1. I wasn't actually "doing" anything with it, per se. Really what I did was take it home to do some major electronics upgrades(converted it electronically from a 5871 to 5972) and all I ever "analyzed" so to speak was PFTBA.
2. I forgot to look at where you were, and it was my mistake to assume US. At 220/240V, you SHOULD be fine with 10A service(which is roughly equivalent to 120V 20A), but double check the specs and make sure the plug fits.
3. If you can get gas, you're probably fine. I'm a big proponent of hydrogen in GC/GC-MS for a lot of reasons, even though it's not a universal solution. Remember that in GC applications, you're running RELATIVELY low pressures(probably not over ~100PSIG in your supply lines, and lower than that in the instrument) and also low flow rates. Hydrogen diffuses so quickly that at say a 5mL/min flow rate(which is really high for a capillary column) that in a decently ventilated space you won't get to 4%. Also, the HP/Agilent EPCs in the 5890 on up will shut down flow and other heated zones if they can't hit the pressure setpoint within a certain amount of time(3 minutes?) so you can avoid a bad leak or a broken column flooding your oven with hydrogen.
4. I have sonicated liners when they weren't super dirty, but for good quality analysis they should be replaced when dirty. In "emergency' situations I've even poked out the glass wool and put fresh in, but this isn't a great idea since even if you're using deactivated wool, you create active sites every time you break a fiber.
As to power cycling-one of the big issues is that, IME, MSs like to be under vacuum as often as possible. The longer they're open, the more crud the insulators and other such pick up(especially water), of course up to a point, while sitting under vacuum there's basically nothing to damage them. Another is that with a diffusion pump, it can be ~12 hours or better for one to stabilize after a pump down, and even turbos advise a minimum of 4 hours. You CAN turn off all the heated zones(in the GC and MS both) and otherwise leave it pumped down, but you have to be super careful with making sure you have let carrier gas flow for a while(to purge O2 and H20 as much as possible out of the vac manifold) but you're still going to need a little while for it to come to equilibrium and drive out any potential residual water. It should just soak up a lot less when under vac.
There are a few things I've done that help a fair bit with keeping idle gas useage down. One is setting up a rest method. For this, you set your flow low(with a 30mx .25mm column, you're going to be at ~1.1psi for a .5mL/min He flow at 50ºC, which is probably the bottom end of where you want to be to actually have the inlet at high enough pressure to keep air out), set your flow to splitless, and set septum purge off. Doing this, in my experience, will keep the gas useage to maybe 100psig off the tank every couple of weeks, which is decently low.
One that I did at my last institute was fit a carrier/nitrogen changeover valve on the line supplying all the GC-MSs(the room was plumbed to supply 2x GC-MSs, 4x SRI GCs w/TCD and FID in series, and a seldom used 5890 with an FID and dead NPD). My GC-MSs in that room always ran helium. They were also used sporadically-sometimes they'd be run a couple of hours a day every day, sometimes they'd run nearly continuously for a couple of weeks, and sometimes they'd be idle for a few weeks. We had two, BTW(one of which I "resurrected" from a corner of another lab), mostly because more than one person would need daily access to them at the same time
. In any case, though, at the end of the day or the end of a period of use, I'd use a valve to switch the carrier flow over to N2. I was paying ~$200 for He(5 9s), ~$50 for H2, and ~$17 for N2, also UHP 5 9s. At those prices, using N2 in the resting times saved a lot of money. If using H2, it also has the benefit of flowing an inert gas when not actively in use, which makes things somewhat safer. It took about a half hour to stabilize back out with He after running N2. I got the idea from the fact that 7890 and 8890 have an add-on to do this automatically.
One last thing-do you have software to run this?