I'm building an EMG circuit and have chosen to use the AD8237 as my front end amplifier.
It's setup as below:

Front End

R20 is actually a digital POT so I can change the gain. The amp is setup with a gain of ~ 80. The following signals are hooked to the following places:

  • VSS - Negative terminal of (4.2V) LiPo Battery
  • VCC - Positive termainal of (4.2V) LiPo Battery
  • GND - Ground electrode (connected on bone on elbow), also see diagram below
  • IN+ CH1 - Positive electrode, placed on bicep muscle
  • IN- CH1 - Negative electrode, placed on bicep muscle about 3 cm away from the positive electrode

The GND signal is also setup as below in my circuit:

Gnd Reference

Just a simple buffered voltage divider to give a voltage half way between Vcc and Vss. Although as a side note this is showing a 0.5V offset from what I expect.

When I have the circuit hooked up as described, with the amplifier gain set to about 80, the output signal (measured at DiffAmpOut_CH1) is a 50-100mV pk-pk signal with a 1V offset like below.


There are more amplification/filtering stages after this however I'm focusing only on this section of the circuit as they are unimportant until this first stage is working. My problem is that I need more like a 1-2v pk-pk signal not a 50mv pk-pk signal. To achieve this I'll need to remove the DC offset. The offset at the output is 1V with 80x gain on the amp so the offset at the input to the amplifier must be around 12.5mV. This amplifier has very good CMRR (106dB) so it seems unlikely that the issue is caused from a common offset. It seems like one of my inputs is always 12mV higher than the other. I'm not sure what could be causing this however.


1 Answer 1


I'm not saying for definite it is the TL084 but it isn't recommended for supplies as low as 4.2 volts. The data sheet recommends 5V as the minimum.

Also, the common mode input range is typically down to -12V on a +/-15V rail and this means you should not expect the device to operate correctly if the input is within 3V of the negative rail.

How does this translate to your circuit - well the negative rail is, in effect -2.4 volts at best so, the half rail resistor network is breaking this rule for the device.

Output from the TL084 is also limited to well-within the supply rails. Under extreme conditions of loading and temperature this might be +/-10V on a +/-15 rail. Clearly this will cause a significant problem when running on such a low voltage as the LiPo.

  • \$\begingroup\$ I'll try with a wider input range and also with a supply generated offsest voltage to replace the TL084 and see how it goes. Thanks, I didn't notice the issues with that chip. \$\endgroup\$ Sep 7, 2015 at 21:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.