DRO Heck.

[PaulL]

I appear to continue to be in DRO heck.
I've never looked at my DRO output vs a dial indicator until I was trying to get my lathe's X axis working better - using just the DRO I was unable to hit diameters by better than ~2-4 thou.
And when I correlate my DRO movement against my DTI, I see a *ton* of jitter and non-repeatability. I've never noticed on the mill because I don't make anything that seems to care about that level of precision on layouts.
I'm using cheap SINO scales and 5 micron tape.
Is my X axis accuracy the thing that drives me to 1 micron tape and a more expensive encoder? Any direct experience with these?

 

[DavidR8]

Is it the cyclic error issue that I've read about?
If so 1 micron scales might help.

DRO Scale Recommendations (for TouchDRO, and other DROs)

Good day, folks. I get a lot of questions about scale recommendations for TouchDRO, so for the last couple of weeks I've been testing a bunch of different scales. I intend to keep an updated list on my website: Recommended DRO Scales, but I think there is value to having a thread here. This is...

www.hobby-machinist.com

 

[PaulL]

Is it the issue described here?

Yes, it may well be.
Looks like I should upgrade my X to 1 micron scales.

 

[Susquatch]

Is it the cyclic error issue that I've read about?

I corresponded at length with Yuriy on this. I was not able to duplicate his findings. He thought perhaps I had a unicorn set of scales.

I cannot explain his results. But I see he seems to have backed away from pressing the concern. I'm not sure if he has doubts or got tired of the flack he got from lots of folks who were happy with their equipment. It's a tough situation for him. I support his efforts to find the truth no matter how unpopular it might be. I'd rather hear the truth no matter how unpopular, than follow a hundred happy blind lemmings.

FWIW, my scales are 1 micron Ditron.

 

[Ironman]

I corresponded at length with Yuriy on this. I was not able to duplicate his findings. He thought perhaps I had a unicorn set of scales.

I cannot explain his results. But I see he seems to have backed away from pressing the concern. I'm not sure if he has doubts or got tired of the flack he got from lots of folks who were happy with their equipment. It's a tough situation for him. I support his efforts to find the truth no matter how unpopular it might be. I'd rather hear the truth no matter how unpopular, than follow a hundred happy blind lemmings.

FWIW, my scales are 1 micron Ditron.

I built two of Yuri's bluetooth scales. One I installed in a wood framed building in Wyoming and it worked on his lathe. I honestly would not waste my time with that system ever again. The one I installed on my lathe jumped around and was completely untrustworthy. I have metal sheathed walls and metal roof...a Faraday cage. I tried using it with all power except the lathe off, and it still jerked around like a BC junky. I have Mititoyo scales on my mill and they work fine. And I believe almost any cheap 200 buck scales would work for me, but I think the standing waves and reflection of the BT transmitter signals are causing this. I used an antenna and a oscilloscope and the amount of noise in the room is impressive to see.

 

[DavidR8]

I built two of Yuri's bluetooth scales. One I installed in a wood framed building in Wyoming and it worked on his lathe. I honestly would not waste my time with that system ever again. The one I installed on my lathe jumped around and was completely untrustworthy. I have metal sheathed walls and metal roof...a Faraday cage. I tried using it with all power except the lathe off, and it still jerked around like a BC junky. I have Mititoyo scales on my mill and they work fine. And I believe almost any cheap 200 buck scales would work for me, but I think the standing waves and reflection of the BT transmitter signals are causing this. I used an antenna and a oscilloscope and the amount of noise in the room is impressive to see.

From what I've seen on Hobby Machinist I think your experience is an absolute outlier.

 

[Dabbler]

@PaulL @Ironman has raised a relevant point - could what you are seeing be electrical noise?

It could also be a faulty reader/install. There is nothing about static 5 micron tape that would cause jitter or non-repeatability. If the reader isn't picking up the magnetic domains (by being too far, or not being sensitive enough) this can lead to jitter. Another source of noise could be a magnetic chip trapped between the tape and the reader. This will distort the magnetic domains enough to see these problems.

 

[PaulL]

could what you are seeing be electrical noise?

It could be. I've adjusted the gap a few times with no difference, and cleaned everything thoroughly. I can easily swap X and Z, so I'll give that a try when I'm back in the shop, to check against a bad reader.

 

[Tecnico]

Just piping up here, I agree with @David_R8 , I have a lathe and a mill with both mag and optical 1u scales and one of Yuriy’s U-solder boards each and I have never seen anomalies in readings.

@Ironman , I don’t know what type of scales you have but if they’re capacitive, I understand that they are not as stable/more finicky than optical or mag scales. You could get support from Yuriy via his forum section over on Hobby Machinist or by PM over there, he will respond to individual inquiries. If you do have a problem with one of his boards he will make it right if it’s the board’s fault.

D

 

[Ironman]

Just piping up here, I agree with @David_R8 , I have a lathe and a mill with both mag and optical 1u scales and one of Yuriy’s U-solder boards each and I have never seen anomalies in readings.

@Ironman , I don’t know what type of scales you have but if they’re capacitive, I understand that they are not as stable/more finicky than optical or mag scales. You could get support from Yuriy via his forum section over on Hobby Machinist or by PM over there, he will respond to individual inquiries. If you do have a problem with one of his boards he will make it right if it’s the board’s fault.

D

My scales are Igaging as Yury recommended in the beginning of his project. I'm figuring micrometers and calipers work well, and I can trust the reading.

 

[slow-poke]

FWIW I'm going to offer my 2c.

I have had zero noise problems with any of my CNC electronics but that might be because I took steps to mitigate it in the first place.

Depending on the particular equipment and how it is wired can make all the difference so just because brand xyz works well on this lathe or that mill, does not necessarily mean it will work well on a abc mill. One might use a cheap 5HP VFD that spews enough EMI to compete with the local radio stations while another has a mains connected 1HP motor. There are EMI standards both for emissions and susceptibility and a lot of the equipment commonly used would fail both sets of tests and likely by wide margins.

I worked in a lab where we tested some mission critical equipment before purchasing or using. Some equipment from global powerhouse companies that you would expect to pass because of the cost and their pedigree would fail. Initially they would tell us no way not possible, how are you doing the test. They would repeat the test in their lab and get the same failed results. So now how do you think some of this equipment that has never been tested to any EMI standard stacks up? Answer in too many cases not very well.

 

[Susquatch]

My scales are Igaging as Yury recommended in the beginning of his project. I'm figuring micrometers and calipers work well, and I can trust the reading.

I recently read that Yuriy doesn't recommend igaging scales anymore but an igaging interface is still available for the touch DRO for those who use them.

 

[PaulL]

Spent some more time with the scales. It's not a TouchDRO problem, it reproduces exactly the same on my old SINO unit.

It's the strangest hysteresis: I set a DTI to measure travel of my cross slide and use a 123 block to measure off a 2 inch increment. As I move away from zero on the DTI, I get about one thou of travel measured in the first 3-4 thou on the DTI. But when I then pull off the block and traverse to the surface the DTI was up against, my DTI lands up to zero right as the DRO hits the 2" mark.
I've repeated this with both my read heads, so either the batch was bad, or the problem lies elsewhere.
I've repeated this at different places in the carriage travel, which means it's less likely to be the tape, though tape problems are still possible.
So now I'm questioning my methodology - does my DTI setup somehow induce some backlash? It's on a Noga arm and seems plenty rigid. And I've done the experiment fairly carefully thinking that might be the case, to approach the measurment only from the same side of the screw.

I'm flumoxed.

 

[Dabbler]

It is prudent to second guess your measuring setup... I know it is tedious about the 'take pictures' cattle call, but there is 100 ways to mess up a measurement, and perhaps that's not all!

Sounds like a great mystery to solve!

 

[YYCHM]

Spent some more time with the scales. It's not a TouchDRO problem, it reproduces exactly the same on my old SINO unit.

It's the strangest hysteresis: I set a DTI to measure travel of my cross slide and use a 123 block to measure off a 2 inch increment. As I move away from zero on the DTI, I get about one thou of travel measured in the first 3-4 thou on the DTI. But when I then pull off the block and traverse to the surface the DTI was up against, my DTI lands up to zero right as the DRO hits the 2" mark.
I've repeated this with both my read heads, so either the batch was bad, or the problem lies elsewhere.
I've repeated this at different places in the carriage travel, which means it's less likely to be the tape, though tape problems are still possible.
So now I'm questioning my methodology - does my DTI setup somehow induce some backlash? It's on a Noga arm and seems plenty rigid. And I've done the experiment fairly carefully thinking that might be the case, to approach the measurment only from the same side of the screw.

I'm flumoxed.

Are you saying the DRO reads one thou over a span that the DTI claims to be 3-4 thou?

 

[PaulL]

Are you saying the DRO reads one thou over a span that the DTI claims to be 3-4 thou?

Yes. But reads 2" over the 2" span, down to one thou. I'll be back in the shop shortly and will photograph the setup carefully

 

[YYCHM]

Yes. But reads 2" over the 2" span, down to one thou. I'll be back in the shop shortly and will photograph the setup carefully

Are you saying the DRO and DTI agree within 1 thou over a 2" span?

 

[PaulL]

Are you saying the DRO and DTI agree within 1 thou over a 2" span?

I set the DTI to zero and the DRO to zero on the 2" side of a 123 block. I remove the block, travel to the edge it was pressed against, and get 2" on the DRO just where the DTI hits zero. I'll take a short video.

 

[PaulL]

Ok, made a short video. In it you can see me bring the 123 block up to the DTI, zero the DRO and advance 10 thou while showing only 4.3 thou of movement. I then I advance the distance of the block, coming in at zero four thou early. It means it's not a messed up radius/diameter thing, and that the overall length is near correct, but the short-scale measurments are off.

 

[DavidR8]

Ok that's completely baffling.