Can any one please verify my design for ADE7953 using Shunt Resistor. I have attached the PCB files here.
I can definitely review your design and provide inputs. Can you please include the schematic file as well- as a pdf?
PFA the schematic of ADE7953. Here I have used all the value given as per the test circuit in ADE7953 datasheet. And Value of Shunt resistor is 20mOhm.
Please go through the same and provide me with a feedback.
Here are my comments on the schematic:
1. You cannot have two shunts in this setup. As you are connecting one of the ends of both R9 and R11 to ground here, that would mean you are tying the phase and the neutral together in the system. Usually, a current transformer or a rogowski coil sensor is used in the neutral channel. Please see the single-phase configurations explained in this app note (page 13): http://www.analog.com/media/en/technical-documentation/application-notes/AN-639.pdf
2. The voltage channel attenuation network has a large resistor (R8). Can you split that into 3/4 individual resistors in series? For example, if you have used 1Mohm as R8, you can consider using four 250ohm resistors in series instead. This helps with creepage on the board.
3. You should have a 10uF, 0.1uF cap combination for decoupling on VDD pin. I know it is missing in the test circuit of the datasheet- we will correct it in the next version release.
4. You have a RC filter hanging off of RESET pin. The pin must be pulled high by default to avoid resetting the IC by mistake. You can also consider sending this line to the MCU, if you wish to do a hardware reset of the IC by controlling the state of one of the MCU's GPIOs. So, tie the floating end of RC filter to VDD.
I haven't looked at your layout yet. After making the above changes, please send me the corrected version. Also, if you can share the complete meter schematic, that will be useful-just to make sure if there are any system level issues that we are overlooking.
I have done all the modification suggested by you.
1) Here we need to do current measurement on only one channel, we wont be measuring neutral current. So I have removed one of the resistor.
2) Also I have added Resistor network for Voltage Measurement.
3) 10uF and 0.1uF filter is applied to Vcc.
4) It was my mistake for reset pin, I have corrected it by adding pull up resistor and switch.
I have attached modified Schematic file.
Here are the Modified PCB files for verification.
The schematic looks good. If you are not using IB channel, it might be a good practice to ground the pins. In the layout, the decoupling can be brought in a little closer to the IC. I have tried to highlight two ground islands that you have made in the layout. I would suggest getting rid of them and bringing the decoupling closer to the IC- you can connect ground pins directly to the ground pad-like how you have connected for DGND pin. Also, try to bring in other decoupling caps.
I have attached a snapshot of ADE7953 recommended layout (from datasheet) below. Please note the priority given to decoupling in all scenarios- AVDD, DVDD, REF, VDD, and Crystal. Teh ground pad isn't grounded here, but it should be-like what you have done.
I have attached the modified files, as recommended by you. Please go through the same and verify the same.
Awaiting your reply.
Thanks and Regards,
The decoupling looks better now.
This Post and the previous post of yours on ADE7816 was very helpful to me.I would like to thank you for that.I am trying to use ADE7953 in my design.I would like to know regarding the performance and accuracy of you got after using ADE7953 with shunts.
Retrieving data ...