Photodetector Pulldown

Thread Starter

robby991

Joined Dec 17, 2007
79
Hi everyone. I have a photodetector in photoconductive mode with an amplifier. It works great, except when I leave it open in the room or shine light into it it doesn't go all the way to 0V, it only drops to 700mV. I need for it to drop to 0V for the rest of my circuit to work. I actually have an LED pulsing into the detector, so each time the LED lights up the voltage drops to 700mV, not 0V. Is there some kind of pulldown network I can add to this circuit to ensure that it will always go to 0V when light hits it? Thanks for any help/ideas.
 

Thread Starter

robby991

Joined Dec 17, 2007
79
here is the circuit. The function generator produces square pulses at 1 Hz driving an LED which is picked up by the photodetector in photoconductive mode. This is then sent through an amplifier which produces square pulses which don't reach 0V, the only reach .7V. This produces a problem because these pulses are teh clock input into a logic block and the pulses need to come back to 0. How can I ensure the pulses reach 0V after being amplified from photodetector. Sorry the picture is so big I don't know how to reduce it. Thank you!
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
You have a photodiode that is connected backwards. Therefore the darlington connected transistors are always turned on. They are shown as NPN, not PNP transistors.

The minimum output voltage of darlington connected transistors is 0.8V.

You should use a photo-transistor that is connected properly to a load resistor and to a transistor. Then the output of the transistor can swing from about 0.1V up to almost the supply voltage.
 

mik3

Joined Feb 4, 2008
4,843
Hi everyone. I have a photodetector in photoconductive mode with an amplifier. It works great, except when I leave it open in the room or shine light into it it doesn't go all the way to 0V, it only drops to 700mV. I need for it to drop to 0V for the rest of my circuit to work. I actually have an LED pulsing into the detector, so each time the LED lights up the voltage drops to 700mV, not 0V. Is there some kind of pulldown network I can add to this circuit to ensure that it will always go to 0V when light hits it? Thanks for any help/ideas.
TTL should accept 0.7 volts as a low logic but anyway.

Move the collector resistor between the emitter and ground and take the output signal from the emitter and connect the collector to 5 volts. Then connect the anode of the photodiode to ground and its cathode to 5 volts via a resistor. Finally, connect the base on the cathode of the photodiode. Ensure that when the photodiode is shinned the voltage drop across the base resistor is at least 4 volts to be sure that the transistors will be off.
 

Thread Starter

robby991

Joined Dec 17, 2007
79
I am sorry, they are NPN, sorry for the confusion. I don't think we have phototransistors here, so I have to build the amplifier. This amplifier should work, I don't understand why it is not. If the LED lights, surely the TTL should work. Mik3, what is the purpose of your new configuration?
 

thingmaker3

Joined May 16, 2005
5,083
Mik3's modifications would create an emitter-follower amp. The output would reach 0V, but highest o/p would be about 4.3 Volts. The o/p will be also in phase with the i/p, if this matters.
 

mik3

Joined Feb 4, 2008
4,843
robby991,

I am wondering why you dont like 0.7 Volts a low input to the TTL chip and you want it to go to zero. TTL logic accepts up to 0.8 volts for logic 0 so it should work. In the modification i told you the output goes down to zero but its maximum output is 4.3 which is again an acceptable high logic from TTL chips. TTL logic accepts voltages down 3.3 volts as logic 1. I am talking about the standard TTL series, here is a table that shows the logic levels of some logic families

http://www.interfacebus.com/voltage_threshold.html
 

Thread Starter

robby991

Joined Dec 17, 2007
79
Mik3. Thank you for the info. I understand that TTL can accept .7 as low, so I don't understand what could be the problem. One thing I noticed is that the shape of the square pulses are irregular. The peaks are fuzzy, I don't knwo if this can have an effect on the logic chips. They look like the picture I attached. Would a simple low pass filter fix this?
 
Last edited:

mik3

Joined Feb 4, 2008
4,843
Mik3. Thank you for the info. I understand that TTL can accept .7 as low, so I don't understand what could be the problem. One thing I noticed is that the shape of the square pulses are irregular. The peaks are fuzzy, I don't knwo if this can have an effect on the logic chips. They look like the picture I attached. Would a simple low pass filter fix this?
Try a simple low pass filter (capacitor and resistor) with cutoff frequency of 0.5 MHz !! The pulses will be not as sharp as now but lot of this noise will be removed.
 
Top