DRO system for micrometer?

[Mike Henry]

DRO system for microscope?

I'm looking for a DRO system for inspection with microscopes for the purpose measuring very small features and dimensions, on the order of a few microns, over distances of 2-6 inches. I have the microscope and an X/Y stage with 0-1" manual micrometer heads but it is a PITA to take manual point readings, record them and then work out hole diameters or spacing by hand, so it would be convenient to find a DRO and scales that could simplify this. Digital micrometer heads are my backup plan to minimize measurement errors. Budget is an issue and I'd like to keep it below $2k.

Does anyone have any suggestions for suppliers? I'll probably be calling some of the DRO sources next week to see what they have to say.

 

[Mike Henry]

Arrgh! - The title was supposed to be "DRO system for microscope". Tried to edit the title but doesn't seem possible.

 

[TGTool]

Arrgh! - The title was supposed to be "DRO system for microscope". Tried to edit the title but doesn't seem possible.

So you meant it to read DRO system for microscope instead of "DRO system for microscope"?

It looks like it works from here.

 

[Milland]

So you meant it to read DRO system for microscope instead of "DRO system for microscope"?

It looks like it works from here.

Check the actual thread title (not the post title).

 

[CalG]

https://www.vcamerica.com/products/encoders/precision-scales/rotary.html

Coupled to a compatible two axis reader display. You "might" stay under $2K.

I've got a sweet little motorized stage if you need another, alas, No DRO ;-(

 

[Archer120x]

So you want to build your own tool makers microscope from scratch?

 

[mhajicek]

What you want is a toolmaker's microscope / toolscope or an inspection vision system.

 

[Mike Henry]

So you want to build your own tool makers microscope from scratch?

Maybe more like a high magnification CMM or optical comparator for less money. At present the intended application is very low volume so doesn't warrant a lot of money spent on it. Perhaps my best bet is just to replace the mechanical micrometer heads with digital ones and deal with the math by calculator and notepad or a spreadsheet. Maybe something like this:

DRO EL550 | Digital Readout System - EMS

 

[PeteM]

Mike, I use an x-y table cobbled together with Mitutoyo caliper scales for just this purpose. Coupled with a cross hair reticle in one stereo zoom microscope eyepiece it does a fine job of measuring most features.

Your present table, if you are happy with the smaller working area, could probably be retrofitted with digital scales. Does it have the 6" travel you noted you wanted?

If you get calipers with SPC outputs, you could place larger displays anywhere you wanted. Ebay might be a source for light but high quality x-y tables (Parker etc.). Some will also have a rotary function, if you need to measure angles.

If this is for very occasional use, you could also make a bracket to hold an articulating arm for a stereo microscope and just use your mill table and DRO for the measurements. Kind of a pain to clean off the chips maybe, but shouldn't take more than a minute to get the scope on or off.

 

[ballen]

I've fitted DROs and scales to a number of my machines over the past few years.

The DRO (or counter, or head, or display) is the easy part, and almost any two-axis display will work fine for what you need.

The more critical part is the linear scales. For those, I am partial to Renishaw, who make lots of optical scale models (both "open" and "sealed")
with resolutions measured in nanometers and accuracy measured in microns or better. One of the standard applications for these is in the (electronics) chip-fabricating business. The prices are reasonable, and when I've written emails with technical questions, I've gotten quick and competent replies from people who are clearly engineers not marketing. Probably if you send them a photo of the microscope stage and tell them roughly what sort of resolution and accuracy you want, they can suggest suitable scales.

 

[Mike Henry]

Thanks for all the feedback and info. DROPros sells their EL400-series 2-axis display with 1µ-resolution magnetic scales that can be cut to length for <$1k. They claim that the Electromekanica EL550 really isn't marketed, which would have been nice for the feature set. I have email into them (India) and will see what they have to say.

I'd like to get more than 1" or 2" of travel with the X-Y stage and am looking into converting a small mill table for that purpose, which would also allow more room to mount the read heads and scales. It's starting to feel like a hack, but that might be all that's needed.

 

[ballen]

Thanks for all the feedback and info. DROPros sells their EL400-series 2-axis display with 1µ-resolution magnetic scales that can be cut to length for <$1k. They claim that the Electromekanica EL550 really isn't marketed, which would have been nice for the feature set. I have email into them (India) and will see what they have to say.

There is a lot of annecdotal evidence that this sort of magnetic scale (polymer with embedded magnetic particles, with magnetic field polarization parallel to the surface) does not provide long-term accuracy. For these scales the pole spacing is 2mm, which means that to get micron precision one must interpolate by a factor of 500! The slightest variation in the magnetic field will wipe that out.

If you want accuracy comparable to your microscope, you'll want glass scales from Renishaw or Heidenhain or another high-end vendor.

 

[Mike Henry]

There is a lot of annecdotal evidence that this sort of magnetic scale (polymer with embedded magnetic particles, with magnetic field polarization parallel to the surface) does not provide long-term accuracy. For these scales the pole spacing is 2mm, which means that to get micron precision one must interpolate by a factor of 500! The slightest variation in the magnetic field will wipe that out.

If you want accuracy comparable to your microscope, you'll want glass scales from Renishaw or Heidenhain or another high-end vendor.

Thanks - I wasn't aware of the potential long-term problem with magnetic scales, though 2 or 3 years worth of use would be the most I'd need.

India never responded to my request for info on their EL550. No surprise there, I suppose.

 

[CarbideBob]

One thing to keep in mind when doing a microscope is the physical size of the scale body and where/how to mount it.
This can get ugly.

Another option here is long travel digital indicators. Most will have a easy interface to a computer and simple to mount.
Watch for resolution here when shopping.
You say 6 inches but are using a 1 inch mic head? Gauge blocks or stop rods in use?

If you need microns be aware the resolution and accuracy can be very different things in this world. I can give you a readout to millionths but only be correct to 5 tenths.
100 percent agree with above that Renishaw or Heidenhain good and top shelf stuff but expensive. One of these has a longer history in this but not sure that matters now.
Keep in mind that I am just a peon with some hope of advice be it good or bad.
Bob

 

[Mike Henry]

Thanks Bob - I'm still trying to figure out how to mount the scales on a platform that will give me several inches of travel at ~ 1µ resolution. The current setup uses a microscope X-Y stage with tenth-reading Mitutuyo micrometer heads but that has only 1" of travel in X and Y so far I've only needed afew hundred µ of travel. Mounting scales and read heads to that would be messy and would still limit travels to 1" without some sort of modification or re-design of the X-Y stage. I may be able to use the table from an old Sherline mill, which would give me the desired travel but far more than 1 µ backlash, though that may not be a deal breaker if the scales are as accurate as claimed - just annoying to use for tedious work. Another option would be to mount the USB scope on the head of a CNC mill and use that as a sort of crude optical comparator in X and Y. That may be the best solution in the end, though I'd have to rely on the accuracy of the CNC mill's readout and that is probably 3-5 µ at best.

This whole exercise may be for naught - the project calls for drilling a hundred or more 300 and 400 µ holes in a thin film and I've yet to successfully drill more than a handful of holes in the material. Lots of challenges, but fun to try.

 

[CarbideBob]

....
This whole exercise may be for naught - the project calls for drilling a hundred or more 300 and 400 µ holes in a thin film and I've yet to successfully drill more than a handful of holes in the material. Lots of challenges, but fun to try.

Challenges will always be in front. Can they be beat into the dirt or is it hopeless.
Who knows when to pursue and when best to just walk away nicely if you can?
Bob

 

[greif1]

Just having accurate scale with 1um or better resolution will not give you a system with accuracy to that level. The stage straightness and offset of the scale on the stage relative to the thing being measured (Abbe error) will cause quite a lot of error. Even with quite good components and alignment, you would be doing well to get 5um accuracy. We have a fancy Keyence microscope that can measure with high accuracy in the camera field of view (5x5mm) and then longer by stitching images together and moving the stage (0.1umm res). The error for the camera 5x5mm view is less than a micron, the error with stage movement is 3 or 4 um.

If you only need very short distances, you can get very high accuracy with a filar microscope. At 100x magnification you get a 800um field of view and accuracy of around 1.5um.

 

[Mike Henry]

greif1 - good points that I hadn't really considered. Filar microscope had me checking Wikipedia - that's a new one to me and looks pretty expensive.

The tolerance might be much looser than 1 µ but I'm waiting some feedback from the client/partner to see where they want to head.

 

[greif1]

I should clarify my note;

100x total magnification, using a 10x objective and 10x filar eyepiece= ~800um field of view and 1.5um uncertainty
200x total (20x objective) = 400um fov and ~0.9um uncert.
400x total mag (40x objective)= 200um FOV and ~ 0.4um uncert

If you did not need digital, old fashioned micrometer screw filar eye pieces are on ebay for cheap. Any decent microscope would work. Since the stage is not moving during measurement (only the crosshair in the eyepiece does), accuracy is much better that moving stage types.
Make sure to get metalurgical objectives or NCG (no cover glass) type (often labeled with a tube length/ cover glass thickness, example 180/0, one that would use a cover glass would be like 180/0.17

 

[Mike Henry]

To finish this out, the person I've been working with, and that generated the need, has found another way of inspecting the parts over a hundred or two millimeters to the resolution desired so I'm putting this way down my priority list. Thanks for all the input; at least I learned a few things.