Payback to this group

How about a film canister, put inside a small mailer? How wide are the strips, and can they be rolled that tight?

A friend and I were just discussing ways to know the orientation of a telescope's equatorial mount. One guy wants to get 12" to 15" brass gears made and use rotary encoders. That'd work, but I was thinking that it could also be done electro-optically with a pattern on a disc or strip.

I'll ping him, see if he'd like to try this. I can generate a pattern in AutoCAD and save it as a .dwg file (R-12), .dxf (R-12) or adobe .pdf but I have no way of accurately printing it.

I wish I could figure out how those Chinese calipers get .0005" resolution. I took one apart. The features on the capacitive scale and reader are not nearly that small: the reader's strips are on .025" centers and the scale's features are .100" wide with spaces between them that look like they were routed with an .0125" dia (might be 0.5 mm dia) endmill.

>

Don, generate your Acad dwg, or I could do same. Be more than happy to plot it for you.

Ivan

Heh.. film cannister. I was wondering a few days ago how I'm going to get supplies of those. They're becoming rare! Maybe he should use cigar boxes.

Dave H> How about a film canister, put inside a small mailer? How wide are the

People that make fireworks use cardboard tubes that are pretty cheap.

The resolution is even higher. It is 1" / 20480. Theoretical. I get a few 1/1000mm out of them with overshampling and averaging.

Have you seen what I have added to my HP in the meantime? I didn't describe how the electronics work*). But I have put a lot of material how the signals look like and about the mechanical arrangement of the couppling and shielding caps.

See:

*) Because I don't know. I have only a vague picture of it: It works with the vernier principle and it seems that it does enhance resolution by messuring the coupled voltage. I _guess_. Still have to laugh when I get remembered how much fun it was to use the HP-54645D Mixed signal scope. :-) Geee, and they sayed that I can keep it as long as I want. What a gem!

BTW: Hopefully at the end of this month, kits for my digital scales interface will be available.

Nick

Nah... you can still get them by the Hefty Bag full at any full-service film processing center.

Some amateur pyrotechnicians make tiny aerial shells with them.

LLoyd

I've got 2 ammo cans full of 'em, but as such, I'm getting low (film cans, not cigar boxes).

Cool. I'll see if the guy responds. If he does, I'd be glad to send you a mailing tube with stamps on it. Just peel off the outer address label.

As a word of advice onthe encoding, look up "Gray code" techniques. the reason it's important is that in Gray code, only one bit changes at any transition. That makes reading the position a lot less prone to reading errors.

Just a thought...

Richard

Encoders these days are incremental quadrature feeding a counter, not Gray code. Gray code made sense when the electronics were costly, but not so today.

Dumpster behind carpet shop or printing plant.

But I think you'd be better off coiling it into something like a 5 x 5 x 5 inch box at $0.18/each.

If you really need tubes:

Carpet store dumpster - full of tubes.

_-_-bear

Richard, thanks, I've tried carpet tubes, way way too heavy. But your link to 5" cube boxes is the obvious answer. Thanks for the link.

Ivan

Ah. I guess I've dated myself. Just wondering if I got lucky...

Richard

Gray code was used on aircraft reporting altimeters because it was easy to determine an error in coding. If any one bit was out the thing would report wierd altitudes in sequence.

John

Why not small boxes. If you don't need 24" why buy them.

Post office has several sizes - but those are not high volume. (and a little pricey) Martin Martin Eastburn @ home at Lions' Lair with our computer lionslair at consolidated dot net NRA LOH & Endowment Member NRA Second Amendment Task Force Charter Founder

Ivan Vegvary wrote:

They were used a lot on simulators too, and the weird read outs were great for troubleshooting...

Gray code is relevant to absolute encoders, not to incremental encoders. An absolute encoder will give correct position even if the encoder was moved during a power-down situation. However, a nonredundant n-bit binary sequence can be arranged so all adjacent states vary by only one bit, thus avoiding race conditions, and it is true that inexpensive silicon (e.g., a microcontroller or FPGA) can now easly convert those state codes to a monotonic sequence of values. Example: adjacent binary states from 0 to 7 would be 0,1,3,2,6,7,5,4. Elex can sort that out. It gets messier if there are 65,536 states (16 bits). Elex can still sort it out.

Mitutoyo uses a really ingenious scheme for absolute position reading of their digital calipers. Google it if interested. It's explained (in patentese) in a patent assigned to Mitutoyo. I think the HF calipers may have ripped that off, don't recall when Mitutoyo's U.S. patent was dated. I think the HF digital calipers, at $16.99 on sale, are an incredibly good buy. I have both HF and Mitutoyo and I see no difference in accuracy, function, feel or finish between them.