Needing more than a spark test?

homebrewed said:
Passport, eh? It's gotta be the output of one of those amusing automated translator programs. Some of the Theremino publications are hard to figure out as well, probably for the same reason.
Click to expand...
Nope - definitely no automated translator funny. Just reach back to your math Greek symbols.
ПАСПОРТ
The П is a Greek capital Pi, as in Raspberry.
The А is exactly what you think it should be
The Р is a Greek Rho, as in "R"
The Т is a "T" of the usual kind, another letter where the correspondence to Latin is the same, as is "O".

Also, remember the USSR was spelt СССР for Sojúz Sovétskix Socialistíčeskix Respúblik.
That was Союз Советских Социалистических Республик in Russian.
..from which I infer that the Cyrillic "C" is pronounced "S"
Not quite code-breaking, but the darn thing unquestionably spells "PASPORT"

It may have been a Soviet administrative thing from their cold-war culture. This sort of kit was perhaps not the sort of thing to be in the possession of just anybody. Maybe they just use the term to mean "identity" in the same sense as with people. Who knows?
Now "all" you need is the HV supply, resistor divider chain, fast A/D and MCA S/W :grin: . The python-based stuff you found may be helpful in the latter case. Just guessing here, but I suspect you have a decent 'scope to look at waveforms. I got spoiled at work so my tastes run on the expensive side there; but, really, a 1GSPS Hantek, Rigol or Siglent should be plenty good for stuff like this.
Click to expand...
One discovers all sorts of very sophisticated resistor divider chain in that Hamamatsu handbook. Everything about extracting the signal without noise, and whether it is to be used for "counting" as opposed to amplititude extraction. How to use DC amplifiers. How to AC couple. We note the careful way they filter and buffer the dynode voltages while not limiting the bandwidth of the output. This stuff has been developed for decades. Still quite easy to implement, but it makes the Thermino version look a bit basic. I will be putting some thought into that part.

For Si(PM), the diode current needs to be supplied carefully, so that the signal from it is from photo-induced carriers, not messed up with diode thermal and other noise.

My scope is not as fancy as a 1GS/s Rigol. It is an "older" scope, a 500MS/sec Hewlett Packard 54520A. Even well beyond the 0.5GHz, it has served well for picosecond risetime measurements. I admit it could be time for something a bit younger, but I hardly use it much. In the last 10 years, the kit was usually spectrum analyzers and VNA's from Rhode & Schwarz and Keysight. They were not mine to own. I do have a R&S analyzer FSH3 good to 3GHz, and Anritsu 20GHz analyzer, both "in need of repair". One has a burned up attenuator because some git connected it to a 150W transmitter, and the other has a display supply problem because some git plugged in the wrong polarity charger. To be fixed "someday" if I really need to.
On a slightly different subject I downloaded the LTspice package and tried it with my version of Wine. It seems to work just fine.
Click to expand...
I just knew it would work for you. The full models and reference circuits for the LT range A/D converters evaulation boards are on the Analog Devices site, but if you just peek in the LTSpice library, I think I saw some there. Now that you mention Wine - > infers you also drive more than one kind of OS.
 
Your electronics lab definitely is better-equipped than mine! Fortunately no RF gear will be needed for this project. I do have a cheap SDR dongle I have used for some troubleshooting purposes but with a baseband bandwidth of only 1MHz it isn't very useful as a spectrum analyzer.

Regarding your comment on Wine and different OSes, Linux is our main platform. These days the only time I fire up a Microsoft product is to do our taxes. At work our desktop computers were all WinXX but the design tools we used were Linux-based. I didn't design IC's, I took 'em apart -- but needed circuit and layout info to track down the failure site.

I haven't looked at the Hamamatsu document yet. It sounds like it was a good find.
 
Last edited:
graham-xrf said:
Hi @homebrewed
Have a little explore of https://madexp.com/
Papadopol Lucian (not sure which is the surname) emailed me with some stuff about the FEC84-5 PMT tube having "low gain". Find what you will, and note the amount of lead and radiation stickers the lad uses.

[Edit --> https://madexp.com/2019/06/09/diy-scintillation-probe-the-easy-way/ ]
[Edit #2 The name is Lucian Papadopol ]
Click to expand...
That's an interesting web site -- thanks for sharing it.

It's nice to see some of the details for making a scintillator/detector assembly. I'm not sure what to conclude regarding the massive lead shielding he's using. Maybe he's looking at much higher energy gammas than we will be?

One of his blogs also has a link to another scintillator vendor, epic-scintillator.com. That could be useful.
 
I am after not so much what he does, but rather, where all his links lead to, especially those of his favourite suppliers. It will take a good long while to read all his site.

Meantime, I have also got thoughts about plastic scintillator + Si(PM). The more I think on it, the more convinced I am that there is value in collecting the light from the scintillator in a light pipe, delivered onto a 3 x 3 mm photodiode. To shape an acrylic plastic scintillator into being it's own light pipe is just a cool thing to try.
I have no UV light source, but I am curious what happens to PMMA (perspex) acrylic if one shines UV onto it.

I know it must seem humdrum parochial thing to say, but I can be a bit tightwad stingy when it comes to eBay priorities. More scintillation crystals and suchlike XRF stuff got stalled because I blew the budget on good green laser level kit capable of working OK outdoors over respectable ranges. This is in the interests of the shed/shop/man-cave/home-for-machine-capable outbuilding project.

Checkin' out epic-scintillator.com
 
Being a hobbyist means choosing your battles, funding-wise. You don't sound all that different from me, in that regard. Too many projects, too little money.....as usual. Here's a toast to all the tightwads in the world!
 
@rwm : Yum! it's so long since I had those. No longer allowed the drizzle of maple syrup on it.

Look at a different kind of workshop
More of a lab. As far as we at HM are into building collections of machines, tools, measuring kit, etc. so is Lucian into spectroscopy. There is a not-yet-organized shared folder he uses for some of his stuff.
Just look at the pictures!
--> https://madexp.ns0.it/nextcloud/index.php/s/n2FRPQmxqnjyjCP
Now that we get some idea of what he does - consider this..

His audio-channel codec A/D version of Thermino already exists!
.. and has been delivering worldwide. He used Kickstarter funding for an audio card based spectrometer about a year ago. That is what this gadget is on the madexp site.

--> PMT adapter
--> Audio card based USB Interface spectroscopy
usb-pmt-adapter-jpg.326923


PMT Adapter V2 schematic.png

Keep in mind - his interest is gamma ray spectroscopy, as opposed to xrf.

As I understand it, his method is to drive the switched-mode power supply FET from a PWM (Pulse-Width-Modulated) waveform generated from an Arduino, in effect doing what purpose-built ICs from Linear Technology/Analog Devices do, but getting into the loop to make the voltage adjustable.
Better resolution circuits pictures are in the project documents.

Why a "thin" crystal?
If you guys don't already know why this is so, then we may need help from @RJSakowski, or @pontiac428 .
For me, the power is about the sheer number of photons, an energy density that can disperse in all directions as photons/sec. I get it that the energy in any photon is locked to it's wavelength, and could travel to the ends of the Universe undiminished.

I believed that a scintillation crystal needed to be "thick enough" to have a high probability that an X-ray photon would get stopped, presumably by encountering a scintillation crystal atom/molecule. Withinn limits, there is no "too thick". A thick crystal would flash, and that gets detected.

BUT.. I quote Lucian..
"The problem is that for XRF you need a thin crystal and possibly one with good resolution in the XRF energy region.
I've experimentally understood that I need to buy a CsI(Tl) and slice it by myself and I've done that... I've to search for the pics and resulting spectra.
The CsI(Tl) could be handled also unenclosed, you can cut and polish it and it's very stable to moisture."
---end quote---

.. from which we learn that he has sliced CsI crystal.
He is even planning a crystal growing furnace for CsI and Na(Tl) with his friend Stanislav.

Lucian had cause to cut a thin crystal - and I don't understand why?

Can we take the short cut?
It seems the low cost hardware may already be available - slightly modified, can we use it?
I am going to try the home-brew Si(PM) diode thing. Do consider that there must be a good reason it has not displaced the use of PMTs.

Lucian's new project
With his colleague.. quote
"The high speed spectrometer.
It's self integrated into a single pipe-probe, usb powered, high speed and high quality (FWHM <8.5%).
I've even made my own MCA software: AmericiuMCA, you can see in pictures. It's at the developing stage but working to test our proto".

As it is with the internet, one discovers that there is very little one could contemplate, work on, spend on, that has not already been thought of before!
 

Attachments

  • USB PMT Adapter.JPG
    USB PMT Adapter.JPG
    54.8 KB · Views: 116
The CM108 looks like a pretty decent audio codec -- according to its spec sheet, with an SNR of 93.6 dB it has 15 bits of "real" resolution. Since he's using an audio codec he also is using the Theremino-style filters to condition pulses coming out of the PMT, but I haven't compared the schematics to see if the time constants are exactly the same or not. Still using a discrete transistor amplifier/buffer scheme, I see.

Regarding the advantages of using a thinner scintillator crystal (other than the fact that it might be less expensive), it may be that more of the generated photons have a direct path to the detector, increasing the odds of being detected. This is speculation on my part so don't take it as gospel.
 
Last edited:
You must log in or register to reply here.
Back
Top