Filiberto, Rich

Motivated by the observation of oscillations on an AA/AI circuit board when loaded by a long cable in the LLO ESD system (and per Rana's suggestion), we performed a stability analysis on the version 6 AA/AI circuit board.

A version 6 AA/AI board was loaded with ~120 feet of typical twisted pair.  The output of one channel of the circuit board was driven through a 470 ohm resistor using a function generator running a 1kHz square wave at 2Vp-p.  The cable-loaded opamp output was then observed on an oscilloscope to watch the perturbed response of the opamp.  By observing the opamp output at the edges of square wave, any tendency toward instability can be detected as excessive ringing.  These observations were done while varying the output DC level to check if the stability is a function of output DC level.  Actually, the DC variation test could be flawed in that a significant load was present from the 470 ohm resistor in series with the function generator.  A better test would have been to AC couple the function generator to avoid being mislead by the excessive loading of the 470 ohm resistor in series with the function generator.

No tendency to oscillate was seen on the version 6 circuit boards.  As a further test, the cable loaded opamp was loaded with an additional 0.22uF fixed capacitor and no instability tendency was noted.

As a sanity check, we took a version 4 AA/AI board and found it unstable with even a moderate cable load (60 feet vs. 120 feet).  Even a ~12 feet of RG58 was enough to get the version 4 board to show instability at 4MHz or so.