we used the evaluation board to test the ADUM 3190. The image on the left side shows the input cirquit of the ADUM 3190.
We removed the resistors between GND1 and GND2.
It works fine with R=1.5k.
The problem with this low R is the low input impendance. Therefore we increased R to 750k. Our source has an impendance of 71.5k.
With this high R the cirquit is no longer working.
Could that be? What do we overlook.
Alexander: The ADuM3190 is a isolated error amplifier and as such was not designed with a high input impedance. The high input impedance source of 71.5K is not a good match for the input impedance of the ADuM3190.
Thank you for the fast response. I will use a buffer in addition.
Is there any hint in the datasheet of the ADUM that I could have known this in advance?
Hi Brian, I also couldn't find this in the datasheet anywhere. I, too, use a voltage divider on the + input of the ADUM3190 (connecting the - to COMP for a 1:1 output ratio), and I have different output voltages (and input, when measured at the + pin to GND2). What information can we use to properly size the resistors in this circuit?
Hi Zac & Alexander:
I redraw the schematic as below:
the relationship between Vout, Vref and Ve is like below:
When R=750K , the Vout would be more than 10 times larger than that when R=1.5K.
We must notice that the voltage range on comp pin is only from 0.2V to 2.7V. by aimplify it 10 times, Vout is probably out of this range.