4 Replies Latest reply on Nov 24, 2017 8:25 AM by BoTa_264741

Problems with adc and differential opamp

user_2867716 Nov 22, 2017 4:11 AM

I'm using CY8CKIT 049-42xx and I'm trying to measure muscle impulses with surface electrodes. I have managed to get some kind of signal, but it is working inconsistently and dc level of the adc is different from run to run. Sometimes I can see a clear signal, sometimes nothing at all.

When it works, the dc of the signal is around 0 and signal jumps around 10-20mV, when it works badly it jumps between 50mv and 0 and it's hard to get anything cohesive out, sometimes spiking to 1024 (possibly due to bad usb connection).

I am wondering if there are flaws in my design or the way I'm using the device or is this due to bad connection and noise? I have attached the project folder. The registers show in top design aren't currently used for anything, they're just placeholder for future.

Joona R

adctesting.cydsn.zip 1.8 MB

1. Re: Problems with adc and differential opamp

user_2867716 Nov 22, 2017 4:18 AM (in response to user_2867716)

I should probably clarify that both of the inputs are connected straight to ECG electrodes manufactured by AMBU and are professional quality.
Like Show 0 Likes(0)

Actions
2. Re: Problems with adc and differential opamp

VasanthR_91 Nov 24, 2017 2:01 AM (in response to user_2867716)

Hey,

Could you tell more about the input signal nature? How is the sensor output varying
Like Show 0 Likes(0)

Actions
3. Re: Problems with adc and differential opamp

EvPa_264126 Nov 24, 2017 3:19 AM (in response to VasanthR_91)

Such a signal as the ECG requires a 1-10μF separating capacitor, otherwise you will see a strong drift.
The reason is the galvanic effect and the unstable resistance of the input circuits.
Install capacitors and wet the skin - this will greatly reduce drift.
In addition, the signal must have at least a small positive bias for correct operation of the amplifiers.
Like Show 0 Likes(0)

Actions
4. Re: Problems with adc and differential opamp

BoTa_264741 Nov 24, 2017 8:25 AM (in response to user_2867716)

If project has some value other than having fun, I would recommend switching to PSoC5LP CY8KIT-059. PSoC5 has Del-Sig ADC which is superior for low-voltage measurements in noisy environment.
I second that ADC input must be differential with one or both inputs decoupled with capacitors (10uF electrolytes will do), and one of ADC inputs has to be connected to 1.2V reference (directly or through 100k - 1M resistor for biasing).
odissey1
Like Show 0 Likes(0)

Actions

Go to original post