AMC1300: AC voltage sensing. Noise at 130-160kHz

Hi Julien,

I would enhance the differential amplifier as shown here:

https://e2e.ti.com/support/amplifiers/f/amplifiers-forum/959797/amc1200-output-fluctuations-during-emi-emc-test/3546191?tisearch=e2e-sitesearch&keymatch=kai%20avik#3546191

Kai

Hi Julien,

Following what Kai said, I would recommend increasing the size of the resistor on the output of the Diff to SE amplifier. 

Thanks Kai! 

Thanks kai.  I am not sure to understand of which amplifier you are talking about. Based on Alexander's reply, we have inceased the 1k & 5k1 to 10k & 51k, whitout success.

Regarding the isolated amplifier, based on the link you provide I am not quite sure of what to do:

1. You violate the common mode input voltage range of -0.16V...VDD1. --> Is it applicable in our case ?

2. The AMC1200 is designed for current measurement across low ohmic shunt resistance. But your R6 is rather high. This decreases the balance of input circuitry and results in decreased common mode rejection. --> We have a 240 Ohm instead of 6k7 on the post. So I think, it is not applicable to our design

3. The AMC1200 provides true differential output signalling and needs to be terminated symmetrically. But you terminate it highly unsymmetrically and too low ohmically. See the AMC1200EVM for a suited output circuit. --> I think we have symetrical output, haven't we ?

Thanks Alexander. See upper post for my answer to both of you.

Hi Alexander,

Thanks for your answer.

Yes INN is connected to GND1. And there is a small mistake on my schematic the INN is connected to neutral and the INP to middle point of the voltage divider.

The other measurements on scope capture are phase currents.

I have tried to look at the input and output of the amc1300 but once I put a diff probe on the measurements, it amplifies the oscillations and I quickly get an OVP from the system.

Hi Julien,

What is the bus voltage that you are attempting to measure? 

Can you measure INP/INN/OUTP/OUTN individually without causing an OVP?

Hi Alexander,

I don't know if it's related to a layout issue. I think the solution will work this time.

However in the future designs, we need to rely on the bandwidth of the datasheet because in our power applications, we often require 100 kHz or more (Overvoltage protection and harmonics distortion mainly).

Thanks for your help!

Hi Julien,

by enhancing your differential amplifier I meant not only increasing the resistances but also to mount suited capacitors (here C13 and C14). Also increase R11 from 100R to at least a couple of kOhms. Adjust the involved capacitor values to your needs.

For best common mode rejection use identical resistors in the differential amplifier (5k...10k). Add another gain stage later, if you need more gain.

Eventually you could take a better (faster) OPAmp compared to the TLV6001, like the TLV9051 or TLV9061 e.g. This could improve the common mode rejection. And you could experiment with a "3 OPAmp" INA instead of a simple differential amplifier. Play a bit with the balance of the passive common mode filters in front of the "3 OPAmp" INA. This can help to enhance the overall common mode rejection.

Also, take an isolation amplifier with a higher bandwidth, like the AMC1300B, e.g.

Kai