PDA

View Full Version : DRO for B-axis?



Jaakko Fagerlund
11-24-2012, 04:49 AM
We have a large and old TOS W100A horizontal boring mill at work that has a B-axis, meaning the table rotates. However, it only has degree markings under the table and it surely would be lovely to get a DRO for it, as the XYZ already have Heidenhains on them.

The table is about 1.25 meters in diameter, so that kind of makes me wonder how on earth would you install a DRO on it and how accurate would you get with it?

Usually the table is held square to the other moves but some occasional work pours in that requires turning the table 45, 90 or some other angle and it would be convinient to have a DRO for that.

Also, does the rotational DRO show the angle in degrees or arc minutes/seconds?

uncle pete
11-24-2012, 05:28 AM
Jaakko,
Newall has a stand alone display for doing this, If I recall it can display either in decimal degrees or degrees/minutes/seconds. While they build the DRO brain box and display to do it, they don't make or sell the rotary encoders to make the system fully work. I'm real limited on my electronics knowledge, so I've yet to fully figure out just what's exactly required for the rotary encoder. It would be real handy if it could maintain an accuracy of say 30 seconds. It would sure speed up divisions and table movements exactly like a standard mill DRO does. Gear cutting would be a breeze without the division plates and it being fairly easy to miscount and make a mistake. Plus the rotary encoder would delete any lead / lag errors in the dividing heads or rotary tables worms or worm wheels. I've been checking around for further information about these set ups for the last couple of years. So far I've found about zero on any of the forums including PM, although I have to admit it's been quite awhile since I last did a proper search.

Hopefully you'll get some decent information that I just don't know enough to provide. I'll be watching this thread real closely myself.

Pete

EVguru
11-24-2012, 06:41 AM
http://www.machine-dro.co.uk/digital-readout-systems/rotary-encoders-and-display-consoles.html

oldtiffie
11-24-2012, 06:51 AM
This encoder which seems tio be the best of them has an accuray of +/- 0.05 degrees which is about 1 in 1,146


The ANRE-SF-1800 encoder is compatible with the ANDC rotary display consoles. The encoder is supplied fitted with a 3 metre armoured cable which will plug directly into these consoles. Only a basic parameter needs to be entered to configure the console. This allows direct angle measurement reading to be displayed, in either decimal degrees or degrees, minutes and seconds.

The ANRE-SF-1800 has an angle resolution of 0.05 degree, when used with the ANDC rotary consoles.

http://www.machine-dro.co.uk/1800-line-count-optical-rotary-encoder-with-quadrature-output-face-mount-type-.html

Jaakko Fagerlund
11-24-2012, 01:51 PM
Thanks for the suggestions, I'll check them out. The biggest problem is probably of where to mount the sensor, as the table is not much accessible in regards of its axel and I think that even a 0.05 degree precision is not enough resolution if direct coupled to the table, as it would mean 0.5 mm of movement without the sensor noticing. So that leaves me only one solution, putting some sort of gear or tensioned belt to multiply movement.


Gear cutting would be a breeze without the division plates and it being fairly easy to miscount and make a mistake. Plus the rotary encoder would delete any lead / lag errors in the dividing heads or rotary tables worms or worm wheels. I've been checking around for further information about these set ups for the last couple of years. So far I've found about zero on any of the forums including PM, although I have to admit it's been quite awhile since I last did a proper search.
Pete, have you checked Division Master? It is exactly what you would need.

Black_Moons
11-24-2012, 03:04 PM
Id wonder about maybe cutting teeth into the rotary table (Or mounting a huge ring gear to it?) then mate a rotary encoder to that.

Or maybe mount a HUGE toothbelt to it. Actualy come to think of it, if your really crazy, Mount a toothbelt to it, then mount a linear DRO encoder to the toothbelt. Assuming you can live with 1 rotation max.

MaxHeadRoom
11-24-2012, 03:34 PM
I have done this type of thing in the past with a up-down Quadrature/Indicator/Totalizer, the ones I used were made by Durant.
http://downloads.eatoncanada.ca/downloads/Specific%20Control%20and%20Display%20%28Durant%29/Instructions/53300-903-04%20-%20Courier%20Model%2053300403%2053302403%20User%20 Manual.pdf
They have more sophisticated models also.
All you need is to mount a small quadrature encoder on either the driven shaft or somewhere off the rotating table.
These can be scaled to suit the encoder, you just need a small encoder that will give you the finest resolution you want, say in degrees?
Something like a Koyo or similar?
Max.

EVguru
11-24-2012, 03:35 PM
I almost forgot about BW Electronics. They have a cable based system that can easily be swapped from machine to machine.

I believe you can wrap the cable around a rotary table and calibrate it to read degrees.

http://www.bwelectronics.co.uk

Black_Moons
11-24-2012, 03:58 PM
Come to think of it, if its a meter in diamter, Considered printing off a G code or similar encoded strip and mounting some IR emiter/detector pairs?
I bet you could get some decent encoding resolution with just a low res printed strip and off the shelf IR pairs.

uncle pete
11-25-2012, 07:27 AM
Hi Jaakko,
Many thanks, but yeah I've known about the Division Master for almost as long as they've been around. While it's obviously digital and a read out, it's more of a power driven semi sophisticated counter that subdivides fairly accurately than what I'd term as a proper machine tool type DRO. All of it and the actual way it operates would be just fine, and it would in fact work really well "if" you've got a perfectly machined and ground worm and worm wheel. Vertex rotary tables and dividing heads are............mostly ok I guess from a HSM perspective. But let's face it, their certainly nothing the Moore Tool Group or a hell of a lot of other real high quality tooling suppliers and builders need to really worry about. In fact I think that those included specification sheets Vertex throws in the box with their equipment for entertainment purposes aren't worth the rice paper those totally bogus claims are printed on. But it's what I can or I am at least willing to spend for equipment like this.

I am real interested in the Newall unit because it totally removes any of the wear or built in inaccuracy's from the display in those worms and worm wheels, and exactly like any half decent DRO would on any other machine tools lead and feed screws. You then end up with only the + - inaccuracies that the encoder itself has. And those can be bought to fit your specifications. I've got both a Vertex R/T and one of their universal dividing heads, I'll probably be adding a larger R/T in the future. If it's possible? Adding a good accurate rotary encoder to each of them internally, and a set of jacks on the exterior so that Newall unit could be plugged in and changed between each one would be a great set up for me. I was in a fairly large machine shop a few years ago having some work done, they had the Newall system on one of the built in R/T's on a maybe meter? meter and a half? somewhat large in HSM terms anyway vertical boring machine. I really didn't have the chance to check it out since I wasn't supposed to even be inside the working part of the shop, and the guy who installed it wasn't available at that time, but I did ask them how they liked it. "We wouldn't be without it now, far faster, less mistakes, and much more accurate in my opinion". I'd say their set up was very much like what your thinking of doing. I do know there's a way to do it since they obviously managed it. So it's possible to mount the rotary encoders internally and around the tables drive spindle. Again it would depend on the exact layout on the internals for the equipment your using. And any tooling that had a sump full of oil for lubrication would obviously add to the complexity of the job.

Right now I just don't know nearly enough about what I need to know.

Pete