Raymond Genovese
- Joined Mar 5, 2016
- 1,653
Toward offering some additional contributions to the thread:
There are many of these types of sensors. Basically, they are all working in the same general way. That is, an analog voltage out that is representative of the light received by a phototransistor (an ambient light sensor). The pulse of blood is going to change the amount of light to the phototransistor and that will, ideally, be reflected in the waveform that is used to calculate beats per minute.
The one from pulsesensor.com has probably become the most common, but the design was improved (what they call “amped”). Looking at the schematic and their text, you can see the addition of the 6001 op amp as a filter.
Reading this thread reminded me that I had purchased a cheap one of these sensors a few years ago and had never bothered to do even hook it up. I did that this morning and it works well, within limits. Mine was some kind of cheap clone.
There is a pulsesensor library for Arduino with a suite of examples and I would strongly encourage the OP to use this library and experiment to get to know the sensor – as was also suggested in an earlier post.
What you may not know, is that the Arduino IDE actually includes a very nice serial plotter so that you can visualize the raw wave form. There is even a getting started tutorial here.
Here is a screen shot of my pulse using the serial plotter from just a few minutes ago and with very little "optimizing". Those peaks are what the program needs to calculate BPM.
To the OP, I would say that there is no point in trying to use additional logic to visualize results on a display independent of the Arduino until you fully understand the sensor characteristics. For example, yes, you will get a response without your finger and yes if you simply cover the sensor with your hand, you will change the response. Movement is going to distort the waveform and you need to hold a fixed and optimal position for a certain amount of time. If you learn exactly the correct pressure to use with your finger, for example, you will start to see an easily interpretable wave form. They do work and my guess is that the one from pulsesensor is likely the best of its kind and they have done the work and development.
The ouput that the OP shows in post #8 is actually quite good. Once you get that reliably, it becomes a simple matter of scaling the threshold value of what the program (the analog in part) calls a beat. Again, pulsesensor has done a great deal to make this sensor easy to use. Additionally, you don’t have to be reliant on the library because you can see all of the code and easily create your own programs for your own uses.
There are many of these types of sensors. Basically, they are all working in the same general way. That is, an analog voltage out that is representative of the light received by a phototransistor (an ambient light sensor). The pulse of blood is going to change the amount of light to the phototransistor and that will, ideally, be reflected in the waveform that is used to calculate beats per minute.
The one from pulsesensor.com has probably become the most common, but the design was improved (what they call “amped”). Looking at the schematic and their text, you can see the addition of the 6001 op amp as a filter.
Reading this thread reminded me that I had purchased a cheap one of these sensors a few years ago and had never bothered to do even hook it up. I did that this morning and it works well, within limits. Mine was some kind of cheap clone.
There is a pulsesensor library for Arduino with a suite of examples and I would strongly encourage the OP to use this library and experiment to get to know the sensor – as was also suggested in an earlier post.
What you may not know, is that the Arduino IDE actually includes a very nice serial plotter so that you can visualize the raw wave form. There is even a getting started tutorial here.
Here is a screen shot of my pulse using the serial plotter from just a few minutes ago and with very little "optimizing". Those peaks are what the program needs to calculate BPM.
To the OP, I would say that there is no point in trying to use additional logic to visualize results on a display independent of the Arduino until you fully understand the sensor characteristics. For example, yes, you will get a response without your finger and yes if you simply cover the sensor with your hand, you will change the response. Movement is going to distort the waveform and you need to hold a fixed and optimal position for a certain amount of time. If you learn exactly the correct pressure to use with your finger, for example, you will start to see an easily interpretable wave form. They do work and my guess is that the one from pulsesensor is likely the best of its kind and they have done the work and development.
The ouput that the OP shows in post #8 is actually quite good. Once you get that reliably, it becomes a simple matter of scaling the threshold value of what the program (the analog in part) calls a beat. Again, pulsesensor has done a great deal to make this sensor easy to use. Additionally, you don’t have to be reliant on the library because you can see all of the code and easily create your own programs for your own uses.
