Kevin was working on characterizing all of our RF photodiodes for the upgrade, and he discovered that REFL11 didn't work, as described in elog 3890. Rana was working on fixing it, but then he went off to Japan.
I visually inspected the components inside the RF cage on the REFL 11 circuit board inside the PD. Most of them were okay, but the connection between L5 and C33 (the big tunable inductor and the next capacitor in the path) was totally flaky. the leg of the inductor had been soldered directly to the trace on the PCB, and the inductor was a little bit tipped over, and pulling the trace off the board. I wiggled it a little while trying to see what was going on, and the trace broke. Since there is nothing going on between L5 and C33, just the trace, I used a piece of resistor lead to attach the two. The connection now seems very robust. I'm a little worried about the connection between the inductor and the board on the other side, but I can't see it since it's under the inductor itself.
Also, the soldering of L4 (a standard surface mount component type body) to the board seemed totally shoddy. I was desoldering the first side, and the whole inductor popped off. It was clear that the inductor was making a physical connection to the board, but not a nice solid electrical connection. So I resoldered it on. (On Alberto's schematics, it is listed as a 633nH inductor. I can't find any of this value, so I just put the same one back on. The best I could do to confirm the component was still okay was measure its resistance, and compare that to a similar inductor of a similar value. It seemed okay.)
After that I powered up the PD, and took an electrical transfer function, just to have a look-see. It seems kind of okay, although the resonance seems to be closer to 13MHz than 11MHz.
Since we would like to remove the capacitor that is in parallel with the diode itself, which will then change all of the resonant conditions on other components, I didn't worry too much about the resonant peak for tonight. We're going to have to look in on this though.
Also, I'm leaving the optical check-out for Kevin, so he will let us know if I magically fixed the PD, or if it needs some more work.
Photos of the circuit board (mostly Alberto's mods) before and after I fitzed with it are on Picasa
More testing. Probably more fixing.