AnsweredAssumed Answered

Amplifier with low phase error DC-5Khz

Question asked by dvetter on Oct 8, 2015

I'm developing a current sensing application that requires some amplification of the sensor output.


One of the requirements is (ideally) zero phase error in the bandwidth of interest (DC-5Khz).  AD8051 looks to be unique among the AD parts I've examined in that it has excellent phase characteristics below 30Khz (hard to see on the graph, but it appears effectively 0 degrees and flat out to that point).  The problem is it's a 5V part, I only have 12V available on this board, and if I have to create a power supply for this part cost and hassle will go up.


So I have a few questions:


1) Is there a 12V tolerant equivalent to the AD8051, i.e. a part with excellent phase error characteristics at low frequencies?


2) More generically, why do most op amps seem to have horrible phase characteristics?  For most the phase response starts off -20 or -40 degrees at DC or the lower bound of the graph (which is sadly not always DC) and falls steadily from there.  And even in the few cases where the phase response is flat, those all seem to start with -60 or -90 degrees at DC and then start to fall well below the -3dB point.


3) What makes the AD8051 so special?  I guess what I'm looking for is any unique property, specification, or topology of this amplifier so I can use these terms to search for equivalents.


4) I read app note AN107 about second order cancellation, which involves using a matched (on die) dual amp and placing the second amp within the feedback path of the first amp providing the forward gain, and this seems to reduce the phase error to near zero, extending the phase-error-free bandwidth by ~10x.  Does this work reliably in practice so long as the amps are matched?