Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Dead time for radiation detector

  1. Jun 17, 2013 #1

    Iam working with a High purity Germanium detector and have an issue with the dead time.
    I let the detector do a background measurement over the weekend, about 55h. When I checked the measurement 60h had passed and the measurement was not finished due to dead time about 20%. At the start of the measurement (up to 3h) there was no dead time so this must have appeared later on.

    Real time = elapsed time
    Live time = including dead time and if dead time >0, then real time > live time.

    My question is, if at time 0 both real and live time are equal, but at later time, real time >> live time, is there a way to interpolate ( I have a set of measurement points) back to where this dead time possibly appeared, like intersection with x or y? Or have I misunderstood the concept of dead time in a detector.

    If this topic is in the wrong part of the forum, please redirect me.

    Thank you very much!
  2. jcsd
  3. Jun 17, 2013 #2


    User Avatar
    2017 Award

    Staff: Mentor

    How do you set those 55 hours? I would not expect that your measurement actually detects the dead time in any way.

    A constant dead time of 20% would mean that every hour, you lose 12 minutes due to dead time. Every minute, you lose 12 seconds (assuming the dead time per event is much shorter than a second). There is no single point in your measurement where this appears out of nowhere.

    A dead time of 20% looks way too large for a real dead time in a background measurement. I would expect that the measurement stopped completely at some point, completely unrelated to a detector dead time.
  4. Jun 17, 2013 #3

    Regarding the measurement time, I have an MCB connected to a software that stops colecting signals when the desired time is reached.
    Anyways, I would like to take back my first question, there seems to be another issue with my detector. My first thought was that the dead time must have appeared later on, but when I think about it, I cannot say for sure the dead time were 0 at the start.
    However, the dead time is still the issue, as you said, it is way to high for a background measurement.
    I have noticed that my MCB (collecting the signal from the detectors preamp/electronics) registers way too many counts (registered pulses from radiation) than possible and therefore the dead time is high, naturally.

    This issue seems to dependend on:

    1) A setting in my MCB under the category "Amplifier PRO" named "Low frequency rejecter", when enabled it raises the dead time to ~90%.

    This issue seems to NOT dependend on:

    1) The voltage applied. When HV (high voltage) is off the system reads about 300 counts and dead time is 20%, when the voltage is applied (~4KV) the counts is 6000 but the dead time remains 20%.
    2) The active channels in the detector (0-16383). I could set collecting channels to 500-501, i.e only two channels collecting signal, however the MCB still registers 6000 counts therefore the dead time is the same (20%).

    Any ideas what could be the cause of this issue with my detector?

    This is a very expensive piece of equipment so I am very thankfull for any help!

    Additional information, I have access to an oscilloscope, is there a test I could make to clarify the root of this problem?

    Edit: I have added a picture from the manual about the "low frequency rejector" setting.[PLAIN]http://i39.tinypic.com/qqpml5.jpg
    Last edited: Jun 17, 2013
  5. Jun 17, 2013 #4


    User Avatar
    2017 Award

    Staff: Mentor

    Probably something in the readout electronics and software. I don't know how your setup looks like, and how the components operate.
    As I expected, it is not the dead time of the detector itself.

    That software should use some clock, independent of any dead times.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook