DRO calibration

[vtsteam]

I recently installed a DRO (thread elsewhere) on my round column mill. To calibrate it, the best I have for equipment is a 6" micrometer (.0001") and its 5" standard. I also have a cheap 8" digital caliper, some 123 blocks, and some old gage blocks won in an auction

The DRO resolution is approximately 2560 lines per inch, which I'd like to refine as close as I can.

What I've done so far is grip a ~6" piece of aluminum in the mill vise, dust cut one end with a new 3/8" carbide end mill, then traverse to the other end 6.375" with the DRO and make a second cut. Then measure between cuts with the 6" micrometer. That reading should be correct at 6.000" in an ideal world. Presently it's 6.008" with the DRO calibrated at 2560. I'm thinking I should reduce to 2557 or so.

So here's the question folks, is there a better way to do this with the tools I've got? I can imagine that if the mill is not exactly .375" (or properly its cut isn't .375) or the spindle wobbling, etc that measuring between cuts might not be ideal.

What do you think?

[awemawson]

Classic way is to mount a DTI as a zero indicator against a stop block, then insert a stack of Jo Blocks of known total length and bring DTI up to zero and compare DRO to know Jo Block length.

[vtsteam]

Thanks, Andrew.  The gage blocks I have are rather small, and frankly, old and unknown. Stacking them might not be any better.

But the 5" micrometer standard is new, and I could probably use that the same way.

[vtsteam]

Since the standard is a rod, it's a little tricky to position, but I think I'll set it in ground Vee blocks and check orientation with the indicator first

Also I think I'll reverse the usual procedure and set it against the stop, first for zero set, then remove it and traverse to the stop for the second measurement. That makes positioning a lot easier.

[vtsteam]

Actually, I could just mill myself a squared off standard block and check its length with the micrometer. Doesn't have to be any particular length except between 5-6". Then mark it for future use.

Then it will be easier to position.

[vtsteam]

Hmmmm (number 4?) I think my parallels are 6" long......

[vtsteam]

Wait, 123 blocks..... two = 6"......... put together, mike, and go.

Back to square 1.... 

[John Rudd]

I seem to be missing something here…..
My  understanding of calibration is to have a zero and span adjustment and possibly a linearity …..( I come from an instrumentation background….think pressure gauges et al)

So for a dro, mine for example doesn’t afford such adjustments…..it’s a standard dro readout and glass scales…

So could you elaborate on the process for calibrating yours Steve please?

[vtsteam]

Okay so here's the tentative procedure:

1) clamp a bar across the table.
2) indicate it square in the y direction
3) lay two 123 blocks together and check their length with micrometer
4) place them against the bar
5) bring the indicator up to blocks
6) zero the indicator, DRO, (and out of curiosity) the mill dials
7) remove blocks and crank to the stop
8) check readings
9) adjust : (DRO reading / Actual length) X Present Lines Per Inch setting = New setting
10) repeat 4 thru 8 as a check

[vtsteam]

John, the above wasn't an answer to you -- just typed while you posted.

My scales are inexpensive capacitative scales of unknown specs, re-purposed into a DIY bluetooth DRO unit. The software for that does allow calibration (it is designed to use many different kinds of scales with different lines per inch resolution) So yes it is possible to calibrate initially to match the scales.

Yes most commercial scales are made to a high standard, so you can depend on the mfr's statement of resolution. But in my case that info is not available. So in a sense I'm reverse engineering.

And apparently these scales have a non-standard (meaning non-intuitive metric or inch) resolution of approximately 2560 lines per inch. I wanted to see just how close I could get by actual measurement to the reality of these scales, as installed.

So well, this isn't what you would normally do, but it is an interesting problem, and will help us (meaning me) learn a little more in the process.

[vtsteam]

Also, the zero adjustment in a DRO is naturally setable (absolute, incremental etc.). So what I'm doing is equivalent to your span calibration. The biggest span I can measure in a trustworthy manner is about 6" with the equipment I have.

[vtsteam]

It seems to work. I got a calibration of 2562 on X and 2563 on Y, and with those numbers I hit the same measurement as the micrometer in thousandths on a 6.000" span.

The 123 blocks were great because I could arrange them to check measurements of 5.000,  4.000 and 3.000 inches as well as 6.000, showing me that linearity was good as well. And repeatability was also perfect with lots of measurements and traverses always returning to an agreed zero with the DTI, as well as repeated setups again to earlier measurements.

You can actually calibrate lines per inch to two decimal places using this DRO software -- I just stuck with whole numbers. I feel like I'm in my "close-enough" comfort zone now, and I'd need a bigger micrometer, more blocks, etc. to go finer

The general 2500+ lines per inch of these scales mean a resolution of a little under 4 tenths, so the nearest thou is really a matter of practicality, and I've limited the display to 3 decimal places. That's also adjustable.

If anyone is interested in this kind of interface, there is a prebuilt version of the interface board (the software is free) -- it's called TouchDRO .

Well, back to the Bluetooth DRO thread, and on with the mounting of the Z axis scale......

Thank you Andrew, again...

[awemawson]

Steve,

I well remember the frustration of using my first 'machine tool', a Taiwanese Mill/Drill that I bought brand new, where the calibration of the down feed although marked as 'thou' was completely arbitrary, and being a 'round column' machine any head movement  lost any datum points anyway.

I got quite good with gauge blocks, engineers buttons, and other mechanical devices to find out where I was!

Glad to be able to pass on something useful for once.

[vtsteam]

Andrew the Z axis scale is going to be a bear. The support for a moving quill scale will have to be cantilevered. And there's no really convenient place to put it. Most people who have done a DRO addition, choose to have a moving read head, and replace the Z axis quill stop in front. But no room there for a moving scale. About the only place that I can find room is behind the drill handle on the right side.

That stop is a mess -- I found 20 thou of vertical backlash, and the pinned side groove in the quill which is supposed to keep it from rotating has major slop. Thus the quill rotates a fair amount. I will probably have to replace the pin with something I make up. That's before even figuring out where to mount the scale. And as you say, all DRO axes are lost when you crank up the head.

I'll probably make one of those side rail attachments some day to keep X and Y aligned when cranking. There seems like a good version shown on, pardon the name, Winky's Workshop on the Tube. includes a motorized lift as well.

[vtsteam]

Oops, hit the edit button on your post instead of mine -and messed it up  -- restored now, but now has edited by sig, sorry! 

[awemawson]

Easily done Steve, I've done it before myself 

My 'Mill Drill' did actually get me into having my first workshop - an 8x12 shed at the bottom of the garden back around 1980 (gasp !) But really it was a horrid machine.

[Sea.dog]

I've started to get quite attached to my Ron Fu, Andrew. Despite its limitations. Once its fully modded I will post a thread.