DRO Accuracy

Topics include, Machine Tools & Tooling, Precision Measuring, Materials and their Properties, Electrical discussions related to machine tools, setups, fixtures and jigs and other general discussion related to amateur machining.

Moderators: Harold_V, websterz, GlennW

User avatar
whisperfan
Posts: 240
Joined: Sun Oct 25, 2009 5:49 pm
Location: Virginia, USA

DRO Accuracy

Post by whisperfan » Sat Dec 19, 2009 10:40 pm

I am installing a used DRO system on an older mill

My question concerns the accuracy of the DRO itself. It reads .0005"

So - the Y-Axis is installed. The scale is parallel with the ways (as accurate as I can get it). Due to it being a used scale, it has small imperfections in the surface that you would expect with a scale that has been on a machine before.

That being said, the scale seems good. It is tightly mounted and I fabricated the mount for the reader head. The reader moves smoothly and it seems to indicate correctly on the display.

So - my question is this ... Is there a way to test to make sure that the scale is reading accurate measurements? I have turned the feed handle in a direction (taking out any backlash) and then I zero the display and zero the indicating ring on the feed handle, and I turn a certain amount and then look and compare the two. They seem to be right, but I sure would like a way to really test to see if it is reading as accurate as it can.

Any suggestions?

User avatar
mechanicalmagic
Posts: 1431
Joined: Sun Nov 04, 2007 12:11 am
Location: Pleasanton, CA Land of perfect weather

Post by mechanicalmagic » Sat Dec 19, 2009 11:15 pm

Zero a good DTI on one vise jaw. Lightly clamp a 1-2-3 block in the vise, check the length. Insert another 1-2-3 block check again.

Approach from both directions for your readings. A loose gib or drag can result in hysteresis.

Dave J.
Every day I ask myself, "What's the most fun thing to do today."
9x48 BP clone, 12x36 lathe, TIG, MIG, Gas, 3 in 1 sheetmetal.

User avatar
whisperfan
Posts: 240
Joined: Sun Oct 25, 2009 5:49 pm
Location: Virginia, USA

Post by whisperfan » Sun Dec 20, 2009 6:41 pm

Good suggestion - thanks

User avatar
GlennW
Posts: 6895
Joined: Sun Apr 15, 2007 9:23 am
Location: Florida

Post by GlennW » Sun Dec 20, 2009 7:50 pm

I check this way, which is similar to Dave's method, but with a little less room for error 'cause I'm anal.

Clamp a 123 block to the table and square it up to the axis of travel. Then place or wring a gage Block (or stack of) against it.

Image

Move up to the Gage Block from about an inch away and touch it to zero the DTI. Zero the handwheel dial or DRO. Now raise the quill to clear it and then move to the 123 block and touch it until the needle zero's. Your DRO should match the Gage Block length.

Image

This way nothing moves or backs up, and the leadscrew is always pre-loaded in the proper direction. If the spindle is square to the table the slight Z movement is irrelevant.
Glenn

Operating machines is perfectly safe......until you forget how dangerous it really is!

User avatar
Bill Shields
Posts: 5996
Joined: Fri Dec 21, 2007 4:57 am
Location: Somewhere in the World
Contact:

DRO accuracy

Post by Bill Shields » Tue Dec 22, 2009 5:46 pm

Most DROs that I have worked with that are glass scales have one of two features built in

1> if they skip a 'pulse' the fail and flash 0000
2> if they skip a pulse, they assume it was just one, add it and keep on going. If more than one is missed in a given distance, they flash 0000

You can get similar problems if the 'reader' isn't aligned properly.

Basically, if a glass scale is not cracked and appears to be reading, it will work with factory accuracy, assuming the pulse count / inch is set correctly in the reader head.

Don't assume that it is set correctly. 3 years ago I purchased a used mill with a DRO on it and the DIP switch was set incorrectly and the DRO was reading off by quite a bit (pulse count / inch in the head didn't match the scale).

It turned out that it had been set wrong for years and the fellow that owned it before me never caught it - said he only used it to 'return to home'...OK...

Another good test is to run the table as fast as you can, one end to the other and check that you don't get an error. Sometimes a funky scale will read OK at slow speeds, but fail at higher (like with a motorized feed on rapid).

lakeside53
Posts: 809
Joined: Tue Nov 06, 2007 12:44 am
Location: Woodinville, WA

Post by lakeside53 » Tue Dec 22, 2009 6:05 pm

..also... try to set the [used] DRO to factory default. If it had linear error compensation programmed in, you want to redo this for your installation.

User avatar
Bill Shields
Posts: 5996
Joined: Fri Dec 21, 2007 4:57 am
Location: Somewhere in the World
Contact:

compensation

Post by Bill Shields » Tue Dec 22, 2009 6:10 pm

Wouldn't you normally find compensation error associated with an encoder / resolver mounted on a screw?

I have never seen / used / known of a glass scale that needed that (never looked either, so that shows my ignorance if it's there).

CarlD
Posts: 504
Joined: Mon Jun 18, 2007 10:33 am
Location: Taylorsville Ky USA

Post by CarlD » Tue Dec 22, 2009 8:01 pm

You can check the DRO over 12" travel with a 12" mic standard, a vise and an edge finder.

I think you can figure out how to use them to check if the DRO is traveling the correct distance.
It's only ink and paper.

gmann109
Posts: 399
Joined: Tue Jun 20, 2006 8:39 pm

Post by gmann109 » Wed Dec 23, 2009 11:54 am

I installed a 2-axis Mitutoyo DRO on my Mill shortly after I got it up and running. I don't know how I got by without one. It is extremely accurate. One day I may test it but for now, it's a lot better than my eyeball or even the dials on the machine.

I'm planning on adding a 2-axis DRO on my Enco 13 X 40 lathe after the first of the year. I'll probably get one from CDCO or some similar seller. They can be bought for around $400.

I believe that they are all rather accurate. I've not heard any complaints of inaccuracy about DRO's. What you hear mainly is when they fail.

lakeside53
Posts: 809
Joined: Tue Nov 06, 2007 12:44 am
Location: Woodinville, WA

Post by lakeside53 » Wed Dec 23, 2009 1:45 pm

I wish I could put a cheaper DRO on my lathe, but physical mounting issue on the cross slide are pushing me to the most expensive types - right now the Newal C80... but I "might" be able to use a cheaper dro/scale for the Z axis and use a high end minature galss scale (Acu-rite SENC 50) for the cross slide.

Funny thing.. last week I started a similar thread on that other board on the same topic. I didn't have much to add, but some of the others posters made for interesting reading. It degenetrated in the old arguement - how much accuracy do you really need... and when?

http://bbs.homeshopmachinist.net/showthread.php?t=38519

User avatar
Bill Shields
Posts: 5996
Joined: Fri Dec 21, 2007 4:57 am
Location: Somewhere in the World
Contact:

dro accuracy

Post by Bill Shields » Wed Dec 23, 2009 1:53 pm

a small shooting star might work

lakeside53
Posts: 809
Joined: Tue Nov 06, 2007 12:44 am
Location: Woodinville, WA

Post by lakeside53 » Wed Dec 23, 2009 2:02 pm

I need a max of 39-40mm top of shield to bottom of reader. The CBX reader might sneak in... if mounted at the back over the taper attachment, but it's not sealed and look like a dirt trap.

Even after reading the contents of the other site thread, the BT series encoding method make me nervous - basically a digital readout from a lead screw. If it as precise as the lathe screws, it should be as good as dials, but... I see they quote worst case accuracy of 2 thou per foot (not resolution)... I'd like to think my lathe is better (ha) but.. what do i know.. right now I have a dial on the Y axis that MIGHT resolve 5 thou IF I can count a zillion revolutions correctly, and if my rack isn't beaten by 30 years of swarf... :cry:

Postives though : likely robust, better than a Stanley tape (kidding...)


Anyone here used the BT series?
Last edited by lakeside53 on Wed Dec 23, 2009 6:45 pm, edited 3 times in total.

Post Reply