Post Go back to editing

switching transient on ADG888

I am trying to use an ADG888 switch as a clamp across the inputs to an amplifier. The signal input is from a pair of electrodes applied to the body adjacent to a target muscle. The clamp is intended to prevent the amplifier saturating when a magnetic field is triggered alongside the nerve activating the muscle- some 300mm from the muscle. The clamp is activated a few milliseconds before the magnetic field is triggered and then released a few milliseconds after the field collapses.

The problem I am having is a spike appearing on the signal co-incident with the ADG888 switching on and off. This is without any muscle activation signal so no input signal. The ADG888 was controlled via a NAND gate, However the issue with the spike also appears when using a simple battery to control the IC. I am suspecting that it is originating within the ADG888 and is caused by the charge injection from the control signal.

In the attached cro shot the blue trace is the switching signal and yellow the  spike observed on one of the switched pins

Is there any way to prevent this or can you suggest an alternate approach to clamping the signal artefact?

Thankyou, Peter.

Parents
  • I have attached a brief schematic showing the connections to the ADG888. The device only connects across the signal input (and does not open the circuit to the amp) and clamps the signal to the amplifier to ground.If I disconnect the signal inputs and have nothing but an oscilloscope connected to S1A, S4B, D1 or D4, when IN1 &2 toggle I get the trace shown in the original Post. When the amplifier is connected this spike is amplified and the amp saturates and takes some time to recover. I do not have time to allow for settling as the signal of interest arrives within 8 milliseconds. I chose this device as I need to clamp both positive and negative going signals. I am starting to wonder if the ADG88 is not suited to this application due to the fast rise time of its switching.

Reply
  • I have attached a brief schematic showing the connections to the ADG888. The device only connects across the signal input (and does not open the circuit to the amp) and clamps the signal to the amplifier to ground.If I disconnect the signal inputs and have nothing but an oscilloscope connected to S1A, S4B, D1 or D4, when IN1 &2 toggle I get the trace shown in the original Post. When the amplifier is connected this spike is amplified and the amp saturates and takes some time to recover. I do not have time to allow for settling as the signal of interest arrives within 8 milliseconds. I chose this device as I need to clamp both positive and negative going signals. I am starting to wonder if the ADG88 is not suited to this application due to the fast rise time of its switching.

Children
No Data