Don't touch that base lead! (Photo Transistor)

I want to play around with some phototransistors and am curious about the base lead. It seems like this lead is treated as taboo . Some sources say it's just there but you won't have to or don't need to use it. Other sources say that it is there so that you can bias the transistor but don't go any further.

What effect does applying a bias current have on the function of the phototransistor? How is the characteristic curve different from a normal transistor when it is exposed to different levels of light? Saturation, active region, cutoff ...

Another experiment that I wanted to try is to cut the top off of a power transistor to make a homemade "power" phototransistor or combine it with a LED to make a homemade "power" optocoupler.

Any help would be greatly appreciated. Thanks

Reply to
jalbers
Loading thread data ...

--
http://www.edn.com/article/CA6378084.html

JF
Reply to
John Fields

Light passing through the silicon produces holes and electrons somewhat like forward biasing the base emitter junction does. So light is just a source of base emitter junction bias current. You can use the base lead to add or subtract from that light generated photo current. If you can't imagine a reason to do that, then don't.

But if you want some ideas, consider a transistor used as a linear amplifier. You bias it into some mid conduction state so that a small additional signal cap pass through the transistor and be somewhat linearly amplified. In the case of a photo transistor, the light signal can be either the bias generator, or the signal to be amplified. there are uses for both cases. Think about how a base to emitter capacitor would affect the response of the transistor to a pulse of light. Or a capacitor connected in series with a resistor. Many useful variations are possible.

Also, a gate combines two or more inputs to produce an output. The light input and the direct base bias can be looked at as two logic input signals and the collector current (or voltage) as the gate output. Think light activated flip flops or electrical enable signals for a light signal.

I think your idea to open a power transistor case (especially one in a metal can) is an excellent educational idea.

--
Regards,

John Popelish
Reply to
John Popelish

John Popelish wrote: (snip)

(snip)

Should read, "You bias it into some mid conduction state so that a small additional signal can pass through the transistor and be somewhat linearly amplified."

--
Regards,

John Popelish
Reply to
John Popelish

Just be careful when you cut open the power transistor, as some of them have beryllia ceramic substrates and the dust can be hazardous. I had thought it was extremely toxic, but the following MSDS indicates that it is not too bad:

formatting link

Paul

Reply to
Paul E. Schoen

A resistor from base to emitter, hundreds of k ohms typically, will reduce sensitivity, reduce leakage current, increase speed, and increase breakdown voltage. I do that on optocouplers sometimes.

An uncapped power transistor probably won't conduct much photocurrent, so really wouldn't be a "power" phototransistor. A regular TO-5 transistor would (I'm guessing) work about as well. Phototransistors are designed to expose a lot of silicon to light, but regular transistors aren't.

LEDs are photosensitive, too.

If you zener the base-emitter junction of some bipolar transistors, they can emit a little bit of white light. Fast CMOS logic emits light when it switches, but not many photons.

John

Reply to
John Larkin

This could be an awesome opportunity to do some experiments, and actually find out! All you need is a DC supply, some resistors and pots, and a meter or two. And maybe some switches. ;-)

Then you'll _really_ learn what the base does!

I once hand-drew the transfer function of a 4N126 or something, and was astonished to discover how linear it was. ;-)

Have Fun! Rich

Reply to
Rich Grise

(snip)

Another thought. If you need real speed, you can tie the collector and emitter together and use them as the cathode of a photo diode and use the base as the anode. Or tie either the collector or emitter to the base to use only one of the junctions. I suspect the base emitter junction generates most of the photo current.

You get no gain from the transistor, only the actual photo current out, but you can either hold the voltage across this "photodiode" very near to zero with an opamp to greatly reduce the effect of junction capacitance on the output, or reverse bias this "diode" to greatly increase the speed that the photo generated charges are swept through the device and amplify the photo current externally.

The rise and fall times go from microseconds (or 10s of microseconds) to 10s of nanoseconds, allowing much faster signals to be coupled through the device.

-- Regards,

John Popelish

Reply to
jpopelish

Two other tricks to speed up a phototransistor are to use a small load resistor (the output drops from volts to millivolts, but gain is cheap; use a comparator), and to connect as a cascode with a second (low-Miller-capacitance) transistor. In both cases, the low impedance presented to the transistor keeps the collector dV/dt low, which is important because of the significant collector/base capacitance of a phototransistor. The base area is the light-collecting limit, so phototransistors have broad base areas, which means high capacitance.

Reply to
whit3rd

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.