IR Heartbeat Monitor -- Weird

[midosi]

Hello,

I need help about some problems i am experiencing with my experiment. What i am doing is a HeartBeat monitor. Sensor is High intensity (2500 mcd) Red LED and photoresistor. I am using a LM358 opamp for amplification, Circuit is coupled with LabView 8.2 via DAQ 9 series. This is pretty simple experiment actually and i dont know what i am doing wrong, i am gonna go crazy. The circuit i use is attached. Circuit is not mine, since my experiment is not based on design i am allowed to use any circuit. In the schematics, i only used the monitoring circuit, since i have DAQ and i dont need pic and stuff. I also eliminated the diode (D7) which is attached to the 330R resistor.

The problem is, i am having a signal output that osciliates very much, a lot of noise. So much that i am not sure if the system can even read the beats. I used a bandpass filter, at some point i may have captured some meaningful signal but it has diminished somehow.

Second interesting (and Weird actually)  thing is that the output signal from pin7 of LM358 is rising continously, it was supposed to be giving constant output while the beam is not interrupted.

(if anyone get curious, my experiment's focus is on environmental effects and characteristics of measurement, so non of the info you will be giving is cheating, please help as much as you can)

Please help, any usefull ideas are welcome.

[jjcimon]

hi

  I would use a signal conditionner (opamp circuit), there is some options to add to the DAQ 9 series

may be your NI  rep could guide you to the right product

the actual problem is op amp are not so easy, trigger and sensivity and lengh of wire will make that experiment a nightmare.....

Jacques 

[midosi]

Hello,

First of all, thank you for your reply. But, circuit already has sginal conditioning features through LM358. I was trying to clearly the noise through potentometer but no good.

Do you have any idea about why pin7 output is not constant? I think the problem lies somewhere there. (In LM358 pin 7 is the outup signal. According to the model light intensity on the photo resistor should effect the voltage output at 7. But even though there is no interruption to the beam, the voltage output is not constant??)

Thanks,

midosi

[jjcimon]

the op amp is may be defect, did you try to replce it?

signal conditionning module like Analog device etc is way different than a simple op!!

what is the voltage range and the signal  impedence of your input voltage you try to measure?  specs please,

[johnsold]

midosi,

The rising output at pin 7 clearly indicates some kind of problem.  If the circuit is wired as posted, one possibility would be DC leakage current in the 470 nF capacitor.  When you eliminated D7, did you leave that branch of the circuit open?  Pin 1 should connect only to pin 5 and R13.

What voltage do you see at pin 7.  How long does it take to stop rising and at what voltage does it saturate?

If you block all light from the photoresistor, what happens?  How much stray light is getting to the photoresistor?  Fluorescent lights are particularly nasty in their optical output.  They have a nominal output at twice the power line frequency along with strong harmonics out to the kilohertz frequency range.

The value of the capacitor seems low for a heartbeat monitor.  The heart rate for humans is approximately 1 Hz (60 beats/minute), but the low frequency cutoff for the R2, R9, C7 combination is on the order of 10s of Hz, depending on the value of R9.  This will attenuate the heart rate signal will passing all of the power line frequency interference which may get into the circuit, either as stray light or as some kind of electrical coupling.

Lynn 

[midosi]

Thank you Lynn, for your reply.

I have replaced the 470nf capacitor, it is not malfunctioning. But you are right about the R2 R9 C7 combination. I have concluded that i should make some major modifications to the circuit. Escpecially to eliminate the Fluorescent noise, which is quite a lot in my lab. When i blocked all light, it just gets the background noise of the whole room. When i put my finger through it gets some signal, but it is probably because my finger alters the noise it gets, not the heartbeat effect.

I will let you know the outcome.

midosi

[johnsold]

midosi,

Your finger can act as a "light pipe" and couple some light from the room into the sensor.  Unless your whole body is in a dark place, you will probably get some coupling.

I would make C7 at least 10 times larger.  Be cautious about using electrolytic capacitors because of their leakage currents and make sure the polarity is correct. Adding a 47 or 100 nF capacitor from pin 1 to pin 2 of the amplifier will reduce the bandwidth substantially and may eliminate much of the power line frequency noise and fluorescent harmonics.

If you are still not getting a heart beat signal after that, measure the DC voltage at the R9-R2-C7 junction.  If that is very close to zero, try reducing the LED current.  If it is close to 5 V, then you may need more light or better coupling to the photoresistor.  You may be able to reduce R8, but check the maximum current rating of your LED to make sure you do not overdrive it.

Lynn 

[AnalogKid2DigitalMan]

How about using an IR LED and photodiode? That should improve your S/N ratio drastically since the photodiode will have very little response to ambient light.

-AK2DM

[aqualeo99]

Did your heartbeat monitor work.I'm working on the same circuit & i'm not getting the required output.Please reply urgently.I am going to do the changes as mentioned in the forum.If u have any more suggestions please reply.

[johnsold]

The person who posted the original question has not been on the Forum since 2008.

Also, this board is for questions about LabVIEW, not circuit design.  Some of us can do both, but you might be better off asking on the Multisim board or the Biomed board (I do not recall the exact title).

Lynn