Fusor Forums

Hi Guys,

I recently aquired a NIM crate and some gear, and now I am spending a bit of time trying to get my electronic neutron detection system working properly, but it's trickier than I first thought, Looks as if I am stuck on the first step.

I have a Reuter Stokes He3 tube RS-P4-0404, coupled to a Cremat CR-110 preamp, on the original circuit board, then 1400V bias voltage from a NIM to the preamp, and a BNC to jack cable feeding into the mike input on my laptop, which is running SignalScope (oscilloscope), see screen shot below.

Now the problem is the whole setup works like a good microphone , it is picking up the slightest sound in the room, but when everything is quiet it picks up nothing. The preamp is in a cast aluminum case.

Richard asked me in a previous post what the bias resistors were...

There are a series of three resistors ...

HV------[10M]-------C-------[100M]-[100M]------D-----[0.01uF]-----PreAmp

Where C is a 0.01 uF filter capacitor and D is the connection to the detector.

Are there some more tricks here that I need to know?

All help welcome..

Steven

Any variable capacitor with DC volts on it will act like an electrostatic microphone. Almost any two things near to one another at HV can satisfy that condition, including coax, loose wire inside the detector tube, and so on. You might try tapping one thing after another (which is tricky if they're mechanically coupled, too) and see where it is the most sensitive. Sometimes if you listen to the sound of tapping something on both an audio amp and "raw in the room" you can figure out which thing is doing it.

The effect used to drive us nuts in the recording studio business, and we wound up buying some special cable (contained graphite between the coax layers) to help get rid of that.

Microphonics can about in a high gain preamp system.

I worry that your microphonics are coming from the 3He tube. I assume it is hooked up when this condition exists. Tap it very gently and see. A microphonic input FET could also be at fault. Tap stuff gently to locate the culprit. One item or component will be found that far out sounds the others.

Admittedly, your setup should be frozen in 3 space, but true microponics in total stasis conditions shows a big issue problem.

Finally, what is the voltage rating of the .01 coupler cap? 2 or 3 kv, I hope.

Richard Hull

Steven
Replace all the caps with film type especially the blocking /coupling cap
RS has these foil caps unfortunately they bigger than the ceramic types.
All biased ceramics are microphonic
wound foil types many orders of magnitude less microphonic

All our portable radiation gauges use foil types on the detectors for filtering and signal coupling
Ceramic caps also work in reverse as speakers, all the noise that comes out of a glassman HV psu during HV breakdown comes from the multiplier caps

Hi Steven-

sorry but sometimes one overlooks the obvious (at least I sometimes do)..a microphone usually is the best mircrophone- for sure!
Did you check if you selected the right input for your scope SW and do not just see what the in- built microphone form your Laptop records?
If it picks up the slightest sound I would not expect it to be a cap or similar even with a sensitive CSA- but I am no expert for sure!

Btw.: I also had microphonic effects but just when I was "beating up" cables to force this effect;-)

Greets and good luck
Roman

I do too, Roman -- that's a good one!

You know, with the FIR filter in most soundcard a/d's, they won't even see the pulses of this width at all sometimes, and that might also be a problem. The AC97 spec most of them meet is pretty bogus, and some that claim higher sample rates don't really do it and merely interpolate...and when using lower sample rates just pick the nearest sample instead of resampling in software, which is cycle-expensive to do correctly. I was a pro in the PC recording studio design business at one time....AC 97 seemed designed to prevent good quality "rips" if anything.

Roman,

Yes, that's right, one can select either the built in mike, or the external mike input. I already figured this out myself (doh..), but as it turns out my pre amp is almost as good a microphone as the built in one.

Now the interesting part is that the preamp is sensitive to sound even without the HV bias connected. This eliminates the Tube as the main source. Of course this would not be a problem if the real pulses were significantly bigger, but at the moment I can't see any real pulses at all.

I am using an iMic USB interphase ($29.95) with Signalscope ($99.00), and it works like a dream when I plug in my CDV survey meter, with a scintillator detector. I can easily pick up the different energy levels of the clicks. I also have some PC software that works like an MCA with a signal from the audio input. Basically anyone can have an MCA for $29.95. The iMic is a terrific Analog to Digital converter.

http://www.griffintechnology.com/products/imic/

Steven

Steven
Its time to go tapping with a plastic knitting needle or similar to find the errant component, or dry joint

Steven -

About the only thing I can add to all the wise advice and suggestions, is to check things with a BNC (or equivalent) shorting (grounding) cap on the input cable, and then work outward to the preamp. Alternatively, but I don't think it matters here, use a 50 or 75 ohm cap. This way you can separate amp noise from cable microphonics.

You might just have a cold soldered joint in the preamp.

Dave Cooper

Thanks guys for all the good advise, I found the main source of microphonics, but don't really understand why it caused the problem ???

I moved the two 9 volt batteries that power the preamp, out of the aluminum box, and the microphonics virtually stopped. I still get a signal if I tap the box sharply, but not if I clap my hands, which is what happened before.

One down, one to go...still not getting a signal from the 3He.

Do my bias resistors look right (200 MR) with 0.1 uF cap ?

Steven

We tested ours with a few megs -- about 3. The pulses from a proportional mode tube are tiny indeed, and you may have them and just not be seeing them. For grins, we tried running up the volts on our tube to the point where you get "gas gain", which in our case was about 2700 volts, and then we got nice fat pulses - volts, but that's maybe not the greatest idea for having the tube live either.

The amount of energy stored in .1 uf cap at those voltages could itself utterly fry a tube, I'd go waaay down from there. No need for response to near DC.

Again, a good coupling cap would be a 500pf or less at 5kv. Some couplers I have seen are on the order of 10pf at 5kv. You are coupling sub-microsecond signals.

Disconnect all BNC's from the tube and supply to the preamp and start tapping. A disconnected HV supply doesn't mean the tube will not still give microphonics.

Elimenate piece by piece

Richard Hull

Steven Sesselmann wrote:

> I moved the two 9 volt batteries that power the preamp, out of the aluminum box, and the microphonics virtually stopped.
What brand are the batteries? We had to switch from Energizer to Duracell's 9V's at work (we use so many Duracell's you may have seen our product in a Duracell commercial).

The Energizers don't have enough compression to keep the stack intact if dropped or moved suddenly. You may have one that is on the edge of opening, which can happen after temperature cycling, so little vibration modulates your power supply.

Thanks again for all the help, I now have a whole lot more respect for those reliable survey meters that just keep on going on, these charge sensitive preamps are temperamental to say the least. once you get it right, "Dont touch it".

Okay, I moved the two Duracell batteries out of the box, and that helped the microphonics, then I resoldered the caps on the circuit board and changed the two 100MR resistors with one 10 MR resistor.

I still had nothing but noise

I then became suspicious of the dual HV NIM power supply that I got on ebay, other than the red indicator light I had no way of testing if it worked. So I dug out an old antique bench style rate meter and used it as a HV supply for the 3He tube, and there were my pulses

Now I still have some learning to do...

There are a number of random small pulses, interrupted by the odd pulse which hits 2.5V on my scope screen, about every couple of seconds. probably too often to be stray neutrons ... I hope.

The problem is pulses never go higher than 2.5V.

Is this an output limit of the preamp, or is it the input limit of my scope?

Now if I am looking for neutrons, I assume I am looking for something bigger, so how can I figure out which is which?

Steven

Pictures:

1) Signal directly from preamp with 3He tube (1400V bias)

2) Signal after it has gone through NIM SCA (single channel analyzer) and NIM amp.

3) The whole mess will look a lot neater when I get it all up in a U rack.

....Have you guys heard the expression

"When you ass'u'me you make an ass out of 'u' and me...

Well Roman Radke was almost right, I just discovered how silly I have been, spending three evenings trouble shooting this preamp problem..

The dual NIM power supply I got from eBay, has two ten turn knobs at the front, marked A and B and two SHV connectors at the back, marked A and B, pretty simple right...., except when the NIM is in the rack, I can't see the back, so I naturally assume that the plug on the left corresponds with the knob on the same side of the NIM, because I am working from the front leaning over the box...

Not so, whoever designed this thing decided to cross the connections over.

Now I am really embarrassed....

Steven

Steven
I think the correct word is
Bugger

you will not be doing that again!!!!!!!!!!!!!??

Two steps forward - one step back, Steven

A moves naturally to B, left to right as in natural alphabetical reading. This is observed on both the front and the back, thus, the seeming reversal to someone feeling they should go straight through. The important thing is they are marked pefectly correctly. This convention is generally observed on all gear.

Without a hot radium or gamma source next to the tube, you can't be sure. The taller pulses after the SCA windowing are probably all cosmics or ultra hot gammas from radon daughters, but it seems about right. I would crank the upper level detect a bit higher until I saw about 5-10 cpm max if I did not have a hot gamma source at hand. Big tubes would see about 10-15 as background and small tubes less than 5. The little Nancy Woods BF3 tubes in rem balls see about 2cpm background.

Richard Hull

Count me in as another hobbyist dealing with microphonic components.

My BF3 tube - charge sensitive preamplifier - oscilloscope benchtop experiment came together this week, as reported in another thread. A major nuisance is trigger events whose displayed waveforms don't look like normal pulses from ionizing-radiation. They happen for sure any time there's a mechanical disturbance.

Before turning to Internet and fusor.net, I dedicated a short lab session to locating the cause.
Took control of the scope Horizontal and Vertical, while tapping and wiggling things.
The most common erratic waveform is a lightly damped oscillation at about 260 Hz.
Looks like the main culprits are the HV blocking cap (a ceramic disk) and the BF3 tube itself.
Fusor.net was more productive than Google, in revealing that both device types are familiar offenders.

NIM bin, bias power supply, home-made SHV connector, and HV coax cable are off the hook. Each can be jiggled without causing a disturbance on the scope screen. And the bias circuit inside preamp box begins with a 47-ms lowpass filter (10 MΩ, 4.7 nF).

Next step: replace the blocking cap with one made of better stuff.
Found a 330 pF, 2000 V silver-mica capacitor in parts bin! Practically ideal, I think, while suspending the desire for 3000 V rating on all HV caps. The reduced capacitance will allow HV bias to be ramped up or down 3 times faster than before, with no more stress on the preamp circuit. To use a plastic film capacitor would require shopping, or squeezing in something electrically and physically too big.