Pancake- works as NOT specified

[myID]

Hi-

my pancake tube should have a bias voltage of 400V.
My ESP 2 has a minimum output voltage of 500V.
To not kill the tube I simply built a voltage divider from 2 10Meg resistors and connected the tube parallel to one of them. I adjusted the bias voltage to 800V and it works fine- incredibly sensitive that thing....
Does this make a difference when it comes to high count rates?

Kind regards
Roman

[Richard Hull]

You have set a distinct time constant there between the tube capacitance and the biasing resitors. You would need to scope your pulse in the ideal case to see what your net dead time would be and correct from there to get a high rate that would be reliable.

Richard Hull

[Carl Willis]

Hi Roman,

>Does this make a difference when it comes to high count rates?

The short answer is that yes, it does. By adding that chain of resistors, you have effectively added a lot of resistance into the charge-collection circuit, which probably had only a couple Mohms to begin with. The time constant t = RC where R is the load resistance on the tube and C is the anode (and cable) capacitance of the tube is usually chosen to be a few microseconds in order to limit pileup effects and improve counting behavior at high count rates. You may have increased RC by as much as a factor of ten. A lot of people accomplish the same thing more casually by putting tubes on long cables. Now whether or not you will run into counting-rate limits because of this, or because of tube inherent dead time, is dependent on the tube and circuit specifics and the source rate you are counting.

Also, you might want to check that you are not loading down the power supply too much with 20M across it. Some such supplies will have a problem holding up voltage (or at the least will eat through your battery much faster). Use of a resistive divider to drop the voltage on a Geiger counter power supply is suspect because the high values of resistance that are acceptable loads for the supply are also probably unacceptable from a time-response perspective.

-Carl

[Doug Coulter]

These circuits are fine, but the limit on count rate is rarely the pulse width due to RC time constant in geiger type tubes, it's the positive ion drift time, tends to be on the order of some 10's to a hundred uS anyway, and during the tube dead time, you can still get a pulse, kind of, but it tends to be lower amplitude if visible at all. All the detector design books mention this, and it's easily verified with a scope on a detector with a hot source -- there's always a gap between the pulse the sweep triggered on and the next full size pulse.

[Richard Hull]

What kind of rate do you want to count? Is it critical? Your circuit will have to be worked out around that. The use of a good o'scope is a must for full knowledge of what you have. Otherwise, you are working in the dark.

I assume this is a portable app. in an already extant ESP-2 counter? If you are using the ESP-2 software to adujust the voltage, then you are not taking advantage of the cal pot ranging. There is a pot that can control the set voltage up or down in the guts of that terrible beast outside of software. If you are brave, fiddling with it might let you set the real voltage down to 400 volts. It will probably screw up future software settings, but that's the price of using locked in software.

I hate software controlled devices that feature 715 individual setting functions controled by 5 rubbery pushbuttons, readout on a less than favorable, tiny, 10 character LCD panel with only 8 dots per character. Do you feel my pain?

Richard Hull