by Doug Coulter » Tue Nov 16, 2010 10:12 am
Ah, in my case I deliberately took the outer envelope off. I do get ozone (still, after a hundred hours or so of runtime) during warmup, before it becomes a high pressure lamp -- most of the mercury is still liquid at that point, and we're running on a little Hg and some Argon (I'm guessing). The color is different during that, a little, less of the output is in the visible range. I only get ozone for the first couple minutes, while the pressure hasn't built up yet, then it's more the longer waves that don't make it. It will sunburn you at any time, though. For ref, that's with a 250w lamp, I also have a 450, but after seeing what this one does, don't fire that big guy up.
My experience is yeah, in a vacuum you're not going to be able to keep it cool. The quartz-halogen bakeout heaters I have in there now are almost a foot long and half an inch diameter for 500w, which is about what it takes, and they get red hot fast -- takes a special clamp mount to thick copper to keep the leadin wires from melting at full power. Kind of fun to run them 30 sec and turn them off and see how long it takes them to stop glowing -- not the filament, the quartz envelope!
But what I noticed with these, which are wired for either half voltage (in series) or full, is that you get a pulse of out-gassing very fast when they are turned on, and that's most of the game. The pulse is faster and has much more H2 and O2 when they are at full volts and "whiter than white" -- lots of UV, but not as much as something meant for that. You get nearly all the good out of it in about 60 seconds of that, after that you're just wasting power and heating the tank in bulk (eg below the surface metal). In that short time, they don't heat up so much, so my thought with an in-tank source was that it wouldn't have to be on very long at all, and the louder the short wavelengths, the better. Then cut it right back off, or perhaps run in a very low duty cycle pulse-mode. I think this is also the reason that running a glow in argon, at fairly high pressure (so you can do it at all) then pumping back down gets you to base pressure so much faster than pumping alone -- even though you had to let in some argon to get up to glow discharge pressures. All that fierce UV just blasts water off the tank walls, and everything else with it. The the argon holds it in suspension while you pump it all back out.
Many people have noticed that one, so it's real -- if there's an alternate explanation, I'm all ears. Seems to work with any inert gas, argon is just cheaper than He.
I've melted plenty of stuff in a vacuum, it was one of those surprises that shouldn't have been -- intellectually I knew it's harder to get rid of heat in a vacuum, but...seeing a 100w bipin lamp running at 70v (about half power) quickly melting the #10 wire I had crimped around the leads was a demonstration of how much the good old air normally does for us. So as the old preachers used to say, there's a difference between "head knowedge" and "foot knowledge". Which is kinda why I put so much emphasis on this board about "doing stuff" rather than "talking endlessly about stuff". Guys who do stuff know things better, complete with the real-world caveats, than people who merely think about the same stuff.
And those are the ones who will be contributing to this particular forum -- it can't almost work until you actually try to make it work!
Note, NASA's "faster cheaper better" stuff has run into this, using what they call COTS hardware (commercial, off the shelf). Some idiot thought they could use things like standard computer motherboards in space apps....that's what happens when you let young academics near hardware design. Gimme a guy who has had to fix it as a tech over that, any day. He'll be far more savvy.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.