Hi!
I have designed an EMG recording system with the ADS1298 and an input filter. The first stage of my input filter is an AC coupling followed by a passive low pass filter with fc=500Hz. I was able to record the potentials at the skin's surface and signals from a Function generator. But I have observed that after some time the baseline of the recorded signal (value provided by the ADS1298) goes slowly to cero. Then the signal crosses the zero voltage line and the negative part of the signal disappears. Finally, the baseline goes further in the negative part and the whole signal disappears and I only get 0 Volts out of the ADS. I can observe the same phenomenon when the baseline is negative and goes to positive values.
If I touch one of the active analog input of the ADS1298, the signal baseline increase, by touching INP or decrease by touching INN and I get again correct amplitude values from the ADS until the signal crosses zero, as I described before. I also use the RLD, but when I connect the RLD electrode, the signal goes suddenly to 0V and the ADS read 0V.
The problem must be in the AC coupling stage. When I measure without it, this problem does not happen. But my question is, why? Is there a problem with the AC coupling or high pass filters and the ADS? I think the problem might be in the Input Common-Mode Range. Is there a way to avoid this problem?
The VDD = 3 V, the input is in differential mode, the registers are configured as follows:
(CONFIG1, HR_SAMP_1_KSPS);
(CONFIG2, 0x10);
(CH1SET, PDn);
(CH2SET, ELECTRODE_INPUT | GAIN_1X);
(RLD_SENSP, RLD2P);
(RLD_SENSN, RLD2N);
(CONFIG3, PD_REFBUF | 0x40 | RLDREF_INT | PD_RLD);
Sorry for my poor explanation but I can only describe what I observe.
