Japanese KamLAND Site


List of Members
OD Status Reports

Apartment Schedule

PMT's status

Outer Detector Installation




On Tue, 8 Jun 1999, Byung-Kyu Kim wrote:

Hi. For the checking the base of PMT, I cut about 5 cm from the top of the 1/5 piece of PVC and top part of glue by using a saw. Then I found another glue part from about 2.5cm depth. And the PMT base is embedded in the glue. Here is the brief cartoon.

             ||+-------------------------------+ <- I cut and sawed
             |||  top part glue (thick=2.5cm)  | <- this part about
             ||+-------------------------------+ <- 5 cm
             ||+-------- another glue----------+ <-
             |||  ==========================   | <- PMT base
             |||         another glue          |||
             |||                               |||
                           ( PMT )

How can we check the PMT base embedded in the glue ?

Byung-Kyu and Bob [at Kamioka]

Two weeks ago we check a group of dead PMT's in the mine that had no signal to try and determine the mode of failure. We were able to check about 1/3 of the PMTs and found there were classes of failure by checking the resistance of the divider chain on the PMT base.

Normal resistance - are nominally 5 MOhm - they probably have a problem with the PMT itself.

Very Low resistance - <1 MOhm. These are probably shorted across a resistor or a capacitor and could be repaired if necessary.

Moderate resistance - (1-4 MOhm) These tend to come up very slowly on the ohmmeter - PMT's like this have been seen in Super-K veto PMT's and are associated with a water leak. Maybe these have water inside?

Infinite resistance - most probably a fried resistor also, we saw one pmt had a clear cathode - so air leaked inside (AKA a dreaded flasher!)

Here is a summary
Dead Veto PMT's
Normal resistance (4-6 MOhm) 4
Very Low resistance (<1 MOhm) 21
Moderate resistance (1-4 MOhm) 9
Infinite resistance (open) 2

Byung-Kyu later removed the outer covering from one PMT, but the electronics is completely potted and not easily accesible. In fact maybe it's impossible to remove the base from the PMT now?

We will try and find out if there is a technique for removing the base without imploding the PMT from the Kamiokande folks...

Anyway - preliminary indictions are that many of these PMTs may be repairable if we need them.

On Thu, 17 Jun 1999, John G. Learned wrote:

Indeed I would suppose that removal of the base is going to be difficult. I would suggest that someone contact the folks at Hamamatsu about this issue. If the potting material is silicon rubber, as I think rather likely, then only a killer solvent will work. If they used something else though, it may be as simple as setting the tube socket in a pot to soak for a while and disolving away the base material. However, I think we once again will need the Hamamatsu people to do their bit with repotting the tube... this is something they have surely developed to a high art by now. Go for the professionals on this one, if we can afford it.

ByungKyu [at Kamioka]

The potting material looks like a silicon rubber. The color is green and it is different from the top part of glue. I think it is not possible to take out the base by hands because it is potted in a material. Tohoku people do not want to use the bad PMTs because there are about 1000 (a sort of) good PMTs, and about 800 PMTs are useful out of the 1000 PMTs (20% are bad from so called good PMTs).

The 800 PMTs are not from Monte Carlo. As far as I know we do not know yet how many veto PMTs we need, right ? If we find a killer solvent, that will be great for preparing more good PMTs. I will try to ask an expert here and/or at Hamamatsu about it.

Today we finished packing 74 good PMTs, tomorrow we will send the 74 good PMTs and pack 69 more good PMTs, then it will be done (800 good PMTs at Tohoku). All Tohoku people will leave on Monday.

Last updated: August 20, 1999