Power Supply Filtering

Question

It appears that the ADMV8526 and ADMV85xx bandpass filters are passive, however, they must connect in some way to the control signals. 
 But I don't see any isolation specs in the datasheets for the power-supply pins or control signals. 
 Since the filters are passive, it seems the isolation may be very large, but without any information I would have to assume none. 
 Is there any guidance on filtering the power-supply pins and control signals? Should they be treated the same as LNA power pins, for example?

enash · Accepted Answer

The general guidance that we would provide for the power supply and ground pins of this part is to treat them as analog, eventhough you are primarily powering digital circuitry. So I would not connect this pins to a bus supply that feeds other digital circuitry. Also, I would connect the part's ground pins to analog ground.

The datasheet recommends specific caps on VDD, VSS and BYP. These values are used on the evaluation board which would have been used to characterize the device.

We have not evaluated whether very sharp digital edges on the SPI and control signals affect performance. Since the settling time of the filter is relatively slow (2us), you could put some small caps on the digital lines to take the edge off of them. However, looking at the schematic of the evaluation board, I see that there are spots for 10pF caps on the digital lines which are de-populated (C82, C83, C84, C85, C86). This indicates that we did not see any sensitivity to sharp digital edges during device Characterization.

ElectronManager · Answer

Is the -2.5V supply on ADMV85xx powering any analog circuitry? 
 Would the RF channel be more prone to in-band noise on this negative supply than on the positive supply?

enash · Answer

Q1. Is the -2.5V supply on ADMV85xx powering any analog circuitry?

A1. The negative supply serves the switches which are used to switch in and out components in the capacitor bank. So I would say that this supply is not far from the analog circuitry.

Q2. Would the RF channel be more prone to in-band noise on this negative supply than on the positive supply?

A2. This is difficult to say as the PSRR of the two supplies have not been closely studied. My recommendation would be to treat both as analog supplies

Popular Forums

Product Forums

Application Forums

Design Center Forums

Highlighted Webinar

Places

Libraries

Challenge Yourself!

What's Brewing

Places

Resources

Visit ContentZone

The Latest Read

New Release

Recent Technical Insights

Products Mentioned

Thread Details

Power Supply Filtering

Thread Summary