looking for emitter detector pair

[rbrandes]

I haven't done any 'lectronics for about 20 years and need to make something.
I used to use an emitter detector pair that was made by GE or RCA. I can't seem to find it on the web.
My application is to mount it where the beam of a balance beam scale will intrupt it and let it drive a meter in order to magnify the scale reading.
I have looked at Jameco's catalog and can't find it. Any help is appreciated.

[Externet]

Hi Ray.
If cannot find, get a discarded 5¼" floppy drive and canibalize the emitter/detector pairs. Fax machines and printers are good donors too.
Miguel

[rshayes]

The emitter detector pairs that you remember were probably made by General Electric. GE bought the RCA and Intersil semiconductor operations, merged them with their own operation, and sold the combination to Harris. Some of the old products are still available as either Harris or Intersil parts. I suspect that the GE optical products were discontinued.

Digikey lists several emitter detector assemblies made by Honeywell and Agilent. These may be similar to what you are looking for.

[philba]

Are you looking for an analog output to measure beam deflection in a narrow range?

I think most interruptors are designed to switch but I'd check them out anyway - depending on the sensitivity you need, you might get lucky. There must be something out that that does this.

However, if nothing turns up, it seems like you could build your own using a phototransistor and led. You might be able to get a graduated filter to attach to the beam to increase the sensitivity. Or use a CDS (slow response, though) which will give you a variable resistance based on the amount of light. Good shielding is mandatory.

Another approach that might work uses a radiometric Hall-Effect sensor and a magnet on the beam. The voltage output by the sensor will be proportional to the magnetic field. Not linear but no need to shield.

In all cases, you'll probably need to offset and amplify the output and then feed it into an ADC or a meter.

[Edd]

I guess that if you have used an optical interrupter in the past, you are savvy with the specific technique that you utilized in mechanical interfacing of the beam balances zero center positioning into the path of the interrupter light path.
Seems like to me that it would require the necessity of quite a very narrow opaque target between the emitter/detector elements.
I would probably think more towards the utilization of a wider target placed between two separate optical/interrupter units and those fed into a differential pair of amp xstrs, thus permitting an extraction of output for a zero center vernier meter driving.

General Instrument Q4815…..line item SWT1031…5/$1. ( Come on down!..... The price is right)

An availability of New Old Stock surplus is available here, along with the dada sheet
http://www.bgmicro.com/pdf/page5.pdf

http://www.bgmicro.com/pdf/giq4815.pdf

Note that this is a tandem pair , built inside one housing, thereby permitting incorporation of either one unit or the both of them.
I have used them for end of limit and reversal of travel functioning in some electro/mechanics that I had built in the past.


73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected].........(Firewalled*Spam*Cookies*Crumbs)

[Chris Smith]

The concept of the "matched pair" back in the 60s came from the fact that emitters and Leds weren’t well known and people would chose the wrong frequency of the Led or the photo transistor. Today all you have to do is match the frequency from thousands of choices and brands, and make sure if you use a Visible Led, that the Receiver also covers that band width.

Most "pairs" for interrupters are in the Ir range, [the lower band] and are easy to shop for and find.

For more specific frequencies above the 900 nm range you have to shop a little.

The most common frequencies are 850, 902, 1100 and 1300 nm with 1500 and 1600 being available but harder to find.

[rshayes]

The "matched pairs" were sets consistinging of an LED and a photodiode. Low output LEDs were matched with high beta phototransistors and vice versa. Actually, GE used three grades each of LED and phototransistor. This allowed then to guarentee a minimum sensitivity for the pair that was two or three times higher than that of a low output LED combined with a low beta phototransistor. The different grades of parts were indicated by paint dots added after packaging and testing.

I suspect that these same pairs were also used in GE's interruption type of sensor, since the pairs were side-looking plastic devices.

There was no need to match for wavelength. Most of the LEDs available in the 1960s were GaAs units which emitted around 940 nm. This is a very good match for silicon photo transistors and solar cells, and very little else was available at that time.

The first visible LEDs were made by Monsanto, and showed up about 1968. These were probably GaAsP. Their power output was lower than the infrared diodes, and their match to silicon detectors was fairly good, but not as good as the infrared diodes. These were mostly used for visible indicators and numeric displays.

The longer wavelength LEDs are much more recent, and are used where their specific wavelength is needed for other reasons. For example, wavelengths in the 1500 nm range are used to match an attenuation minimum in low loss optical fibers. Shifting the wavelength to this range extended the range of optical fiber links by a factor of ten or more. The photodiode material also had to be changed, since this is well beyond the spectral range of silicon.