Optical Sensors

[Tom Line]

Has anyone found a low-cost optical switch (say, for detecting the presence of an orbit ball) that works with the Crio control system? The only thing I have access to are switches for manufacturing environments, and they tend to be quite expensive.

[Al Skierkiewicz]

Check in Digikey under opto isolators.

[Tom Line]

Al - I'll admit I'm an ME not an EE, so when I look at these things they make little sense.

I've got photo-switches here at the plant that have a reflector and check for part present. We also use laser-switches. We also have photo-prox switches.

These all send simple on/off voltages to either digital or analog inputs.

So when I see something like this: http://www.toshiba.com/taec/componen...eet/Tlp351.pdf

I'm haven't the faintest how to turn that into ball detecting sensor. Any hints?

[N7UJJ]

Caveat: I usually don't know what I am talking about.

The opto isolator is a device that has an electrical output that mimics the electrical input. Internally, it couples the output to the input via light emitter/detector, thus there is no direct electrical connection, hence electrical isolation using optical means. Anyway, it is not what you want to sense or detect light.

[MrForbes]

There are "opto isolators" that have a separate source and detector, and these might be used for an optical sensor, but they are usually sold as optical sensors, not as opto isolators. It's just a terminology thing. Radio Shack sells an IR source and detector, but you need some additional circuitry to make it work.

That 8 pin Toshiba opto isolator can not be used as a sensor.

[Al Skierkiewicz]

Tom,
The part you reference does not have an external light path. In the types of devices being discussed here, the opto isolator prevents the introduction of errors due in a large part to circulating currents. When the controller is far away from the controlled part, they are often on different power supplies and sometimes different phases or transformers of the main distribution system. When connection is made between systems that are powered this way, any voltage differences cause this circulating current that may compromise or injure the equipment or user. By having the input circuit be an LED that is on one power supply and the detector on another power supply, the current is not able to flow between these devices but the signal is coupled by the light path internally. Opto isolators are usually specified for their isolation measured in volts, kilovolts as in the case of my DTV transmitter 32 kilovolts. The LED can be supplied by a 5 volt power supply while the detector is powered by a similar power supply that has it's ground reference terminal at -32,000 volts.
If you can tell me what your criteria would be, I will try an look through the catalog and pick a few that may work. I really like the Banner Sensors that were supplied in the KOP in past seasons. They were able to be set for distances up to a few feet and could be extended with the use of reflective tape. They were easy to interface to digital inputs and were able to run from a breaker on the PD for the supply.
http://www.bannerengineering.com/en-...M-QS18-Series/

[Andrew Schreiber]

Tom, the problem with optical sensors for detecting orbit balls is that the sensor may not see them. I would suggest using a limit switch with a light weight paddle on it. Not nearly as fancy but it works and Im sure you know how to work with them.

[Tom Line]

We considered that - an even easier way is an ultrasonic since they have a width inherent in their beam that is generally 6 - 12 inches.

But there are a couple other applications we are looking at optical / laser sensors for, and I've been unable to find them on the internet inexpensively. Still looking though.

[emersont49]

You might consider using a photo sensor connected to an analog port. Radio Shack used to have an assortment of them for a few bucks. It's been a while but I think I connected one side to the red (+5volts) and one side to the white. The resistance of the sensor should vary with light and that will, in turn, vary the voltage you see on the analog port (white lead).

Prior to connecting this to your expensive cRIO, you would want to do some bench testing on it. Connect the solar cell to a multimeter and read its resistance. Shine a light on it and see how that changes. See how much it changes if you block the sensor from ambient light.

This approach is cheap but it requires someone with some basic electrical skills (a knowledge of Ohm's law).

The problem will be that fact that the "balls" won't entirely total block the light.

[JasonF]

Originally Posted by Tom Line

We considered that - an even easier way is an ultrasonic since they have a width inherent in their beam that is generally 6 - 12 inches.

Our Team tested using an UltraSonic Device.. the Devantech one that now has an example.

It worked iffy at best during our testing, but we eventually found a decent place for it above where the ball was going to be pointed at an angle.

I would think optics would be a poor choice of sensor considering the nature of the orbit balls (Not a solid face) since the beam could go through on of the many gaps on it.

Or a mechanical Microswitch / Tactile Switch to have the ball run over it.

[EricS-Team180]

We just tried these IR proximity sensors last weekend, and plan to use them in Atlanta:

proximity sensors

I also agree with Al. The Banner sensors have always worked well for us, too.

Eric

[Gdeaver]

I'll second the choice of Sharp IR sensors. They come in other ranges in addition to the above linked sensor. Sharp also makes fixed distance on off sensors. Check Digikey and the other electronic distributers for other models. If the out put is not stable especialy with long leads, add a 100 uf cap across the 5 and ground.