Heart rate sensor device, what method use ??

Hi all!
I'm studing the best way to pick-up a rate of a heart. I found two ideas or two ways to do this, one is a voltage amplifier and other is via infrared at the fingers tips.

Now, which of both ways got more precision, i do not only need the pulse i want to get a wave form from this sensor because i want to plot it.

At polar devices, i think they needs contact with the skin, i think internaly this devices are a voltage amplifier with some band-pass filter to get only the signal from a heart beat and not noise or others signals. :o

So for start my study i need some experience from who have it!!! :wink:

i would say the infared on he finger tips hospitles use that mesthod for a reason. im guessing because more accurate and instant and you could also maby use the concept of the infared and maby try it somwhere elts. the ideas are open

(1) Electrical pulses for the stimulation of muscle groups in the heart itself
(2) Blood pressure in the arteria system
(3) An echo of the latter in the capilars

I have used both methods and I find they both are subject to noise generated while the patent moves. One is due to muscle movement and the other is due to the sensor movement.

The IR is less intrusive, I found the best results were obtained with an angled emitter and receiver mounted on the same board with the IR bouncing off the bone in the finger. I did write an article about it in April 84 of a computer magazine, but I only have access to the text at the moment I could look out the schematics over the weekend if you want.

As a medical doctor, I get hands on with both methods daily. As I see it, everything is dependent of what sort of information you want to plot. Both methods is good for frequency measuring (pulse) if the person you monitor is healthy and have a normal frequency. In some cases like Atrial fibrillation, ther can be a pulse deficit that reads out as a false low pulse when measured with a pulse oxymeter. There is no other information to be read from a pulse oxymeter than saturation and pulse; the electrical behaviour of the heart cannot be recreated from that information. Measuring voltage over the heart is more exact (shows the true frequency even in AF) and can plot as an ECG curve that can be analyzed, but is more sensetive to noise from muscles in the thorax and reqire a little more complex setup than a pulse oxymeter (at least 2 electrodes in contact with the skin).


in the fitness industry, the companies POLAR and SALUTRON are two major suppliers of boards to sense heart rate. The goal is to provide the rate per minute, not the actual waveform although obviously a waveform is picked up and filtered by the boards.

The most accurate way in fitness machines is to wear the chest strap and use telemetry as this avoids a lot of the interference from muscles. contact sensors are also used and for this we need a minimum of 3 pickups but we normally have 4. Two on the LH and two on the RH. I guess that a differential signal between the LH and RH sensors is amplified and filtered but I am not an EE so I don't know much of the details.

one way to learn is to use google patents or other patent search tool and search for patents by these two companies. of course you cannot use what you learn for commercial purposes...and I assume that's not your intent.

It would be helpful to know what you are looking to gain by plotting a waveform.

Also, it is possible to get the pulse by electromechanical methods as well.

Very thank all for the responses!.
Now i'm reading a little more about heart monitoring, i want to digitalize all the signal i can. I mean not only the frec of the beats some other details of the pulse signal will be great.

What i want to do is build both sensors, an IR and a voltage amp, i'm to start with the IR but i want to test what happend if i use it at the finish of my fingers, similar to a ring, and using relflected IR and not a Ir barrier.

What signal form give me an oxymeter?, i mean it's only a simple pulse or you can read or see some tips at the signal plot?

Best Regards!

For an oximeter, you need two wavelengths (red and IR) and compute the ratios of the peak-to-trough amplitudes between the two.

The time difference between the ECG QRS peak and the detection of the pulse at the finger is called the "Pulse Transit Time", which is the time it takes the pulse wave to travel to the periphery after the heart beats. It's related to vascular stiffness and obviously distance.

Look up photoplethysmogram on wikipedia and you'll be better informed about a pulse wave, which is very different from ECG (but does have some physiological relevance).

Hi all! .

I build a simple IR rx amp, a red and ir transmiter. the pulse from my fingers looks nice but i must work i little bit more to remove all the noises, i'm think use a signal carrier and not only switch on/off the leds.

Now reading about the electrical method, i discover a Op Amp configuration that i never use, a instrumentation config/amp. Some build it using severals OA, but someone knows a easy-to-find part that is already a IC for this application? ;D

Best Regards

Yeap you are rigth..
I want to use conventional ICs, seeing some project i found my favorite OA, LF347/LF343.

My next steps will be build the amp/filters and then study how set-up it to use an analog in from the arduino!.

Best Regards

Late to the party, but I just found this on YouTube and thought it might help.

I want to use conventional ICs, seeing some project i found my favorite OA, LF347/LF343.

You can do what you want, but I would strongly advise against using anything other than an instrumentation opamp for anything like EEG or ECG/EKG use. I may sound overcautious, but I rather value my heart, brain, and living...so be aware of exactly what the risks can be before you start strapping on electrodes and conductive gel...


I used an IR sender and diode - matched wavelength and they worked well. I built them into something rather like the finger of a glove, but made of cardboard and duct tape.

I had to do some smoothing of the readings before I could find the local max an min.

The script can be downloaded from

download the zip and then look at Project 12

Humm...paint the IR filter with a marker.... yeap that guy have less party time than meee! :wink:

My IR device looks more cool! is a paper sensor from a ticket printer:

This piece is a reflection optocupler (TCRT1000), it already have an IR filter and the diode/transistor are matched at 950nm.
The problem using this is that i can use RED light to build a complete oximeter but i think for a easy heart beat sensor will be cool. You have other reflection couplers like the CNY70:

This CNY is used a lot like speed sensor and position sensor, for rotary encoders and angle detection.

So other case to study, if someone want to test and do not have a place to buy one, remember is inside your computers printer! :o

Discard the CNY70 (at least if you're planning to use the 10-bit ADC of Arduino). I've tested just now and and it doesn't work.

Hum.. use an amplifier to get the IR signal from the receiver?, remember that for get the pulse you need a OA as comparator to get the triggered signal.

Best Regards