ECG/EKG circuitry help

[jnehc]

I have done my research and recently read this thread https://www.edaboard.com/threads/128582/. Props to aredhel and audioguru! I am also trying to do the same thing, but I am using TL084 as my op amps. Anyways, I'm looking at the schematic provided by aredhel, and was wondering what is the gain of the IA in this schematic . Since the resistor connected to the AD620 sets the gain, and there is a DRL connected to the AD620, is the gain still from just R1, or the sum of R1, R17, R18? And isn't gain from the sallen keys 1+(R14/R15)? I'm asking this because in the thread I've provided [post #50 page 3], audioguru was calculating the gains and it seems from his equation that he did 1+(R15/R14). For EKG, I would want the total gain of the circuitry to be at least over 1000 right? Also when connecting the electrodes, do I put the V- in the left chest (near armpit), and the V+ on the middle of my chest? Also! I'm planning to connect this circuit to arduino, and to do proper ADC i would need a DC level shifter right? Much appreciated for the help, and once again thanks for the thread aredhel!

I'll be testing the results on a 'scope, and realized that it might be different once it goes through arduino and onto a screen, right? From my understanding, the gain of the schematic provided by aredhel is very low.

 

[thylacine1975]

Hiya jnehc - I designed a simple ECG amplifier using cheap/garden variety parts (such as the TL074/084...) a handful of years ago - here's the schematic

It's fairly self explanatory and similar in concept to the .gif you referred to - but without the hassle of a dedicated instrumentation amplifier and with the added bonus of patient/PC isolation. (Dedicated DC/DC converter bricks are so commonplace now that you could dump the ringing choke converter I used and use something packaged and easy, such as https://au.element14.com/murata-power-solutions/nma1212dc/converter-dc-dc-dil-1w-12v/dp/1021438).

Like the previous design, yes, it's intended to be connected directly to an oscilloscope. To connect it to a microcontroller ADC, add a constant DC voltage (ideally ~Vref[ADC] / 2) to the output before sampling and you'll be set!

Cheers

 

[jnehc]

thanks for the response! I'm trying to use a circuit design similar to the one i posted. So for patient safety, I would need to include the two diodes and each electrode correct? where would I put the two diodes in my schematic? would it be at pin 3 and 2 on the ad620 ? How do i design a constant DC voltage? I thought I needed a level shifter so that there's no negative values for ADC. And on the thread I included, audioguru states how I must keep the gain of the filters low, is that true? I'm planning to set the gain of the ad620 to 20, would that be too high? Thanks for the info!

 

[thylacine1975]

Hmm - two back to back diodes between pins 2&3 of U1 would go some way to protecting the patient from a failure in the input stage of the AD620 that presents a differential voltage across its' inputs, but it won't do much for a a failure which raises LA and RA above ground in a common mode fashion. Perhaps back to back diodes from LA - ground and RA - ground might be a better plan. Ideally though, the whole circuit should be electrically isolated from the PC/microcontroller via optocoulpers, isolation amps etc... (but I never bothered with such details when I was the only one my circuits were being connected to!

You could implement a level shifter (= fixed DC offset) easily by making U4 a summing amplifier and adding a voltage at that point. Alternatively, you could impose a small DC offset right at the start of the amplifier/filter chain by making R12 terminate at a small voltage ~1/34th (the reciprocal of the gain of the subsequent stages) of the desired output offset instead of ground. i.e. if you want a 2.5V static offset at the output, feed the lower end of R12 with ~73mV (from a voltage divider) instead of ground. While a potential source of drift, it's easy to implement

A gain of 20 sounds reasonable for the first stage to me - I really don't have the experience to comment on optimal gain values in this application - but I'd start with that and see. Changing it is only a resistor!

 

[jnehc]

thanks for the help so far! much appreciated. i meant back to back diodes on pins 2 and 3 each, thats what you mean too right...? lol sorry. for the level shifting which would be the most simplest way? I'm not trying to buy more parts than i already have :-( . im confused by when you say add a voltage. would a voltage regulator do anything?