OPA111 as PMT preamp any good?

Why?

Obsolete opamp; mediocre performance.

Vladimir Vassilevsky DSP and Mixed Signal Design Consultant

On a sunny day (Sun, 09 Oct 2011 11:22:38 -0500) it happened Vladimir Vassilevsky wrote in :

Found a diagram with an OPA124, looked up the datsheet, it referred back to OPA111, found used OPA111 on ebay for 10$.

So what would you use?

It depends. I would start with application requirements. Usually, noise current is most important parameter for that kind of application; this narrows down the choice.

Vladimir Vassilevsky DSP and Mixed Signal Design Consultant

PMTs put out a lot of signal. Dump it into a 50 ohm resistor and follow with some voltage gain. There's no reason to use a FET opamp.

What are your requirements... gain, bandwidth, drive?

John

They can be run at high gain. It's often not a good idea to draw too much current from the anode or the last few dynodes.

Sauerbrey did a comprehensive study back in 1972

Unless you want a more or less linear response, a respectable life from the photomultiplier tube ...

A 50R load can be fine for single photon counting, if there aren't too many photons. If you are collecting data at a rather higher rates, 50R can be on the low side.

-- Bill Sloman, Nijmegen

That depends. What kind of PMT and generally not if you are counting single photons.

VLV

On a sunny day (Sun, 09 Oct 2011 11:37:10 -0500) it happened Vladimir Vassilevsky wrote in :

Right, and the application requirements are quite open here. I am still experimenting. Apart from the fun of detecting cosmic stuff, and all sorts of gamma radiation using a scintillation crystal, I am also faced with different PMTs I have now with different characteristics, some sort of unknown. I just received 2 PMTs with individual reports and test values in an alien language, to decipher that ... Somebody had a go and came up with 500 V supply for the PMTs, but now I am translating the stuff myself and come up with

On a sunny day (Sun, 09 Oct 2011 09:56:12 -0700) it happened John Larkin wrote in :

See my other posting to Dr Hobs, with pictures and video, where I put the signal directly from the anode into a PNP darlington to drive a LED.

See my reply top Mr Vassilevsky

On a sunny day (Sun, 09 Oct 2011 12:58:52 -0500) it happened Vladimir Vassilevsky wrote in :

But 'photons' do not exists :-)

Jan Panteltje wrote: : Vassilevsky wrote in : >That depends. What kind of PMT and generally not if you are counting : >single photons.

: But 'photons' do not exists :-)

Replace the word 'photon' with whatever you like to describe the phenomenon that, at decreasing illumination level, the current from the PMT begins to be granular, i.e. comes out as bursts.

Regards, Mikko

Well, it's not as if we know what he's trying to do.

John

Too true. Any number of automatic gain control schemes for photomultiplier tubes work very badly at low light levels, when the photons only show up from time to time.

-- Bill Sloman, Nijmegen

You know someone who wants *more* Pre Menstrual Tension???

1750 is high for pmts - it could burn it out or limit its lifetime, you should limit voltage to 20% below max at all times.

Also the SNR changes with voltage, at 200 volts SNR may be near 0, the best SNR may be around 600 volts, as one goes higher in voltage other noise sources happen, PMT glass scintallation, electrons missing gates, and SNR goes down to 0 again as voltage goes up near max.

There is too much background to ever run out of gamma rays, I get about 20 per second here

You need a fast scope to see stuff, 100 MHz is just enough to see pulses using 8 MHz bandwidth on pmt, using plastic scintallator, in a dark place. (poles on pmt are one or two at most)

If you have a pmt with no amp, pulses may be very narrow, very fast, depends on your scintallator, plastic may be 5 ns, Natl is about 100ns, etc.

If your looking for single photons, you need large BW, 8 MHz is quite fast hard to resolve pulses, and 20KHz is just fine for gamma, muons etc.

here is a manual, there are a fuew more great ones out there

not much need for a low noise amp at all, the tube will put out 5 or 10 volts easily with a muon or a strong gamma.

Once you get it all working, and a speaker puting out the pulses, show your wife how radioactive her granet counter top is.

About 25 years ago, the OPA111 used to be the standard for tunnelling microscope front ends. It was easy to remember--1 MHz bandwidth, 1 pA bias current, 1 mV offset, 1 uV/K drift, reasonably low noise.

As Vladimir says, there are better choices these days, at least in all of these specs separately. Whether they're better in all respects for a given application is a judgment call.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

People do sweat blood over photomultiplier tubes. They do tend to be idiosyncratic beasts, and how best to use them depends very much on the application.

Jan Panteltje is slowly revealing more about his application, which seems to be detecting cosmic rays hitting some kind of scintillator crystal, for which I'd be tempted to use a single-photon avalanche diode detector operating in Geiger mode, which can't detect as many events per second as properly set-up photomultiplier tube, but many more than you need to keep track of cosmic rays.

The OPA111 used to be popular for following slower processes.

-- Bill Sloman, Nijmegen

On a sunny day (Sun, 09 Oct 2011 21:36:29 -0400) it happened Phil Hobbs wrote in :

Thanks, I may order some to play with.

On a sunny day (Sun, 9 Oct 2011 18:17:23 -0500) it happened "holyhigh" wrote in :

This is interesting, thank you for the great feedback. The alien documents that came with the latest 2 PMTs I obtained were sort of translated by one that obviously knew a bit about electronics, but not about PMTs. He stated 500 V, and of course I could get no signal at 500 V (anode voltage). Fortunately the PMTs came with individual test data, in alien language, so I started doing some deciphering of that language. Now this document proved to be a sort of Rosetta's stone, it even contained a simple diagram in the style I always draw those myself, low contrast, on what the aliens use for paper, some yellowish sort of paper like stuff. That diagram proved to be one of the keys to decipher the language, as it contained their word for 'voltage' in a place were I would expect it (on the voltage divider). Looking for the same hieroglyps (for voltage) did lead me to an other part of the document were there were numbers. Now some think aliens have seven fingers, unlike us with 5. so if they started counting the same way as we learned it, on their fingers, we, at base 10, then they would use base 14. Now my theory is that all DNA based lifeforms have some sort of symmetry, and that all these lifeforms use a even base for counting, just a theory. Anyways after finding the base those numbers were in, I could write them out in our own decimal system, and found 950, 1250, 1750 for the anode voltage. Now why did this alien translator come up with 500 V? This I had to solve too, to make sure I was not imagining things. Well, now I knew the base, I could decipher other numbers in the document, and all the way at the top I found a place where it said: 350-650. So, what would that be? Well looking at my own XP2432 PMT (a very nice Photonis one, 10 stages), it has a spectral sensitivity peak at about 420 nm (nano meters). So I figured the alien took the first 2 numbers, took it for a voltage (voltage limits), took the average (500) and was kind enough to specify that as the anode voltage. Anyways I have deciphered much of the alien document, and and learning that language (at least the technical part) in the process, as now I knew the text next to those numbers had to be nano and meters. If all this is a bit too much, I really enjoyed the cryptographic challenge, and really felt like the guy who must have used that Rosetta stone. Lovely. Cross sciences, cross universe, electronics as an aid, I have always seen it as a aid, never as a means into itself. The power is to know a little bit about all those fields of science. I already knew about them before I did the time warp thing, but this is the first time I have done business with them.

Exactly.

But photons do not exist. You mean enough energy to knock an electron lose from the cathode.

Oh yes, I have that handbook, and have read it too. Already long time ago :-)

This is with just a darlington driving a LED, no amps.

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:

And this PMT: And this crystal:

Well stainless steel here..., here last years project, before the time warp experiment:

On a sunny day (Sun, 9 Oct 2011 23:21:30 +0300 (EEST)) it happened Okkim Atnarivik wrote in :

Exactly, that strength of EM waves that is big enough to knock an electron free from the PMT cathode.

And that EM energy may be zero at times, for example consider that zilloins [tm] of neutrinos travels through it too, some may sometimes interact, does not have to be EM waves at all.