Needing more than a spark test?

[WobblyHand]

All the "rubbish" unwanted photons emanating from the material around the source source cannot get back to the detector.

In an idealized world, this is true. Is the instrumentation conforming to an ideal setup? Probably not yet.

There's a lot of stuff, perhaps rubbish, perhaps resultants of Mark's signal processing or sampling. This is not meant as a criticism, there's usually a consequence to what ever choices we make in processing. Nonetheless, is seems there's still a lot of junk in the output. Going to be hard to do quantifiable spectral analysis with the existing data. Qualitative might not be easy either. Mark is definitely out front of us, actually trying things and trying to make improvements, I know I feel like I am on the outskirts looking in.

The question I have, is this the best we can do? Or a more pointed question, what can be done to improve the spectral data collection? Is this a limitation of the jitter due to the triggering, which might cause smearing, or is there something else that is more fundamental?

Why are the apparent count rates so much lower than expected? Are our thresholds too high? Something else? Feeble/old/not up to specification sources could be one reason. I'd think that conjecture could be tested somehow. Any other things that might account for the low count rates?

If I understand correctly the photons are scattered spherically? But we only get the photons that are contained within the solid angle in our direction? And we have a probability, depending on the energy of the photon, of the detector even seeing that photon? I believe this is what you were calculating earlier. Just checking if I have grasped the basic ideas, not the totality of the solution just yet.

 

[graham-xrf]

In an idealized world, this is true. Is the instrumentation conforming to an ideal setup? Probably not yet.

There's a lot of stuff, perhaps rubbish, perhaps resultants of Mark's signal processing or sampling. This is not meant as a criticism, there's usually a consequence to what ever choices we make in processing. Nonetheless, is seems there's still a lot of junk in the output. Going to be hard to do quantifiable spectral analysis with the existing data. Qualitative might not be easy either. Mark is definitely out front of us, actually trying things and trying to make improvements, I know I feel like I am on the outskirts looking in.

The question I have, is this the best we can do? Or a more pointed question, what can be done to improve the spectral data collection? Is this a limitation of the jitter due to the triggering, which might cause smearing, or is there something else that is more fundamental?

Why are the apparent count rates so much lower than expected? Are our thresholds too high? Something else? Feeble/old/not up to specification sources could be one reason. I'd think that conjecture could be tested somehow. Any other things that might account for the low count rates?

If I understand correctly the photons are scattered spherically? But we only get the photons that are contained within the solid angle in our direction? And we have a probability, depending on the energy of the photon, of the detector even seeing that photon? I believe this is what you were calculating earlier. Just checking if I have grasped the basic ideas, not the totality of the solution just yet.

This is a fanciful ideal - but anyway..
I have a a speculated sequence I would probably try. It starts with the ADC, and achieving the data sheet specified dynamic range. The short-circuit input had better not have any unexplained counts. Probably it won't be zero, because it's quite hard to get that to happen. The common point of the digital and analogue return currents is at the ADC. Limiting the amount this can get bounced around by computer clock noise is what I want. Screening this from the computing stuff, and everything to do with any power supply voltages both can see, is important. Even screening the amplifier is tricky, I won't be wrapping it in copper tape, but I will be doing something.

Then I go back through the gain stages test points. I would probably hook up some kind of artificial signal here. I want amplitudes measured by the ADC to track these, pretty much exactly. Very tricky is to measure artificial signal generated at the input, but I would try.

Then, with the diode on, but lead blocked. I would want to know about dark current noise counts.

With some Am241 aimed straight at it, right up against the paint. I would want to see where 60KeV should plot. I would want to display the pulses from test points. I would want to know how many per second are happening.

Finally, just like Mark did, I would start showing it materials. Essentially, I would want to know exactly what the electronics is doing. It has to be good enough to to discriminate between the energies we hope to see as separate peaks.

One thing we do know. It cannot reasonably be feeble sources. Even if the Am241 was created in a 1950's reactor. The half-life is 432 years. The little speck is going to keep pinging away. It's just that the total flux energy we have out of that little half a millionth of a gram is so low, compared to a high voltage X-ray tube only inches away.

 

[WobblyHand]

Good bring up of the circuits. Very methodical. Well done.

One thing we do know. It cannot reasonably be feeble sources. Even if the Am241 was created in a 1950's reactor. The half-life is 432 years. The little speck is going to keep pinging away. It's just that the total flux energy we have out of that little half a millionth of a gram is so low, compared to a high voltage X-ray tube only inches away.

We are assuming that Mark has received high quality Am241 sources from some unvetted AliExpress supplier. How do we know that we are getting a full dose? It might not be true that the sources meet standards. Or the ones for export may not. If we had a way to test them, then we could remove this as a possible cause. It's admittedly pessimistic, but the only way you know for sure is to measure it or infer it somehow. At the moment, I do not have a way to measure the sources.

I will admit, if they are compliant, there's still a bit of a problem (or three) to solve.

 

[homebrewed]

@homebrewed how does one measure vRef? There's no pin. It's an internal node.

Sorry, I wasn't specific enough. The T4's just short Aref to the board's +3.3 so measure that.

 

[homebrewed]

Have you tried literally putting a small piece of lead on the PIN diode and checked the background level? Is it low, or is there scatter with stuff coming in from the edges or from behind? Trying to figure out the nature of the problem...

Are you talking about doing that with the sources in place? I think that's the case, just want to make sure.

 

[graham-xrf]

Good bring up of the circuits. Very methodical. Well done.


We are assuming that Mark has received high quality Am241 sources from some unvetted AliExpress supplier. How do we know that we are getting a full dose? It might not be true that the sources meet standards. Or the ones for export may not. If we had a way to test them, then we could remove this as a possible cause. It's admittedly pessimistic, but the only way you know for sure is to measure it or infer it somehow. At the moment, I do not have a way to measure the sources.

I will admit, if they are compliant, there's still a bit of a problem (or three) to solve.

The process of making them is expensive, long, and complicated. The origin of the material is very special, made from plutonium left in highly radioactive spent nuclear fuel. There are not so many places that re-process spent fuel. The making of the little punched out buttons is still very highly regulated, simply because it's about the only radioactive isotope widely used among the general public. That the amount is always 0.9uCi, regardless the "brand", or whether it came via AliExpress is telling.

I think the source will, to high probability, have the properties printed on the casing, and fundamental physics pretty much forces that they will not change much, even over decades.

 

[homebrewed]

Good bring up of the circuits. Very methodical. Well done.


We are assuming that Mark has received high quality Am241 sources from some unvetted AliExpress supplier. How do we know that we are getting a full dose? It might not be true that the sources meet standards. Or the ones for export may not. If we had a way to test them, then we could remove this as a possible cause. It's admittedly pessimistic, but the only way you know for sure is to measure it or infer it somehow. At the moment, I do not have a way to measure the sources.

I will admit, if they are compliant, there's still a bit of a problem (or three) to solve.

You have an excellent point, and one I've worried about as well. That's why I also have been working on an ionization type detector to characterize sources. The biggest problem is how to calibrate the thing. It will be most sensitive to alpha particles so the best approach would be to find an alpha source with a known emission rate. Worst case is we mail the same detector to participants to see how close we all are, in terms of Am241 source activity.

I have attached a schematic for the ioniization detector. I also have designed a board for it, but it's on the expensive side because it's on the large side and I used a 4-layer board -- it was a lot easier to properly shield the TIA that way. The charge pump is used to provide a fairly high voltage to sweep carriers to the TIA. Untested, I don't know how well it will work. I specified TL072's just as a placeholder so something else will most likely be used instead. With a 1Gohm feedback resistor the amplifier will need to have very low bias current, at least down in the low pA range. And I just realized I didn't include a capacitor in the TIA's feedback loop. Not too hard to add & actually best to hand-add, due to the guard ring.

 

[RJSakowski]

I am hoping @RJSakowski may be able to help us with what might happen with plastic.
I get it that the energies from carbon, hydrogen, oxygen, etc are so low, we cannot detect them.

The problem is, what ever low energy got subtracted when the plastic got hit, should also release the non-expended energy as photons, still quite high level x-rays, able to go on and hit more stuff, with entirely new random directions.

It is that - or we don't understand what happens to the exit energy.

Regarding the future of an x ray photon after entering plastic, momentum must be conserved along with energy so I would expect that the interaction with a nucleus resulting in the elevation of an electron to a higher level would be the x ray photon would continue on close to the same path but with less energy.

 

[graham-xrf]

Would one of these help? -->https://www.aliexpress.com/item/1005004817721983.html?
Of course, it says "detector", as opposed to "measurer", and it seems full of whizz stuff like "vibrate", and "glitter", whatever that may be?
It's even got a "spectrum" display. ??
Still, I think it might have more in there than the $16 bucks "dosimeter".
It claims to see gamma!!

 

[graham-xrf]

Regarding the future of an x ray photon after entering plastic, momentum must be conserved along with energy so I would expect that the interaction with a nucleus resulting in the elevation of an electron to a higher level would be the x ray photon would continue on close to the same path but with less energy.

Thanks RJ. That means @homebrewed Mark can expect that plastic can reasonably replace making the important part out of aluminium.

But .. there you go having me think about a photon having momentum! It's like light, is it not? Momentum is m*v, and v is speed of light.
Does a photon have mass?