In the ADE9113 datasheet, Figure 40 shows the connection method for DC metering.
In this diagram, the shunt resistor is connected to the +DC side.
If I want to place it on the –DC side instead, would it be correct to connect it as shown in the attached file?
Hi KNH,
With a shunt measurement, we want the AGND connected to one side of the shunt for a pseudo-differential configuration. The isolation between primary and secondary sides of the ADE9113, along with the AGND connection, enable the ADC to stay at line potential whether its 3-phase, DC+/-, or neutral. That is what you have above. 
The next consideration is the voltage sense connections. Although we have fully differential ADCs on the ADE9113, we need to keep the voltages presented to all the ADC input pin relatively close to one another. This means when the voltage divider is setup, the small resistance goes to one side of the shunt, and the large resistance goes to the other line. This is shown correctly above also.
When it comes to actual implementation, there may be some adjustment of the voltage sense divider values. Maybe a bit of tweaking for good use of the ADCs dynamic range vs accurate measurement of overvoltages, for example. On the higher resistance side of the divider, the resistors may need to be implemented as resistors in series to meet required safety specs for creepage/clearance.
There is also the subtle difference of which side of the shunt the voltage divider is placed. I would infer that the load is on the bottom of the diagram. The small voltage drop across the shunt will be part of the measurement. That may be fine, and easily reconfigured if its not.
Regards,
Jason
