Here is the relevant page from the SNO HV Interface Card schematic showing the way the signal is treated on its way to the custom IC's - signal in on the left (PMT IN and PMT Return) with a small standoff resistance from the return to system GND (so we could, crudely, measure the current on the low side of each PMT) and then the HV feed (FRM_RLY) on the top with the HV decoupling cap (4.7nF to match the capacitor in the PMT base and, first order, cancel the pole) then a transformer to send a pseudo differential signal to the ASICs). The actual termination of the signal is done on what is called the daughter card and is a complex termination (RLC) so that energy first goes to the discriminator and then all the charge is eventually sloshed to the integrator for measurement after a few ns of delay - but all looking to the cable like 75 Ohms up to about 500 MHz.

I have also included a picture of the PMTIC (used to be called the HVC for High Voltage Card until we decided that the mine might get upset about the name and changed it to PMTIC). The picture does not show the cables coming in, but four groups of eight cables came in from the top of the pix and were clamped onto the big aluminum bar for strain relief. Eight cables got cable tied to a simple pcb that plugged into the for sets of two black single row sockets about in the middle of the board and then you see the giant 4.7 nF blue caps (film caps for low radioactivity - not that we cared about the radioactivity in the rack but these exactly match the film caps on the bases where we did care about radioactivity) and then some circuitry for individual test pulsers for each channel and the ERNI connectors that took the PMT signals to the front end card (so this is a Rear Transition Module in modern crate speak). The HV feed for these 32 PMTs comes in on the little white Molex connector in the bottom center and then there are places for individual tuning resistors (and HV disconnect relays so we could remotely shut off any 8 PMTs) in the upper part of the picture.