Advice/comments on DRO

Resent via a different ISP :( Obviously wanadoo are killing legal emails :(

My take on DRO's given what experience I have had with them.

First off I always thought DRO stood for Damn Rules Outside

First one I got was the BW cable type to fit my CVA lathe in two axis and just one axis on the big TOS.

At that time he was advertising the accuracy at 0.001" but I believe it's now up to 0.002" On a lathe this doubles because of the radius error so worst case scenario is

0.004" on diameter, hardly a decent bearing fit but in all fairness that original advertised 1 thou is still nearer the point. As I have said accuracy is variable but once set repeatability is good. Good points is they can be fitted to machines needing low profile as you can stick the sensor box out the back and run the cable alongside the feed screw. Advertised as being interchangeable I found the clip in clips didn't hold tight enough for repeatability and finished up with clamps but they can still be moved with not a lot of fuss. At one time they were very cost effective with proper DRO's costing around a grand and £330 for two axis was a lot better but as the professional scales and systems have come down these have stayed the same.

Reading is limited to three digits and a light for the 1/2 thou movements or two digits in metric.

Next foray into DRO's was an old Heidenhain system that came with the Bridgy. Glass scales, 1/2 thou resolution and 1 thou accuracy. Good system, reliable and accurate but this is what was being touted at the time for a grand so it should be.

Moving on I bought a Shumatech with Chinese scale for the cross slide of the 14 x 40 TOS. Not entirely happy with this as the scale jumps about a bit even when parked. tried the capacitor and new batteries but still has this problem. Lester reckon's it's the scales but as they are so cheap what can you do to guarantee a good scale. Funny they don't jump when used as a scale. Was going to fit a glass scale to the long axis but again like the rest the lack of 4 digits means that it's wasted.

I find them handy to use instead of counting turns, on the 14 x 40 TOS the cross slide screw is very fine and when you are plowing down from say 100mm to 25 it's easy to be a full turn out. It doesn't worry me roughing off and finishing with micrometers as I feel happier that way. Another point to bear in mind is unless you are running constant flood coolant, you have to stop and wait for the final cut as the work has expanded with heat. Plowing straight down to above size, taking a reading and putting the last few thou on will guarantee an undersize shaft when it's cooled so that swap over from Vernier or DRO to micrometers is a good time to let the work cool before the finish cuts.

Next step will be to move to 4 digit displays as without these a lathe DRO is useless. Have to wait and see what's on offer at Harrogate. .

-- Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

Gary

That would be extremely helpful if you don't mind, my e-mail address is jontom_1uk (at) hotmail (dot) com Thanks very much for taking the time and trouble to help.

Best regards

Keith

Gary,

If I could possibly trouble you to send me a copy as well, that would be most useful.

david at dlittlewood dot co dot uk

Thanks,

David

In article , John Stevenson writes

This is an important point (for me, at least). For generating curved surfaces, I would need a DRO which, if I put on a facing cut of say

2.000 inches for half a thou, I could be sure that the next cut of say 2.002 inches was that, not 2.004 inches. From your experience, would you expect the BW scales to do that reliably?

Do you mean three decimal places and a light/ two decimal places in metric? (I sure hope so).

That is a really good point, and of course applies however we are measuring. I'm sure I have forgotten it many times! I just estimated that for a 100K (100 deg C) rise in temperature there would be a 1.25 thou per inch expansion. That sound about right?

Hm, a DRO in tenths that firs easily on a Super 7, would be worth a lot of money....

Thanks,

David

Sounds like closed loop controllers can't handle large amounts of backlash in the screws, but controllers to get to closed loop are available if you upgrade the machine to decent ball screws?

So still loads of money to get to "proper" CNC and not worth the effort for low end chinese mills.

The first CNC mills I was allowed to play with (literally) were £1m a piece with Fanuc controls and that was 25 years ago - so I guess I was being a bit optimistic...

I just don't like the idea of "open loop".

Thanks for yout replies, very useful.

Steve

I read somewhere that the Chinese scales have their bodies connected to the +ve supply for the serial comms (odd, I know). Have you tried mounting it so the scale is completely insulated from the lathe ? (Ie nylon screws and insulating standoffs between scale and lathe). Just a thought.

...snip...

I didn't see anything much of interest at the London MEE last month, is Harrogate normally much better for cnc / dro type stuff ?

I think the Shumatech is not very good in filtering the data. To my knowledge, it is averaging over 3 samples. The Chinese scales do have a jitter of +/- 5 digits and a resolution of

20480 steps per inch. So the jitter is ... calculator ... 0.00025". I found that averaging over 10 .. 15 samples works good for a resolution of 0.001". With 0.0001" the last digit gets unstable. Don't expect more from the Chinese scales. Anyhow, there is always a position with a 0.001" resolution where the last digit changes.

Reminds me of a friend who had retrofitted his Deckel FP1 at work with a Heidenhain and a resolution of 0.001mm. He then discovered, that the last digit is unstable, when the mill is running. :-)

Nick

I've created a slide show (at least, I hope I have - its the first time I've used it), of several pictures of the BW unit on my lathe, and one picture of the Newall unit on my mill. Here's the URL...

Gary

That's superb, thank you very much. I've downloaded the pictures so that I can have a good look as my "absorption" rate is slower than it used to be. Good setup you have there and I suppose I have to ask as you have both the BW and the Newell, should I dig deeper and go for a Newell system for the lathe? Does it offer any advantage for the extra cost? I suspect given John Stevenson's comments and those of yourself and Chris, that the extra money might not be justified given the current state of DRO development for the lathe. Given the technique you have all outlined for accurate use I guess the BW system will do all that I want and be fairly unobtrusive to fit as well as being considerably cheaper.

I have to say again that this forum never ceases to amaze me, so many of you willing to help and advise and go to a lot of trouble to do so, thank you all.

Best regards

Keith

Gary, Not wanting to appear negative but I couldn't use that setup as you have it. I run in a commercial environment where time is money and make loads of chips and swarf. In your slide show the cables have no protection and if I was to run like this within 2 minutes I'd have swarf wrapped all around the cables.

On my CVA I have mounted the sender at the back close up to the end of the cross slide with the cable running inside along the screw. I'll get some pic's later The long axis runs at the front and clips onto the carriage but when not in use [ most of the time ] its un clipped and lives out the way because of this snarling up. In fact I have had to shorten the cable a bit because of accidents.

Even in a non commercial environment it's something to take into account.

-- Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

use [ most of

Hi John, I take your point. In fact, only one cable is exposed: the one on the long axis is actually underneath the rear way and well out of harm's way and has never yet been clogged with swarf. I agree that the X-axis is somewhat exposed, but in practice hasn't actually suffered because of it, even though whilst not being a commercial environment I still occasionally generate lots of swarf. I agree though, it would be better off being protected.

By the way, in answer to David's query, the BW display is 2 decimal places in metric, ie.01mm.

Regards, Gary Wooding (To reply by email, change feet to foot in my address)

Local Co near me had to change all their DRO's to come into line with Bullshit

9,000, swapped all the Heidenhain units from 3-1/2 digits to 4 at mucho cost. When I went in a few weeks later all the DRO's were sporting a band aid plaster over the last digit.

There is a real reason for hands on as opposed to theory.

-- Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

URL...

> --

Keith

I've taken the liberty of sending you some pictures of the BW set on my Myford. As you will see, the wire on mine is completely enclosed in the telescopic environmental cover which I think is now standard. If anybody else wants them, just shout.

--

Chris Edwards (in deepest Dorset) "....there *must* be an easier way!"

!" - Hide quoted text -

Chris

Thanks very much, superb pictures and a great help to me. I've replied off list but just wanted to let any other interested parties know that they are well worth seeing if you are thinking of putting a DRO on a Myford. Hope this doesn't "set fire" to your inbox Chris and thanks again for taking the time and trouble to send them to me.

Best regards

Keith