Hi All,

Ive got a question on phototransistors. Im attempting to shine a flashing led onto a phototransistor and get the digital output from the transistor. Is there a maximum frequency that the phototransistor can react at reliably ?. For example i I did 5 flashes a second, can I expect it to output the 5 pulses accurately ?

 

Depending on how it is biased, a photo transistor should follow the flash rate up to hundreds of kHz and maybe even low Mhz.

 

Depending on how it is biased, a photo transistor should follow the flash rate up to hundreds of kHz and maybe even low Mhz.

Thanks thats better than i expected. So if I connected it forward bias, And I flashed an LED at it at a rate of 20khz say, representing 010101 etc. Would the output of the transistor be outputing 10101 at the same freq of 20khz?

 

The data sheet on the components can give you all the operating characteristics. If this is on a breadboard you might encounter some stray capacitance that would have an effect on your circuit.

 

The devil is in the details such as circuit configuration.
The maximum frequency will depend also on the frequency response of both the transmitter and the receiver. Frequencies in excess of 1MHz or even 100MHz is not unheard of.

 

The devil is in the details such as circuit configuration.
The maximum frequency will depend also on the frequency response of both the transmitter and the receiver. Frequencies in excess of 1MHz or even 100MHz is not unheard of.

Thanks. Would you expect the output of the phototransistor to be the same frequency as the led ?

 

Thanks thats better than i expected. So if I connected it reverse bias sorry not forward bias, And I flashed an LED at it at a rate of 20khz say, representing 010101 etc. Would the output of the transistor be outputing 10101 at the same freq of 20khz?

 

Thanks. Would you expect the output of the phototransistor to be the same frequency as the led ?

Of course the frequency would be the same and in synchronization? Why would it not be?

 

A phototransistor is a simple energy converter. It has no signal or data processing capabilities. In the standard common emitter configuration it is inverting. That is, as the light intensity increases the voltage at the collector decreases. If the transmitter is a simple LED driver, then a logic 1 turns on the LED, which turns on the phototransistor, which makes a logic 0. Other than this simple inversion, there are no logic signal changes.

Note that while what I described is a very simple system, biasing the phototransistor and shielding it from ambient light interference are critical to getting it to work properly. For example, a phototransistor sees a fluorescent light as a 120 Hz strobe light.

ak

 

Of course the frequency would be the same and in synchronization? Why would it not be?

Thanks. I just had a feeling the transistor might not switch its output quick enough. I might want the distance between the LED and the phototransistor to be quite significant around a meter. Do you think transmitting it through a fibre would still mean the led and the photodiode can reach each other well enough ?

 

A phototransistor is a simple energy converter. It has no signal or data processing capabilities. In the standard common emitter configuration it is inverting. That is, as the light intensity increases the voltage at the collector decreases. If the transmitter is a simple LED driver, then a logic 1 turns on the LED, which turns on the phototransistor, which makes a logic 0. Other than this simple inversion, there are no logic signal changes.

Note that while what I described is a very simple system, biasing the phototransistor and shielding it from ambient light interference are critical to getting it to work properly. For example, a phototransistor sees a fluorescent light as a 120 Hz strobe light.

ak

Hi thanks for your response. Do you think inserting the photo transistor into a fibre would prevent it from being affected from ambient light?

 

Depending on how it is biased, a photo transistor should follow the flash rate up to hundreds of kHz and maybe even low Mhz.

Can't remember the speed thresholds, but as a general rule; photo diodes are faster but less sensitive, while photo transistors are slower but more sensitive.

If you can still get them - photo Darlingtons are even more sensitive - but even slower.

 

Can't remember the speed thresholds, but as a general rule; photo diodes are faster but less sensitive, while photo transistors are slower but more sensitive.

If you can still get them - photo Darlingtons are even more sensitive - but even slower.

I think as a general rule for discrete parts this is pretty true. In the integrated world, however, it is entirely possible to use photodiodes to do single-photon event counting.

 

I think as a general rule for discrete parts this is pretty true. In the integrated world, however, it is entirely possible to use photodiodes to do single-photon event counting.

That doesn't appear to be what the TS is trying to do.

 

Agreed. At the speeds that the TS seems to be talking about, it's hard to imagine any photonic device that can't respond quickly enough.

 

Agreed. At the speeds that the TS seems to be talking about, it's hard to imagine any photonic device that can't respond quickly enough.

Years ago I was messing about with a 10.7MHz ceramic filter and decided to add an FM radio IFT to make a proper oscillator.

It didn't seem to be doing a lot, so I added an inverse-parallel pair of red LEDs across the secondary pins - there was a barely visible glow on the die inside each LED, and it blotted out my FM reception.

 

Agreed. At the speeds that the TS seems to be talking about, it's hard to imagine any photonic device that can't respond quickly enough.

Right thanks. My data light source is 20khz. This however gives a wavelength of 15km which is huge. Looking at datasheets for some phototransistors, most of them are optimal at around 1000nm which is way off 15km. Hence why I am concerned . Comments?

 

Hello,

The 20 kHz is most likely the modulation frequency, wich is completely different to the IR wavelenght.
The carrier ferquency will be the one of the IR wavelenght.

Bertus

 

Hello,

The 20 kHz is most likely the modulation frequency, wich is completely different to the IR wavelenght.
The carrier ferquency will be the one of the IR wavelenght.

Bertus

Hi,

Yes the modulating frequency is 20khz. Im going to connect this to an LED and insert the led into a fibre. The carrier will then be light. Do I need to calculate the wavelength of the carrier then ? Which in this instance is light

 

Right thanks. My data light source is 20khz. This however gives a wavelength of 15km which is huge. Looking at datasheets for some phototransistors, most of them are optimal at around 1000nm which is way off 15km. Hence why I am concerned . Comments?

A light source being modulated at 20 kHz in now way means that the wavelength of the light is 15 km. You are confusing very different concepts. The wavelength of the light is related to the color of the light. If you take a red laser pointer and flick it on and off, does the color change? Also, according to your reasoning, if you had a light source that was on continuously it would be DC and have an infinite wavelength. Does that make sense?