Pulse Oximeter


[Mike] is building his own Pulse Oximeter which uses light to measure the oxygen saturation in blood. One collateral benefit of this measurement is that pulse rate can be calculated from the same data. The parts used for the detector include a red LED, infrared LED, and a TSL230R light intensity measuring chip. As explained in the video above, each LED is shined through the tip of your finger and onto the light sensor. The IR LED is used as a baseline and compared to the red LED, which has some of its intensity absorbed by the red blood in your finger. This is a pretty approachable biometric concept so you may want to start here before moving on to more involved biometric interfaces.

[Thanks Russ]

15 thoughts on “Pulse Oximeter

  1. Too bad he didnt finish the oxygen reading part yet. As its the complexer of the two to measure.

    Also, Isn’t this part of a group-project in an electronics class/college/uni? I had this at College/Uni. Fun to work on such projects though!

  2. Mistake: Blood is never blue. Veins in the skin appear blue for a variety of reasons only weakly dependent on the color of the blood. Light scattering in the skin, and the visual processing of color play roles as well. But blue is never blue.

  3. Thanks for pointing out my blue blood mistake. Good thing that I’m in electronics instead of health :-) My post a few posts back from the one linked here has links to the background reading that I did.

    If blood does not turn blue after being consumed and being on its way back to the heart, why does one see what appears like blue veins in ones’ arms and such?


  4. Mike-
    Nice work. I have a spare TSL230 sitting around and now I’ve got something for it to do.

    This Is Spinal Tap-

    NIGEL: This is exact… my exact inner structure, done in a T-shirt. Exactly, medically accurate, see.
    MARTY: So, in other words, if we were to take all your flesh and blood and every….
    NIGEL: ..take them off…
    MARTY: …and you’d see..exact…
    NIGEL: This is what you’d see…
    MARTY: It wouldn’t be green, though?.
    NIGEL: It is green! You know, see, see how your blood looks blue?
    MARTY: Yeah, well, that’s just the vein, I mean the color of the vein, the blood is actually red..
    NIGEL: Oh, mabye it’s not green…anyway, this is what I sleep in sometimes.
    MARTY: Yeah.

  5. mike im workin on pulse oximeter as well.. im not using tso230r though. my circuit is almost ready. my problem is that im getting a lot of errorous reading when im tryin on different people with different skin colours. are you having the same issue? and if yes then what are you planning to do about it?

  6. Aryan,

    I have not played with it enough to try different people. It works on the various people in my family. It does seem to work a lot better on me than my wife. Perhaps that is an indication that the application is sensitive to personal variations.

    I did get stuck trying to calibrate the red and ir light sources in order to calculate the oxygen saturation. My plan was going to be to calibrate it with no figure by adjuting the brightnesses of the LEDs to give the same reading from the sensor. Unfortunately, I don’t have enough adjustment. Not enough current and the LEDs aren’t bright enough to go through a finger. Too much current and the LEDs are over driven.

    Best of luck!

  7. mike
    im from india so people are dark as well as very light skinned here. and the readings are not just sensitive to skin color but also varies in accordance to the thickness of the skin. that is something that i found out. so i still have to take care of that fact.

    i was initially having the same problem with leds. red was bright enough but ir wasnt all that satifactory. so im using a good quality red leds(i forgot the name of the manufacturer)but instead of ir leds im using a laser. it works just fine.

  8. Dear Mike,
    Hi. I was making a pulse oximeter along with my group and we seem to have hit a roadblock with the software. Can you please tell me which microprocessor/microcontroller you have used and if you have the code or even the algorithm for the project. We are using the PIC18f4520. Please do reply.
    Thanking you,

  9. My design project for this semester is to design a oximeter using an Atmega328p. I’m having a problem with the Led driver. I am using two different circuits, 1 is for the oxygenated blood (IR Led) and the other is for the de-oxygenated blood (bright LED). I wanted to know if Both the IR and Bright LED needs a switching frequency so that the photodetector can detect the light? Please be easy on me. Its my first design project

    Thank you

  10. What about using a discrete diode like the one from a laser diode sensor, and a separate V-F IC? Or for that matter something like an LM567? That might work. Obviously it would need temperature compensation.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.