery..... :)

CarveOne

As I put in my log, I tried some solid wood dampers on my Y-axis and the program that was experiencing the resonance defiantly ran better.. I can only think that more massive dampers like you have would have changed it more.

I'm not a mechanical engineer but I remember that generally spinning mass was a bad thing in a dynamic system (depending on the moment of inertia and rotational speed)..

It will have the mechanical effect of making the whole gantry appear more massive as it's speed increases. Probably a bigger effect on screw based systems than R&P models, Maybe not because most R&P systems have two drive motors on a gantry.

I wonder, at what ipm, does the energy stored in the spinning dampers approach the energy stored in the moving mechanism (gantry or z-axis)..

My gantry right now is nearly 80 pounds.. But I have two motors driving it.. It would be an odd conflict of the linear and symmetric ramps in Mach having to apply to asymmetric behaviors of both low and high speed accelerations.. Both of the bad effects , stepper torque fall-off and rotational inertia increase at high slide rates... Do bigger industrial controllers allow more control on the motor ramps?

I guess my other mechanism-ignorant concern is.. "are dampers really removing the resonance or just changing the natural resonance of the system to some other frequency to be discovered at a later time? (Murphy says the worst time)" I suppose that is a good argument for making the dampers tunable...

... I don't mean to sound negative here.. after last night's test I think I'll plan on adding some if I find bad speeds that I want to operate at or pass through.. I was just wondering...

Mil

It would be an odd conflict of the linear and symmetric ramps in Mach having to apply to asymmetric behaviors of both low and high speed accelerations..................................... Do bigger industrial controllers allow more control on the motor ramps?


There is no difference in Mach3 for low or high speed accelerations. accel is linear, so require the same force regardless of the current velocity. Some Commercial machines use an S curve accel ramp, which is much smoother. If you want to try it, download and install Mach Quantum.



I guess my other mechanism-ignorant concern is.. "are dampers really removing the resonance or just changing the natural resonance of the system to some other frequency to be discovered at a later time?



They remove it at all frequencies, from what I see in my testing. In the video I posted, the first part without dampers shows a lot of resonance at 115ipm. I didn't show it, but at 90ipm without dampers it runs as smooth as with them. I had resonance at several different speeds, but there were sweet spots where there was none. Using a very slow accel will show where they are. With the dampers, it's smooth from 0 to max ipm. And with accel set slowly, you can hear the dampers rattle at different rpm's as they encounter resonance.

Gerry,

I just found this thread and read through it from page one. Your machine gets my vote for prettiest mdf machine! That gantry looks like aluminum!!

Those dampers seem to make a great difference in the smoothness of your steppers. I noticed some resonance frequencies on my machine as well. Because it is all metal, it rings like a bell when it hits those frequencies. Dampers may be in my future as well, but I'll have to get the machine cutting first.

A bit belated, perhaps, but congrats on having the beast moving and cutting.

I guess my other mechanism-ignorant concern is.. "are dampers really removing the resonance or just changing the natural resonance of the system to some other frequency to be discovered at a later time?

After running a few different programs, I've seen that the dampers remove resonance at many different frequencies. They rattle quite a bit when removing resonance, and they rattle at many different rpm's, from very slow to very fast. Actually, the rattling sounds a lot like the resonance, except that the machine is running smooth.

Also, after running it more, I've run into some issues. I originally had my gantry running at 175ipm, but have had issues with one side stalling, so I currently have it backed off to 155ipm. This is actually faster than I had ever hoped for, but I think I can fix it. I'm pretty sure that one screw is a larger diameter, which is making the nut much tighter. It's noticeable when turning the screws by hand with the power off. I have two options. One, file down the diameter until the nut loosens up. I did this to the Y axis screw, to get 1/2" bearings on it to grind the ends. It worked fine. Two, cut a slot in some scrap screw material and run the nut over it to loosen it up. I'll probably go for filing down the screw, as I just need to pull it out and chuck it in the drill and it'll only take 5 minutes once I get it out. And I think I can pull the screw in about 2 minutes. It's gotten colder here, so maybe that's made it worse. I've been spraying LPS on it and it get's rud of the stalling, but I only need to lube one side. So I'll get that fixed this week. I'll also check the VREF on the Xylotex and maybe dial them up a little. The motors only get slightly warm after an hour of running.


Second thing, was that I lost steps on the Z axis today. I've actually cut a few parts, and run a bunch of code without noticing a problem. BUt today I was V Carving and the rapid Z moves got me. All I needed to do was give it a quick shot of LPS and it was fine again.

I'm very happily surprised that I get a solid 180ipm out of the Y axis without stalling. When I start doing heavier cutting I'll need to determine how fast I can go, but for the light cuts I'm starting out with, I can cut at about 150, which is fantastic. Remember, when I started this and bought the parts 5-6 years ago, 60-100ipm was pretty good. And I'm getting double that, so can't complain about anything.

Here's a pic of my first V Carving. Not too bad, considering my table is not quite flat, and I threw it together pretty quick. Attached the .dxf if anyone want to play with it.

Your machine gets my vote for prettiest mdf machine!

Most of it is plywood, not MDF, but thanks. :)

Thank you for posting that spiral cut dxf. That looks like an excellent test to see if a router is aligned and working properly.

Now I just need to finish my router and figure out how to convert a dxf to gcode. :)

Need to catch the dust, so today I'll be mounting a dust shoe. Cut it out last night and put a quick coat of paint on it. It needs another, but probably won't get it.

Having a machine to cut parts on sure makes things easier. Especially the elliptical hole for the hose flange. :)

I bought 3 ft of strip brush from McMaster Carr, 3" long and 5/16" wide. Routed a 5/16" x 5/16" slot in the bottom of the plate, and will epoxy the brush in with 5 minute epoxy. I need to make 2 small aluminum angle brackets, which I'll mount to my motor holder. I'll epoxy two carriage bolts into the dust shoe, which will let me mount it with 2 wingnuts. Hopefully I'll be able to get it on and off pretty easily.

I'm hoping that I'll be able to also get my Igus track installed today as well. Then I'll just need to throw a box together to get all the electronics off the table. :)

Finishes the dust shoe. After I finished it, I realized that I maybe should have gotten the 4" brush. They measure from the top of the metal, so the 3" brush is really just over 2-5/8". When I drew it up, I also drew a spacer, but didn't cut it yet. After mounting it, I see that I'll probably need two, so I'll need longer carriage bolts. Heating them with a torch will break the epoxy bond and they'll pop right out. :) In the meantime, I raised the router as high as it'll go to get the bit even with the brush bottom, but I'm nearly out of Z travel doing it that way. I'm going to cut the spacers right after I post this, so that problem will be quickly fixed.

A bigger problem I noticed, is that the dust show exposes some flexing in the Z axis. Part of it is the whole carriage flexing a little, but it also appears that the bushing on the shafts have a bit of play in them . I believe it's because they are the adjustable type that allow for some misalignment.I can tighten them up by shimming with tape, but I don't really want to take it all apart right now. So far, I haven't seen any flexing issues showing up in the quality of cuts I'm getting, but I've been taking fairly light cuts.

I believe the dust hose, though, may cause a bit of flexing if it pulls side to side, so I think I'm going to need to make some type of bracket to hold the hose in place, mounted near the Z stepper. Also need to figure out how to run the hose, as the machine is under my garage door when it's opened. I think I'll run some PVC pipe and hang it just below the open door, but I still have to work that out. I jut need the dust collection so I can mount an MDF spoilboard and surface it.


Also got the cable carrier mounted along the gantry. Didn't realize that the mounting brackets were only $7 at Mcmaster Carr, or I would have ordered it. Instead, I made a wood bracket and tied the carrier to it with wire ties, with some double faced tape to keep it from moving. The other end I just screwed it to the aluminum through the pivot hole. Seems to work fine.


Next step is the control box. I need that done to see where to mount the X carrier track.

Today I need to cut a few parts before I go play golf, and tomorrow I should be able to start on the control box. I'd like to get it done quickly, as I have a possibly lucrative paying sign job lined up. :banana:

Dust shoe works awesome, no more dust. My concerns about it moving the spindle around are probably not much of an issue, since any pressure applied by the hose is relatively constant. I've got a few ideas on what to do with the hose, so I'll see what I can come up with.


A little heads up. When programming a part and setting Z zero, make sure the part is as thick as you think it is. I was making a spacer out of what I thought was 3/4" plywood. Programmed to cut .73 deep. Unfortunately, the ply was only about .72 thick, so the part came loose at the end of the cut and got a little gouge in it. Since I'm painting it, I'll bondo it up first. :)

Six-year construction aside, you seem to becoming a fast worker! :).

Great work on the dust collection shoe - would love to see a vid of it in action. Did you make any attempt to isolate the router exhaust fan from the vac pick-up? I'm trying to come up with a shoe design myself, and it sucks not being able to make any of the cuts with my machine. Daddy-o says no cuts 'till dust protection is in place :(. It's the classic catch-22: need the machine to make a shoe, but can't use the machine 'till I've got a shoe on it (nuts)!!

Hope see more details soon.

Great work on the dust collection shoe - would love to see a vid of it in action.

You really can't see anything when it's In Action. :) Right now the hose barely reaches, and is hanging off the end of the table. I need to hold the hose while it's running so it doesn't get caught on the edge of the table.


Did you make any attempt to isolate the router exhaust fan from the vac pick-up?

No. It's not a vacuum, it's a 600cfm dust collector with a 4" hose. I've seen a lot of people try to deflect the fan blast from their routers, but imo the only reason to do that is if you have poor dust collection. The dust collector sucks more than the fan blows, so there's no problem. I actually wanted to add a compressed air line to blow the chips out of the cut and help cool the tool, but that'll wait til the next machine.

The key to good dust collection is to have a good seal. It's important that the brush is lower than the tool, or the tool will throw chips out beyond the brush. In some cases there's not much you can do to prevent it, especially carving thick stock. But for cutting flat sheet goods, you can generally catch well over 90%. There likely will still be chips packed into the cut, and a few chips may get away, but you should catch just about all the airborne fine dust, and only have larger chips to vacuum up.

If I had wanted to spend more time on this, I'd have made it so the side brush came off, to both change tools easier and to locate a zero point. But for a quickie, this came out really well, and I can pop it off in less than 30 seconds. It's much easier to design something after you've already built it. ;)

Here's a bad video. I'm going to make a board to hold my router bits in a few minutes, so I'll try to get a better video.

YouTube - dustvid

Better video, although you can't really see anything. :)
YouTube - Cutting a Router Bit Holder

Looking good Gerry, wish mine were that far along.

CarveOne

Got all the electronics in a box, now. Two power supplies, motherboard, hard drive, Xylotex, and 5 fans.

After the picture was taken, I wired everything up and added the cable track for the X axis. In the next day or two, I'll cut the cover for the box, and some parts to mount the air filters.

Next item is to try and straighten out the one X axis screw.

Still undecided on workholding, but I'm leaning towards a table full of tee nuts.

After spending a few weeks trying to decide on a hold down method, I'm going with T-nuts. First, I screwed 3/4" MDF down to the bed of the machine, with the screws going into the plywood ribs of the torsion box for the most holding power.

I carefully laid out a grid of nuts, 3" x 3" for half the table, and 3" x 4" at the other half, making sure that they don't interfere with the screws. Also have 2 rows of nuts for mounting fences along the X and Y axis at the origin. Next, I temporarily screwed down an oversized sheet of 1/2" MDF, which will hold the T-nuts.

I then had the machine peck drill the 120 holes 1.2" deep with a 1/4" bit. The T-nuts are 1/4-20, so this will let the hold down screws/bolts go though the nuts about 5/8", which gives some flexilbility in regards to screw length. After the 120 holes were done, I went back and enlarged them to 5/16", at a depth of .55" This is the barrel diameter of the nut, and it's a very nice fit.

Today, I'll remove the 1/2" MDF, and pocket all the holes 7/8" diameter x 1/16" deep, for clearance for the nut flanges. Once all the nuts are installed, I'll glue the 1/2" down, using screws to hold it in place until the glue sets. I'll probably use Gorilla glue, for more working time. I may use epoxy, but I'm concerned that squeeze out might fill the clearance holes. After the glue has set, I'll remove the screws, and fill the holes with an epoxy/sawdust filler mixture. Once that sets overnight, I'll be able to surface the table tomorrow.

I plan on making some cam holddowns to hold the work against a fence, without the need to screw them down. Similar to these.
http://www.vectric.com/forum/viewtopic.php?p=18096#p18096
I'll also be able to screw parts directly into the nuts. And make dedicated fixtures that can bolt to the table. Maybe even a vacuum table at some point.

Installing 120 T-nuts is not the most pleasant task. :( But they're in there now. Now, to surface the table and make some stuff. :)

Made some zero clearance inserts for my table saw. I used arcs in the XZ plane (G18) to cut the clearance slots for the blade. Unfortunately, I found a bug in Mach3 which limited the speed of the G18 arcs. Brian says he thinks it's fixed in the next version, but will check tomorrow. What happens is the arc feedrate is limited to my Z axis velocity setting. So even though the feedrate is F125, they only get cut at 55ipm.

Only took 3 times to get these right (or close enough) :) Have to remember when using G42 that if you want the part bigger, tell Mach3 the tool is bigger. :( Also, when you cut parts upside down, make sure they are drawn upside down. :(

Batteries in the camera were dying, so I had to cut the video a little short. :)




YouTube - ZC Plates

Hi Ger,

I have read this whole thread - congratulations on a cool build. I am just starting out, and am trying to decide between EMT and regular old black pipe. I did read one thread where you said the use of EMT was troubling you as the galvanizing was wearing off. Do you still feel that way...or would you use EMT again.

Since I've been using it, the galvanizing is slowly flaking off, but doesn't seem to be a problem. I don't think the steel is going to wear out. :)

I made a small sample of some crown moulding for my kitchen. It's going to get veneered to match the cabinets I made. I can only cut about 46", so I'll need to run each 97" piece three times to get the full length. I think I need about 10 pieces, so they'll take awhile. I used a 1/4" ballnose bit, with .02 stepover, one pass, 140ipm. When I run the actual parts, I'll do a quick roughing pass with a 1/2" straight bit, and probably use a .03 stepover. Should save a bit of time. The veneer will cover and smooth the small steps.

Gerry,

Please make a video of the veneer being applied to the molding.

It would be interesting to see the procedure.

Jeff...

Yeah I was just going to say that looks like it would be hard to veneer. Would love to see your process though if it works out.

I keep the radius' to 1/8", so the veneer can easily bend around them. Lightly dampen the veneer to make it more flexible, and using a female caul, just clamp it down. It's really pretty simple. I've done it once before, but this profile is a little more complicated. I probably should cut a female sample and glue up the piece I cut last night, just to make sure I won't have any problems.

Gerry,

I was guessing a number of pipes or dowel rods clamping down the veneer.

Does the veneer need to be soaked in water prior to being applied?

Jeff...

No, since it's only about .02" thick. Just wipe both sides with a damp cloth. Depends on the actual veneer, but some species would probably bend around that dry. But the figured birch I 'm using is a bit stiff.

Yeah I was just going to say that looks like it would be hard to veneer. Would love to see your process though if it works out.

I've done this before, but the profile was much simpler.

http://www.cnczone.com/forums/showthread.php?t=8427

There's a close-up pick in my kitchen thread.
http://www.cnczone.com/forums/showpost.php?p=450241&postcount=23