Project: Bio Amplifier

Several organs of the human body are known to generate electrical signals that tells a lot of things about our body functions. For example, our heart generate periodic pulse patterns that can tell how well our heart is doing. This is the basis of ECG/EKG analysis. Our brain cells generates multitude of signals collectively called EEG signals. By electronic means of acquiring the patterns of EEG signals, brain activity can be monitored and studied, without the need to open someone’s skull.

Acquiring bio signals is not exactly an easy task. Extremely weak signal generated by the human body necessitates the use of amplifier with very high gain.  And that is the easy part. What is really troublesome is the existence of stray signals, notably the strong 60Hz from the power line that is coupled to our body in many ways, and are usually far stronger than the body signals we want to view. Special circuit is needed  to pickup the bio signals from among the unwanted signals.


Safety First

Figure 1. The AD202 isolation amplifier has inputs and outputs physically and electrically separated from each other.

Probing a patient with AC powered equipment is a dangerous activity. Any AC side fault can cause AC mains current to leak through the probing amplifier circuit. The unlucky patient will be subjected to his shock of a lifetime, and heavens forbid, could cause serious injuries, even death.

Leaving nothing to chances, I built the bio amplifier using a component with built-in safety – a AD202 isolation amplifier. Isolation amplifier are especially built OPAMPs with input circuits electrically isolated from its output. Although the isolation OPAMP essentially works like an ordinary OPAMP,  the lack of physical connection between the input and output makes it easy to build circuit safe from AC mains faults. Oh yes, you are right, this component is expensive, but inarguably worth it.




BioAmp circuit

I mentioned that one of the problems we must deal with when measuring bio signal is the existence of unwanted signals that mixed with the weak bio signals. Fortunately for us, the unwanted signals appear as common mode noise signals. This makes our job much easier. What we need is an amplifier that has a high common mode rejection ratio CMRR, and through differential probing technique, we can actually extract the weak bio signal even if it is swamped by common mode noise!

That’s how our prototype circuit came into existence. The partial schematic showing the isolated input conditioning circuit  is shown in Fig. 2. The output part, on the other hand, is shown in Fig. 3.

Figure 2. Partial schematic of the BioAmp showing the input circuitry.

The AD202 input is equivalent to a one opamp block, with an inverting –input, non inverting +input, and output FB. Together with external dual OPAMP LM358/NJM2904, a classic instrumentation amplifier circuit is formed. A properly built instrumentation amplifier gives the high CMRR performance we wanted, and its gain can be easily adjusted,  using only one trimpot for the gain adjustment. To get the best CMRR performance, the resistors used in the circuit must be very closely matched. Hence, 1% precision resistors are used. Actually, even with the use of 1% precision resistor, we can not get the best CMRR the circuit can offer, 1% precision is still too loose. But we have little chance of getting higher precision resistors from our local (Philippines) electronic shops. Nevertheless, even with 1% resistors, the circuit performed respectably, as will be shown in the later part of this blog.

Figure 3. The other half of the BioAmp schematic, showing the output side and power circuit.

Power to the external OPAMP is tapped through the isolated power source –Viso and +Viso, with the isolated common ISOCOM  giving the isolated GND reference. Power to the whole circuit is supplied through the +15V and COM pin of the output side of the AD202 isolation amplifier. It is important to note here that the input side isolated supply lines have no galvanic connection to the output side (i.e. input circuit is completely isolated from the output circuit). The output side can float to as high as 2000V relative to the input without harming the circuit or posing danger to the patient.  Do not connect the input isolated common to the output common, you will loose the isolation protection offered by the isolation amp circuit. In fact, do not connect any part of the input circuit to the output circuit.

Figure 4. The BioAmp prototype.
The BioAmp prototype is built using a perforated prototyping PCBs with uncommited (unconnected) pads. Although this uncommitted pads makes wiring somewhat more tedious, it does not impose a restriction in the placement of parts and wirings.

Figure 5. Prototyping board with uncommited pads allows the experimenter to arrange freely the components.

Figure 6. Disposable ECG probes ensures good connection between the probe and the patient’s skin.

Probing

The importance of probe for a successful bio signal acquisition cannot be overstated. The probe is the front end interface between the circuit and the patient. If the probe does not do its job well, nothing useful will ever come out of the output.

Use probes manufactured for this purpose. Disposable and ready to use ECG probe contacts have highly conductive and sticky gel surface that gets the job done quite nicely, but will set you back for 60 pesos each. I actually experimented with several homebrew contacts, but without the conductive gel, not one gave satisfactory result. Poor contacts introduces so much imbalance at the  bio amp input, degrading the CMRR badly to the extent that all you can see at the output is some barely recognizable bio signal riding on a strong 60Hz interfering signal.

Figure 7. Trying out the BioAmp on myself to produce a view of my own ECG pulse pattern.

ECG amplifier

The Bio Amp, without any modifications, can be used as a simple ECG/EKG amplifier. Testing it on myself (I could not convince the ladies in our office), I attached the P1 probe on my right chest and the P3 probe on my left. P2 is attached on my right leg. With this setup, the BioAmp gave a clean output as shown in the oscilloscope trace in Fig. 8. Be warned that my probe placement may not be a correct one. As far as ECG probing is concerned, I don’t have the slightest idea to what am I doing. I did this to just to test the BioAmp only.

Figure 8. This is the output of the BioAmp captured with an oscilloscope showing pulse patterns that looks like the ECG pulses. I hope this pattern indicates a heart is in good shape.

Known Issues

I picked a low cost general purpose OPAMP for the input conditioning circuit to keep the cost low. General purpose OPAMP do not excel when it comes input offset voltage drifts, and I gambled with the hope the drift stays within tolerable range.

Gain adjustments causes a shift in the DC output offset, you have to check to ensure the output does not go in saturation whenever you adjust the gain.

What’s Next

I will make a simple DAQ system that will allow you to view the BioAmp output waveform using a PC. Visit this blog again for updates.