Fusor Running, No Neutron Detection

[Alan Sailer]

So after spending more time trying to get precipitator supplies working than anything else* I finally got a chance to try a fusion run today.

I spent much effort trying to make sure that the D2 was good. I partially filled a syringe with the gas, then vented it after 10cc (to push out
the air originally in the syringe)and filled to 90 cc overall. I pinch off the tube with a hemostat clamp and attach it to the vacuum system.
The vacuum line has a cylinder of DryRite to remove moisture. I then start the turbo/fore-pump and bring the system down to less than
1 micron. The needle valve and the main valve are opened. This clears all air from the DryRite/valve/tubing assembly up to the clamp.
I then close the needle valve and remove the clamp. This lets D2 into the line.

At this point I adjust the needle valve to get about 40 microns of D2 flow. Turn on the precip and run up the voltage. It it tricky to get the thing
stable (as many have noted). But I had several stable runs with 25-30kV and between 20-40 microns of pressure. Current was 2-4 mA
depending on the run. I have looked at other successful fusion runs and these numbers are in the same region.

However the neutron count is stone cold background. I have counted background neutrons (moderated and unmoderated) using this tube
and a scope so I know the tube is working. During my fusor runs I'm using a Ludlum 12 meter. I'd love to use the scope because I know it
works but the electrical noise during operation makes this impossible.

So three questions.

Does anyone have any pointers for making sure my counter is working without a check source?

All I have done is use the basic technique using a hot U source. Turn down the discrimination. Crank up the bias until the tube goes nuts counting gammas.
Then turn up the discriminator until the counts mostly go away. I have measured the bias being supplied to the tube and it's right in the
high end of the tubes bias spec. That's all the information I've found.

I have read hints that people claim some sort of conditioning of the set-up is required. Is this true?

Does the visual appearance of the grid give any information?

Cheers.

* I know that they have worked well for others other, but results may vary.

[Mark Rowley]

Things are looking good. Being a 2.75” system, those numbers should provide a very healthy neutron count rate.

The only difference in your system is the addition of a drierite column in the high vacuum portion of the line. I haven’t heard either way if those plastic columns are rated for high vacuum OR the high vacuum is pulling some other type of gaseous crud from the drierite itself. Typically, the columns are put in-line immediately after the electrolysis or PEM cell which then feeds dry D2 into the syringe. The syringe is then moved to the fusor leaving the drierite column behind. It may be possible you’re poisoning the D2 with the drierite under high vacuum. Try removing the entire column from the picture and see if that changes anything.

Aside from that, I can’t add much regarding the Russian H3 tube. Conditioning isn’t anything I’ve heard of nor have any of mine needed that to operate. You may be able to run it in corona mode which may help if the drierite removal doesn’t help.

Regarding your input power...
There’s no need for you to go beyond 2.5mA at 35-40mTorr. Pushing it up to 4mA is just overtaxing your supply in this early stage of learning and getting it going. I have found that letting it churn out neutrons at 30-35kV, 35mTorr, 2mA, for about 5 minutes will stabilize the conditions and allow you to slowly raise the current flow (and subsequent neutron production). These things just don’t fire up and spit out good numbers. They really need to be run for awhile before the inner chamber conditions become optimal.

You’re getting close! Just proceed slowly and expect more hiccups. In time you’ll get the neuts.

Mark Rowley

[Mark Rowley]

Almost forgot, what diameter is your grid? Too big could pose some issues.

Appearance wise, it should be glowing brightly with some rays stemming from it. The 2.75’s generally don’t produce the picturesque stars seen in the larger systems.

Mark Rowley

[Alan Sailer]

Mark,

Thanks for the reply and the encouragement.

The DryRite is in a small stainless steel column I built. Threaded brass barbs sealed with TorrSeal epoxy.
Quite vacuum tight. Outgassed under vacuum with a heat gun. But I will try a run with the column
mounted on the PEM cell.

The grid is ~3/8" modeled after your later efforts. It uses a stainless steel stalk insulated with a length of alumina.
The technique used was the Rosenthal (SP?) construction.

I suspect my neutron counter, the Ludlum 12 is the problem. If so, I have no idea what to do.

Cheers.

[Richard Hull]

The problem is either the gas or the counter. It is reasonable to suspect the counter if it has never counted in a known neutron field. I have been loath to respond as I have no experience with cross fusors, yet.
Neutron counters are another matter.

What is your detector tube? What is its size"
What is your bias voltage?
What was your gamma source?
What kind of moderator?
Photos of the setup? Everything needs to be in the photo of your detection system.

Diagnosing, at range, is always a crap shoot.

Richard Hull

[Bob Reite]

I've had no problems with having the desiccant in line between the over/under pressure vials and the mass flow controller leading to the fusor. So I would suspect an issue with the neutron counter first, if you have already checked for leaks causing air contamination of the D2 source.

[Alan Sailer]

Richard,

The detector is an SNM-18 proportional tube. The tube is about 18" longer with and active length of 10.4 inches.

The bias voltage is about 1350 volts. Since the Ludlum 12 has such a low output current my meter may be loading it down.
According to Carl Willis and electrometer style meter is best to measure. I don't have one.

The gamma source is a piece of autunite (50% natural uranium and all the daughters) with a NaI scintillator tube count of ~7000cpm.

Moderator is 30 pounds of paraffin inside a cardboard cylinder with a Lexan tube in the center to hold the tube. The wax thickness is
3 inches.

The tube and the wax were used to measure background using external bias and a scope to manually count pulses. The background was
about 7cpm outside the wax and ~14 inside. This result convinced me that I was actually counting neutrons.

I tried to attach a picture but I kept getting HTTP Error. There is nothing special, a He3 tube hidden inside a white painted cardboard
tube filled with wax. Coax cable going to Ludlum 12.

The only unknown to me is if the Ludlum 12 can count neutrons. I plan to spend an hour today manually counting background in and
out from the moderator.

Cheers.

[Alan Sailer]

Bob,

I also suspect the counter.

It's good to know that, at least in principle, the desiccant does not contaminate D2 gas.

The D2 source is a new 100ml syringe. It's always possible that there is a leak. Since the syringe
is on the non-vacuum side of the needle valve the pressure there is pretty high. I use the syringe
right after the PEM cell fills it so there is not any real time for air to diffuse in.

It might be a useful exercise to let the vacuum system totally drain the syringe and see what
vacuum I can get to.

Cheers.

[Mark Rowley]

Unless they were crushed or damaged during shipping, I can’t imagine the syringes being so leaky that the D2 is contaminated to the point of not generating one neutron. I’ve used 3 cheapo syringes from Amazon without a hint of problems. I’ve also let the syringes go empty under full vacuum and obtaining sub micron levels has been 100%.

Have you pre-charged the chamber with D2 before turning on the power? Prior to any runs I’ll backfill the chamber to 500+ mTorr of D2 and then bring it down to 40. At 40 I’ll ignite the plasma and walk it down to 35 or less.

Bob, you speak of the “over/under pressure vials”. Picturing your PEM set up, I’m not exactly sure we’re referring to the same thing. Just to be clear, is the drierite in your system subjected to the same deep vacuum found within your Fusor chamber during operation? If so, and it’s not outgassing, this is good to hear. But I don’t see how water impregnated drierite wouldn’t outgas under deep vacuum conditions.

Mark Rowley.

[Richard Hull]

Your background count sounds about right. Sadly, that is only encouraging. Too bad you don't have access to a warranted neutron source. A good scintillator will always count far more counts from any U mineral source than a GM counter as gammas will be counted from the guts of the mineral, while the GM counter will only count the betas from the surface and virtually none of the gammas. When I say a blisteringly hot gamma source I am talking about a 100,000 cpm source on a GM counter and much more on a scintillation counter. I have a lot of different counters, but have never used a Ludlum model 12 or the Russian tube. Do you have data on the 3He pressure in the tube? My 4 atmosphere Reuter and Stokes works around 1450-1500 volts bias. Wish I could be of more help.

Richard Hull

[Jim Kovalchick]

I'm not convinced it's your counter. 2 to 4 mA at 25 to 30 kV is not necessarily a lot of fusion in a small cross fusor. My cross fusor started popping neutrons out at around 8 to 10 mA and above 30. The results were variable too with one day neutrons and next none. Have you tried different gas flow rates? If your chamber hasn't had a long heating period or is leaky, you may need to flow more d and not throttle your outlet as far closed.

[Alan Sailer]

Jim,

I'm not convinced either. I'm working on that.

As far as the higher current I'm worried that my much smaller grid will no take that amount of power.
I've got a small two loop tungsten grid. I was looking at one of your threads and the grid you are using is much bulkier.
Ti vs W as I remember.

Cheers.

[Mark Rowley]

Jim,
Until I see pics of Alan's arrangement, his system seems to be very close to mine (minus the neutron detector). Using a 60kV precipitator, 2-2.5mA, 35mTorr D2, 35kV will make a respectable amount of neutrons. In my case ~300k n/s isotropic. BTI results, indium activation, and my Ludlum 2221 / SNM11 all support this output. At 20kV / 2mA Alan should be getting a very noticeable amount of counts over background.

Alan,
If the input and pressure readings are what you say they are, then you are 100% making neuts I contend there's no debate on this. Concerns over high amperage grid destruction is a non-concern. You will not need to go that high with your system.

As stated in an earlier thread, I'm not a fan of the Soviet 3He tubes as my experience as well as others show them to be much more tricky and suseptible to electrical noise than the boron counterparts (ie SNM14 or 11). I dont know why this is, perhaps something to do with the tubes pressure. And to reiterate, my previous attempts with the Soviet 3He tubes have all been with heavily shielded cabling, etc etc. Still no go. For detecting Po/Be neuts they worked great! This however is not the problem with Alan's system as he's not getting any reading at all.

Alan, if your picking up the 1-2cpm background count then I'm still somewhat in the gas integrity camp. Unless the background counts are some type of spurious noise there's no reason why you cant detect a higher number of neuts over what you can detect from background.

Mark Rowley

[Mark Rowley]

Alan,
If it helps, I can make a complete startup to fusion to shutdown sequence youtube video with my fusor to alleviate any concerns you may have about the required input power.
Let me know.

Mark Rowley

[Alan Sailer]

Mark,

Thanks for the information and offers. I am regrouping and considering options. Unlike Apollo 13, no lives are on the line and failure is always an option.

I have definitely been using your system as a model from the beginning. I'll try again to post a picture.

As you mentioned your issue with the He3 tubes was noise. As far as I understand the output of a He3 tube is lower than boron tubes. But with the scope
I am seeing clean pulses that fit neutron detection perfectly. One option that I hesitate to turn to would be shielding my scope so that I can eliminate
noise form the fusors and see if neutron production changes during operation. EMI shielding is painstaking though.

Back in the 80's when I was starting to work in engineering an older engineer told me that if you see it on a scope you can pretty much bet it's real.
With very few exception he was right.

Cheers.

[Mark Rowley]

Alan,
Check out Dennis Brown’s posts in 2016/2017. He went through similar issues getting his detection system up and running. Not completely the same, but it should offer some insight.

Mark Rowley

[Richard Hull]

A very sensitive charge amplifier is demanded on all 3He tubes as each pulse is just a whisper of current of a strength that should tickle an FET front end for sure. This is why Carl noted the electrometer which is easily capable of registering 10e-13 amp or lower currents. Many special FETs are quite capable of doing this as are a number of special FET operational amp ICs. Typically, in all proportional pre-amps in the nuclear biz, a single high impedance FET is the most commonly used front end component, due solely to the minuscule currents delivered.

Richard Hull

[Alan Sailer]

Mark,

I'll do that.

Richard,

I have considered rolling my own electronics for neutron detection. I just hadn't thought it would be needed.
Carl Willis's video showing a Ludlum 12/He3 detecting a moderated PoBe source was my model. It's why I bought
the 12.

Cheers.

[Richard Hull]

Believe me, If Carl Willis got the model 12 to work with his 3He tube, then you can too! It is a matter of having 1. real neutrons to count and 2. a known good 3He tube. Add a bit of skill in setting it up.

Richard Hull

[Alan Sailer]

I finally got a successful photo attachment. I will be surprised if there is anything unusual.

The white cylinder is 30+ pounds of wax in a cardboard tube. A lexan tube is inside. An NM-188 He3 tube is inside that.
It is attached with a short BNC cable to a Ludlum 12 counter.

The grey box is a Convectron tube readout. The 2.75" cross is north of center. HV feedthru is to the right, viewport to the
left and D2 input top. Viewport seen by means of 45 deg front surface mirror. Additional shielding with 2 inch thick quartz window,
Plywood at top holds 1.5mm lead sheet forming radiation shadow. Vacuum pump (turbo-rough) under table holding the cross.

Precipitator supply hidden behind assembly along with HV meter (100Meg and uA meter). ~60kohm ballast. All wiring done
with 40kV silicone wire further insulated with plastic tube.

Th egrey box to the right has nothing to do with the set-up but is a homebuilt HV supply for a microflash unit. Positive output
or I would be using it for the fusor. It has a Variac like good HV supplies should have

It's a very minimalist set-up.

[Mark Rowley]

Alan,
I think your detector is too far away from the grid.

For this type of arrangement, the detector tube distance should optimally be about 15-18cm / 7 inches from the grid. Especially since your initial neutron rate will be low until you become familiar with driving the system.

To be clear, the detector tube itself (moderator excluded) should be around 18cm from the grid. As it looks now, I don’t think you’re getting enough neuts into the moderator to be thermalized for guaranteed detection.

Mark Rowley.

[Mark Rowley]

Here’s a couple pics of my geometry for reference. The red cap on the screwdriver is equal with the location of the grid within the chamber. The brown cap in the paraffin moderator is the end of the detector tube. Distance between the two are 18cm/7”.

The front picture is just for orientation purposes of where the cross point is located.

A couple more thoughts.

The precip supplies generally don’t need a ballast resistor as they have a built in current limiter. You may be choking off your supply. Also, is the current monitoring circuit installed before or after the ballast resistor? Should be after, but I’d suggest removing it altogether.

Lastly, your detector has a ton of moderator. You may be absorbing your thermalized neutrons before they get a chance to get to the tube. But then again, possibly not. Depends on the diameter of your tube. I think 3.5-4” of moderator around the tube is what’s normally called for. I’m sure others here can weigh in on that better than me.

Hope this helps

Mark Rowley

[John Futter]

I would have thought 2" thick wax moderator all round would have been enough
Too much absorbs all ie thermalized to zero. I know our Cosmic neut He3 counters only use 6" thick moderators and the cosmics are much higher in energy than DD fusion

[Richard Hull]

Carl Willis put up diagrams related to moderator thickness years ago. I found a couple of images that I captured years ago. They are related to radiative capture in Silver, but you will note that simple high density polyethylene which is close to Paraffin wax peaks near 5cm or 2 inches of moderator any thicker and it degrades rapidly. I use a 3" water moderator surrounding the 3He tube. (centered in a 6" water cylinder.)
The second image relates to Paraffin specifically. Closer is, indeed, better with any moderator/detector neutron counting situation for new fusor operators.

The upshot is.....I had to say this.....

use moderation in moderation.

No apologies offered.

Richard Hull

[Alan Sailer]

Mark,

The distance issue is valid. I'll look into changing this. I'm at work right now but I'd guess that my grid to He3 tube
distance is twice yours which would reduce the flux by a factor of four.

The voltage is being measured after the ballast. Your observation about the precipitator having current limit is true.
In my reading of the FAQs/threads I got over-sensitized by the insistence on a ballast. In retrospect the power supplies
being ballasted are probably heavy iron monsters. It's probably not an issue but I will think about removing it.

John,

Since the only neutrons I have been measuring are background maybe the 3" paraffin was enough.

Richard,

Thanks for the curves. From the bottom one my 3" (7.6cm) moderator would be "killing" close to 50% of my theoretical D-D
neutrons. Coupled with Mark's distance observations this may be the reason I'm not seeing anything. I work in microwave amplifier
engineering and 9bB of neutron "power" loss is a lot.

My goal will be to reduce both the grid/detector distance and the thickness of the moderator. Should give me a better chance.

Cheers.