Post Go back to editing

ADG419 Potential Problems

Hi team,

I'm having problems with using the ADG419. It is very inconsistent sometimes and I'm not sure what I'm doing wrong. My first problem is that the voltage drops off at the drain output despite making no connections to other circuits. In this circuit, I am feeding 3Vdc from S2 and using port 1(drain) as the output. The switch is configured to output whatever is in S1 to port 1 as port 6 is set to 0.

It is a bit hard to see, but the output starts at 3V and immediately drops to 2V. I tried conducting a test with the exact circuit in a new project and the results matched up with what I have anticipated. I'm not sure why this is a problem. 

My second problem is that when I connect port 1(drain) to the circuit on the right above, port 1 becomes 0V. It seems that port 1 matches whatever voltage I put in the positive terminal of the AD8676 as I tried putting 3V and the voltmeter measured approximately 3V at port 1.

This is what happens when I connect the port 1 to the negative terminal of the AD8676. The circuit is connected similar to the schematic shown below.

My third problem is that the switch sometimes dampens the output signal. For example, in this simulation I am running a sine signal with an amplitude of 3V to port 1(drain) as the input. I am using S1 as the output because I set port 6 to 0V. However, as shown in this graph, the switch seems to dampen the input signal. 

If you could help me clarify these problems it would be greatly appreciated.


Norman Guo

  • Hi Norman,

    You seem to have spotted an issue with our SPICE model. Just to be certain, all your observations were done in simulation, correct? Or did you test a prototype of this circuit in the lab?

    The first issue you are seeing with the ADG419.cir file seems to be caused by the simulation startup (a limitation with the ADG419 model). One way around this is to toggle between S1 and S2 at the start of your simulation, like a 1us pulse. Another solution is to use LTspice, the ADG419 model in there is more accurate, and this issue was not spotted in there. I know that the AD8676 is not in there, but the AD8675 is and it should work for you.

    It seems you have connected S1 to the op-amp circuit for your second issue and the ADG419 drain is left floating. Could this be the reason you are seeing the voltage put in the non-inverting input on the ADG419 drain?

    Concerning the third issue, Vin toggling for the DC input does not seem to work with an AC signal in my tests. For this, you should introduce a 0.2ms delay on the sine wave input. Again, this issue is not present in LTspice.

    I will discuss this issue with the team and see how we can fix the model. In the meantime, I recommend using LTspice for the highest performance simulation with ADI Analog Switches and Multiplexers.

    Yours sincerely,

    Igor Ono.

  • Hi Igor,

    I did all of these simulations in the lab with pSpice. About the op-amp circuit I accidentally posted a different circuit then what I had, my apologies. This is the circuit I was referring to. 

    By itself, 3V will appear at the output. 

    However, in the presence of other circuits the drain becomes 0, similar to this. The red line is the drain. 

    I guess this is a just a problem with pSpice. I will see if I can use LTSpice.