First light on SNM-17 3He tubes

This is bound to get mixed up with things in Electronics, check both. Physics-specific stuff here, mostly.

First light on SNM-17 3He tubes

Postby Doug Coulter » Wed May 18, 2011 11:57 pm

Jon Howard has been visiting here and we've been busy fusing and making and testing things this week. When we got those Russian B10 tubes we luckily also scooped up a couple of 3He tubes as well. Y'all keep eyes peeled -- we paid -- $20 for each. Not quite like the whining on fusor.net. Sadly, there were only two in that batch, and Jon's getting one.

Still working on the "ultimate preamp" design, with some success. This is a two transistor asymmetric risetime low quiescent current design, 700 uA on a 9v battery that makes 5v or so pulses without clipping, and has feedback for great linearity. I will probably be using this on phototubes as well, as it's really fast. In this design, both transistors are biased barely on, and the incoming pulse turns them on further. This gives very fast response, with a slower turn off time, perfect for a signal that does the same thing. The signal gain for AC is roughly 50, set by the ratio of the npn collector transistor to the emitter resistor on the pnp input stage. Input transistor is a 2n5087, npn is a 2n3904. I took some care in design and testing this to have it work decently over the whole voltage range of an alkaline 9v battery -- from about 9.5v down to about 5.8v or so.
Here's a sketch of the preamp design. It's neater than I usually post, because I didn't draw it, Jon did.
NewDPre.gif
New preamp schematic, thanks Jon.



Here's what the assembly looks like. We tested in the same moderator as used for the CHM-14, which is a little short for this, but it works fine anyway.
WholeDetector.jpg
Whole assembly


And here is what the top looks like with the coffee can shield removed. We'll punch some holes in the can for the connectors for HV and signal, and the power switch of course (and paint the thing).
DiscretePreamp.jpg
The preamp and wiring


In a test on the main fusor here, this tracked our 22" 3He tube quite well -- at about 90% of the same count rate, when positioning the new one where we usually put the neutron oven. That's only a little bit closer to the action than the big guy, so sensitivity is more or less equivalent. Here's a scope shot of all three of our detectors going at once.

new3he.gif
Top is big 3He, middle is big B10, bottom is the new Russian tube with our new preamp design.


The new preamp in combination with this tube gives much shorter pulses. We took this shot without the can shield, because the radar wasn't hammering us so badly, but did pick up a bit of noise on the baseline due to that -- and this scope sees that due to sampling at 2.5 ghz in peak detect mode.
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: First light on SNM-17 3He tubes

Postby chrismb » Thu May 19, 2011 1:36 am

FYI:

(The 3He tubes with * are regular proportional counters.)

b10_data.jpg
he3_data.jpg
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: First light on SNM-17 3He tubes

Postby chrismb » Thu May 19, 2011 4:49 am

Doug, I have two of the SNM-18-1 tubes [lucky me!] which are much the same electrical settings as the 17, I think, just bigger. The issue I have had is, simply, that I do not [yet! :) ] have a 'known' neutron source. So in testing it, I have simply looked out for background neutrons. Frankly, can't see a thing in the noise, so kinda gave up.

I would like to commission you to build me a circuit to do this. Yeah, sure I could easily copy your diagrams into a board. but then what? Is that good enough? An unverified circuit attached to an unverified tube without a known neutron source! At least if the circuit is already tested with a tube then I've gotten rid of one of those variables. Will you be creating a circuit for these tubes with a CCFL driver, that all I might need to do if you can send me one is attach the tube and a battery? Maybe you can take the wires off, or put wire terminals on the boards, then just send boards and I'll wire it all up? Do you think this is a good idea, or just as well make my own circuit?

Too many variables in this experimental soup. Need to start pruning them all down!!
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: First light on SNM-17 3He tubes

Postby Doug Coulter » Thu May 19, 2011 10:11 am

Depending on the amount of hurry you're in, I do plan to do a PCB for this preamp, and another for power supplies as time permits. The board layout for the perfboard version wouldn't need jumpers on a single sided PCB - it's pretty tight. The CCFL supplies radiate a lot of 50khz noise, so they have to go in their own shielded box - packaging is key on these kinds of things.

For those sharp-eyed who didn't see the input protection diodes -- they aren't there in this prototype, due to my junk box lacking *good* ones -- the ones I tried were leaky and noisy so I cut them back out (for now). Without them, the first transistor will inevitably go pop at some point when someone hot-plugs power or on some odd other event. The trimpot was in there at prototype time to get that bias resistor value quick and easy, but wound up at a standard value -- nice. Due to the large emitter R in the NPN, it's not so critical as it might otherwise be.

We ran these tubes (tested two) at 2330vdc, with the same 50 meg series R to the HV as in the other circuits.

I'll get some more scope captures up showing the response to background/cosmic rays, which are actually a pretty decent self test if you're lucky and have a DSO scope to examine the waveform whilst waiting for another trigger to occur. We see a few per minute from that, and for that, moderator doesn't really matter so much - cosmics are LOUD and really plow the gas.

I'll also get some pix of any signals we see with a hot rock on the tube -- gammas and such, if I can. We'd have to take it out of the moderator so it wouldn't shield from most of those to get a really good picture of it. I suspect that it would be in the tube noise, but it's interesting enough to give it a go if we get to that point. Right now, we're concentrating on what Jon and I can get done together before he has to go back to your side of the pond.

None of these really count well enough to eyeball on our staticmaster/Be neutron sources -- you could see it out of background with a counter and timed runs probably, but hard to be sure without that. Luckily, even a very poorly operating fusor will count any of these tubes pretty fast. We guess our sources are something under 30 neutrons/second. That's just not a heck of a lot.
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: First light on SNM-17 3He tubes

Postby chrismb » Thu May 19, 2011 10:31 am

Well, I guess if you're that confident about it then I'd better get on and bash out a circuit for myself for my tubes. The worst that can happen is that it still doesn't work!! ;)
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: First light on SNM-17 3He tubes

Postby William A Washburn » Thu May 19, 2011 2:28 pm

Doug,

When I tried to order the diodes I noticed that they are germanium diodes. Apparently these
must vary in quality according to your comments.

I suspect you want the germanium for its low forward voltage loss and even a low-noise
silicon device would not work here.

What brand and model of germanium diode would you suggest for this app?

Thansk, Bill
User avatar
William A Washburn
 
Posts: 93
Joined: Fri Oct 15, 2010 8:12 am

Re: First light on SNM-17 3He tubes

Postby Doug Coulter » Thu May 19, 2011 2:58 pm

They *are* slightly odd. I wouldn't have the confidence I have if not for having the neutron source, and especially the nice storage DSO to see rare events with. We've seen one anomaly so far, to be checked in a few -- the thing continued to count at a low but much higher than background rate after shutting off the fusor....

As promised, here are some pix of scope captures in various conditions. The long and short -- no natural rad source we have has any effect on this at all. We tried torbermite and Ra226+ daughters since WWII -- there are spectra of these elsewhere on the board, and there's some hot stuff there - no counts at all from those.
So, first, some cosmic ray responses. We didn't have a big amount of these, perhaps every few seconds, we didn't time it. They come in all different pulse heights, just as the real neutrons do.
new3heCosmic1.gif
Cosmic ray hit w/o shield can
new3heCosmic1.gif (8.9 KiB) Viewed 12460 times

new3heCosmic2.gif
Same setup, different time
new3heCosmic2.gif (9.59 KiB) Viewed 12460 times

You'll note the shapes are the same other than amplitude. We see that with real neutrons too. Someone asked if these see gammas...here is what we saw with a chunk of torbermite that counts our standard counter at ~147,000/min.
new3heTorb.gif
With a very hot rock resting against the side of the tube

Nothing, nada, zip. Same result with a very hot Ra-226 source (in glass, no alphas). So that's not a worry at normal gamma energies.

Now for the fun part. Here's what this looks like with the scope in "roll mode" and with fast sampling (peak detection). This "gets" noise that your electronics can't see up to 2.5 ghz...which is why I'll show you a baseline with no fusion first.
new3heNoFusionroll.gif
No fusion baseline
new3heNoFusionroll.gif (5.59 KiB) Viewed 12460 times


Now, with the fusor running:
new3heFusionroll.gif
Fusor running
new3heFusionroll.gif (8.21 KiB) Viewed 12460 times



But of course, "you had to be there" so here's a movie.
http://www.youtube.com/watch?v=_7VY137L5hY
And of course inline.


After awhile, the extra counts post run went away, but we saw increased background from cosmics. Cross checking with our standard geiger counter, yup, it's that time of day. I'll try to get a good count at night when our other counters are also counting normal. We have a big extra background around dawn and dusk here for whatever reason.

William, any quiet low leakage signal diode with decent current will do there -- it has to be fairly fast (eg not a 1n4001) but that's about the only requirement. Silicon is fine (probably better).
They'll take a beating better than a transistor be junction, and are cheaper to replace if needed.
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: First light on SNM-17 3He tubes

Postby Doug Coulter » Sat May 28, 2011 5:25 pm

For what it's worth, I've been having occasional troubles with this 3He tube "going berserk" once in awhile. Just making counts when there's nothing - on the scope the waveform is often a little different (louder and flatter) than the real deal, but it's annoying. It has happened a couple of times so far, and it's pretty obvious -- a khz or so of garbage. It seems to "recover" if you just leave it powered and wait half an hour. It has the same symptoms as "conditioning". For example if you start out with 2.5 kv, then reduce it to 2.2 -- the crud goes away for awhile and it counts neutrons fine like that. If you leave it on long enough -- say half an hour, it seems to recover the original low background count. Perhaps the fact that I've moved it all over the lab while finalizing it shook something loose (dust in there?), no way to know.

Taking the other Russian B10 tube, and putting it in the exact same setup -- adjusting the HV so its happy -- it's reliable as can be. I think I'm starting to like them better, even if they're not as sensitive.

JonH got the other 3He tube, so we can see if he runs into the same thing. At this point, the jury is out, but I'm liking all the B10 tubes I have more and more -- they are a little less sensitive, but rock solid, a big bonus in any measurement system. I'm only going to need max sensitivity for the beam on beam stuff, so the old texan 3He is going there, but for the fusor, a nice reliable loud neutron source, the B10's are just fine -- maybe better as they don't have the pulse pileup issues the more sensitive tubes do.
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: First light on SNM-17 3He tubes

Postby Joe Jarski » Sat May 28, 2011 8:25 pm

Doug, are you using the same (new) preamp on the B10 & He3 tubes? Or are you using a different design for the B10s?
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Re: First light on SNM-17 3He tubes

Postby Doug Coulter » Sat May 28, 2011 9:08 pm

Same stuff down to the atoms it's made of -- I just swapped tubes in the same moderator/preamp box. I also tried just pulling the plug on the 3He, leaving HV and all intact -- bad noises went away, and came back with the tube connected again. The only change between tests was which tube, and what I set the HV supply for (the B10 tube wants less HV than the 3He).

This looks like a little more watching and waiting is going to be needed before a final verdict. When the 3He works right, it's much more sensitive, but...if it's going to be flakey...not so useful.
I tried cleaning off the insulator with acetone (it's teflon) and so on and so forth -- anything to make it work right I could think of. Just waiting was the answer, this time.
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Next

Return to Metrology

Who is online

Users browsing this forum: No registered users and 2 guests