Post by thread:[OT] Vernier scales

Talk of printing encoders reminded me of a project at the back of the cupboard. I got part-way through an optical ruler, which I wanted to have very fine resolution. Originally I tried a printed transparent scale but couldn't get the lines small or sharp enough. Is it possible that I could use a sliding Vernier scale with the fixed scale, and if so can anyone suggest how to achieve the best results with a micro ? I'm thinking of a ruler with a pointer at the '0' position and a pointer on the movable section, the distance between the two displayed on an LCD. Perhaps I'm re-inventing the wheel -- is there already a product around like this ? Now that laser diodes are so cheap do you think it feasible to measure short (500mm) distances on a ruler using light interference techniques, and again, if so, what detection method could be used for a micro ?

Jinx wrote: > Talk of printing encoders reminded me of a project at the back of > the cupboard. > > I got part-way through an optical ruler, which I wanted to have very > fine resolution. Originally I tried a printed transparent scale but > couldn't get the lines small or sharp enough. > > Is it possible that I could use a sliding Vernier scale with the fixed > scale, and if so can anyone suggest how to achieve the best results > with a micro ? I'm thinking of a ruler with a pointer at the '0' position > and a pointer on the movable section, the distance between the two > displayed on an LCD. Perhaps I'm re-inventing the wheel -- is there > already a product around like this ? > > Now that laser diodes are so cheap do you think it feasible to measure > short (500mm) distances on a ruler using light interference techniques, > and again, if so, what detection method could be used for a micro ? > You can buy digital verniers over here pretty cheap these days, I've seen a pair for as little as 35 UKP. Mike

> Perhaps I'm re-inventing the wheel -- is there > already a product around like this ? Yup. plenty of cheap digital calipers around in the import tool catalogs. Up to 6" span, they are cheap, and accurate to around 0.001". ------------ Barry King, KA1NLH NRG Systems "Measuring the Wind's Energy" http://www.nrgsystems.com Check out the accumulated (PIC) wisdom of the ages at: PIC/PICList FAQ: http://www.piclist.org

Not as cheap, but there are also digital calipers with serial interfaces built in. -- Rich Barry King wrote: {Quote hidden}> > > Perhaps I'm re-inventing the wheel -- is there > > already a product around like this ? > > Yup. plenty of cheap digital calipers around in the import tool > catalogs. Up to 6" span, they are cheap, and accurate to around > 0.001". > > ------------ > Barry King, KA1NLH > NRG Systems "Measuring the Wind's Energy" > http://www.nrgsystems.com > Check out the accumulated (PIC) wisdom of the ages at: > PIC/PICList FAQ: http://www.piclist.org

Have a look at the Shimadzu site (and similar) for their non-contact extensometers Bye. -----Original Message----- From: Jinx [SMTP:spam_OUTjoecolquittTakeThisOuTCLEAR.NET.NZ] Sent: Wednesday, 10 May 2000 19:38 To: .....PICLISTKILLspam@spam@MITVMA.MIT.EDU Subject: [OT] Vernier scales Talk of printing encoders reminded me of a project at the back of the cupboard. I got part-way through an optical ruler, which I wanted to have very fine resolution. Originally I tried a printed transparent scale but couldn't get the lines small or sharp enough. Is it possible that I could use a sliding Vernier scale with the fixed scale, and if so can anyone suggest how to achieve the best results with a micro ? I'm thinking of a ruler with a pointer at the '0' position and a pointer on the movable section, the distance between the two displayed on an LCD. Perhaps I'm re-inventing the wheel -- is there already a product around like this ? Now that laser diodes are so cheap do you think it feasible to measure short (500mm) distances on a ruler using light interference techniques, and again, if so, what detection method could be used for a micro ?

Hi, wrt using a laser and interference techniques, the short answer is, you can't. Not with diodes bought off the shelf at least. The coherence length of the light must be 2 x L where L is the max length you measure plus any optical overhead. At least. The practical way to do it is to use a high quality laser (usually infrared, as it is cheaper), and build an interferometer with two detectors offset by lambda/2, a half transparent mirror, and a total reflector attached to the moving part. The outputs are sizeable if you only need to measure to lambda, and no preamp is required. You can feed photodiode outputs directly to a PIC (digital inputs). The system acts exactly like a quadrature encoder, except you can reach MHz output rates by just moving the vernier with your hand and the uC must keep track. There is also need for a zero point (close the caliper and measure 'closed' with an accuracy of 1/2 lambda ;-) ). The mechanical construction must be such that the device does not have 'play' larger than about lambda/2 if you need precision. Not your average ruler, and I don't think that the average laser diode will work here ;-) Peter

From: Peter L. Peres > The mechanical construction must be such that the device does not have > 'play' larger than about lambda/2 if you need precision. Not your average > ruler, and I don't think that the average laser diode will work here ;-) > > Peter > > Yup. plenty of cheap digital calipers around in the import tool > catalogs. Up to 6" span, they are cheap, and accurate to around > 0.001". > > ------------ > Barry King, KA1NLH The catalogues I've looked in show inside, outside, depth, height but not length per se and definitely nowhere near 500mm. Extensometers come close, but no cigar. What I was thinking of was a quick and easy way to measure a line/distance on paper/photograph to 0.1mm or better and then perform a scaling calculation. Once I get a read head working the length of the ruler is immaterial, it's only printing The idea was to use a simple slider, not a screw mechanism, and optically measure the distance by counting the fixed divisions from a zero point, like you say, by transmission or reflection and then finding the best coincidence between the fixed and the sliding vernier markings to get the decimal place. I figured with 1.0mm fixed and 0.9mm Vernier markings a reading to .1mm would be doable. As for interferometry, thanks for the mechanical advice, my Rubbish Workshop is not in that league at all at all. I think I'll have to fall back on Plan B, which involves some degree of precision. The PIC would look at 9 miniature interrupters, one across each Vernier mark, possibly using lenses to magnify the effect, to determine which has the lowest output. "Might" ;-) be tricky knocking up a prototype but it'll be fun trying.

> The catalogues I've looked in show inside, outside, depth, height > but not length per se and definitely nowhere near 500mm. Check under accessories for milling machines and lathes. There are DRO (digital read-out) units that give position data to better than 0.001" (0.01mm) over lengths of up to a meter. A friend was looking into them at an industrial machinery show. In one I remember, the position sensor (slider) was optical against pattern laid out on a fixed "ruler" (bolted to the ways). Problem is they are not cheap -- several hundred dollars for used ones up through thousands for the better grade of nice, new high precision units. Lee

>> You can buy digital verniers over here pretty cheap these days, I've seen a >> pair for as little as 35 UKP. >And mine even had a serial (TTL) data port hidden inside. 1200 Baud I take it that you bought a cheap version without interface, and found the expensive version is the same but at twice the price, with the only difference being the serial connector?

Depends on what you're looking for - Mitutoyo has a "SPC" line of serial output measuring devices that starts at around US$125. I figured out the protocol so it can't be too hard! The cheap stuff resolves .0005", .00005" available last time I looked. Very low power. Main disadvantage is speed (Quite slow) and the non-existent tech support from Mitutoyo USA, unless they smell a big order. Maybe things have gotten better... regards, Jack Lee Jones wrote: {Quote hidden}> > > The catalogues I've looked in show inside, outside, depth, height > > but not length per se and definitely nowhere near 500mm. > > Check under accessories for milling machines and lathes. > There are DRO (digital read-out) units that give position > data to better than 0.001" (0.01mm) over lengths of up to > a meter. A friend was looking into them at an industrial > machinery show. In one I remember, the position sensor > (slider) was optical against pattern laid out on a fixed > "ruler" (bolted to the ways). > > Problem is they are not cheap -- several hundred dollars > for used ones up through thousands for the better grade > of nice, new high precision units. > Lee

Alan B Pearce wrote: > > >> You can buy digital verniers over here pretty cheap these days, I've seen a > >> pair for as little as 35 UKP. > > >And mine even had a serial (TTL) data port hidden inside. 1200 Baud > > I take it that you bought a cheap version without interface, and found the > expensive version is the same but at twice the price, with the only difference > being the serial connector? No. I just bought a cheap unit for $50.00 for general use around the lab, and found that it had a PCB style edge connector hidden inside (behind a sliding cover). Poking around with a scope showed that it was sending data every 180ms. Just dumb luck that it had an async port. I have since found another no-name caliper that also has a port, but it sends out 2 24bit SPI style words instead. Presumably for an external dumb display (Shift register -> 7 segment decoder). I haven't bother figuring out the format since the serial one was easier to use. I guess you just have to go down to your local tool shop and take a look inside the case (battery compartment). -- Robert.Rolf-AT-UAlberta.ca