A not-so-old expression is "one man's vacuum is another man's sewer" (N. Milleron, 1970). Richard correctly states that no cold trap (in fact, no trap) is required for a diff pumped fusor especially if the oil is one of the synthetics with a very low vapor pressure. But, if you were doing mass spectrometry or operating an electron multiplier detector, you would need a good trap or, preferably, a completely different type of high vacuum pump.
A couple of hints:
A baffle above the diff pump chimney is recommended as some oil may condense on the top jet and rapidly evaporate, sending a stream of oil up toward the chamber. (This is not backstreaming which is a molecular flow process.) The baffle will get in the way of the oil puff and keep it from getting toward the chamber. The baffle should be "optically dense" i.e. you can't see through it. Chevrons and stacked plates with cutouts are examples. Don't go overboard or conductance will degrade excessively. Many diff pumps have a baffle built in and newer pumps will have a chilled cap right over the top of the chimney. These are useful features to look for in a surplus pump.
Never vent a diffusion pump while it's working. The oil jet will become an oil cloud and will go everywhere making a hard to clean mess of your neat apparatus. Everyone who's ever used a diff pump has probably turned the wrong valve at the wrong time and done this (great learning experience). I've only done it once. So far.
Let the pump cool before venting to preserve the qualities of your oil. The old (cheap) hydrocarbon oils (butyl phthalate, octoil, etc.) will oxidize and crack leaving a thick mess. The synthetics range from much better to impervious to this.
An untrapped pump may eventually contaminate your ion gauge. I like cold cathode gauges for untrapped pumps because you can take them apart and clean them up if they get contaminated. Hot cathode gauges get thrown away.
Steve
Created on Wednesday, March 21, 2001 8:23 PM EDT by Steve Hansen