Linear Encoder Read Heads.

[HackerMechTech]

Hey All:

I have an old set of Sargon linear glass encoders. I've verified that the glass is still in one piece (By eye), and I have the read head. The part with the LED and photo sensor.

I don't know if I get quadrature sine waves or just quadrature square wave off the detectors. But I need the physical part of the read head that relieves strain for the cable and holds the detectors around the glass.

It'd be easier to just buy new read heads instead of starting yet another development project. But I can't seem to find anyone who sells individual read heads. Can anyone give me any pointers to vendors?

Also if this is the wrong forum sorry.

HackerMechTech

[Al_The_Man]

You may be virtually committed to obtaining the read heads from the OEM, as these are matched to the scale, the head carries a small grating identical to the scale grating itself.
If the company is still in business they should be able to supply them.
Al.

[HackerMechTech]

Ouch! There was no second grating built into my read heads. Just two PCB's one with 3 LED's and a current limiting resistor, and the second PCB has three photo sensors. I'm not sure Sargon is still around, so I may be SOL. Hummmmm

Thanks for the advise.

HackMechTech

[Al_The_Man]

I would be surprised if it did not have it, unless it was extremely coarse grating, AFAIK the only way an optical scale can be read with fine graduations is to use what is called the moiré effect which has the effect of widening the grating to block the LED.
Al.

[HackerMechTech]

I'm hip man, , that makes perfect physical sense. Sounds like someone already hacked these encoders. I got them for free so no money lost. I wonder if I can scam a hologram from my friends in Op-Sci. .... Heck maybe I can buy a holographic grating. I just need to find out the physical model my glass was based on.

HackerMechTech

[HackerMechTech]

.... Check this link out. So it seems you can get the quadrature with an offset pattern as in this link.

http://mechatronics.mech.northwester.../encoders.html

I'm still going to chase down the probability of having to have a moving grating. But I just looked at the geometry again, and there's only enough volume for one grating to fit in. Time will tell....

HackMechTech

[Al_The_Man]

Many manufacturers include the simplistic explanation, but in fact do use the Moiré effect to enlarge the 'shutter' effect.
Manufacturers that I can think of that use this are Baush & Lomb, Acur-rite, Heidenain, Renco, BEI, and many others etc.
Page down in this link to Moiré fringe.
http://claymore.engineer.gvsu.edu/~j.../cont_sens.pdf
Al.

[HackerMechTech]

Great link. Thanks! Even if mine is not based on a interference pattern, I may want to take it in that direction.

HackerMechTech

[HackerMechTech]

Okay took a good look at the pickup. Yes, there is a piece of glass bonded to the photo-diode PCB. And they look intact. Thanks for the pointers. It looks like it may be worth the effort to rescue these.

HackerMechTech

[One of Many]

Sargon has gone the way of the wind so to speak. Here is one company that may be able to provide you with a bad read head to use the hardware parts. Ebay might be another source if you have the patience.

Can Do Machinery

I know of one Sargon DRO that has got to be pushing oh 20+ years old and still working great daily. I don't think it has been turned off much. They weren't all that bad for an inexpensive Enco DRO in my experience.

It has been many years since I had one of those apart. I cannot recall how the read head board itself was attached to the mounting block. Often there is a piece of spring wire that keeps it pressed against the glass scale riding on small bearings or sled blocks. The spring wire allows a little float to the mount while keeping the read head aligned to the glass.

Have you tested the read head as-is on the scale to verify that it works before going to the trouble of getting parts?

Having another scale assembly with the required parts would help if you can swap out the boards for testing and sending pictures to make certain you get the correct parts for that particular scale type.

DC

[HackerMechTech]

Thanks One_of_Many:

I found them, Can Do Machinery that is. I kinda figured Sargon must have gone away with the few references I could find. The DRO itself still works on the digital end. I used a function generator to make sure the axies still counted up and down, and reset. But I'm not going to use it. After I make a new pickup structure I'll sample the raw output and decide where to go.

If the output signal from the moire pattern as it falls on the detectors is pretty stable, I'll gussy up a real instrumentation amp on each output and present that sig to a high level A/D and oversample the signal to boost performance. I'll probably use a small 14 bit processor and FPGA to keep track of real time count. Then I'll telemeter the data to a PC and use a standard graphics package to make a real display with multifunction. I never liked the idea of using 7 seg displays these days.

But you are correct, I found a spring wire dangling to what must have been some sort of block that slided along the bottom and took up tension for the cable connection to the DRO. It sounds like you have tones of experience with these things, if know how the bottom engaged with the glass container that would be helpful. I was debating with myself whether I needed to constrain the read head in two axies, there on the slide, or is that done with the mating structure to the mill?

Thanks for all the help guys!

HackerMechTech

[One of Many]

You tested the display, but have you connected this scale to the display to see if it is still functional? A crude setup with the sensor on the glass scale could at least verify the sensor and light source is still working. It is always suspicious when a device like this is removed from service and has pieces missing.

Typically, the cable restraint and read head mount are integral with the only coupling between the sensor head and mount being that spring wire fastened to the mount. That spring wire forced the read head against the face and edge of the glass scale. Yes, the cable restraint and mout are fastened solid to a stationary object and the spar moved with the axis. If you had another complete Sargon Scale, it would become clearer.

Normally linear encoders are shipped with plastic clips holding the read head to the spar, not to be removed until the spar is aligned to the axis and then the read head mount aligned to its mounting surface without any stress applied.

Without those clips, the read head flops all over the place and can stress the glass scale to the point of SNAP!. Or be mounted in such a way that the read head lifts off the glass scale failing to sense and therefore no output.

That sounds like a round about way to get 5v signals into a PC, but I know nothing of your application. There are a couple freeware or sharware DRO programs that use the parallel port direct to do the same thing.

DC