First, here's the new grid under construction, in a jig I made for keeping it straight while setting the rods.
- Grid and jig. .020" rods, Ti backend, carbon front ring
The Ti base was fabricated here from a 1" rod of Ti, cut down on the lathe and drilled in the mill with a circle table. This was done so we drilled through the Ti, and partway through the graphite ring beneath it, so for better or worse, all the holes line up perfectly. I actually used a 5x stereo loupe to observe the exact marks on the circle table to get it as close to perfect as I could. The Ti here has been anodized to reduce sputtering and arcing - this has shown to work very well with Al, so I thought I'd try it with Ti too. The Ti, where the rods go through is about .1" thick, or thinking another way, the holes for the rods are about 5 calibers deep. This was one tough job of drilling - .020" drills are on the fragile side, and breaking one off in the work is almost always a disaster. The rods, once set in place, are staked with a spring-loaded center punch. This only works with a strong punch, and holes that are right at the edge - these are about 40 mils inside the rim.
Obviously, there's no hope of having the rod straight post staking if it wasn't before, so I built the jig pieces shown. There's a tapped chunk of steel to hold the back level, and a cut piece of 1" ID copper pipe to hold everything perfectly aligned (well, kind of) while doing the deed. This leaves the carbon ring at the other end just a push on, and with forces we normally see, that's not good enough, so that end gets a girdle. In this case, I used .020" pure tantalum wire, because I had it (Thanks BillF) and its ductility makes this easier than using, say, type C thermocouple wire (W-Re alloy).
- Ta girdle - without this, you can count on the end flying off due to E field forces.
We've been playing around with a shield for the feed through glass behind the grid. This seems to surpress some sputtering and arcing, while also making it harder for hot ions to hit the glass and reduce it to silicon metal - very bad, that almost always results in shattered glass. Here's the old prototype, and the new one I'm installing today:
- Feed though protection shields, old and new. The new one is anodized aluminum, and a lot slicker generally (and lower sputtering) than the copper prototype. This can be used as a faraday probe or antenna for driving fields as well.
And it looks like this with it all installed:
- Installed in tank, with shield, ion source grid, and 2.45 ghz antenna. Note SS screen right up at the window - turns out to be important for more than shielding 2.45 ghz.
This time, since getting too eager tends to mess things up, I let it pump down overnight before even thinking about doing the conditioning. This AM, the pressure in the tank was about 5e-8 mbar, pretty good, so I proceeded to condition the grid. I set the supply at a reasonable (for me) current limit - 15 ma, and 35kv initially, but didn't turn it on till there was enough D in there to hold it at 5kv or so, gradually taking out gas as the little arcing spots went away at each voltage level. I did push it a little fast today - above about 40kv, it's going to take a lot more time and care to get this really broken in and all the tiny whiskers and contamination gone.
While doing all this, a new phenomenon was observed. Normally if I turn on the ion source at a low enough pressure to want it, we see the central beam out of the end of the main grid bend toward it. While I could't see that very clearly this time (but I'll look more closely next run) what we DID see is that there is now a spot where something very energetic is hitting the viewport protection screen, originally put in there to keep 2.45 ghz out of the shack when we inject that. Turns out it was a good move for entirely other reasons, as this movie shows. You can also see electrons hitting the glass through the screen (faint blue-gray glow pattern), but wow, what a variability - something is "almost oscillating" or "getting chaotic". No interpretation yet - all I have is data, not information.
http://youtu.be/vkWyikizEbg
Note you can see the spot move *away* from the ion source at left when I turn it on. Doesn't make a lot of sense just yet. You can prove this by putting your cursor on the spot center before I turn on the ion source.
Overall, I noticed getting to fine focus at lower voltages and higher gas pressures than normal. The jury is out on neutron generation at this point - no accurate measurements have yet been made on this one - I need to do some PC work for that as the dedicated machine I use for this is losing its USB interfaces one by one (it's old, and this isn't a nice safe place for a computer, in general). Ah modern computers with GUI-only and USB for everything - without a mouse working, it's kind of tough to collect data!
Edit: Looking closely at the video, by the end the carbon end-ring is showing orange hot in the video. However, in real life it wasn't visibly glowing at all, and was perhaps a very dull red when I turned off the HV (couldn't see it till then). I guess this means my camera sees into IR a little better than the human eye does.
