I'm very close to having to make this decision. I am making a 4'X8' plasma that may be used from time to time as a router. So now I'm wondering, do I use ballscrews or rack and pinion??? I know there are advantages either way so is money the only real consideration at this point? Ballscrews seem easier and their not that expensive. What about racks, how do they compare in price, how difficult are they to get on the machine correctly??

Thanks for any advice you may have
Allen James

8' is too long for a ball screw, it can be done but the dia. would be up there at that length to retain rigidity and cure whip etc. and precision ones are very expensive, a proper tension pressured rack & pinion would be the way to go and much cheaper, but due to the large travel/motor rev. some kind of reduction is required, either planetary gearbox or timing belt/pulley.
There are several post in these forums with examples.
Al

I also have a similar question...

Would a belt drive work in lue of a Rack & Pinion?

I was wondering the same question as "Freak Brain" and had decided that a ball screw for the 8' movemnet (on a 4 x 8 table) would create too much whip. instead of a Rack, I wanted to use a gear reduced drive motor with a belt drive. To eliminate "strech" or slippage, I was going to include an eight foot linear Encoder (Yes I know they are expensive) instead of a rotary Encoder.


Thanks...

I have used belt drives up to 8ft you need to use the high quality reinforced non-stretch belts, but alignment, support and tension is critical. The rack & pinion is usually the accepted way today by commercial tables, especially for plasma where working to really fine tolerances are not crucial. This method usually involves a low backlash planetary gearbox.
I would hate to see the cost of a 8' linear encoder.
Al

A non-precision 1" ballscrew (I assume rolled, not ground) can be purchase for around $250-$275 area with two ball nuts. Actually by the time it's all said and done would most likely be around $300. Thats with two nuts without flanges, you could save $25 or so if you only needed one nut. So, I have to wonder how much a person would save by using the alternative methods, rack and pinion or belt. Any idea?

-Allen

1" dia seems a bit skimpy for that length, If I was using a ball screw that length I would think at least 2"dia would be in order, especially if you are using anything like 500"/min or up for rapid.
Al

Al,

I called Allen at Roton and he told me that a 1" dia. screw for my needs would be fine. I'm figuring no more than 350"/min. and 8ft of travel. on my machine. Do you think that's bad advice? I'm kind of at the mercy of everyone else as I'm not very knowledgeable on this subject.

-Allen

For a .25" lead screw, thats 1400rpm at 350"/min, I would have thought whip would be a problem, maybe someone here who has used a long ballscrew will jump in.
Al

Too long IMHO. I used 3/4" acme at 4" and it whiped at 1000 rpm.

Eic

Two possible solutions to ballscrew use are, either
use a fixed screw and rotating nut which makes screw
whip a complete non-issue or "borrow" Thermwood's
idea of spring loaded plastic dampers that press against
the screw at a couple of points and flip out of the
way when the carriage passes.

Kind of funny. I was setting at work tonight watching my machine run and I realized that my "X" axis screw doesn't rotate. I then remembered that I had talked about this before and I came up with the plan of belt driving the nut. (I'm sure I heard it here first)I like that idea better for "X" anyhow.

-Allen

Al,

I called Allen at Roton and he told me that a 1" dia. screw for my needs would be fine. I'm figuring no more than 350"/min. and 8ft of travel. on my machine. Do you think that's bad advice? I'm kind of at the mercy of everyone else as I'm not very knowledgeable on this subject.

-Allen

Hi, you might want to check out:

http://www.nookindustries.com/ball/BallCalculators.cfm

Especially look at the column strength calculator.

From what I can tell using a 1" diameter (~.8 root dia) screw is iffy in this situation no matter what type of end conditions or nut/screw rotation scheme you employ due to buckling danger. According to the calculator, shaft whip could also be a problem depending on what lead you choose.

So, if you are going to use a ball screw, you might consider a larger diameter one. Even the 1.5" dia ones might be pushing it if you go by the book, depending on how you fix the ends.

Another good resource for this sort of thing is the thomson ballscrew catalog (pgs 50&54 of the 2004 ed has charts that say the same thing as the nook calculators).

Good luck!

Mitchell

Thanks for all of your help

That changes things just a bit for me. Now I have to think about this just a bit more. Where is the best places to get information on rack and pinion setups? How do people drive from the center or drive both sides with a rack and pinion? Anyone have any pictures or websites that they would like to share?

-Allen

Did you do a search here first?http://cnczone.com/forums/showthread.php?t=6815&page=2&pp=8&highlight=rosta
Al

My setup (which I just got up and running) uses rack and pinion to drive a plasma torch. It hit 600 ipm before my slower computer started to have following error problems, so I don't really know what the upper speed limit on the machine really is. I have two 450 oz-in X-motors on the gantry, with one motor on each end. The steppers are driven using the same step and direction inputs off of seperate motor controllers. The Y and Z axes use on motor and rack and pinion as well.

The motors were a bit of a trick to get adjusted and in sync on the gear racks. Basically I mounted one side, energized the system, then aligned the second motor to the rack and finished tighening it into place. The motor torque holds the shaft still while you bolt it dwon. If the motors aren't aligned to the racks properly, the steppers can get stuck oscillating between gear teeth at tight spots on the racks, like mine did the first time I fired it up.

I don't have any pictures yet, but if you would like some let me know.

Sure! I would like to see pictures of your setup. I plan on slaving two servos with a rack on each side (end) of the gantry. Every pic helps.

Bloy

I too would love to see some pics. Thanks for all the info

-Allen

I just finished converting (was line follower) one with racks and steppers. I could get 1200ipm easily. We slowed it to 700ipm because 1200 was scary. 916 oz motors with Gecko's @ 65v. Gantry is ~90#. I used .875" gears & 4:1 reduction. Runs GREAT. MIKE

You use the gear box to give the motor a mechanical advantage over the weight of the axis, right? So an .857 gear dia. has a linear move of 2.750" per motor rev. if hooked direct. Divide that by 4 (4:1 reduction) gives you .687 (call it 5/8")linear move per rotation of the motor. That is actually fast compared to a ballscrew. 1/4 lead on a screw gives you 1/4" linear travel per rotation of motor.

Am I understanding the information correctly?

Thanks again
-Allen

Ok, I'll get some pictures taken and post them tonight if I remember.

My setup uses 1.5" pitch diameter gears direct drive, so I get 4.71 inches of motion per revolution of the 1000 counts/rev steppers. That gives me .0047"/step which is more than sufficient for the art work I intend on cutting. The acceleration is set slower than it would be with gear reduction, but then I'm only cutting at about 100-200 ipm with the torch anyway so it doesn't matter much.

My steppers are 434 oz-in running at about 65 vdc off of Gecco drives also. One of the timing settings in my software must be the culprit for the following errors. I'll have to fix that, then test the velocity and acceleration to see how high I can crank it up. I'm using EMC on a Linux system because I like the windowing it gives me along with the real-time operating system. DOS and TurboCNC won't do that, at least not on an older PC anyway.

Thor

Here are the pictures, mostly of the gantry with a pen holder on the tool mount. Everything moves on sealed roller bearings obtained at the local army surplus store, except the Z axis which uses a linear slide. All of the motors, racks, and pinions are identical. The gantry is a 4" square steel tube with a 1"x1" cold-rolled steel bar on the bottom and the rack on top. The right side rail has the 1"x1" bar on top, with the rack welded onto the inner face, gear teeth down to keep crud out. I only have about 1" clearance below the motors, but I will only cut up to 5/8" stock with the torch anyway. I may move the X motors up highher later and use a belt drive to couple it.

You use the gear box to give the motor a mechanical advantage over the weight of the axis, right? So an .857 gear dia. has a linear move of 2.750" per motor rev. if hooked direct. Divide that by 4 (4:1 reduction) gives you .687 (call it 5/8")linear move per rotation of the motor. That is actually fast compared to a ballscrew. 1/4 lead on a screw gives you 1/4" linear travel per rotation of motor.

Am I understanding the information correctly?

Thanks again
-Allen

Yes. My own machine uses 4-turn screws. 1/4" per turn. The rack/gear machine moves ~.687" per turn. I am only using 187oz motors to move my 65# gantry 90ipm (425rpm). The rack/gear machine has 916oz motors and made 1750 rpm. I could get it to accellerate to 1200ipm in less than 2", but it was hard to watch without taking cover. We settled on 700ipm in 2". We have been cutting in the 60-90ipm range, 700 ipm rapids are fun to watch though. Good luck, MIKE

Thanks for the pics and all the added info. Also thanks to you Mike, both of you have been a big help. I guess I need to go with the rack and pinion...I priced the racks today for a 1/2" wide rack it's around $12 foot, is that about what you guys paid? Thats from boston gear distubutor

-Allen

I paid a little less, about $62 for 6 ft from McMaster Carr (www.mcmaster.com) and search for "gear rack".

My table has a 4 ft x 5 ft work area, so the 6 footers worked well on mine if that helps any.

Thor

Thanks for the pics, Javelin. There's a lot to look at in those!

Bloy

Hello Guys, several questions regarding your transmision too, I want to use direct shaft on a 4x8 gantry, with double X motorization;

What pitch and gear angle are you using? I was thinking of pitch 24, 14.5º.

What number of teeth are you using on the pinnon? I was thinking of 14-15 teeth.

Hi There,
I used 16 pitch 24 tooth pinions, 14.5 deg.

Without a gearbox, I can only use it to move a torch where the torque isn't required. We have a second table under construction where we are using gearboxes, so torque won't be a issue so we can use a router. My motors are frame 34 by the way.
Thor

Thank You Javelin;

I need an extra help, I looked at the page you reccomended www.mcmaster.com, and they have what I need, but I want 8' racks, did you had any experience joining their rack sections?

Anyone that can help me:

I've got another question, do you know the standard pitch and degree on machine torchs, I don't know how to calculate it and just received a new oxy-a victor without any instructions.


Regards
Antonio

Thank You Javelin;

I need an extra help, I looked at the page you reccomended www.mcmaster.com, and they have what I need, but I want 8' racks, did you had any experience joining their rack sections?



Use a spare piece of rack teeth to teeth to line up the joints.

Use a spare piece of rack teeth to teeth to line up the joints.

Also, unless the rack is precision ground at the ends, i.e. if its random cut, the easiest way to trim the ends is to cut off a fraction short of the last valley on each end, this leaves a small gap at the bottom of the valley when mounted, which does not affect the operation due to the gear tooth never bottoming out.
Al.

Excellent tips on getting the racks extended. Thanks guys, I need that tip too. I'm building a second table which is about 10 feet long and I hadn't figured that part out yet.

I'm sorry, but I don't know the oxy-acetylene question.
Thor

Hello guys again a question from the newbie, (Havent told you I'm in Mexico, and havent found anyone else that's in a project like these that could help me)

Several questions:

My controller has two internal opto relays, I find risky to turn the plasma on/of with them but I want to do the activation remotely. I know I can slave biger relays to these ones, but I don't know what type of relays to use. Can you help me please.

I'll keep the other questions for another thread.
Regards

Allen,
Are you looking for speed for repetitive production, or are you looking to do some one off parts. If your machine is going to make you money, you make a lot of it in rapid traverse period. This is the beauty of ball screws, they can handle it in the right enviroment but will get expensive depending on the application. If you are doing some one off parts from time to time, a rack will be cheaper to maintain if you don't punish it. Racks don't seem to stay tight & accurate with high speed linear changes over time with out adjustment where a ball screw will. Tom