I've been around and around with Yuriy. He is totally convinced that magnetic is inferior to glass when it comes to absolute accuracy. His methods may not be perfect but they sure are compelling.
However, as
@Dabbler says, they go against the grain and therefore are worthy of scrutiny. I'm not saying that Yuriy is wrong or right. Just that scrutiny and objectivity is required. At the end of the day, we all want the same thing - an advance in our knowledge of the subject. Advancements almost always come from different ways of looking at things. Challenges, reviews, and healthy debate are a part of that process.
Wife was gone to play pickleball the other day, so I headed out to the shop to do some testing of my own that I should have done ages ago. My bad........
In general, my results do not reflect what Yuriy found. I've sent him a copy of my results to see what he says. He has not yet replied. (He might be tired of me..... LOL... and I wouldn't blame him). So I don't have any advice for
@David_R8 yet.
In the meantime, here is the new video that Yuriy said he was going to post.
Hi folks, I just uploaded a video that covers scale calibration on a lathe. Some of the stuff is TouchDRO specific, but the principles apply to other DROs as well. Hope you find it useful. Regards Yuriy
www.hobby-machinist.com
It's a lathe calibration, but he talks about the cyclistic errors that are the subject of this thread and he discusses ways to deal with them. (On a side note, I will say that I think his videos are awesome! Coming from me, that's a heck of a compliment!)
But I am not yet convinced.....
Consistent with the comments provided by
@Rauce &
@Dabbler, the testing I did covered a number of approaches. Some of which have advantages and disadvantages. I'm not promoting any one approach, just presenting them for what they are worth.......
I began my testing using a 0.002 metric indicator (2 micron or about 1 tenth of a thou) and had my Ditron DRO set to 4 significant digits. Each measurement was consistent. I noticed no random character to the measurements - certainly not cyclic within the range of my indicator. However, I did notice a very slight slow creep upward in the DRO Readout as the distance increased. I need to evaluate this more in the future. Its been a while since I trammed my mill so it might be a simple tram cosine error. Lastly, I suppose it might also be slightly lower rigidity in my otherwise awesome FISSO indicator holder than in my spindle.
My sensor and engineering background have both engrained in me a very good grasp of the concept that everything moves - it isn't if - it's only how much. Doesn't matter how big it is. It will deform under load. I think this played a bit of a role in the results to follow.
After this, I mounted a digital edge finder (that is fairly consistent to a tenth of an inch) and compared various random measurements of Grade B gauge blocks from 1/2 inch to 4 inches. Again, no cyclicity to the measurements - just a very very slight slowly growing error with overall distance. However, the error was less than the growing error in the prior test which suggests a combination of tramming, scaling, and rigidity for the prior tests. In all cases, the readings were within the specified tolerance of the gauge blocks. So they might even be right.
I also mounted a Mahr Millimess 1 micron indicator in my Fisso holder and got:
10 vs 10 microns on the DRO
15 vs 15
31 vs 30
46 vs 45
61 vs 60
Again a tiny bit of creep if you can even call one micron creep. The usable range of the indicator is only +/- 30 microns.
Lastly, I mounted an inexpensive 2 inch length 1/2 thou dial indicator and got:
10. 9.9 on DRO
20. 19.9
30 29.8
40. 39.8
50. 49.9
60. 59.8
70. 69.8
80. 80.0
90. 90.1
100. 100. 0
125. 124.7
150. 149.8
175. 175.1
200. 199.8
250. 250.2
300. 300.3
350. 350.2
400. 400.2
500. 499.8
600. 599.7
700. 699.9
800. 799.9
900. 899.8
1000. 999.7. 999.8. 999.6. 999.6
1500. 1499.9. 1499.8. 1499.9. 1499.7
2000. 1999.9. 1999.8. 1999.7. 1999.8
At the end, as you can see, I decided to do some repeat comparisons. I did this because I thought my cheap indicator might be sticking. It is also only a 1/2 thou indicator. Although one can discern better resolution than that, it's still outside the usable range of the indicator and can't be trusted (not that it's all that trustworthy anyway....). The numbers sort of validate that but it might also be holder flex in my Noga Indicator Holder. Who knows for sure. I am planning to do some holder comparisons in the near future so something might come up then.
All in all, my testing did show a few interesting anomalies that arise from the methods and equipment used. But the good news (or bad) is that I don't seem to have the cyclicity errors that Yuriy experienced.
One of the things I didn't do in the testing I did was to deliberately test the scales at the magnetic boundaries. I plan to do that sometime in the next few days (perhaps even today). But I'll have to think a while on exactly how to do that. How do I know where the boundaries are! Or do I just do a full curve using two indicators simultaneously - one to track location and one for precision..... LOL!
I've told Yuriy that I would be happy to do any further testing he might want me to do. But I don't have good video skills, I have no data access to my DRO or my measurement equipment, I am also old, blind in one eye from a stroke, and I shake like a leaf, so it won't be fancy!
May the force be with you David......
What does it all really mean?!