2) Size (Griffin)
3) Operational practicality
Due to a large number of large square o-ring of poorly specd diameters and several flanges that do not connect simply (i.e. must be slid sideways) vacuum of the chamber is poor. This leads to thermal cycles being necessary more often.
The active part of SPICE is actually very small. Almost the entire aluminium chamber serves only for getting signal cables out and holding an over-sized cold finger and the dewar. Because it is so large it does not currently fit on the GRIFFIN beam-line on the upstream side.
3) Operational practicality
In the way it is currently designed SPICE requires a crane to remove the dewar and cold finger to access the Si(Li) detector and FET board practically. Additionally to make any changes to S3 cabling, which is frequently necessary, the side panel must be removed which requires removing the large side panel, this involves a sideways slide and leads to frequent vacuum problems. Additionally the side mounted turbo pump and 4 legs makes alignment challenging as the weight is always poorly distributed.
The plan for SPICE II centers around a move a to conflat cube, such as this https://www.idealvac.com/product.asp?pid=6499
An adaptor flange on one side would connect to the same Delrin magnet assembly. By making the largest sealing surfaces round, removing any unused panel from the original design and switching to conflat copper gasket fitting for any rarely used flanges the vacuum can be significantly improved. The turbo pump should be moved to the bottom panel (with grill) to achieve better balance and a three point alignment system use instead of the present 4. A new A-frame will be needed but due to the change in size the G10 can be reused. The top panel will contain only the preamp feedthroughs, the side panel will have only the S3 feedthrough. The rear panel will contain a narrow beampipe, target wheel rod, cooling feedthrough and PT100/warming resistor feedthough. By this arrangement everything can be easily accessed, repaired and removed while on the beam line.
Many size and practicality issues with SPICE are spawned from the dewar.
The solution I suggest is as follows:
The SPICE Si(Li) and FET board should be mounted on insulated stand offs from the upstream or downstream flanges. Attach a heat block with Swagelok fittings to the FET board and use a something like the following to connect to a small LN2 compressor:
If the SiLi is mounted on the downstream side the support stand offs will be simpler and there will be more room in the chamber for work but LN2 connections would need to be removed more often. Where as if SiLi were mounted on the upstream side rigid piping could be used. Either way having the SiLi/FET assembly removed in the beam direction makes practical sense.