Can I use a logic IC instead of a transistor?

I made a sensor consisting of:

an inverted photo diode connected to the NPN base with a feeback resistor of 470k (between NPN base and collector), and the NPN collector pulled up to 5V with a 10K resistor. NPN emitter is grounded.

Could I achieve the same results if I connected the output of a 74HC00 nand gate to the two inputs of the nand gate through the same feedback resistor and omit the pull-up resistor at output? or are there properties to this gate that won't allow the same effects?

Basically I'm making a sensitive light sensor.

Don't think that would work, unless there is a very large change in light level. For a certain range of light levels the gate would just oscillate at high frequency. You certainly wouldn't get the same results.
If your sensor output only needs to be on/off, I'd try using a suitably-biased gate from a 74HC14, 74HC132, CD40106, or CD4093, all of which have Schmitt trigger inputs. The feedback resistor wouldn't be required.

Another option to consider might be to use an op-amp. This would allow for pretty high gain/sensitivity and give you greater flexibility around feedback and/or hysteresis ranges, or even be configured as a comparator depending on the kind of output you need.

alec_t said:

Don't think that would work, unless there is a very large change in light level.

Click to expand...

There will be a large change in light because I'm making a game where the scene is dark with maybe a few dim lights (not bright enough to be able to read a book). and the things that will activate the sensors are lazers and IR emitters.

For a certain range of light levels the gate would just oscillate at high frequency. You certainly wouldn't get the same results.

Click to expand...

That can be a bad thing for me because I'll also be transmitting low speed data with this method as well. The speed will be at most 300bps.

If your sensor output only needs to be on/off, I'd try using a suitably-biased gate from a 74HC14, 74HC132, CD40106, or CD4093, all of which have Schmitt trigger inputs. The feedback resistor wouldn't be required.

Click to expand...

Hmm... would schmitt trigger prevent data from going through the sensor at 300bps?

mik3ca said:

Hmm... would schmitt trigger prevent data from going through the sensor at 300bps?

Click to expand...

Not at all. Most logic chips should have no trouble carrying data at several MHz, let alone in the hundreds of Hz. I'm guessing the limiting factor in this case will be the response of the photodiode.

An unbuffered CMOS inverter makes a pretty good high-input-impedance inverting amplifier, but usually for AC-coupled signals only.