How to improve my photo cell

[Pax Writer]

Hi all

I've built a couple of projects which involved photo cells, and so far, I've just used an IR-led with a current-limiting resistor as a light source and a photo transistor with a pull-up as the light sensor - Sometimes with a buffer before the processor and sometimes without one.
This works okay, however, I made a test-bench where I had a propeller spinning through the photo cell, and I noticed that at around 100Hz, the circuit started having problems and the counter display attached showed spontaneous counter resets or no counting at all. This did not occur at lower frequencies.
Since the processor runs at 8MHz, it shouldn't have any problems keeping up. Nor should the display, as it consists of 7-segment displays driven by 4511s.
So I suspect the design of the photo cell needs optimisation, and I was hoping some of you users would advise me a bit.
I have attached a snip of schematic containing the cell.
What would you change, and why?

 

[Rolf]

Hi all

I've built a couple of projects which involved photo cells, and so far, I've just used an IR-led with a current-limiting resistor as a light source and a photo transistor with a pull-up as the light sensor - Sometimes with a buffer before the processor and sometimes without one.
This works okay, however, I made a test-bench where I had a propeller spinning through the photo cell, and I noticed that at around 100Hz, the circuit started having problems and the counter display attached showed spontaneous counter resets or no counting at all. This did not occur at lower frequencies.
Since the processor runs at 8MHz, it shouldn't have any problems keeping up. Nor should the display, as it consists of 7-segment displays driven by 4511s.
So I suspect the design of the photo cell needs optimisation, and I was hoping some of you users would advise me a bit.
I have attached a snip of schematic containing the cell.
What would you change, and why?

Try a different shaped interrupter, something that gives you more of a 50% light to dark ratio and see what happens.

 

[Pax Writer]

Could you be more specific? I'm not sure I understand your reply...

 

[philba]

I think he's saying play with the shape of the interruptor. are you getting full light and then full darkness on the PT?

I would play around with your pull-up resistor. That PT doesn't produce a lot of current and may be having trouble pulling the voltage down. try 47K and 100K. That will increase the sensitivity. The Schmidt triggers need a pretty good swing to overcome hysteresis.

 

[Pax Writer]

Ahh, okay. Thanks.
Currently the PT has a 1k pull-up and there's no buffer before the int-pin on the processor, so maybe that has something to do with it.
Thanks for the tip.

 

[Roff]

Ahh, okay. Thanks.
Currently the PT has a 1k pull-up and there's no buffer before the int-pin on the processor, so maybe that has something to do with it.
Thanks for the tip.

Your schematic shows 10k PT pullups, with Schmitt triggers and a diode OR gate. Is this not what you actually have? If not, why did you post it?

 

[tonyharker]

Are you doing this in the dark, ambient light or under a bench light/fluorescent?

 

[Pax Writer]

About the schematic: The posted picture came from a schematic for a chronograph, it that one does have the buffer, however, I also made som circuits for confined spaces where the buffer was cut away to save space. Sorry for the mixup, I see why that could be confusing.

About light level: The photo cell will be placed in a small tube at the end of an airsoftgun (its for a pellet counter), so in practice there may be a small amount of ambient daylight.