PDA

View Full Version : Magnetic DRO scale



EVguru
02-05-2012, 06:28 AM
I recently replaced the aged 2 axis Mitutoyo DRO on my Bridgeport clone with a 3 axis Sinpo unit from The DROstore.

Whilst I was at it I decided to add a DRO to my Harrison L5. I really wasn't sure I was going to find room for a glass scale, so I upgraded to a 1 micron scale that could be employed on my cylindrical grinder instead.

As I suspected I wasn't happy with any placement of the glass scale, so adpoted Plan B and went with a magnetic system from Allendale Electronics
http://www.machine-dro.co.uk/digital-readout-systems/magnetic-tape-linear-encoders.html
Although not on the website, they do a 1 micron encoder too, but I went for the cheaper 5 micron.

http://www.compton.vispa.com/Pictures/magencdr.jpg

The tin leadscrew cover was replaced by a bit of Aluminium plate and a recess machined into the top slide to take it. Mill an nice slot for the magnetic strip and the job's a good'n. As a temporary measure, the strip it protected by a layer of clear tape. I have a tougher plastic cover to go on, but it's too cold out there!

I had to make some modifications inside the taper turning attatchment to make clearance for the thicker leadscrew cover. I flycut the cast top cover and made the slot for the reader head which is shimmed for clearance over the magnetic strip.

It turned out pretty well I think.

There's a cheap system for those who'd like to experiment;
http://store.makerbot.com/magnetic-linear-encoder-v1-2-kit.html

It's based on a chip from;
http://www.austriamicrosystems.com/eng/Products/Magnetic-Encoders/Linear-Encoders

IanPendle
02-05-2012, 09:27 AM
Hello EVguru,

I had been looking at the exact same magnetic strip with a view to using it as the basis for a DRO system on my small TOS lathe. My issue was its resistance to coolant. If it ever warms up and you ever get back out to your lathe, I would be interested in your further experiences with this strip, especially it's resistance to coolant and oil. Can you completely seal-in the strip with epoxy, for instance?

Ian .

EVguru
02-05-2012, 11:33 AM
You can cover the strip with anything non magnetic.

They sell a support profile in two versions
http://www.machine-dro.co.uk/two-part-magnetic-tape-support-profile-300mm-length.html
With an aluminium cap.

http://www.machine-dro.co.uk/magnetic-tape-support-profile-300mm-length.html
With a stainless cap and rubber seals.

Silicon sealant would do the job.

philbur
02-05-2012, 12:54 PM
I looked at the site a couple of months back. the data sheet for the tape wouldn't open and it still doesn't.

DRO-MG-TP-05 linear magnetic tape- technical data.pdf

The resolution is good but possibly the accuracy at 0.02mm (that's 20 micron) is not so good, at least for a cross slide or a mill table. The trouble is without the data sheet it's not possible to know the basis for the 0.02mm.

Phil:)

EVguru
02-05-2012, 01:26 PM
I looked at the site a couple of months back. the data sheet for the tape wouldn't open and it still doesn't.
Why not send them an email to point that out and ask for the file.


The resolution is good but possibly the accuracy at 0.02mm (that's 20 micron) is not so good, at least for a cross slide or a mill table. The trouble is without the data sheet it's not possible to know the basis for the 0.02mm.

Phil:)

A accuracy of +/-10 micron is what Heidenhain quote for their LS600 series scales.

If your DRO has the option you can always do a linear or segmented compensation.

alanganes
02-05-2012, 06:12 PM
Nice setup there. Something like this would be great for my old Sheldon. There is not a lot of space to mount a typical encoder on it. Anyone know if these reader heads and tape are available from any US suppliers?

Davo J
02-05-2012, 11:00 PM
You did a nice job on installing that scale, well done.

philbur
Here is a copy I kept. I saw it on HMEM and then posted about it on MadModders. Maybe this is why they have taken it down.


Dave

http://i702.photobucket.com/albums/ww21/1top720/comparison-1.jpg

philbur
02-06-2012, 03:36 AM
+ or - 0.015mm doesn't seem to be too good for a precision machine tool, especially when it is only a typical value, meaning you could end up with worse than this. I think it is mainly intended for positioning systems that don't require quite the same accuracy as a precision machine tool, otherwise glass scales would have disappeared long ago.

Phil:)

EVguru
02-06-2012, 05:20 AM
Nice setup there. Something like this would be great for my old Sheldon. There is not a lot of space to mount a typical encoder on it. Anyone know if these reader heads and tape are available from any US suppliers?

DRO Pros sell a magnetic system, as do Sony (quite expensive).

philbur
02-06-2012, 06:34 AM
Thanks for the table Dave.

I have to say it is still not clear to me what the stated accuracy means in actual performance terms.

Hypothetical example: I turn a part to a diameter of xx.xxx mm. and then zero the cross-slide DRO. Now I wind in the cross-slide to a reading of 1.000 mm. on the DRO. Is the new tool position uncertainty (relative to the previous xx.xxx mm position) + or - 0.015mm. If the answer is yes then what if I wound the cross-slide in only 0.025mm, surely the new position uncertainty is not still + or - 0.015mm, or is it?

If the cumulative error is relatively linear over the whole length then that would be much easier to live with on a cross-slide than a scale that has localised errors of the same magnitude.

The table seems to imply that it is localised error as the stated error doesn't change with scale length. In addition isnít there likely to be some additional installation error due to unintentional stretching or compressing of the tape.

I think the answer to these issues would make all the difference as to whether the magnetic strip is good or not so good for a lathe cross-slide for example.

Phil:)

EVguru
02-06-2012, 06:37 AM
+ or - 0.015mm doesn't seem to be too good for a precision machine tool, especially when it is only a typical value, meaning you could end up with worse than this. I think it is mainly intended for positioning systems that don't require quite the same accuracy as a precision machine tool, otherwise glass scales would have disappeared long ago.

Phil:)

0.03mm is 1.18 thou of an inch.

When applied to the cross-slide of a lathe the method of working is going to reduce the error anyway. Take a clean-up cut on the diameter, measure the diameter and program that dimension into the DRO. Machine to required diameter. You''re only working over a short section of the scale, so the linear error is much reduced and any critical diameter will most likely be checked with a micrometer anyway because of springback.

Once again I'll point out the ability on modern DROs to apply compensation to improve the accuracy.

Many people these days are finding it an advantage to fit digital dials. They they don't account for leadscrew error, wear, or backlash, but the big digit displays, multi turn counting and metric/imperial conversion are still quite a boon.

Magnetic scales are made to high accuracy (Sony for example) and survive conditions that glass scales would not.

These cheap magnetic scales are a good step up from the capacitive (digital caliper derived) scales that so many people have fitted as a cost effective DRO.

philbur
02-06-2012, 06:51 AM
On a cross-slide the scale error results in double the part diameter error.

I'm not saying the cheap magnetic scale is no good. I'm just trying to understand where it truly fits in the overall scheme of things. The web site seems to try and hide the true case, which in itself raises suspicion.

Phil:)

Davo J
02-06-2012, 07:48 AM
I just had a look and the chart is still up on the site.

You need to go to "Magnetic Tape Encoder - 5 micron resolution" page,
http://www.machine-dro.co.uk/digital-readout-systems/magnetic-tape-linear-encoders/magnetic-tape-encoder-5-micron-resolution.html

Then click "High Accuracy Magnetic Linear Tape", then scroll down to the "Linear encoder comparison.pdf"
http://www.machine-dro.co.uk/high-accuracy-magnetic-linear-tape.html

This is the link
http://www.allendale-stores.co.uk/dro/info/linear_encoder_comparison.pdf

Dave