I have been trying to use the OPA657 but I have ecountered many problems that I need help with.  First off, I need to understand several things that are listed in the data sheet.  First off, page 16 of the datasheet says " Wheredouble-side  component  mounting  is  allowed,  place  the  feed-back resistor directly under the package on the other side of the board between the output and inverting input pins. "  The only way that I could put the feedback resistors under the package is to drill two holes through the board and have the signal go through one of the holes to get to the bottom of my board and the other hole to get back to the top of my board.  These holes would effectively act as an inductor that the signal must pass through.  In my freshmen level physics courses I learned that inductance in a an AC circuit significantly slows down the circuit.  The whole point of this amplifier is for it to operate at high speeds; so why does the data sheet suggest designed the circuit in a way that will introduce inductance and therefore slow down my circuit?

Second question about the data sheet: it says also on page 16 "For short connections, consider the trace and the input to the next device as a lumped capacitive load. Relatively wide traces (50mils to 100mils) should be used, preferably with ground and power planes opened up around them."  I can why making the trace short would be beneficial and why it would be beneficial to clear ground and power planes around the trace, but what is the physicall reason that I should use a wide trace?  I did a whole series of calculations and found that a wider trace actually introduces more capacitance into the circuit and the OPA657 has a very hard time driving capacitance.  Perhaps there is some other reason to use wide traces that I just am not thinking of, but the wider the trace the higher the stray capacitance of the trace is.

My next questions are about the ability of one OPA657 to drive another OPA657.  First off, is the OPA657 designed such that it should be able to drive another OPA657?  I have no trouble getting it to drive just a fifty ohm load, but when I try to drive an OPA657 with another OPA657 I run into significant problems.  If the resistance between the two is small (say 50 to 200 ohms) the chips behave just fine when they have a signal they are amplifying but when they have no signal, they oscillate.  If the resistance between them is large, then they don't oscillate when they have no signal but they don't perform well (i.e. the output signal rolls off very quickly) at speeds higher than 40 MHz (and this is when I only have them wired up for a gain of 10).  Can you explain this behavoir?