adum3190 simulation model

Hi, 

I moved your question on the ADuM3190 to the Interface and Isolation community. Someone here should be able to assist you. 

Thanks, 

Kris

 FEIYANGZHU:

We have a simple model for the ADuM3190 that I could share with you by email if you like.  It is designed to show the behavior of the ADuM3190, but is not designed to show the accuracy of the amplifier, so the characteristics of gain, offset and linearity are not shown as they would be in the datasheet.  The datasheet specifies the gain, offset and linearity, which have more than 10% error when used as an amplifier in an open loop, which is very different than when used as an isolated error amplifier in a closed loop, which has 1% accuracy.  What is your application?

Regards, Brian

Hello,

I'm still confused with the ADuM3190 accuracy figures, even after reading many of your replies.

In this one however, you mentioned something that added more to my confusion . You said 

The datasheet specifies the gain, offset and linearity, which have more than 10% error when used as an amplifier in an open loop, which is very different than when used as an isolated error amplifier in a closed loop, which has 1% accuracy. 

Won't making a feedback between EAout and -IN break the isolation, which is the whole purpose behind using this Op Amp for isolated voltage monitoring? Even the test circuit you used for the 1% accuracy figure had their GND tied together.

Hatimoooo:

The circuit with the feedback from EAOUT to -IN is a test circuit without isolation only used in production test where the GND1 and GND2 are connected together.  This circuit is for testing the +/-1% accuracy specification, and it does have a closed loop where the high gain of the amplifier will correct the amplifier and output errors to maintain the accuracy. In an actual application circuit, the GND1 and GND2 would not be connected and EAOUT would not be connected by feedback to -IN. There would be a circuit more like the Application Block Diagram figure 32 where EAOUT connects to a PWM controller which drives a power stage and isolated feedback would be maintained.

Can you share the application circuit that you need to use?  This can be done by email if you want.

Regards, Brian

Yes, I understand. But why do I need to use an isolation Op Amp in applications where isolation is not a issue since I'll be connecting the EAout and -IN through a resistor? I could use a traditional Op Amp. I understand that you connect GNDs for your tests but it doesn't provide any info which would be relevant in real applications of this component.

By the way, we're going to use this component in isolated voltage monitoring, but a 10% gain error isn't a pretty figure.

Thank you

Hatimoooo:

You are right, you would not need to use an isolated Op Amp in an application where isolation is not and issue and you connect EAOUT and -IN through a resistor, you could use a traditional Op Amp. In some applications where isolation is not needed but the application has noise from having different ground locations, isolation could be used to reduce the noise.

For an isolated voltage monitor, isolation may be needed for safety with high voltages. If the case of a voltage monitor, the ADuM3190 datasheet has specifications that need to be considered, as shown below.

Regards, Brian