help for sensor

[sam288]

I want to build an optical sensor using photodiode/phototransistor pair (photoplethysmography) for measuring heart rate...
In this, the photodiode and phototransistor will be kept on either side of a finger or ear lobe........
Can anybody suggest which photodiode/phototransistor will be suitable?.......
What signal conditioning will i require?.......
Where can i get the circuit?......

 

[KMoffett]

There are lot of circuits out there. A quick Google got this: **broken link removed**

For more, Google: IR heart rate sensor circuit

Ken

 

[duffy]

Wow, this is like the 3rd or 4th post about this in the last couple of weeks, isn't it? Must be a school project.

 

[blueroomelectronics]

It's the #3 project in 3rd world schools it seems.

1. GSM to 8051
2. UPS
3. ECG

 

[KMoffett]

Seems like there should be a "School/University Projects" forum to simplify their searches. Wait! What am I talking about? They don't do searches!

Ken

 

[audioguru]

Look in Google for HEMON (HEart MONitor circuit).

 

[sam288]

Yes its a semester project we selected ourselves because it interested us....neway..i was wondering which opamp should we use in the sensor...
LM295 is too costly....
Will lm741 do the job...as we just want pulses and precision is not very important..what about LM358 ?
I was thinking of giving the pulses to a schmitt trigger(to get a square wave) before giving it to the microcontroller for counting...
I hope the idea will work..

 

[duffy]

The LM358's a fine chip, the LM3162 is better. Don't use an old 741 for anything, time to put them out to pasture.

 

[audioguru]

The dual LM358 and quad LM324 are extremely slow because they are the first low power opamps ever made. At 28V p-p their max output is up to only about 2KHZ BUT THEIR DATASHEET DOESN'T SAY SO. They have severe crossover distortion.

MC3317x opamps have the same low power but have a full power output to 35kHz and NO crossover distortion. MC3407x opamps have full power output to 100kHz and also have no crossover distortion.

 

[duffy]

I can confirm the lousy frequency characteristics of the LM324 and its lying spec sheet.

And I think maybe I'm confusing the LM358 with the LF353. Been a few years since I bothered with either.

 

[audioguru]

An LM358 dual opamp is half of an LM324 quad low power opamp.
Their datasheet shows that with a 15V supply and an output of only 13.5v p-p the highest undistorted frequency is only 5kHz and at full output of about 24V p-p the highest undistorted output is only 1kHz.
The datasheets hint about their high crossover distortion which is measured at 3%.

Better opamps have a full output bandwidth of at least 100kHz and no crossover distortion.

 

[sam288]

Hey i managed to build the sensor using an IR led and phototransistor... its working fine...and giving me pulses as output...but still there is a lot of high frequency noise...
Can anyone suggest what should be the cut off freq if i pass the o/p through an LPF....so as to get fairly accurate pulses??
Also should i use a schmitt trigger or a simple comparator...before giving it to a microcontroller...?

 

[audioguru]

its working fine...and giving me pulses as output...but still there is a lot of high frequency noise...

The lousy old LM358 is one of the the noisiest opamps ever made.

Can anyone suggest what should be the cut off freq if i pass the o/p through an LPF....so as to get fairly accurate pulses??

A lowpass filter cuts harmonics. Then pulses become a sine-waves. Not accurate.

Also should i use a schmitt trigger or a simple comparator...before giving it to a microcontroller...?

You need to use a newer better faster opamp.