Need Help!- NEW CNC BUILD! Brilliant Minds needed within!

[RotarySMP]

Nice rendering.

The machine will be stiffer if you turn the table around 90°. The stiffness of the table axis will be the same, but the shorter gantry will be much stiffer.

You will soon need to make your spindle decision, so that you can finalise the gantry position.

Have you looked at those HF spindle + VFD sets that are becoming common (Keling has them).

[diyengineer]

Originally Posted by RotarySMP Nice rendering. The machine will be stiffer if you turn the table around 90°. The stiffness of the table axis will be the same, but the shorter gantry will be much stiffer. You will soon need to make your spindle decision, so that you can finalise the gantry position. Have you looked at those HF spindle + VFD sets that are becoming common (Keling has them).

KL-2200 Milling Spindle $849 Special sale: $699 Specification
OD=80 MM
Voltage: 220V
Speed: 8000-24000 R/min
Power: 2200 W (3HP), Frequency: 400 Hz
Water Cooling, ER20 collet chuck
Grease Lubrication, NSK Bearing,
Water cooling about 100 gal/hr

KL-VFD22 Specifications: $258.95
LED Display
Input: 220V 50/60Hz, Single Phase
Output: 0-240V 2.2Kw 11A
Frequency Range: 0.1-400Hz
0-10V Analog interface capability
Variable speed potentiometer

3Hp, water cooled and 24,000rpm. It seems to fit my price range as well. Have you heard good things about them? Are the bearings replaceable?
Do they or anyone else sell a mount for that spindle?

In regards to the table when you say flip the table 90 degrees would i swap the shorter 37" rails to the top and the longer 48" rails below. I'm kind of confused. Or would the table instead of currently sliding front and back, the table would slide side to side? The concept sounds like it would be better though, shorter gantry equals more stiffness by putting the longer axis on the ground under weight. Just cant picture it at the moment so help me out haha.

Thanks RotarySMP!

[RomanLini]

I think you will have plenty of stiffness for cutting 0.063" sheet which you'll probably do in 2 passes anyway.

Suggestions;
Remove the front gantry angled support, they will interfere with your spindle and rigidity won't change much if you lose them.
Make the main gantry beam twice as high (thick) to get those rails much further apart. That also stiffens it.
Make the table bearings slightly further apart (6"?), this probably means making the machine a bit longer to still get 24" travel there.
Consider lowering the gantry a bit, not everyone needs 4" of Z travel and it hurts rigidity a lot. A good sheet cutting machine has a small Z.
I would use a 10mm alloy table to save cost and screw a 12mm acrylic sheet to the top of it. That's a good tradeoff of rigidity and cost, and the acrylic is replaceable, self-tapping for holddowns and will dampen the table ringing. It also won't swell like MDF.
Table under-supports are good (Bartuss1 already suggested).

[diyengineer]

Originally Posted by RomanLini I think you will have plenty of stiffness for cutting 0.063" sheet which you'll probably do in 2 passes anyway. Suggestions; Remove the front gantry angled support, they will interfere with your spindle and rigidity won't change much if you lose them. Make the main gantry beam twice as high (thick) to get those rails much further apart. That also stiffens it. Make the table bearings slightly further apart (6"?), this probably means making the machine a bit longer to still get 24" travel there. Consider lowering the gantry a bit, not everyone needs 4" of Z travel and it hurts rigidity a lot. A good sheet cutting machine has a small Z. I would use a 10mm alloy table to save cost and screw a 12mm acrylic sheet to the top of it. That's a good tradeoff of rigidity and cost, and the acrylic is replaceable, self-tapping for holddowns and will dampen the table ringing. It also won't swell like MDF. Table under-supports are good (Bartuss1 already suggested).

Thank you RomanLini for your suggestions.
Front gantry 45 degree support: Removed.
Rear gantry angle support: Beefed up.
Vertical gantry beams: Increased sized, from 3"x3" to 3"x6".
Width of machine: Increased by 1" to place space between moving table and vertical supports.

The linear rails are already *fully* supported from the underside. I DO need to add bracing in between the 80/20 rail supports, to make sure the table doesn't sway or walk side to side.

I would like to at least keep a usable 6" Z axis, so i can use a variety of bits and tools, while keeping the door open for overall multi purpose use later on down the road. Hopefully by beefing up the gantry it should be fine.

The table thickness ultimately will depend upon the price of the plate, which i have yet to call Alaskan Copper in Seattle.

I will finish up the new modifications and repost better renderings with multi angle views so everyone can lend there knowledge! :P

Thanks again to everyone that has helped me out!

Edit/Update:
Eliminated the need for all 45 degree angle supports used for the gantry. Where the main vertical gantry intercepts the base perpendicularly it is now fully boxed in on all 4 sides using, (2) 90 degree double gussets, and (4) monster bolting plates (6 parts total on each gantry leg + hardware). The verticals are now 3"x6" instead of 3"x3". Ran a few basic static and dynamic load tests vs the before and after and it has made a HUGE difference, plus looks cleaner. Also added horizontal supports for the underlying rails, so the table wont walk horizontally.

Hope to have pictures of the updates tonight, and a fly around video of it tomorrow night.

Then i can finally focus on the Z axis, motor mounts, ballscrew mounts, wire carriage, etc.

[diyengineer]

Uploaded 5 pics.

Excuse the random bracket in the picture solidworks photoview 360 was giving me a ghost bracket and i didn't feel like going through each bracket and deleting them and there mates to find it at 4:11AM.

What you see so far is weighing in it 259 pounds (the table being roughly 60 pounds of that weight).

I still need to finish putting brackets on the underside table support. The 4 small pieces of 80/20 underneath (the ones with no brackets yet) will act as stiffeners to prevent the table from walking left and right.

Overall its a beast and will most likely tip the scales @ 300+ when its all said and done.

To do tomorrow list:
Delete ghost bracket.
Apply brackets to underside stiffeners.
Delete all fasteners and use dome head socket bolts instead of protruding socket.
Start on the creation of the Z axis.

Days to come:
Model ballscrews, support ends (fixed and free), and brackets.
model motor mounts.
find a cable carrying system and model.
Model spindle, + find spindle bracket.

Overall it is coming together nicely.

[packrat]

You could stiffen up your table by flipping your rail system. Bolt the rails to the bottom of the table and the pillow blocks to the frame. The rails then act as ribs to stiffen the table.

cary

[Drassk]

Rails aren't structural members, so attaching them to the table would actually make the whole system less stable rather than more. The frame is there mainly to stiffen up the rails and keep them straight, so they actually need to 'take' stiffness from the system to work rather than 'giving' stiffness to it.

[jalessi]

Diyengineer,

Your rendering looks awesome, the machine looks as rigid as a tank!

No doubt it will cut .063 like soft butter.

Jeff...

[diyengineer]

Thanks for the compliments! Kelinginc wrote me back and said the ball screws they offer DO NOT have end machining BUT they do it for cheap. SO that makes me happy. What kind of end machining do i want if i plan on running Nema 34 steppers/servos? Please advise.

Ballscrews
SFU1605-C7-1055 (41.5") Flanged ( for X): $86.95.00/ea

SFU1605-C7-775 (30.5") Flanged ( for Y): $70.95.00/ea

SFU1605-C7-305(12") Flanged ( for Z): $52.95.00/ea

Those are "overall lengths" above. How much usable thread do you think there is on each one?

End Blocks
C7 Grade Steel Ballscrew Support Specification

BK12-C7 (Fixed End): $58.79

BF12-C7 (Free End): $29.95

BK15-C7 (Fixed End): $62.95

BF15-C7 (Free End): $33.95

Thanks everyone!

UPDATE

Which coupling is the best? The zero backlash one?
http://www.kelinginc.net/CNCPackage.html

[packrat]

Originally Posted by Drassk Rails aren't structural members, so attaching them to the table would actually make the whole system less stable rather than more. The frame is there mainly to stiffen up the rails and keep them straight, so they actually need to 'take' stiffness from the system to work rather than 'giving' stiffness to it.

The O/P, the last I knew was using supported Thompson style rails. They are capable of being used structurally. They are many times stiffer than the THK style linear rails.

[RomanLini]

That's starting to look quite good.

Just a couple of points;
1. Its weak cutting at the front and rear edge of the table, unneccessarily so. You can move the bearing trucks further apart, say 4", and make the machine 4" longer from front to rear. This also fixes the problem where your table hangs over the front and rear frame of the machine at the limits, ie at home position. That could also cause issues with your motor and table interference and would look clumsy and unprofessional having table and job hanging out.

Also (again this was Bartuss1 suggestion) don't be shy to glue and screw some box section under your table, even some 1" square tube glued and screwed under that table in the right places will stiffen it enormously. One of the weakest links in your design at the moment is that the table will start behaving like a sounding board, and the last thing you need with cutting thin sheet is the table deflecting downward.

2. Really think hard if you need 6" Z travel and 9" clearance between table and gantry. This is the weakest part of your design now in my opinion. If you have 0.001" slop in each gantry bearing, and the cutting tool is 9" below the bottom bearing, you get about 0.003" movement at the tool! And there is a lot of flex in Z, the tool will always be 9" below the closest Z brace point which will be the lower Z bearing.

If you were to reduce the gantry clearance and Z travel to about 3" the payoffs are enormous, maybe 3 times less slop at the tool and many times less flexion. It lets you get the lower Z bearing much closer to the tool also.

If you look at this picture of my machine the lower gantry bearing, and lower Z bearing are both around the same level as the router collet when it is cutting flat sheet. It's enormously stiff in all XYZ. The table-gantry clearance is 60mm and Z travel is 65mm, BUT I can easily unbolt the rotuer plate and lift it up to 100mm so I can get a job up to 160+ mm under the tool provided it doesn't need to go under the gantry. The tool is well forward from the gantry so I can easily fit a 4" rotary table vertically in there for cog machining (the rotary table never needs to go under the gantry).

http://www.cnczone.com/forums/attach...3&d=1267261555

It comes down to a tradeoff if you want to sacrifice a huge amount of accuracy and rigidity ALL the time to get a couple more Z inches you might use sometimes. Remember the decisions you make now are the ones that payoff for the life of the machine or haunt you for the life of the machine.

At the very least, regardless of the Z travel you choose I would increase the gantry beam vertical thickness from 6" to 9".

[diyengineer]

Originally Posted by RomanLini That's starting to look quite good. Just a couple of points; 1. Its weak cutting at the front and rear edge of the table, unneccessarily so. You can move the bearing trucks further apart, say 4", and make the machine 4" longer from front to rear. This also fixes the problem where your table hangs over the front and rear frame of the machine at the limits, ie at home position. That could also cause issues with your motor and table interference and would look clumsy and unprofessional having table and job hanging out. Also (again this was Bartuss1 suggestion) don't be shy to glue and screw some box section under your table, even some 1" square tube glued and screwed under that table in the right places will stiffen it enormously. One of the weakest links in your design at the moment is that the table will start behaving like a sounding board, and the last thing you need with cutting thin sheet is the table deflecting downward. 2. Really think hard if you need 6" Z travel and 9" clearance between table and gantry. This is the weakest part of your design now in my opinion. If you have 0.001" slop in each gantry bearing, and the cutting tool is 9" below the bottom bearing, you get about 0.003" movement at the tool! And there is a lot of flex in Z, the tool will always be 9" below the closest Z brace point which will be the lower Z bearing. If you were to reduce the gantry clearance and Z travel to about 3" the payoffs are enormous, maybe 3 times less slop at the tool and many times less flexion. It lets you get the lower Z bearing much closer to the tool also. If you look at this picture of my machine the lower gantry bearing, and lower Z bearing are both around the same level as the router collet when it is cutting flat sheet. It's enormously stiff in all XYZ. The table-gantry clearance is 60mm and Z travel is 65mm, BUT I can easily unbolt the rotuer plate and lift it up to 100mm so I can get a job up to 160+ mm under the tool provided it doesn't need to go under the gantry. The tool is well forward from the gantry so I can easily fit a 4" rotary table vertically in there for cog machining (the rotary table never needs to go under the gantry). http://www.cnczone.com/forums/attach...3&d=1267261555 It comes down to a tradeoff if you want to sacrifice a huge amount of accuracy and rigidity ALL the time to get a couple more Z inches you might use sometimes. Remember the decisions you make now are the ones that payoff for the life of the machine or haunt you for the life of the machine. At the very least, regardless of the Z travel you choose I would increase the gantry beam vertical thickness from 6" to 9".

1. I guess i will bite the bullet and extend the table so it wont overhang, I was just trying to save $ but your right, might as well make it the best i can the first time around or live with it forever lol.

2. The gantry was planned to get lowered i just have yet to model the spindle and the Z axis so i just left it alone for the time being at a higher height then needed. I also need a bit of clarification which i'm sure you can help me out on. When you define Z axis travel, that is basically from the table top to the spindles collet face correct? If it was 6" above, and it had 6" of travel it could put the collets face flush right on the table correct? Hopefully I'm on the same page. I really want to add a 4" 4th axis rotary positioning table in the future. So as long as i can easily add that i will go lower (4" z axis?). Also in general, how long are most bits that fit in the spindle? I need to take that into account as well.

3. After its all modeled i will run static and dynamic load tests as well as deflection tests on all the beams as well as gantry. I will then engineer accordingly and increase rigidity to locations that are prone to excess structural deflection. =P

4. Thank you for your suggestions! Get back to me regarding that Z axis definition, and general bit length.