Choosing Scales

[John_Dennis]

I am wanting to add a Touch DRO to my Bridgeport, I am wondering if Chinese glass scales or Igaging style would be best. What are the advantages and disadvantages and are glass scales more complicated?

Thanks
John

 

[ycroosh]

John,
Glass scales are in every way superior to iGaging scales. Without going into too deep of a rabbit hole: glass scales have 2x to 10x the resolution (5 microns per "tick" or 1 micron for high-resolution version), they are immune to machine shop noise and refresh rate is pretty much real-time (they don't have any digital logic, so pulses are sent to the DRO head/TouchDRO adapter without any delay).
iGaging EZ-View or Shahe "Remote DRO" scales are less expensive, though - they cost about 1/2 to 2/3 of what a set of comparable glass scales would cost, and can be easily cut, which is important if there are space constraints. The tradeoff is a lower resolution (10 microns per tick). iGaging scales are fast enough, so no issue with the refresh rate/lag, but suffer from the flickering the last digit that can be bothersome on some specimens. Shahe scales are much more stable but are a bit slower, which can be noticeable in some cases. I would not bother with iGaging Absolute scales - they cost about as much as Glass scales and don't offer any tangible advantage over iGaging EZ-View
Basically this is cost vs. performance tradeoff. For a smaller machine capacitive scales are somewhere between "perfectly fine" and "adequate"; for a Bridgeport, I would want glass scales. If space is an issue, you can use a magnetic scale on one of the axes. This is how my mill is set up - glass scales on the column and Y-axis and magnetic scale on the X-axis.
As far as complexity goes - neither is more complicated than the other. If you are thinking of using the scales with TouchDRO, you can buy a pre-made adapter for either type.
Hope this helps
Yuriy

 

[RJSakowski]

Thanks Yuriy, for the extensive comparison of scale types.

My mill has 5 micron glass scales, circa 2005, with resolution of .0002". This is more than enough for any work that I can do on the mill. On my Grizzly G0602 lathe, I opted for the iGaging scales with the TouchDRO because at the time, my only othe option was the glass scale which wouldn't fit for my cross slide. The resolution is .0004" and with the cross slide set for diameter mode, that works out to .0008". I set the display resolution for four decimal places so I see the .0008" flicker but I mentally average the reading so it doesn't bother me. For critcial diameters, I rely on final measurements by micrometer anyway.

Were the current choices available, I would have gone with magnetic scales, preferably, 1 micron. For all but the smallest mills, glass scales should be no problem. The quill DRO is the most difficult if you are trying to avoid loss of functionality. I tend to use custom mounts rather than the provided hardware.

 

[John_Dennis]

Can you recommend a brand or supplier? I am also curious to see how scales are attached to the quill.

I want to order some scales before the Wuhan apocalypse shuts down all of the suppliers.

 

[RJSakowski]

Can you recommend a brand or supplier? I am also curious to see how scales are attached to the quill.

I want to order some scales before the Wuhan apocalypse shuts down all of the suppliers.

Here is what I did on a Grizzly G0755. The OEM depth stop was removedandthe scale attached to the spindle bracket. The pickup was monted to the case 2with fully adjustable mounts. This was a full size glass scale.

 

[ycroosh]

JD,
For the most part all Chinese Glass scales appear to be made by the same manufacturer. I've taken apart Ditron, Easson, Sino, Jino and a an unbranded scale and the internals are identical, with the same markings, same carriage castings, etc. If you're looking at eBay, I'd find the best price from a seller with decent rating and pinout diagram.

 

[RJSakowski]

And here is how I handles a RF style mill/drill. The depth stop was replaced with a 5/8" stainless steel rod to provide support for the scale at the upper end.The lower end mounts to the spindle bracket. The OEM depth stop screw was moved and is still functional.

 

[JimDawson]

My personal preference is magnetic scales. Very compact. Have them both on my lathe and mill. 1 micron resolution.

 

[JFL4066]

Yuriy,

Is there any reason to go with 1 micron scales vs 5 micron scales on a BP mill? Software rounding errors or backlash compensation etc?

 

[ycroosh]

The simple answer is that 5 micron scales are more than good enough for any milling machine. You get 0.0002 resolution, and even brand new, unworn, perfectly setup machine will hard time holding those tolerances (despite what some might say). Even if you can measure that distance, things like cutter deflection, thermal expansion of cast iron, etc. will affect the measurement. There is no significant downside, though besides the cost.
In practice there are a few things to consider: the quality of "name brand" Chinese glass scales is good-to-excellent. The scales that you get on eBay, AliExpress etc. for under $300 for the whole set are that inexpensive for a reason. Best case scenario, the vendor might've skipped some QA and adjustment steps, used lower grade ball bearings (the carriage rides on a set of miniature ball bearings inside the scale) or included only one set of rubber seals. Worst case, the scales are QA rejects from better brands. This can be simple scratches on the aluminum, but might also mean bad encoder strip with bad etching, missing marks, etc. If you get a set of "reject" micron scales, they still will be better than "good" 5-micron scales simply because the error will be proportionally smaller. That said, unless you are doing aerospace work, even "bad" 5-micron scales are more than good enough.
One thing that can be a negative for micron scales is the 5x pulse rate. On a mill with manual feed this is not a problem but on a lathe, where the apron can move pretty fast and unevenly, or on a mill with fast power feed the scale might move too fast for the pulses to fully rise over the threshold. Also, some DRO heads might not have enough processing power to handle the pulses. I tested TouchDRO firmware with 100KHz square wave input into four axes simultaneously (100mm/second with 1 micron scales). With proper input chips this was not a problem, but at that point any line noise would wreck havoc on accuracy. 100mm/second is more than achievable on a lathe...
Hope this helps
Yuriy