Good point indeed, lately in Sept'15 in the context of revamping electronic s of our pollutant analysers (NOx, CO, O3 ...) i begun a design with a Roye r based CCFL (Copper Bussman) CTX210605-R - XFMER ... etc. i finally decid ed to bring up the same with the C4900-01 from Hamamatsu (for only 100 EURO ) and believe me i did not regret such a choice (small, efficient, very qui et). I'm too old to being amused with these stuff and time is money.
In a PMT, very nearly all are. There's a big peak at a few eV, and then basically nothing till the backscatter peak up near the incident energy. See the Philips PMT book, Fig. A1.11. So it might be 103 or 123 eV instead of 100 or 120.
Okay, 150V gets you a whole 0.025 c. The tube in view here is running between 55 and 90V per dynode (with tapered bias).
Nope. 2% at very most. 8% of c requires an energy of
E = 0.5 * 0.08**2 * 511 keV = 1.6 keV, or 2.5 keV for 0.1 c.
You aren't getting anywhere near that in a PMT. You're probably confusing it with a SEM, where you can get up into the tens of percents of c. (10 keV gets you about 0.2 c.)
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
hobbs at electrooptical dot net
http://electrooptical.net
My wife likes that, I prefer the original dark Guinness. One of my bike ride "gas stations" has it on tap. But last time I had an Oak Park Russian Imperial Stout that clocked in above 9 volts. Sometimes they even have Pliny the Elder.
Sounds good. As long as they did not inseminate it with lots of chocolate. I do not like spiced beer. Pumpkin Ale is the absolute horror to me.
For a nice evening at home the Belgian beers with a cork in the bottle are my favorites. Where smoke comes out after popping the cork.
I am eyeing the CTX210655-R which is its 2.5 watt brother, should give me 1.3kV out at around 6-8V drive. The coil in the DC intake is chosen so that the drive current is as close to sinusoidal as possible over the control range I need. But there's going to be a linear regulator behind it which makes the output very quiet. Since the CCFL transformer and surrounding ciruitry will reside in the instrument's base it should also be magnetically quiet.
Are you going to used a resistive divider chain for the dynodes? Why not do the C-W trick and filter the outputs? That doesn't need any high-voltage parts... everything is stacked in series.
The divider just wastes power. Go to a lot of trouble to make high voltage, then burn it up.
I recently did a 1400 volt C-W supply which drives two separate 0-1200 volt linear regs.
formatting link
You wouldn't need the pulldown part.
If your regulation range is narrow, below 500 volts, you could use an opto and a LND150 depletion fet as the pass element, which would be super-simple. Win says that the LNDs avalanche nicely somewhere north of 500 volts, so that would be asfe.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
Yup, I like those Belgian "Abbey" ales too. And I agree I usually don't like fruity or other stuff in my beer. The chocolate stout has a lot of chocolate and alcohol taste... bordering on a liqueur maybe.. but underneath it's still beer! Nice once or twice a year during the holidays.
I picked up some Ale Syndicate "Van de Velde", a 6.4% "Belgo-American Ale". Unlike the usual banana tint of Belgian styles, I swear it's more like....blueberry? Very interesting, and moderately refreshing.
Only for the younger dynodes. The last ones get emitter followers, with it all burning around a watt under no-load. It is based on the recommendation of an expert we both know, we did sort of a relay run where the baton just got handed over :-)
Not sure what C-W is but it's all filtered and scrubbed until SPICE could no longer resolve the residual noise.
The only thing I can't have is a PFC choke on the secondary side because that would develop St.Elmo's fire over time.
Nice. Designs like that can be fun. Until debug time comes. One wrong move with a finger and ... ZZZAPPP. Happened to me on a large Pockels driver where I had the HV warning on the board in bright yellow, in English and Spanish. But not in German, so ...
I only have a token pull-down so that stuff won't float. But at HV almost every resistor is in reality a resistor chain because the working voltage even with 1206 parts is very limited.
I have to use PNP (or P-channel) because in a PMT the energy must be sourced from negative, from the cathode side. 500V is the max you can get in reasonably obtainable small semiconductors.
My favorites are Chimay, La Chouffe, Delirium Tremens, Westmalle. Belgians always knew how to make good beer, especially the Trappiste monks. I lived in the Netherlands for years and it was only a short bike ride into Belgium. Sometimes on the way back I had to push the bike.
C-W is a Cockroft-Walton multiplier chain. Connect each stage to one dynode.
formatting link
The cool thing is that you can make a huge voltage from a modest AC source and low-voltage diodes and caps. And you get all those intermediate voltages with no power loss.
Here's my 1400 volt supply:
formatting link
A skull and crossbones is universal.
formatting link
Well, sort of like that. That sticker is from the Zeitgeist Ski Team. What a concept.
A depletion fet and an opto is a 2-terminal floating conductance. It can go anywhere.
Speaking of anywhere, I've got to pack the Audi with presents and stuff and head up to Truckee. It's sunny today, so we can sneak in between the storms.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
That's where the ball-park 10% of c came from. My patented wide-voltage ran ge beam blanking scheme had to work from 15kV to about 300V beam energies, with blanking plates (actually a pair of round wires) about 18mm long, whic h was more than 0.5nsec of transit time at the lower beam voltages.
The core of the patent was having the wires parallel to the beam at 15kV, a nd rotating them until they were normal to the beam at about 1.5kV and lowe r. Obvious enough, but it still get a US patent.
That all got worked out in some detail - I've not had to worry about electr on transit times in any detail since then, and it does show.
The divider lets you adjust the voltage drops across the various stages. Fo r perfectionists, the photocathode to first dynode voltage should always be the maximum the tube can support - this maximises the gain at the first dy node, which gives you the least possible extra multiplication noise. After that the interdynode voltages can vary to give you your variable gain.
The voltage from the last dynode to ground doesn't have any direct effect o n the gain, and should be stabilised to make sure that it won't change as t he current flowing through from the last dynode to the anode changes. This avoids changing the voltage drops across the rest of the dynodes with anode current, which is a well-known source of photomultiplier non-linearity.
Setting this sort of stuff up with a Cockroft-Walton multiplier as your hig h-voltage source can be tricky, though it's perfectly do-able. John Larkin doesn't seem to have done it, but he likes his systems simple and easily co mprehensible.
For perfectionists, the photocathode to first dynode voltage should always be the maximum the tube can support - this maximises the gain at the first dynode, which gives you the least possible extra multiplication noise. Afte r that the interdynode voltages can vary to give you your variable gain.
on the gain, and should be stabilised to make sure that it won't change as the current flowing through from the last dynode to the anode changes. Thi s avoids changing the voltage drops across the rest of the dynodes with ano de current, which is a well-known source of photomultiplier non-linearity.
igh-voltage source can be tricky, though it's perfectly do-able. John Larki n doesn't seem to have done it, but he likes his systems simple and easily comprehensible.
Nice, thanks... I was thinking about this and I think the C-W pickup we were seeing was north of 200 kHz... 220? I redid the front end on the pmt... nothing special. but it could have been a ground thing/ pickup. The hamamatsu sockets may have gotten better too. Getting the ground right is like the other 1/2 of electronics.
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.