![]() | |
| Home Page | Mark Forums Read | Today's Posts | My Replies | Classifieds | Reviews | Photo Gallery | Web Links | Share Files | Advertise With Us | Ad List |
| |||||||
| Linear and Rotary Motion Discuss ball/Acme screws, R&P, linear slides and theory here. |
| This forum is sponsored by: |
![]() |
| | LinkBack | Thread Tools | Search this Thread | Display Modes |
|
#1
| |||
| |||
I'm CNC'ing a new 14" lathe. I've decided on 5/8" ballscrews, .2" pitch, and anti-backlash double nut setups. I have a collection of servos, mostly in the 300-400 watt area, 75v, 5-7A, 2500-3000 top RPM. A lot of them Sanyo-Denki. So I'm at the point of deciding pulley ratios, encoder count, belt size, etc. My off hand, wild ass guess, is to use 2 to 1 reduction with 500 (2000) count encoders. I've used that before and it works. But since in this case I must buy encoders and buy or make pulleys, I'm wondering if someone with some engineering smarts would agree with this setup. I really have no idea what forces are involved in moving the axis of a lathe while cutting and no idea how much power or force these motors have when moving the small distances involved in most CNC turning. Also is the XL belt size sufficient. If it makes a difference I'll be using Geckos and Mach3. I cut steel, aluminum, mostly small parts up to 2-3" diameter, up to 8" long, but I might like to turn a pool cue, wood. With small metal parts, big time rapids are not so important and if I decide to cut wood I could change pulleys. Also at this point it's a hobby usage machine. Comments requested. Ozzie |
|
#2
| ||||
| ||||
| Hi Ozzie, I just went through the same engineering excersize for my new router/milling machine. I'm using 900oz/in servos and according to my calc's, XL size belt is sufficient for the x,y,z axis'. I used H sized for the spindle just becase I had the room, and it will do the most work. If you are worried about it, go to H or L sized pulleys and belts, they don't cost much more, but can take much more space, especially if you want to have a gear reduction through them. The Martin website has some info on sizing, but it's easiest if you actually get a Martin or Dodge catalog (lots of page flipping), they have all the info you need to design a drive system. If you ever want to have fast rapids, I would suggest that you use a lower CPR encoder, maybe 250 or less. Of course this will have an impact on your resolution, but will probably be negligible. Paul |
|
#4
| ||||
| ||||
| Torque is a function of distance. Drive systems are usually designed by power, which dosen't rely on lever arms. Torque is a force at a distance, and can be measured with a scale and lever arm off the motor. You already know the power capability of your motors. This makes it easy to design the drive system. You have 1/2 horsepower motors. You need to pick pulleys and belts that can handle 1/2 HP * your service factor, which should be 1 for moving an axis. Martin Sprocket and Gear - You wil have to google this, I can't find their real webpage, but I know that it exists. Rockwell Automation - Dodge Gearing :http://www.dodge-pt.com/products/pt_...ts_drives.html - no info here really. SDP/SI uses the same conventions as the above two. XL,L,H belts are the same across the board, and so are the pulleys. http://www.sdp-si.com/ I got frustrated trying to pick pulleys that met a well defined set of parameters on their website, that's why I ordered my components from our local Bearing and Transmission and used the paper catalogs, it took several hours after I figured out where I was.... I got Browning pulleys and GoodYear belts. Nice stuff. Paul |
|
#5
| |||
| |||
| Thanks Paul, The site is www.martinsprocket.com . I'm pretty sure I want to end up with 20,000 steps per inch so I can figure the encoders once I know the ratio I want to use. The other possibility is put encoders on the end of the screws; probably a more accurate setup anyway but a lot more work. Regards, Ozzie |
| Sponsored Links |
|
#7
| |||
| |||
| If you're using servos, putting the encoder on the ballscrew is a bad idea. This will result in a very low system stiffness or you'll get some nasty oscillations at a high stiffness. As for pulley sizing, I am using tiny little Nema23 servos with a 3:1 reduction and a 10mm lead 25mm diameter Hiwin ballscrew. Maximum axis thrust was calculated to something like 900lbs with a continuous thrust of over 300lbs. This will be going on a 12" lathe and I think these values are more than enough for what I will be doing. I'm only using 2mm GT-2 belts which are very small but have very low backlash. Belt loading is only 12lb at 300lb thrust so I don't think this will be a problem (although pushing it). Also remember that a well tensioned belt spreads stiffness across both sides of the belt, so the actual belt stretch at 12lb belt load is equal to 6lb per side. One side of the belt relaxes by "x" while the other side stretches by "x" so they share the load. Anyway, you can see below how far I've gotten. Should have it finished sometime next month |
![]() |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | Search this Thread |
| Display Modes | |
| |
Similar Threads | ||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Gripper Finger Forces | ANKU | Mechanical Calculations/Engineering Design | 12 | 11-18-2007 06:39 PM |
| cutting forces | struthersdou | WoodWorking | 1 | 10-10-2007 06:23 PM |
| cutting forces milling | guido | General Metalwork Discussion | 28 | 01-24-2006 11:48 AM |
| Pinion Gear forces | JerryFlyGuy | Mechanical Calculations/Engineering Design | 6 | 01-18-2006 06:49 PM |
| thrust/feed forces | pixbroker | General Metal Working Machines | 4 | 12-27-2005 11:59 PM |