Build Thread- And So Build Number 3 begins......

[thkoutsidthebox]

Hi all, its that time again!

I've started sourcing parts for and designing build no.3. I'm trying to learn from past mistakes, and although I have much better woodworking equipment now, I'm going to go with a metal build. I want this to have the performance I've never achieved yet, and to be of high enough quality that it can be in my shop long term.....about 5-10 years. Its also practise for my future large metal build.

So....I've been working on a design as per the pics below. I'm switching back to a single stepper for the X axis instead of a second slave to reduce the width of the footprint because of liimited space, but also mainly because I don't see the need for two when one will work and leave me more flexibility for a rotary axis later.

I'm going with ballscrews for all axis. Its going to be mainly aluminium. I say mainly because I haven't decided if I need steel anywhere yet. The size of the aluminium is not definate yet, for example I dont know if the box section will end up having an 8mm, 10mm, or 12mm wall etc. I have to drop in to the supplier soon and figure out what he can order for me. I rang him today for the first time since my old aluminium supplier closed down. The new guy was pretty helpful.

I still dont have a welder so its going to be a bolt together construction. The aluminium angle used to join pieces in the pics is all 6mmx38mmx38mm. I'm wondering what people think about this....will it be strong enough....I have my doubts?

Supported rails for the X axis with open pillow blocks. Unsupported 20mm hardened steel rails, with closed pillow blocks for the gantry. The gantry rails I won on e-bay today from MarchantDice and they're 800mm (32"). Do you think this will be ok unsupported...I hope so...I already bought them?!! The X-Rails I'm going to go with about 1550mm (61").

The ballscrews I'm eyeing are 16mm single start 5tpi....thoughts?

I'm throwing around the aim of getting Gecko G203V's for this, but the cost is already going to be spread out until about Christmas, and they'll push it back a few months. I want to have a NEMA 42 for the X-Axis which would need the Geckos. I'm trying to decide between the 2974 oz/in and the 4036 oz/in:
http://www.anaheimautomation.com/man...ec%20Sheet.pdf

Otherwise I might go to the gecko G540 which would limit me to a NEMA 34 at 1700 oz/in which I think would be underpowered for the X axis.....thoughts?

For the Y Im also thinking either a 1700oz/in or a 2974oz/in. Z I'm not so sure....probably somewhere around the 1200/1300 oz/in mark.

The base is going to be supported from the angle on either side. i'm thinking either steel for this, or lengths of 12mmx150mm aluminium across the width....perhaps with angle bolted to the underside to give some more strength. Perhaps 1" or 2" tubing across to support the base? Its not very wide remember only 752mm.

I haven't finished the design. I think the Y axis stepper will be forward of the posts to the rear of the Z carriage back plate....does that make sense? I'm not entirely sure about this....I have my eye on a ballscrew auction on e-bay, but its only 875mm, which is going to be just a bit tight for my gantry....another 50mm would be just right....

The Z is my major hurdle I think. Can someone suggest the best type of rail setup for the up/down movement on the Z carriage to remove unwanted movement while not making the carriage wider than necessary? Ithink that for the Z a smaller ballscrew...maybe 10mm or 12mm would probably be best...ideas?

I haven't put legs onto it yet....I have to figure out the best place for them to not interfere with the movement of the X axis....

I would like a spindle on this....maybe 3Hp, and am currently investigating prices but damn! they are expensive! I came across this asian one: http://www.homeshopcnc.com/HFspindle.html any thoughts?

Thats it for now. I'll post design changes/updates as we go and as I'm arranging build materials.

Main areas of concern:

1. Unsupported Y axis rails.....will they be ok for 752mm span?
2. Pillow blocks for X axis and height of posts above them. Having never used pillow blocks before, will I have a lot of 'rocking' of the gantry?
3. Type of metal pieces to span the base while maintaining ridgidity?
4. 250mm long 12mmx150mm flats on either side of the X axis transferring movement from the screw....will these work out ok under high acceleration/decelleration ?
5. How to design the Z axis movement, and type of rails to use?

Thanks in advance for any advice......

[ger21]

I'm throwing around the aim of getting Gecko G203V's for this, but the cost is already going to be spread out until about Christmas, and they'll push it back a few months. I want to have a NEMA 42 for the X-Axis which would need the Geckos. I'm trying to decide between the 2974 oz/in and the 4036 oz/in: http://www.anaheimautomation.com/man...ec%20Sheet.pdf Otherwise I might go to the gecko G540 which would limit me to a NEMA 34 at 1700 oz/in which I think would be underpowered for the X axis.....thoughts? For the Y Im also thinking either a 1700oz/in or a 2974oz/in. Z I'm not so sure....probably somewhere around the 1200/1300 oz/in mark.

You probably shouldn't need more than 600 oz motors for that machine. The bigger the stepper, the lower it's top speed is. You want them to spin fast with your 5tpi screws. Guys are getting 300+ipm with 400 oz nema 23's and 1/2-10 5 start acme.

With the G540, you'll probably get the most out of your motors by using 400 oz or less motors. You really need to go to Gecko's website, in the support section, and read the part about choosing motors. Before you buy motors 5 times bigger than you need that will actually give poorer performance than correctly sized ones.

[thkoutsidthebox]

Originally Posted by ger21 You really need to go to Gecko's website, in the support section, and read the part about choosing motors. Before you buy motors 5 times bigger than you need that will actually give poorer performance than correctly sized ones.

Thanks ger, I'll go read it again. But do you really think a single 400 oz/in motor could move that all-metal gantry at 300 ipm cutting speed?

Do you think I'd be better sticking with ACME over Ballscrews?

[LeeWay]

My router is using 425 oz running now a double nut, but was a single nut Thompson ball screw for a long time. It is center driven on the X. I use roller skate bearings on cold rolled steel. I am using a 62 VDC power supply running at 3 amps. My gantry weighs at least 100 pounds. I easily rapid 300 IPM. It's pretty quick and plenty strong. I can't cut lexan that fast and get good cuts, but I can do 150 IPM on straights while cutting. Just using a PC 690.

I would at least use ball screws on the X. You will likely need to preload it pretty well to get rid of racking. A steel frame will help you load that nicely. The other two lighter axes would do fine with acme and anti backlash nuts I think.

[ger21]

I didn't really read your first post or look at the pics, I just skimmed through it before. Now I've looked a little closer. Here are some observations, jmo of course.

1) The unsupported gantry will flex too much.
2) Unless you have a really rigid gantry, it'll rack if driven from the center. And your gantry is far from rigid. Spreading the blocks farther apart on the X axis will help with the racking a bit, but even a little racking can cause binding.
3) to get 300 ipm with those ballscrews, you'll need to spin those steppers at 1500 rpm. You may have a hard time spinning any motor over 600 oz-in that fast. What you really want are 2 tpi ballscrews, but the price for those will skyrocket.

[Khalid]

my first metal machine 5' x 3.5' was with 200Oz-in steppers Nema23..and i did rapid 200IPM+..but i limit my cutting to 150IPM...

[thkoutsidthebox]

Thanks all. What type of screws were you using Khalid?

Just to clarify.....I want to cut at absolute minimum 250ipm, in hardwood at minimum 6mm DOC. I know this is quite fast. I did mention that this machine was more of a long term investment than previous builds. I also want very good accuracy ( Nice clean cuts like this: http://vectric.com/forum/download.php?id=6576 ) which is a large part of going for ballscrews.

If I need to, I can fit my 3Hp router to the Z carriage after the build while I save for a spindle.

ger....I heard back, and I can get ballscrews from the same supplier at 2.5 tpi....uhmm...300 ipm @ 750 rpm. Can we suggest a suitable size stepper, with a margin for error since I don't know the gantry weight yet? Is this close enough to the 2 tpi...300ipm @ 600rpm (I'm not sure of the average speed of rotation for a stepper) ? Otherwise I can start sourcing 2 tpi elsewhere online......

I imagine having a low tpi like 2 / 2.5 will mean for cut accuracy I'll have to run my steppers at 1/8 or even 1/16 step, is this correct ( Or just a brain fart! )?

Ballscrews I'm thinking minimum 16mm diameter X and Y axis....but maybe 20mm or slightly larger for the X axis to minimise whip, but I've only checked pricing for 16mm so far.

I'm working on some gantry alterations at the moment. I'll post pics for opinions later. I also have a pretty simple Z-Axis idea with pillow blocks which I'll post later.....but I'm not sure if pillow blocks are the best type of rails for a Z axis....?

Thanks folks.

[thkoutsidthebox]

I'm watching a 1950mm long (71"), 1.2 tpi, 20mm diameter four-start ballscrew for €275 'Buy It Now' on e-bay with free shipping.....sound good/bad ?

It's slightly longer than I intended, but I think I can manage the space if I reorganise my shop....once I dont get wider on the Y axis.

[LeeWay]

If you are wanting that kind of accuracy, I would absolutely go with either supported rails or preferably flat linear rails and trucks. Vee bearings are also a good option if done correctly. The supported rails you show on the X axis won't be able to be preloaded into or away from each other very much. That is likely the best orientation for them though.

There is a cable method floating around the Zone that helps eliminate racking, but the speeds and depth of cut you are looking for won't do too well with all that hardware in the mix. That part is opinion only though.

That big ball screw might do nicely for this.

[thkoutsidthebox]

Hi again,

Ok, so I've been looking at this....

I took the advice and changed to Hiwin Rails on the X-Axis. I may use them on the Y, but I'll wait and see how stiff my unsupported rails are when they arrive. I have a question about preload....do I specify it when buying, and if so how do I decide how much?
Here are my preload options from the website.....I think (Where 'C' is the basic dynamic load rating):

1. Very Light - 0
2. Light - .02C
3. Medium - .05C
4. Heavy - .07C

Pic.1 :

• I'm going to try to get larger aluminium angle pieces for these joints to allow me to place two parallel lines of bolts to help resist movement. Thoughts?

Pic.2 :

• To resist bending in the base piece which will support the flanged X ball nut I have two lengths of aluminium angle bolted to the flat. This may not be necessary when I see how rigid the 150mm x 12mm thick flat bar is, but I need to prevent the ballscrew having to support any weight when at the center of travel.

Pic.3 :

• I moved the X-Rails to the rear of the gantry beams in order to better distribute the weight of the spindle over the X-Trucks.
• In addition, I lowered the top one to bring them closer together nearer to the weight when the spindle is lowered for cutting.
• I also re-inforced the base of the gantry beams to help resist racking. I'm still not happy with this and am going to investigate larger angle pieces in addition to those mentioned for Pic.1 above. Thoughts?

Pic.4 :

• I moved my X-Trucks further apart. They are now 250mm from outside end to end. Thoughts...too much/too little?

Pic.5 :

• I'm thinking about Hiwin rails for my Z-Axis, but I've been having a brain freeze and really can't figure out where to attach a flanged ballnut for movement .....ideas please? Maybe Hiwin rails are bad for this application?

Last but definately not least! I received an e-mail back from PDS Colombo today. They gave me a very favourable ( I think ) quote for a 1.5Hp spindle. 60% duty cycle. Cooling by shaft drive fan. Including VFD. Its quite a bit more expensive than the Chinese one I mentioned previously, but I think its worth it because of the good things I've heard about Colombo on this site, and also because I'll have a decent warranty, spares availability, instructions, after sales support, parts included etc

The only problem I have is that the max shank is 10mm in an ER16 collet, which I'm not happy with, so when I have the money saved I'll probably shoot of an email to them and ask what the nearest available model to take a 1/2" shank is. More relevant right now, the spindle weights 9lbs.

I have requested a quote for a 2.5tpi 16mm ballscrew for the Y axis.

ger's points:

1. Reduced racking with unsupported gantry with new rail layout.
2. Reduced racking with Hiwin rails and increased truck spacing.
3. Do you have a suggestion for where I drive the gantry from if not the center.... preferably without buying two ballscrews ?

Leeway's points:

1. I switched to Hiwin rails.
2. I'm watching that 'big ballscrew' , although I have reservations about the diameter not being large enough for the speeds I want at its nearly 2m length? But unless someone tells me not to?...(The bidding ends in 5 days) I'm going to get it for my X-Axis.....

Thats it for now, any thoughts and help are appreciated. Thanks so far.

NOTE: IGNORE Z Orientation in Pics 1,2,4.

[thkoutsidthebox]

HEY! What about putting the X-Rails on the underside of the beams to reduce racking, instead of the top. I think you can get Hiwin rails for such a setup, although not too sure about weight considerations......good idea/bad idea?

[ger21]

Unless you have a very rigid gantry beam firmly attached to the two sides, your still going to get racking imo, unless you use two screws. You mentioned that you reduced the racking, but you haven't eliminated it. Any is still too much.And the unsupported gantry shafts are still going to flex.