Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
DRO encoders- Rotary or Linear?
|
Roger Linscheid
Hi, I am ready to begin adding a DRO to my G4003 lathe. What is the
current thought on the best method to accomplish my goal? IIRC, when I was looking into this before, the linear strips from U.S. Digital were only available in 250 CPI and this was not considered high enough to provide the necessary resolution. Using rotary encoders meant the possibility of losing counts if the drive cable slipped or stretched. Now U.S. Digital has linear strips in 300 and 360 CPI. Will these provide a resolution of 0.0005 or better? I have a completed DRO board from Camtronics and several HEDS on hand, so I just need to scrounge an old PC and start the installation. Thanks for any input! |
|
alex
This may be a stupid question, but why not to use a normal bicycle
toggle quoted message
Show quoted text
chain\sprocket combination for driving an encoder? Is it a backlash issue? Alex ----- Original Message -----
From: jcc3inc <jccinc@...> To: <CAD_CAM_EDM_DRO@...> Sent: Tuesday, October 29, 2002 9:44 AM Subject: [CAD_CAM_EDM_DRO] Re: DRO encoders- Rotary or Linear? --- In CAD_CAM_EDM_DRO@y..., "Roger Linscheid" <r.linscheid@a...>aol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if you have trouble. sister site to the CCED group, as many of the same members are there, for OT subjects, that are not allowed on the CCED list. DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........ bill |
|
--- In CAD_CAM_EDM_DRO@y..., "Roger Linscheid" <r.linscheid@a...>
wrote: Hi, I am ready to begin adding a DRO to my G4003 lathe. What isthe current thought on the best method to accomplish my goal?Roger: Since you seem interested in accuracy, why not use rack and pinion for rotary encoder drive. A 1/10 inch circular pitch rack with a 20 tooth pinion requires a 500 count encoder to read .001". Precision rack is available from PIC Instruments, Berg, and Allied Devices. They can also provide you the pinions. Less expensive rack is available elsewhere. Linear encoders eliminate backlash and encoder mounting errors and are probably the better of the two approaches. A 250 line scale will give you (with 4x multiplication) a .001" resolution. A 360 line scale yields 1440 steps per inch or .6944 mils per step. Good luck, Jack C. |
|
Chain and sprockets are non-linear.
At 09:34 AM 10/29/02 -0500, you wrote: This may be a stupid question, but why not to use a normal bicycleaol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if you have trouble. sister site to the CCED group, as many of the same members are there, for OT subjects, that are not allowed on the CCED list. DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........ billRegards, Hoyt McKagen To prevent virus propogation, don't put this addy in your book Belfab CNC - US Best MC - Camping/Caving - Two-Wheel-Tech List - If you torment a serpent, it will cough up a dead rat |
|
John A. McFadden
Instead of using a rack and pinion, would it be possible to use a rigid bar
(like 1" or 1.5" drill rod) and a rubber roller mounted to a rotary encoder? With a solid bar, wipes could be used to clear the surface before the roller passed over it, unlike the rack and pinion where clearing it would be a difficult measure at best if something were to get in the teeth. John |
|
Raymond Heckert
Alex, It's not merely a backlash issue. (Although, a
properly designed roller chain transmission *must* have some slack built in. As any kind of roller chain meets the sprocket, and changes from linear to rotary motion, the chain jumps up, and down at each pitch. This is more pronounced with low tooth numbers, and double pitch chain. Immediately, upon full engagement with the tooth, the linear section of the chain begins to swing downward, as the disappearing tooth 's rotary motion moves it downward. When the next tooth engages, the chain is brought up to tangency with the pitch circle again. This is only minute movement, but it sets up a quite a vibration. Timing chain is about the only chain that doesn't exhibit this quirk, which is why they're called timing chains, because that's what they're used for. Virtually vibration free, and they take up their own slack by climbing to the highest available form on the tooth. RayHex ---------- From: alex <telecomt@...>normal bicycle chain\sprocket combination for driving an encoder? |
|
Raymond Heckert
Check out the "Shooting Star" DRO's. They use a
(plastic-loom, covered) rack, driving a pinion, coupled to an encoder, and are very reliably, and pretty accurate, as long as you mount them properly. RayHex ---------- From: John A. McFadden <mcfadden@...>to use a rigid bar (like 1" or 1.5" drill rod) and a rubber roller mountedto a rotary encoder? With a solid bar, wipes could be used to clear thesurface before the roller passed over it, unlike the rack and pinion where clearingit would be a difficult measure at best if something were to get in theteeth. |
|
Matt Shaver
On Tuesday 29 October 2002 06:39 pm, you wrote:
Instead of using a rack and pinion, would it be possible to use a rigid barThis is essentially what a Trav-a-Dial does, except the roller is steel, not rubber. The company that makes them, Southwestern Industries, also makes a CNC control called a ProtoTrak which uses position sensors based on the same principle. See: Matt |
|
alex wrote:
This may be a stupid question, but why not to use a normal bicycleSprockets and chains, unless very carefully designed have an inherent cogging effect that causes cyclical errors every tooth, and then the chains have variations in pitch from roller to roller, which varies as the chain ages, and the rollers and pins wear. There may be instrument-grade chains, but normal bicycle chain, besides being heavy enough to break an encoder, are certainly not manufactured to the tolerances needed. W. M. Berg has all sorts of miniature instrument-type cable-chains that use plastic molded over aircraft cable. These are not instrument-grade accuracy, but might work better than the bicycle chain. I use them in my mill, but only for tachometer drive, not the encoders. Jon |
|
John A. McFadden wrote:
Instead of using a rack and pinion, would it be possible to use a rigid barYup, this is basically a trav-a-dial. I think they use a brass or bronze roller under a fair amount of pressure. They just use the mill table or lathe ways as the 'track'. A rubber roller won't work, unless it is extremely hard, as the pressure will cause a flat spot to develop as it rolls, and you won't get a consistant turns/inch. Also, when sitting for a while it would get a more permanent flat spot. Jon |
|
alex
OK, thank you.
toggle quoted message
Show quoted text
----- Original Message -----
From: Raymond Heckert <jnr@...> To: <CAD_CAM_EDM_DRO@...> Sent: Tuesday, October 29, 2002 11:18 PM Subject: Re: [CAD_CAM_EDM_DRO] Re: DRO encoders- Rotary or Linear? Alex, It's not merely a backlash issue. (Although, aaol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if you have trouble. sister site to the CCED group, as many of the same members are there, for OT subjects, that are not allowed on the CCED list. DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........ bill |
|
alex
Thank you.
toggle quoted message
Show quoted text
Alex ----- Original Message -----
From: Jon Elson <elson@...> To: <CAD_CAM_EDM_DRO@...> Sent: Wednesday, October 30, 2002 2:09 AM Subject: Re: [CAD_CAM_EDM_DRO] Re: DRO encoders- Rotary or Linear? aol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if you have trouble. sister site to the CCED group, as many of the same members are there, for OT subjects, that are not allowed on the CCED list. DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........ bill |
|
John A. McFadden
Matt:
Thats probably the system my dad saw at Lockheed (one of his friends works for Lockheed in a machine shop and my dad asked him for advice on my DRO idea), it was a brand new CNC mill, just delivered the week before. I guess the problem would be putting the load on the roller, it doesnt seem like it would be easy to do. John |
|
While the linear encoders may be better for the purposes of this list I
toggle quoted message
Show quoted text
think the question is really: 1 - Is my machine such that I would notice any difference? 2 - Even if I could notice a difference is the cost/benefit at a point that makes sense for me. I run my Bridgett my Bridgeport series II on Gecko G320 drives with rotary encoders on the motors. The accuracy and precision with the rotary encoders is more than adequate for my purposes. It will easily hold to less than .001" and I can sell all the stuff I produce on the machine. Spending the money to upgrade to linear strips would buy me nothing as the items I am producing would not become any more valuable and no one would pay more money for then. It is all a balancing act to get the accuracy and precision needed and not waste money getting more. Also remember what the rest of your machine is capable of. Putting on super high resolution linear encoders on a mill drill conversion is probably a waste as the rest of the machine will have enough play and flex that you will not likely see an improvement. Also remember that setup of the part can have a huge impact on machining precision. A weak setup will introduce more inaccuracy than the difference between linear and rotary encoders. Ditto with thermal expansion. So, if you are not using flood with the coolant being held to a relatively constant temperature than you will also have inaccuracy. In summary, my point is that you have to consider the whole and not just focus on one part. Like so many things in this hobby it is easy to get into a pissing match where the claims of what is required no longer fit the reality of most machining much less hobby machining. So if you really feel your setups will be rigid enough to allow machining to .0001" and you will control the temperature, and your machine is capable of moving that small an increment, than maybe the linear encoders will make sense. Tim [Denver, CO] ----- Original Message -----
aol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if you have trouble. sister site to the CCED group, as many of the same members are there, for OT subjects, that are not allowed on the CCED list. DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........ |
|
gittt2000 wrote:
Linear - without a shadow of doubt. Here's why :Well, yes, in $100,000 CNC machining centers, no argument! Notice, though, that the graphs are in uM units, the ballbar plots are 5 uM/division! To be of much use, the linear scales must have resolution on that scale, or these errors would lie undetected underneath the basic resolution of the scale. What is the standard resolution of typical DRO scales? .0005" ~= 12.7 uM if I computed it right. So, the backlash errors they showed would be barely one quantum on the typical DRO scale. Linear errors due to screw expansion depend on how hard you run the machine. Note the thermograph of the hot screw was run continuously for 6 hours at 24 M/Min. That equals 945 IPM. Are you planning to keep your machine running at almost 1000 IPM for 6 hours? If I tried to move my Bridgeport at 1000 IPM, if would hurl itself over so hard it would break the floor slab! So, while the things stated in the Heidenhain article are real concerns for makers of dedicated, HIGH-precision CNC machining centers, and more important as size and speed increase, they may not be so significant for small machines. If you price linear scales with sub-micron resolution, you'll find them pretty expensive. Jon |
to navigate to use esc to dismiss