Designing Pulse Oximeter Circuit

You said "... i am designing a pulse oximeter circuit" - what have you got so far?

Hello there, I appreciate the reply. There are three sections, here as follows.

1. A 15V supply (for testing will be 9V for battery use) supplies a resistor (1K2) and a bright red led (1.2V across) and connected to ground.

2. The red light triggers a phototransistor which is in series with a resistor (100k).

3. The output current rises and voltage drops (oscilloscope shows ideally a 0V to 15V drop which is what I want, scale on oscilloscope is 5V per square).

4. The signal on a digital oscilloscope show and output once my finger is placed inbetween the red led and the phototransistor (in close proximity i.e. close peg with drilled holes).

5. My altering blood flow allows light intercepted by phototransistor to vary i.e. current rises and falls.

6. The output signal is sinusoidal with noise, it calculates at 1.25Hz which is like 800ms i.e. about 1.25 x 60s is like 75beats per minute.

7. The high frequencies can be ridden by a passive low pass filter before the signal is intercepted by a schmitt trigger for a pic to pic up.

Now I am finding it hard to design the low pass filter, I have the noise at 20mV so do I now need to calculate the signal and obtain a signal to noise ratio in order to calculate the values of resistance and capacitance for the Low pass filter?

It sounds easy but that part is doing my head in ha, if you can help that would be great or perhaps give a better alternative like an IMGF filter.

A burning desire to have someone else do his work.

See todays question. It's really a hoot!

" Hi, i am designing a pulse oximeter circuit"..
What kind or which type do you have in mind? Is it the Fractional or the Functional type?

Hello there, I appreciate the reply. There are three sections, here as follows.

1. A 15V supply (for testing will be 9V for battery use) supplies a resistor (1K2) and a bright red led (1.2V across) and connected to ground.

2. The red light triggers a phototransistor which is in series with a resistor (100k).

3. The output current rises and voltage drops (oscilloscope shows ideally a 0V to 15V drop which is what I want, scale on oscilloscope is 5V per square).

4. The signal on a digital oscilloscope show and output once my finger is placed inbetween the red led and the phototransistor (in close proximity i.e. close peg with drilled holes).

5. My altering blood flow allows light intercepted by phototransistor to vary i.e. current rises and falls.

6. The output signal is sinusoidal with noise, it calculates at 1.25Hz which is like 800ms i.e. about 1.25 x 60s is like 75beats per minute.

7. The high frequencies can be ridden by a passive low pass filter before the signal is intercepted by a schmitt trigger for a pic to pic up.

Now I am finding it hard to design the low pass filter, I have the noise at 20mV so do I now need to calculate the signal and obtain a signal to noise ratio in order to calculate the values of resistance and capacitance for the Low pass filter?

It sounds easy but that part is doing my head in ha, if you can help that would be great or perhaps give a better alternative like an IMGF filter.

Are you trying to re-invent the wheel solve your problem with an analog solution? Think BIG. A Digital Signal Processor (DSP) will do the job with fewer components and (probably) a more accurate result.

You need the LED & sensor at the front end, but thereafter it's easier to go digital.

_IMHO

Comparing the analogue and digital approach is the end goal but using an analogue approach is what i am struggling with, guess i better get reading.

Based on your circuit description, what you will get is a typical waveform representing variations of a typical blood pressure wave which is called as Plethysmography.

Pulse oxymetry is the measurement of oxygen saturation in the blood and to do that, you need to be able to distinguish, measure then quantify the differences between the red LED light with that of the red blood after being exposed to the LED light. Aside from being able to modulate your red LED circuit, you also need to differentiate and quantify also the amount of red cells caught in that particular waveform segment. Now you also need to add into your algorithm separately, comparing those that are good cells against those that bad red cells.......Bearing in mind that not all red cells are capable of carrying oxygenated blood to the different organs of our body! ....

Excellent, I will take that into consideration too. But I am simply recording heart beat and wanting it to display on an LCD 4digit segment display. And I want to get rid of the noise.

"I am simply recording heart beat and wanting it to display on an LCD 4digit segment display"

--which means you made a mistake labeling your post as; "Designing Pulse Oximeter circuit"?? Either that or you just realized from the responses that oximetry is way beyond your pre-conceived notion?

Finger Pulse Oximeter With LED Display that site is http://www.amazon.co.uk/s/?ie=UTF8&keywords=pulse+oximeter&tag=googhydr-21&index=aps&hvadid=8294806209&hvpos=1t1&hvexid=&hvnetw=g&hvrand=11702103891487732630&hvpone=&hvptwo=&hvqmt=e&ref=pd_sl_7f3ksvw7wv_e

I am designing a circuit a simple circuit that does that using the elements that I have mentioned. So basically it is pulse oximeter on a small scale. I don't know how else to put it vsar?

No. You are designing a heart rate monitor. It has already been explained that an oximeter is a different animal.

Basically it is "Plethysmography" Not Oximetry! Pulse Oximetry is the determination how much oxygen perfusion as carried by blood in our body.. as explained in the following sites:

www.medicinenet.com › ... › asthma az list › oximetry index

or;

Pulse oximetry - Wikipedia, the free encyclopediaen.wikipedia.org/wiki/Pulse_oximetryPulse oximetry is a non-invasive method allowing the monitoring of the oxygenation of a patient's hemoglobin. A sensor is placed on a thin part of the patient's ...

HTF you measure that C anyway ... i mean - certain points in the body would give you electrical "Heart" even through the single terminal connected to a test anymal ? human F:) ... the [led-photosensor] would be "unknown error" source in the system ... there is always an environment noise near power grid

i designed a keyboard trigger years ago that "Roger'ed" event.KeyIsDown on first valid signal level and such hold for the timeslice dt (to trgger the key) dt=T/N ... T -- scan repeat cycle , N -- number of keys to scan for isDown <= this i purely digital of course

ok, it starting to get complicated, my results on the digital oscilloscope shows noise but it is still a heart beat pulse I just need to get calculate the resistor and capacitor to get rid of the noise. how i am going to do that is the problem.

Its best to just experiment with different values of resistors and capacitors and to also build and IGMF filter too to see if the noise lessons.