Need solid advise, for 60"Y axis, screw or rack and pinion Whip is of concern!

[Born2dive9702]

Hello everyone I have finally decided to build my own CNC for a project I am working on.The work area of the CNC must accomidate a work piece 60" long, by 24" wide, hence the reason I would like to build.Now I have read up a bit, and it seems that most folks do not use one this long as the DIY builds, but I am going to give it a try any way.What would be the best for this system. I planning on running dual screws or rack and pinon one on each side of the Y axis. NO center thrust as I have found out on my small store bought unit this causes binding and twisting sometimes of the deck.So the questions are 1. which is more accurate screw with a upper and lower v rider rails? or a Rack and pinion with v rider rails?2. What If I went with a screw what would be the best one to prevent whip.I would like to jog the gantry a little quicker as I am milling foam. and balsa wood.I will start here with just these 2 questions.

Similar Threads:

• New Machine Build- How can do ""manual stepper motor sizing for "rack and pinion"calculation""???
• Newbie- Rack n pinion or ball screw
• New build concern lead screw whip?
• need advise on metric rack & pinion selection
• first post can the rack an pinion be used to drive x axis rather than screw on lathe

[RCaffin]

There is no question that proper ball screw & nut will be far more accurate than anything else.
But, if it is a thin (eg 10 mm) ball screw you may (will) have problems at speed. It will flap around at times something awful.
You can solve this by going to a more solid ball screw - say 40+ mm. This will however be more expensive.
However, the funny thing I have noticed is that once you are running in production mode, the original cost seems to just disappear.

Single screw vs double screw: your call. A really solid carriage and a really solid single screw would work fine, but it would need to be Really Solid. You could get this with heavy linear bearings. Mach3 can handle double screws just fine.

Cheers
Roger

[Born2dive9702]

am looking to get speeds from say 10Ipm to 600 IPM while cutting.I was thinking on using a 1/2" acme with a bearing block on each end, of the table, then use 2 bearing blocks on each end of the gantry where the screw passes.I was thinkin of using a 1/2" acme but have been told to stay away from them on this length and speeds regardless of how many round guide rails are used in combination of V rails.I have been told that rack an pinion are more accurate, but are lot slower...What are your thoughts.

There is no question that proper ball screw & nut will be far more accurate than anything else. But, if it is a thin (eg 10 mm) ball screw you may (will) have problems at speed. It will flap around at times something awful.You can solve this by going to a more solid ball screw - say 40+ mm. This will however be more expensive.However, the funny thing I have noticed is that once you are running in production mode, the original cost seems to just disappear.Single screw vs double screw: your call. A really solid carriage and a really solid single screw would work fine, but it would need to be Really Solid. You could get this with heavy linear bearings. Mach3 can handle double screws just fine.CheersRoger

[RCaffin]

I think 1/2" anything would be a bad choice. I said 40 mm: that is more than 3x the diameter.
Acme - OK for a manual mill where backlash does not matter. Never any good on a CNC.

Cheer
Roger

[ger21]

For a screw, I would use no less than a 2525 ballscrew. (25mm dia., 25mm pitch)
A 1/2" acme screw won't be able to get anywhere near 600ipm, unless you used an 1/2-8 8 start screw, and even with that, it will whip like crazy.

I have been told that rack an pinion are more accurate, but are lot slower...What are your thoughts.

Actually, it's just the opposite.
When comparing acme to R&P, accuracy is similar, but R&P is much faster.
Good ballscrews are probably a little more accurate. But accuracy will really be a question of how much you spend. More accurate parts are more expensive. Much more expensive.
But realistically, accuracy should not even be a concern.

[ger21]

regardless of how many round guide rails are used in combination of V rails

I just saw your post about machining 1-1/2" aluminum for CF molds.

I would definitely go the 2525 ballscrew route, and forget about round rails and V rollers. You really need to use profile linear rails and guides.

[Born2dive9702]

That is what I was beginning to think was to use a 2525 ball screw, but how long can I go with dual screws (one each side of the table on the y axis) and at what speed can I run it. The pieces are 50" so I was thinking I would need to use screws that are about 60" Does this sound right??

What profile linear rails and guides do you suggest? I was thinking of getting 1" rounds for the Y axis (top and bottom of the screw Screw in the center ) and then running V rails on the deck to make sure that there was no movement on the deck.

Which profile linear rails and guides would you suggest?? You said the v rails would not be needed, but would it not help reduce the loads on the screws and the (if used round rails) as some of the weight would be on the decks??

I just saw your post about machining 1-1/2" aluminum for CF molds.

I would definitely go the 2525 ballscrew route, and forget about round rails and V rollers. You really need to use profile linear rails and guides.

[Born2dive9702]

Hey Ger
Would something like this suffice you think? it is a 2525 https://www.ebay.com/itm/2-set-HSR25...IAAOSwOA1Z6JcJ

I just saw your post about machining 1-1/2" aluminum for CF molds.

I would definitely go the 2525 ballscrew route, and forget about round rails and V rollers. You really need to use profile linear rails and guides.

[ger21]

600ipm would be about 600RPM. 2525 screws should be fine with that. If your parts are 50", you'll probably need screws that are 66"-70" long.

I would use Hiwin or THK 20mm rails.

[Born2dive9702]

So in your opinion the 2525, at 600 IPM would be fine with little or no whip, correct?
I also want to run it at very slow speeds for very fine polishing/ cutting of the molds, something like 5 -10 IPM so I assume that the 2525 ball screws will work for this too?

Concerning the profile rails and guides I am finding 2 types The ones you recommended and these https://www.ebay.com/itm/SBR25-2400m...UAAOxy1VlREP1v

What is the advantage/ disadvantage of the supported round rails vs the HiWin or the THk 20's

600ipm would be about 600RPM. 2525 screws should be fine with that. If your parts are 50", you'll probably need screws that are 66"-70" long.

I would use Hiwin or THK 20mm rails.

[ger21]

So in your opinion the 2525, at 600 IPM would be fine with little or no whip, correct?

I would think so, but I've never used them. There are a lot of factors involved. Most screw manufacturers have critical speed calculators.


Profile rails like the Hiwins are about 10x more rigid than supported round rails.
Other than some cheaper chinese machines, or machines for the hobbyist market, you'll never see round supported rails on a commercial machine.

[footpetaljones]

So in your opinion the 2525, at 600 IPM would be fine with little or no whip, correct?
I also want to run it at very slow speeds for very fine polishing/ cutting of the molds, something like 5 -10 IPM so I assume that the 2525 ball screws will work for this too?

A 25mm ballscrew with a 25mm pitch having one fixed end and one floating end will have a maximum speed of ~820 IPM.

However, your motor selection becomes very important with a high pitch screw. The speed will not be an issue, but the resolution might. For example, if you were using steppers set to half step, 400 steps per revolution would give you motion in increments down to .0025". I've never seen a hard rule of what resolution is required for a certain finish, but as a general guide I've seen many recommendations to keep your resolution a minimum of an order of magnitude below your finest step.

[Born2dive9702]

As far as cutting resolution, on the alumimum to make carbon fibre molds, I can get by with an error of .008" across a span of 43" I would like it less than that say more like .003" for the same 43"

I am thinking of copying Gabrials setup here
https://www.cnczone.com/forums/diy-c...294-cnc-5.html
I really like how he has the system, I would just make it longer using as he recommencs the 25mm ball screw and the 20-25mm profile rails.

Do you think this would be achievable??

I would think so, but I've never used them. There are a lot of factors involved. Most screw manufacturers have critical speed calculators.


Profile rails like the Hiwins are about 10x more rigid than supported round rails.
Other than some cheaper chinese machines, or machines for the hobbyist market, you'll never see round supported rails on a commercial machine.

[Born2dive9702]

Hello Foot,

For the work that I will be doing on the alumimum molds for the carbonfibre, I can live with an accuracy of between .003" and .008" over a span of 43-45"

The molds do not have to be ultra precision as they will be (more than likely) manually buffed to a mirror. and also there will be some variations in the thickness as the molds are warmed during the curing process for the carbon fibre parts. there is on average about a .01-.015" distortation from one end to the other from the heating. Up to 200'F

Hence why I was looking into both the profile system and the Side mounted round linear rails. I really like Gabrials setup seen here
https://www.cnczone.com/forums/diy-c...294-cnc-5.html

What are your thoughts.

A 25mm ballscrew with a 25mm pitch having one fixed end and one floating end will have a maximum speed of ~820 IPM.

However, your motor selection becomes very important with a high pitch screw. The speed will not be an issue, but the resolution might. For example, if you were using steppers set to half step, 400 steps per revolution would give you motion in increments down to .0025". I've never seen a hard rule of what resolution is required for a certain finish, but as a general guide I've seen many recommendations to keep your resolution a minimum of an order of magnitude below your finest step.

[ger21]

When using steppers, you always have to trade resolution for speed. This is because steppers lose torque rapidly as rpm's increase, and because they have a fixed resolution. You're max RPM will usually be between 600-1000 rpm. So, the higher the pith of the screws, the lower the resolution.
You might want to consider AC servos.
A 400W AC servo, geared 3:1, will have far more power than any steppers, and give you over 1000ipm rapid speeds, with much higher acceleration.
The downside is that they'll cost about 2-3x what a good stepper system costs.