Image Removed
Image Added
Model
label: EntranceQtrWavePlateAssbylabels: DetChannelAssby:1, DetChannelAssby:2
Assembly instructions
Oil method
The goal is to couple the lens onto the PMT face with oil
Seal around the edge of the PMT face to the PMT housing window that is on top of the PMT photocathode, so that oil can not leak into the body of the PMT, with 5 min epoxy - . Be careful not to soil the plate above the photocathode photocathode.
Add 30 - 50 ul of high viscosity microscope immersion oil.
Place condensor lens on oil–touch to oil on one side first so as not to trap large bubble underneath. Make sure there are no bubbles and carefully remove overflow oil from the housing.
Place a thin ring of quick seal Kwik-Sil (WPI) on the PMT housing around the lens
Add PMTMountPlate before Kwik-Sil sets, such that sealant squishes between the lens and PMTMountPlate. Hold lens down so that it does not move up while mount plate is being pushed down. Attach PMTMountPlate to PMT with two M3 screws.
Allow ~30 minutes at least for Kwik-Sil to set (oil mixing with Kwik-Sil will interfere with the setting process).
Hold PMT in vise with window side down for several minutes to check if there is oil leakage. Blow gently around seal with canned air to see if there is movement of liquids/uncured Kwik-Sil.
If there is leakage, seal further between the condenser lens and PMTMountPlate using epoxy, or other sealant. It is OK if the sealant touches the face of the condenser lens a couple mm above where this lens meets the edge of the PMT mount plate.
Hold PMT at angle it will be used in microscope (depending on red or green channel) overnight, and check again that oil coverage between the condenser lens and PMT window is intact. There should be no obvious bubbles or inhomogeneities when looking at the photocathode through the condenser lens.
UV cement method
Note: applying UV light to cure UV cement will not harm the PMT photocathode. There have been problems when using UV cement with thick layers of cement (few hundred micron), resulting in permanent bubbles present between the lens and PMT, but the strategy below should not have such problems. It is currently untested, however.
A relatively foolproof way to do the PMT condenser lens bonding with UV cement would be to fill the gap between the condenser lens (flat, optical surface) and the PMT window with another piece of glass (index of refraction ~1.5), and then only use thin UV cement layers, like when bonding a doublet lens, between all the glass pieces. Pieces of coverglass would work fine for this. It would just be necessary to find the right thickness of the gap-filling glass, such that there aren't any air gaps, and the condenser lens doesn't have to rise up much beyond its normal seating position (shoulder touching the PMT housing). Two to three #1.5 coverglass pieces ought to do it. After assembling everything dry to test the thickness and fit of the filler, the filler pieces could be cemented to the condenser lens first, and then whole assembly would be cemented to the PMT.