Hi,

how can i make a infrared proximity sensor with the Psoc Creator (And Cy8kit 001 dev-board). 

Take a look at this project done in PSOC 1, you should find it relatively easy to

port to PSOC 3/5.

http://www.cypress.com/?rID=360

Regards, Dana.

Hi

I make my first trial to make a IR proximity sensor.

I drive a IR sending led go on off.

with a adc I messure value from a Photo IR LED.

but the value of on and off are always same.

I think you want D1 flipped in its connections.

http://www.eetimes.com/design/analog-design/4009969/Understand-and-apply-the-transimpedance-amplifier-Part-1-of-2-

Regards, Dana.

I retract the prior post, I was wrong, for your bias conditions should work.

I did notice the TIA conversion gain is low, eg the fdbk R is fairly low, what is the

diodes current for its typical light input ? The sensor part number ?

Regards, Dana.

Also, the impinging light flux determines the changer in photo diode current from dark current

level to illumination, and this depends on distance, and angle entering diode, which I am sure you

are aware of.

You can set up diode and emitter on bench, reverse bias diode by a 1V, and for the distance

and angle measure its illuminated current. Then x 120K you have the voltage you could expect.

Just as a check.

Value of R_2 should establish a base current of ~ (Vdd - Vsat)/R_1 to get Q_1 to sat.

Regards, Dana

Hello Dana,

the Photo diode is a Sharp PD100MF0MPx..

the IR LED is a SFH4650 from Osram.

Some issues -

1) Diode detector looks like its repsonse is min .4 uA at 100 lx. So that

implies from dark to 100 lx a change of ~ 48 mV with a 100k TIA fdbk R.

2) To drive the emitter, it is speced at full output for 100 mA. That implies

to sat Q1 it needs a base drive of 10 ma. Specs are 4 mA for pin. So you

can either use a low Rdson MOSFET instead of a bipolar, or use 3 pins

paralled to drive Q_1 base. Each pin is rated min at 4 mA source.

3) You will have to do the optical calculations to see how much incident flux

you get on the sensor, to see if it is enough. If not either you get higher output

diode, larger area detector (more sensitive), or both.

4) You could take more gain in TIA, but offsets go up as you would expect

due to TIA bias current and dark current of diode and temp effects, bias

currents doubling every 10 degrees C.

Regards, dana.

Just for the heck of it I used a glass 1N914 as a detector, and a flashlight as a

light source, and TIA fdbk R at 1 Meg. 1N914 is far smaller detector area than

your solution, less sensitive. I got ~ 500 mV of change, but worse was pk-pk

noise which was ~ 1.2 V. Filtering would be a good idea with high valued TIA

fdbk R's.

Regards, Dana.