Light ampl;ification by experimentation

Light amplification by experimentation;

Making the cosmic rays visible:

cosmic rays and what have you | \ / +5V [ NaI(Tl)] | [ xtal ] |

Reply to
Jan Panteltje
Loading thread data ...

PMT's have a 'Dark current' or leakage. I think your circuit needs a modification.

Cheers

Reply to
Martin Riddle

On a sunny day (Tue, 4 Oct 2011 21:30:08 -0400) it happened "Martin Riddle" wrote in :

My circuit works like a charm. You seem to have some concept about PMTs. I have seen others struggle here with PMTs... It is simple. A good PMT has very little dark current anyways. Where would the electrons come from? Planck would ask you. Try measuring something.

Maybe you connected the resistor divider to the input? The base of the transistor connects directly to the anode of the PMT. The voltage divider goes from -1000 V to +5 V, and the decoupling caps in the last few stages before the anode also to the +5V. I have 2 setups like that working here, and about a year ago I published the diagrams here.

Reply to
Jan Panteltje

e"

degrees blue LED

he +5V.

the diagrams here.- Hide quoted text -

Any sort of discriminator in your circuit Jan? I can believe that cosmic rays give 'big fat' current pulses. (Much bigger than the thermally generated dark count that Martin is worried about.) Probably most of what you are seeing are muons.... Here's a fun experiment, look for double pulses from your PMT, separated by ~ 0 to a few micro seconds.

George H.

Reply to
George Herold

On a sunny day (Wed, 5 Oct 2011 06:28:52 -0700 (PDT)) it happened George Herold wrote in :

The way the circuit works is that the pulse is fed into a comparator in a PIC, and at the same time into the PIC AD converter. The 10 bit AD converter is used in its fastest mode, using only the 8 most significant bits, and is started when that comparator sees the pulse go high. It is all optimised for speed, the fall time of the pulse is long enough for it can be digitised reliably. As soon as digitising ends then the software shorts that PMT output with a transistor, to discharge any charge, and it is then ready in the fastest possible time for the next pulse. (that is my invention). The output bytes are fed via a serial link as binary 8 bit values, and displayed with my software, here is what it looked like a year ago, many things have been added and changed, that was a first quick hack and test:

formatting link
Now there is an LCD output, signal processing, and it will identify many of the isotopes. That is all I am allowed to tell you:-)
formatting link
Of course in these times of US aggression there are many states working on nuclear weapons etc. and this is a commercial project. US got its rear-end kicked today in the UN by Russia and China, Rice was 'outraged', well now if the black soul in the white house also gets 'outraged', he may trip over that button, and then we all need to check what radioactive isotopes are in our food and drinks. A market is being created as I write this. 'Outraged' sounds a bit like a kid that lost its temper because it could not steal the icecream [1] from the little guy (Assad).

[1] oil.

As far as co[s]mic rays are concerned, some think I use those to communicate with the aliens, the code is really simple, it is the time between 3 pulses of the same hight, divide it t2 / t1 and write the digits out as a ASCII string, there is much in there. The experiment you are referring to could be using 2 crystals and a coincidence detector to test muon speed? According to Einsteins relativity the muons can only reach earth because for them time runs slower. Anyways if you see 2 pulses in the detectors at the [almost] same time, you may have a co[s]mic ray, or a muon. I have not had a lot of time to play, been doing hardware design and writing soft all day.

This Obama and his cronies like Soros are very dangerous people, even before I heard the news that the UN kicked his rear-end I was wondering if he would, in a moment of insanity, attack Russia for its oil. Weak people like him are very very dangerous, they will over-react on anything. OK :-)

Reply to
Jan Panteltje

ddl=3D

)

max

o t=3D

ed =3D

t significant bits,

a transistor,

e for the next pulse.

Oh, you are using whatever capacitance exists on the PMT / amp output as a integrator? Pulse height is proportional to total charge.

splayed with

n added and changed,

of the isotopes.

You've got some radioactive isotopes that you are using to calibrate it? (Oh maybe you can't tell me that :^) I've never done much nuclear stuff.. pulse height analysis and all that...

n nuclear weapons etc.

'outraged',

y trip over that button,

nd drinks.

not steal the icecream [1]

ate with the aliens,

ght,

ch in there.

idence detector to test muon speed?

No, If you were just looking at 'cosmic rays' then most of them are muons. (at least on the surface of the earth.) Every muon that goes through your crystal/scintalator gives a pulse. Once in a while a muon gets trapped in the x-tal. And then decays in ~2us (on average). This decay gives a second pulse of light.

George H.

for them time runs slower.

ou may have a co[s]mic ray, or a muon.

ing soft all day.

ore I heard the news that

sanity, attack Russia for its oil.

thing.

Reply to
George Herold

+5V.

diagrams here.

A good blue-sensitive bialkali PMT has about 70 electrons/s/cm**2 coming off the photocathode. (See e.g. Hamamatsu R6353P, 30 counts/s, 0.44 sq cm photocathode area.) Most are quite a bit higher than that, maybe a factor of 10 or so, and red-sensitive ones with negative electron affinity (NEA) photocathodes are not as quiet.

At a gain of a million, 30 counts per second is about 5 pA, not counting the background from the dynodes, which is easily removed by thresholding.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

On a sunny day (Wed, 05 Oct 2011 12:27:56 -0400) it happened Phil Hobbs wrote in :

+5V.

diagrams here.

Thanks, and 5 pA x 250 (for beta) in my circuit is 1250 pA or 1.25 uA. for 'dark' in the LED, so it is visibly off. Of course this differs widely from PMT to PMT, and depends on how much anode voltage you apply (gain). With the circuit I described in the other tread, I can either continuously change anode voltage from 0-1000V (or more if must be), or do it in 32 steps from the software. The analysing software then does the 'AGC' thing if needed. I was wondering what would happens if I added one transistor and made a darlington. That would give you 300uA 'dark' in that LED, with the advantage of pulse-stretching. The Nichia LEDs came with a warning not to look into these...

15 degrees is very narrow, and blue really looks bright. I have some more smaller PMTs on order, given some time I will play a bit. The lab is well equiped these days, amazing, even a pizza oven now :-)
Reply to
Jan Panteltje

+5V.

diagrams here.

Assuming your transistor has any beta down there, it's actually 1.25 nA.

darlington.

pulse-stretching.

Nice. I now have an espresso machine, but no pizza oven. Something to aspire to. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

On a sunny day (Wed, 05 Oct 2011 18:50:20 -0400) it happened Phil Hobbs wrote in :

+5V.

the diagrams here.

Ooops!!!!

But that LED is still off:-)

darlington.

pulse-stretching.

I buy them frozen, 'Dr Oetker', 12.5 minutes to prepare.

Reply to
Jan Panteltje

I did some experiments, and observed some interesting effects. First I changed the simple one transistor PMT anode current amplifier to a darlington: +5V | PMT e cathode anode e-----------b BC557B -- K A ------ b BC557B c PNP

-1000V c-------------| | | | | | | | PNP |

- R R R R R R R [ ] 1k voltage | | divider /// | --- / \ / / LED blue --- / | ///

This flashes the LED on those BIG pulses that happen every now and then when [the aliens] send us their high energy particles.

formatting link

What I wanted to know was if there was a coincidence between the big pulses counted with a low sensitivity geiger tube, and the ones detected as light flashes in the crystal in front of the PMT. Using this geiger counter:

formatting link
And this PMT:
formatting link
And this crystal:
formatting link
formatting link
That is 4 inches long, picks up a lot. If you scope it, then you see a whole 'grass-field' of pulses, most too small to light the LED. This was filmed in very bright daylight, if it is dark you can see many more pulses from the PMT. Or you can turn up the PMT high voltage, and get it lighting most of the time (not from dark-current though). This was recorded with about 900 V on that PMT, it is specified at 1250 V. Its spectral sensitivity seems to match well with these crystals. What I see, is that there *normally*, so most of the time, is no coincidence. But sometimes the Geiger sees what seems like a lot of flashes at the same time, I call this a 'shower' of particles, and then the PMT sometimes also sees a flash. I think this is a co[s]mic ray, that then hits the earth atmosphere and causes a shower of particles, some of which hit both detectors.

Reply to
Jan Panteltje

I did some experiments, and observed some interesting effects. First I changed the simple one transistor PMT anode current amplifier to a darlington: +5V | PMT e cathode anode e-----------b BC557B -- K A ------ b BC557B c PNP

-1000V c-------------| | | | | | | | PNP |

- R R R R R R R [ ] 1k voltage | | divider /// | --- / \ / / LED blue --- / | /// Here the same thing, recorded in the almost dark:

formatting link
You can see zillions [tm] of pulses, some strong, and the occasional huge one that shows up in bright daylight. I want to point out again that this has nothing to do with 'dark current', if you remove the crystal then the LED stays off.

So zillions [tm] of particles hit the crystal and cause tiny flashes of light, some are only nanoseconds long flashes. The hole storage time of the transistors darlington configuration stretches the pulses, or in more common language, the transistors are slow to switch off.

I am now playing with the idea of a parametric PMT preamp, making diagram sketches...

I wrote, relevant to this subject:

flashes

to light the LED.

pulses from the PMT.

(not from dark-current though).

flash.

Reply to
Jan Panteltje

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.