[erikjgreen]

That'll work, but if you're going to the trouble of doing a fill like that, you'd want the best possible vibration damping, which I recall reading somewhere is achieved using multiple gravel sizes. I can't find the reference now, of course

Here's a relevant pdf I ran across though, which should provide some good information about machine design, rigidity, and vibration damping.

http://www.hardingeus.com/usr/pdf/Ma...lStructure.pdf

Erik

[steve123]

Why not try sand mixed with resin? its dense and waterproof, and easy as hell to mix. I have used it in both my CNC builds.

I've been reading that short little thread http://www.cnczone.com/forums/epoxy_...ete_frame.html
I do see them talk about just sand/resin for damping. I'm not opposed to adding some larger pieces along with the sand. The really serious EG bases (without a metal shell) use many grades of aggregate, down to really fine powder. That makes it hard to mix, and I don't need that strength.

Here is the deal for me with EG vs Concrete:
I don't think I want to weld on something with epoxy inside! For the base frame, the horizontals will be filled before they are welded together. So those would be pre-filled with concrete. I would fill and cure them while they are vertical (with the bottom end sealed) to keep them from curing with a sag. Each of the four A-frames for the base would then be welded together separately. I would also fill the X support beams with concrete. Last, I would weld all of the base together in place, leveling everything as I go. I'm looking into low-shrinkage concrete. It looks like it is the kind of aggregate that makes it low-shrinkage. As far as getting the concrete into the tubes, I would have the tube suspended vertically, with a vibrator running, and a funnel on top. I think the concrete would just liquefy and go right down.

The Gantry:
I would weld that together with no pre-fill. I would have it stress-relieved and possibly the front faces ground flat (or epoxy it later). Then I would fill the two beams with EG (play sand,pea gravel,epoxy). I could have a PVC pipe running down the centers to cut down on weight, but maybe not (weight is good). I see that EG is better than concrete for damping, so I see that it's a good idea to use it where I can.

Thanks for the links Erik. More to read! It's amazing that after more than a year, there is so much to learn.

[steve123]

I have to say up front that I have no relationship with Precision Epoxy. I'm sharing this information I found out so you can easily use them.

I got another call back from Mike Ramy at Precision Epoxy to answer my questions. They have updated the web site to make it clearer about this product.

Their product: SC-15P "Steel Casting Epoxy" is specifically designed for self-leveled beds for linear rails. It has the viscosity, steel-bonding, and toughness we want.
All you have to do is tell them the area to be covered, and they will make up a kit to do the job. That includes a part A can with enough room to hold the part B. You pour in part B, mix with your drill and the supplied mixing blade, and pour. They recommend a 1/4" deep pour.

He said not to skimp on the width of the pour. My beams are 3" wide, and the linear rails rails are about 3/4" wide. He told me not to make the effort to have a smaller width that 3". He doesn't know what the minimum width would be, but more width gives better flow and leveling. He also recommends three temporary bridges 3" wide between my two 5' X beams. This allows cross flow so that the two beams to be leveled with each other. The bridges at the ends would be 6" in from the ends of the X beams. You end up with a glass-like flat surface. No sanding required!

Lead time: They recommend a 2-week lead time. They get busy at times, so it may take a while to get your order in, but they like to call you back and talk to you about exactly what you need (Mike prefers the phone to email).

Support: Mike gives you his cell phone number. You can call him 24/7. That seems pretty impressive to me! I also didn't get any "that's too small a job" attitude. I told him it would be a couple of months before I order, and he was fine with that. :cheers:

To sum it up, this seems like a no-brainer to me. Exactly the right product, and excellent support.

P.S.
Mike told me about another method you could use if the bottom rail surface isn't perfectly flat (I assume my profile rails will be). You bolt down the rails with a gap using shims/washers perfectly flat using some reference. You have the rail itself coated with Vaseline. You pour epoxy to fill in underneath. You can take the rail up again if you need to. I told him that wouldn't work for me for two reasons: 1. I don't have a reference, 2. The profile rails are not straight. Anyway, I thought I would pass along that method.

[erikjgreen]

Okay, I'm confused then.

If you're going to weld the frame and not stress relieve it, aren't you going to have to deal with it warping? Or are you just planning on your table slab (concrete, granite, whatever) being rigid and flat and wide enough to mount your gantry rails on it? If you mount the rails to the (welded, poss. warped) base, they won't be aligned with the table...

Erik

PS: I highly recommend the book "High Performance Machining" by Arnone. Excellent discussion of many of the design/construction issues that machine builders deal with.

[erikjgreen]

Annoying fact:

Linear profile rails aren't flat or straight. The manufacturer requires the installer to attach them to a flat/straight surface, which pulls the rails into true. It says this in the documentation from at least two manufacturers I've read (NSK and Thomson, I think).

Erik

[steve123]

Okay, I'm confused then.

If you're going to weld the frame and not stress relieve it, aren't you going to have to deal with it warping? Or are you just planning on your table slab (concrete, granite, whatever) being rigid and flat and wide enough to mount your gantry rails on it? If you mount the rails to the (welded, poss. warped) base, they won't be aligned with the table...

Erik

I wasn't going to stress-relieve the base, because I don't know how to move over 1200 lbs somewhere and get it done. It would still not be exactly level with the rails anyway. I plan instead to weld all of the base in short beads to minimize warping. Then I will wait a week or two, and pour self-leveled X rails AND self-leveled base beams. The grid of base beams will be a pain to make all the dams for, and keep leak-free, but I have some ideas.

PS: I highly recommend the book "High Performance Machining" by Arnone. Excellent discussion of many of the design/construction issues that machine builders deal with.

Thanks Erik. I just purchased that from Amazon.

[steve123]

Annoying fact:

Linear profile rails aren't flat or straight. The manufacturer requires the installer to attach them to a flat/straight surface, which pulls the rails into true. It says this in the documentation from at least two manufacturers I've read (NSK and Thomson, I think).

Erik

Yep. I know it. I will have the flat surface. For sideways I will use a piano wire + feeler gauges or something similar. I thought of gluing down temporary backer blocks against the wire, then just clamping against the wire. I will do the 2nd rail by using the gantry with two blocks on the reference side, and one block on 2nd side. That method is described in the linear guide docs.

[erikjgreen]

Hmm, doesn't sound too bad, as long as you get the base beams and rail bases "close", although you'll need to check that the rail base beams are straight using a taught line, laser or straightedge. Self leveling epoxy should take care of a level base for everything, but I'd still check to be sure... surface plate and prussian blue, or a laser.

Another way to handle straightness of the linear profile would be to obtain a rectangular or square tubing base for them, have it machined/ground flat with holes drilled for the rails, then bolt the rails to it, which will force the rails straight and true. Then you just have to worry about mounting the tubing level and the rails/tubes parallel to each other. Same thing with the base beams, have 'em checked and machined or ground flat with reference or bolt holes. As long as you don't weld 'em into place after, they'll be straight.

Although hauling that much metal to be stress relieved is a pain, it won't be hard to find a shop with a furnace that could fit it.

What you could do if you use granite for your machine base (A used surface plate would be ideal for this... perfectly flat, solid, will never warp) is just weld the base up. Skip self leveling the base beams because no matter how you mount the granite, it won't warp. If you have a big enough granite piece to use the top surface for rail mounts or the sides are also ground (usually they are) then you don't need to self level anything... just attach the rail to the granite, it'll be on a perfectly flat surface, and if you use a mill to drill holes for mounting them or just take care to drill the rail mounting holes in a straight line, they will be perfectly straight, too.

If you don't want a granite surface to mill on, that's no problem either. Just use the granite as a mounting base and get some tooling plate or a vacuum hold-down set up or even cast iron with T-slots and mount it. As long as the top and bottom surfaces of whatever plate you mount on the granite are parallel, then the top surface will also be parallel to the rails.

One more thing... if you haven't, then look up "Madvac CNC" on here and on google. He used a very nice construction technique to get accuracy for his table... basically all his joints from the floor up are replicated with epoxy and dowel pins for centering. Lots of work but he has nice results.

Erik

[erikjgreen]

One more goofy idea I wanted to try is worth mentioning... instead of stretching piano wire to check straightness, stretch the tubing itself with a hydraulic cylinder, then fill with polymer concrete. Once it hardens the tube won't move, and will be straight as an arrow. Scrape a bit to get it perfectly level, then use a laser line or mill to make holes for mounting the profile.

Been wanting to try this for a while.. unfortunately the rails for my next project are pre-mounted

Erik

[steve123]

Wow, this is interesting: Micro Flat Inspection Granite Surface Plate 48" x 36" x 6" | eBay
Too bad it's in Florida.

Micro Flat Inspection Granite Surface Plate 48" x 36" x 6" | eBay
Brand new, in California, 20 miles away. Ebay is quite a bit cheaper than their catalog price. 1200 lbs!

Very interesting. I need to research drilling granite. As you say, no self-leveling of the base required.

I haven't had any luck finding a 3x4" cast-iron t-slot table.

[erikjgreen]

Yeah, shipping on those things is killer. I'd try looking at local surplus places/used tool shops, on craigslist, or do an e-bay search for surface plates in your local area. You can limit by a radius in miles around your ZIP code. The shipping weight has two sides to the issue... you don't want to pay to ship it, but neither does anyone else, so dealers have a hard time getting rid of them depending on how close they are to industrial areas.

For what it's worth, even the least accurate grade of plate is as or more accurate than a self leveled surface.

Also, surface plates are mostly granite now but occasionally can be found in cast iron, keep your eyes open.

You might also want to look for someone parting out an older mill, surface grinder, etc. since they might have a table or work surface that's in good shape. Even if it's got scratches or rust, it can be cleaned up and (if it's otherwise ideal) ground/machined to a perfect finish.

Lastly, tooling plate comes in a variety of sizes/types. I have one plate that's about 7 inches by 16 and is cast iron with three T slots. I also have a piece of steel ground plate that's 18 inches by 36 and an inch thick. If you are willing to pay for stress relief and grinding, almost any suitably sized steel plate will make a fine table for you.

Erik

[steve123]

I spent the last couple of weeks searching Ebay and Craigslist. This granite table appeared 11 miles from me. It is 36" x 48 x 8, with two ledges. The surface is 38" high (higher than I really wanted, but I will go with it). It appears to be an old grade A plate. I don't have the tools to be sure.

Attachment 188220

It ended up costing $500 + tax plus $120 to ship it to me. The price was higher than some I have seen, but it was so close that the shipping was much less. The company that sold it to me are riggers Advanced Riggers & Millwrights About They get tables like this (and VMC's and giant mills) as cast offs when they move companies around.

The granite looks pretty good. I was bummed to discover that there are lots of really tiny divots you can feel. There are also some scratches where you can see silver or brass colored metal. I ran a test indicator over it attached to my Noga holder. The needle shivers within a range less than .0001. The scratches don't show any depth! :wee:

It has bolts for leveling... Err. No. It has one bent, rusty bolt for leveling. The other 3 legs have a rusty bolt hole.
Attachment 188222

I have a nice bolt supplier nearby, and picked up 4 new 5/8-11 bolts.
Attachment 188224

I installed the bolts today. That was a pain. No, I didn't have a 5/8-11 tap.

I now need to make some feet for the bolts to bear on. I don't think I want to concentrate that much force on my concrete floor. I'm thinking of using 2x2" x 1/4 aluminum plate with a divots drilled in the centers.

I'm working on a design that uses the existing metal table, with beams bolted to the sides of the granite plate, and lots of bracing. I still want a design that has a lower beam in front for a vice or 4th axis. I'm also starting reading "High Performance Machining". Fun times!

[wizard]

You will want so etching thicker than 1/4" thick aluminum. Aluminum isn't ideal in contact with concrete anyways so if you do use aluminum isolate it from the concrete. You will likely be just as well off buying castiron pads designed for usage as feet.

I spent the last couple of weeks searching Ebay and Craigslist. This granite table appeared 11 miles from me. It is 36" x 48 x 8, with two ledges. The surface is 38" high (higher than I really wanted, but I will go with it). It appears to be an old grade A plate. I don't have the tools to be sure.

Attachment 188220

It ended up costing $500 + tax plus $120 to ship it to me. The price was higher than some I have seen, but it was so close that the shipping was much less. The company that sold it to me are riggers Advanced Riggers & Millwrights About They get tables like this (and VMC's and giant mills) as cast offs when they move companies around.

The granite looks pretty good. I was bummed to discover that there are lots of really tiny divots you can feel. There are also some scratches where you can see silver or brass colored metal. I ran a test indicator over it attached to my Noga holder. The needle shivers within a range less than .0001. The scratches don't show any depth! :wee:

It has bolts for leveling... Err. No. It has one bent, rusty bolt for leveling. The other 3 legs have a rusty bolt hole.
Attachment 188222

I have a nice bolt supplier nearby, and picked up 4 new 5/8-11 bolts.
Attachment 188224

I installed the bolts today. That was a pain. No, I didn't have a 5/8-11 tap.

I now need to make some feet for the bolts to bear on. I don't think I want to concentrate that much force on my concrete floor. I'm thinking of using 2x2" x 1/4 aluminum plate with a divots drilled in the centers.

I'm working on a design that uses the existing metal table, with beams bolted to the sides of the granite plate, and lots of bracing. I still want a design that has a lower beam in front for a vice or 4th axis. I'm also starting reading "High Performance Machining". Fun times!

[steve123]

I got some 3x3x 1/4" steel for feet, made a divot for the bolt-ends to sit in, and leveled the plate to within .0002 per 10". I spent more than 4 hours on it by myself (next time get a friend to help read the level!). The problem was twist in the plate. I could back off the high-corner (front right) bolt until it rocked on the other diagonal bolts, but it still wasn't level on that side. I jacked it up and looked at the hard rubber pads underneath. These pads have threads in them like tire rubber. I saw that the plate has 3 points of contact. Two on the left side, and one in the middle on the right side. The one that was for the middle-right was not centered (probably moved in transport). I moved it back into the box marked on the bottom of the plate, and tried again. Better, but still not quite there. I moved farther it back several times and got it all level. The pad is half-way out of the box marked on the bottom of the plate. By this time it was midnight. The next day my quads are killing me.

I'm not satisfied with this setup for several reasons:
1. There is still very little weight on the back-left and front-right legs.
2. You can put weight on any corner and watch the level move (probably the rubber pads squishing).

I'm thinking that in order to use this as part of a machine, I would want 4 hard points of contact that I can adjust. That way I can take twist out, and have it be solid with lots of weight on any corner. I'm thinking of something like four 4"x6"x1/2" steel plates epoxied to the bottom, with a connection to the structure at the middle of each plate.

[steve123]

Here is an updated design. This has a frame that wraps around the granite surface plate. There are leveling bolts under the plate. Once the plate is level, I will apply epoxy filler between the plate sides and the metal pads there. I will bolt the sides against the plate.

Attachment 191278 Attachment 191280 Attachment 191282

This design now has 3 x 5 x 1/4" rail support beams. They will be filled with EG.

The beams don't stick up as far as before, and they are bolted into the granite. That should be very stiff and dampen vibration very well.

The gantry now has short risers on the sides.
Attachment 191284 Attachment 191288

Here is the full assembly.
Attachment 191286

The Z can slide over the x rails. There is about 35" of x travel with the 1" steel spacer blocks. Attachment 191290

I'm still working on sourcing ball screws and servos. I looked for quite a while for C5 grade screws, and there isn't much available on ebay. There certainly are no pairs. I think it will be rolled ball screws. I'm thinking of mounting glass slides on all axes (at least the Z). Servos at the moment will be kellinginc.net 1kw brushed.

There is a hacker space opening in my city. This is going to be extremely useful. VOCADEMY ? The place to learn and make ANYTHING!

[steve123]

I have seen a lot of advice on ball screws. The milling machine guys say "use 2005 or 2505". They want the 5mm lead for accuracy and torque. The router table guys say "use 1610 or your rapids will suck" (paraphrased). My machine spans the divide (or is it the no-man's land?). It will cut metal, and it's as big as many router tables. It's much larger than the small vertical mill designs.

My design requires reasonable rapids like 400 IPM to cover the 35" wide by 55" long area. I have had advice to move away from stepper motors to servos. This allows more power throughout the rpm range, and allows much higher RPMs for rapids. I'm OK with that.


Here is a chart I made by using the Nook critical speed calculator: Metric Critical Speed Calculator | Nook Industries
Note that the Nook calculator has a 20% safety margin, so you can probably get higher rapids than shown. Also note that the 2010 size is not sold by LMB2008. They say the ball nuts for that size are not sturdy enough. It's good to know that they have some standards!

Attachment 192126

Things to note:
1. The critical speed for ball screws is extremely dependent on length. The short ball screws in small mills don't whip! The only issue on these mills related to rapids is the loss of stepper motor torque at higher RPM's.

2. My Y (about 41" of screw) would be fine as far as rapids with 1610, or 2005. I'm going to try fitting a 2005 into the Y. I know a 25mm ball screw won't fit. I would be spinning the 2005 at about 2000 RPM! I need to be sure that's OK.

3. I can see that my 65" X really needs 2510's. 2505 only gives 184 IPM! I would have to cut 20" off that axis to use a 2505! I'm considering servos and 3 to 1 reduction to get the torque.

One of the common things mentioned by the milling guys is ball screw stretch. I haven't been able to find any numbers for this! I'm upgrading X and Y to bigger screws, so I hope they will be OK. The only axis left with a 1600 series screw is the Z. I could drive it with a servo and get 1200 IPM rapids! ' Rapids are not a limiting factor for Z. Only ball screw stretch would be an issue. The space is VERY tight in the Z.

[erikjgreen]

I'm replying inline.

I have seen a lot of advice on ball screws. The milling machine guys say "use 2005 or 2505". They want the 5mm lead for accuracy and torque. The router table guys say "use 1610 or your rapids will suck" (paraphrased). My machine spans the divide (or is it the no-man's land?). It will cut metal, and it's as big as many router tables. It's much larger than the small vertical mill designs.

My design requires reasonable rapids like 400 IPM to cover the 35" wide by 55" long area. I have had advice to move away from stepper motors to servos. This allows more power throughout the rpm range, and allows much higher RPMs for rapids. I'm OK with that.

First, it's worth noting that choosing ballscrew sizes is mostly driven by, in order of priority: 1) Mechanical advantage desired 2) Precision of positioning needed 3) Stiffness of ballscrew required (based on load).

The individuals saying "your rapids will suck" are only partly right... whether your rapids are "zippy" or not depends on the rest of your drive train. RPM limits permitting, a 5 lead ballscrew can be just as fast as a 10 if rotated quickly enough, and provides twice the mechanical advantage.

More importantly, for any non trivial load of table and work piece or axis and spindle torque is far more important than speed for a ballscrew. Yes, it sucks to have to wait for your spindle to rapid from point A to point B, but the ability to rapidly accelerate and decelerate the axis affects how long programs take to run much more than rapid speed, as well as (more importantly) how accurate your results are. It also affects acceleration, which you need to make use of big rapid speeds. Here's a math problem to work out: If your machine has a 4 foot wide table, is capable of 800 ipm rapids, but can only accelerate (torque) at 1 inch/second/second, how long will a rapid from one side of the table to the other take, and how fast will the spindle be moving when it gets there?

On the subject of whip... if your ball screw is supported at both ends as it should be, it won't whip much unless it's very undersized. For comparison, my mill has a 44 inch long X axis. The ball screw on this axis is a 3605 ground screw... that's 36 mm diameter, and is supported at both ends.

The reason you don't see ball screws on router tables is that they tend to get really expensive really quickly.

Ultimately, whether your machine can cut metal depends successfully and/or well depends less on the ball screw than it does on the stiffness of the machine frame.

Here is a chart I made by using the Nook critical speed calculator: Metric Critical Speed Calculator | Nook Industries
Note that the Nook calculator has a 20% safety margin, so you can probably get higher rapids than shown. Also note that the 2010 size is not sold by LMB2008. They say the ball nuts for that size are not sturdy enough. It's good to know that they have some standards!

Things to note:
1. The critical speed for ball screws is extremely dependent on length. The short ball screws in small mills don't whip! The only issue on these mills related to rapids is the loss of stepper motor torque at higher RPM's.

FYI, whip is dependent on a combination of things... length of the screw vs. diameter, metal composition, support type, etc. Each ball screw maker has recommendations to minimize it, as well as different RPM limits, so make sure if these are critical that you know what you're getting. Nook in particular is known for inexpensive ball screws. I'd buy one of theirs for a tool to cut wood or plastic, but for metal I'd look for an NSK or similar, since most likely your tolerances are higher.

2. My Y (about 41" of screw) would be fine as far as rapids with 1610, or 2005. I'm going to try fitting a 2005 into the Y. I know a 25mm ball screw won't fit. I would be spinning the 2005 at about 2000 RPM! I need to be sure that's OK.

3. I can see that my 65" X really needs 2510's. 2505 only gives 184 IPM! I would have to cut 20" off that axis to use a 2505! I'm considering servos and 3 to 1 reduction to get the torque.

2000 rpm is probably borderline depending on the screw. For most of the ball screws I have, that would be much too fast.

If you are using servos and you don't have mechanical reduction (at least a couple timing pulleys) then you'll probably be short on torque and have too much RPM. As mentioned above, fast rapids aren't really the goal.

If on the other hand you're using steppers, then you can use direct drive, but you'll probably want to use microstepping drivers so you can make small movements of the spindle. That also changes your power curve some, you'll lose torque more rapidly as RPM increases.

FYI, 180 ipm isn't bad. My mill, which is an older machine, was originally designed for 250 ipm, which is plenty for its purpose. The most it would ever have to move for a rapid is 20 inches, and most of the time rapids are less than 10 inches. There are some VMCs released in the last few years with 1000 ipm rapids, but they tend to have enormous torque as well as high HP spindles to permit them to make full use of that speed.

One of the common things mentioned by the milling guys is ball screw stretch. I haven't been able to find any numbers for this! I'm upgrading X and Y to bigger screws, so I hope they will be OK. The only axis left with a 1600 series screw is the Z. I could drive it with a servo and get 1200 IPM rapids! ' Rapids are not a limiting factor for Z. Only ball screw stretch would be an issue. The space is VERY tight in the Z.

Unless you're dealing with very tight tolerances, ball screw stretch will not be an issue. Tolerances of less than 0.0001, I mean. To be blunt, if you cared about tolerances that small you wouldn't be buying anything less than new precision ground ball screws, and you'd be building your whole machine in a climate controlled room. A ten degree temperature change in the metal your tool is made from will cause more inaccuracy than ball screw stretch.

As mentioned above, designing for fast rapids is a fool's game. That said, if you want fast rapids, the Z is the place to have them, since for many types of hobbyist operations (lithophanes, contouring) Z speed is the limiting factor in program run time.

Erik

PS: If I haven't recommended it yet, Miles Arnone's book "high performance machining" explains a lot of these issues quite well.

[wizard]

…

There is a hacker space opening in my city. This is going to be extremely useful. VOCADEMY ? The place to learn and make ANYTHING!

Nice! I do wish we had a similar Hackerspace around here. I know of nothing within the local area with similar features.

[wizard]

I have seen a lot of advice on ball screws. The milling machine guys say "use 2005 or 2505". They want the 5mm lead for accuracy and torque. The router table guys say "use 1610 or your rapids will suck" (paraphrased). My machine spans the divide (or is it the no-man's land?). It will cut metal, and it's as big as many router tables. It's much larger than the small vertical mill designs.

A fundamental problem with machine tool building is finding the right compromises for the planned uses. Tat is part of engineering our design and biasing that design towards what you need from the machine. Nobody can really help you there.

My design requires reasonable rapids like 400 IPM to cover the 35" wide by 55" long area. I have had advice to move away from stepper motors to servos. This allows more power throughout the rpm range, and allows much higher RPMs for rapids. I'm OK with that.

You are getting close to the dimensions where rapids might actually be realized and sustained assuming you can accelerate fast enough. But here is the real problem with fast rapids, how often do your programs really leverage those rapids? If you have parts that require several hours of feed rate moves and a couple of rapids it just isn't worth the effort in many cases to design for high speed rapids. This is especially the case on smaller machines though in your case you are probably past that point.

In the end think long and hard about how much energy you want to put into designing in fast rapids. It may be a big benefit for your intended usage though obviously I can't say one way or the other. In the end you will send a lot of money to get there.

Here is a chart I made by using the Nook critical speed calculator: Metric Critical Speed Calculator | Nook Industries
Note that the Nook calculator has a 20% safety margin, so you can probably get higher rapids than shown. Also note that the 2010 size is not sold by LMB2008. They say the ball nuts for that size are not sturdy enough. It's good to know that they have some standards!

I'm not a big fan of high speeds on ball screws. In my mind it just leads to premature wear.

Things to note:
1. The critical speed for ball screws is extremely dependent on length. The short ball screws in small mills don't whip! The only issue on these mills related to rapids is the loss of stepper motor torque at higher RPM's.

Actually the big issue on small machines is actually ever reaching maximum rapid speeds for any length of time. Ask yourself this how often on a small mill do you slew an axis the full length of travel. On a Z axis for example that may only happen upon a tool change if then. This is especially the case on a router sized mill handling sheet goods where the Z cold run all day moving barely an inch.

I'm not trying to say rapids aren't important just don't over stress the feature in a design if it isn't needed. High rapids require a lot of torque meaning bigger components throughout. Bigger components means more money.

2. My Y (about 41" of screw) would be fine as far as rapids with 1610, or 2005. I'm going to try fitting a 2005 into the Y. I know a 25mm ball screw won't fit. I would be spinning the 2005 at about 2000 RPM! I need to be sure that's OK.

I don't like that RPM myself, if you do go that route I'd suggest some sort of automatic lubrication.

3. I can see that my 65" X really needs 2510's. 2505 only gives 184 IPM! I would have to cut 20" off that axis to use a 2505! I'm considering servos and 3 to 1 reduction to get the torque.

Which in the absolute worst case is like 20 seconds. How often wold the worst case happen?

One of the common things mentioned by the milling guys is ball screw stretch.

Do you mean thermal expansion? Stretch implies something else in my mind.

I haven't been able to find any numbers for this! I'm upgrading X and Y to bigger screws, so I hope they will be OK. The only axis left with a 1600 series screw is the Z. I could drive it with a servo and get 1200 IPM rapids! ' Rapids are not a limiting factor for Z. Only ball screw stretch would be an issue. The space is VERY tight in the Z.

I'm not sure what you are talking about with respect to ball screw stretch but it isn't a problem on the Z axis either if you mean thermal expansion. In any event your velocity profiles would look like inverted V's with the axis never reaching full rapid speed in most cases. High speed rapids on a Z axis just aren't a big deal for most routers. Just make sure you have good stiff acceleration that is reliable (no lost steps).

[steve123]

Thanks guys for your replies. By "ball screw stretch" I mean lack of stiffness. The screw compresses or stretches when pushing/pulling the load. Some people that have done RF45 conversions have complained about screw stretch.

I did get the High Performance Machining book. It has provided some valuable insights. I am definitely building an "under $100k" machine that the book talks about. Ha! But seriously, I will do another post about what I learned in that book.

I switched my design to using servos because the stepper's torque tops out <1000 RPM, and going with any larger ball screws puts you over what the steppers can do. If that wasn't an issue, everyone needing a powerful push could just use 1600 oz/inch steppers, but they are terrible at higher RPM's.

I was going to argue about total mechanical advantage including the step down pulleys, but OK, I get it. You want as much torque as you can get for quick small movements, which is what the machine is doing most of the time. I see both of you telling me the same things. I will go with 5mm pitch screws. It looks like:
Y (65"): 2505, max of 220 IPM. A C7 pair from LMB2008. I haven't seen anyone stocking larger ones.
X (41"): 2005, max of 470 IPM. A single screw, so I can try for a C5, or C7 LMB2008 if I have to.
Z (18.3") 1605, max of a gazillion IPM. A single also, C5 if I can.

I am planning on glass slide encoders for X and Z. The 2 60" for Y are quite expensive, so maybe not yet for Y. I will do another post showing the slides in my design. They should help precision even with C7 screws.

I did some research on maximum ball screw RPM. Looking at a PDF, they show that the max rpm varies based on the nut design.
The general rule is (root diameter in mm) * RPM < 50000. This max speed is allowed for "short periods". Here is the max speed for the common sizes based on that formula:
16mm = (50000 / 13.3) = 3759 RPM
20mm = (50000 / 17) = 2941 RPM
25mm = (50000 / 21) = 2381 RPM
At these high speeds the lubricant starts getting thrown off of the screw, and you have to monitor the lubrication and nut temperature.

This answers the question for my RM2005 Y. 2000 RPM for short periods is OK.

Here's a math problem to work out: If your machine has a 4 foot wide table, is capable of 800 ipm rapids, but can only accelerate (torque) at 1 inch/second/second, how long will a rapid from one side of the table to the other take, and how fast will the spindle be moving when it gets there?

Assuming it didn't have to slow down, It would be going 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 = 45 inches in 9 seconds. Traveling at 9 IPS or 540 IPM.
Yes, I have to look at accelerations. Servo sizing is next! I will be looking at gear ratios, kilowatts, brush vs brushless, motor inertia, encoders, drivers, tuning, and accelerations. I had it made when I was just looking at steppers!