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

Nanosecond pulsars

  1. Sep 8, 2003 #1

    wolram

    User Avatar
    Gold Member

    why cant we use nanosecond pulsars to gauge expantion of
    universe? even if they are irregular ,if the "tendancy "of
    pulses measured in a second is always downward then at
    least it would be a partial proof.
     
  2. jcsd
  3. Sep 8, 2003 #2

    LURCH

    User Avatar
    Science Advisor

    One possible problem with this might be that pulsars are already known to slow down over time. Since this deceleration is considered to be a result of the pulsar losing energy and spinning more slowly, any measured increase in the rate could be the result of expansion, or it could just be the pulsar slowing down more. I can't think of any way to tell the difference.
     
  4. Sep 27, 2003 #3

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    wolfram:
    (my emphasis)
    Do you have a link describing observations of such bizarre denizens of the cosmic zoo?
     
  5. Sep 27, 2003 #4

    wolram

    User Avatar
    Gold Member

    sorry nereid, i do not have a link its something i read
    these nanosecond pulses are comming from a small area
    of a microsecond pulsar, a matter of a few squar METRs
    if i remember correctly the smallesr thing so far detected
    in space,
    i hope one of the more informed members can jump in
    and save me.
     
  6. Sep 27, 2003 #5

    wolram

    User Avatar
    Gold Member

    In 2003 observations of the Crab nebula pulsar's signal revealed "sub-pulses" within the main signal with durations of only nanoseconds. It is thought that these nanosecond pulses are emitted by regions on the pulsar's surface 60cm in diameter or smaller, making them the smallest structures outside the solar system to be measured.

    --------------------------------------------------------------------
    found this in wikipida something similar to the article
    i read. only this refers to CMs.
     
  7. Sep 27, 2003 #6

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Thanks wolfram.

    So they are nanosecond pulses rather than pulsars.

    To turn to your original question:
    I'm not clear on what you had in mind; something special about the pulsar frequency? or simply that the frequency can be measured with extraordinary accuracy?

    One difficulty to overcome would be what LURCH pointed to: how to disentangle any 'expansion of the universe' signal in the data from other causes of any slowing that might be observed.

    Have you done a calculation of how big the 'expansion of the universe' effect might be?
     
  8. Sep 28, 2003 #7

    wolram

    User Avatar
    Gold Member

    Have you done a calculation of how big the 'expansion of the universe' effect might be?
    ----------------------------------------------------------------------
    well im sure not qualified in that area, maybe someone has
    some figures.
    i have 80 cubic kilometers per sec per mega parsec, for
    expantion.
    cant even remember where i got these from, i may have look
    on google later.
    the origonal post wasnt well thought out, i just posted it
    as a "maybe if".
     
    Last edited: Sep 28, 2003
  9. Sep 28, 2003 #8

    wolram

    User Avatar
    Gold Member

    http://antwrp.gsfc.nasa.gov/apod/ap960513.html

    which Freedman and team have used to identify some 50 Cepheids. Their distance and velocity measurements determine Hubble's constant to be about 80 kilometers per second per megaparsec.
    ----------------------------------------------------------------------
    i don t know how upto date this is, but the accuracy of a
    nanosecond pulse would give an indication, if only an
    average over time?
     
    Last edited: Sep 28, 2003
  10. Sep 28, 2003 #9

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Assume that there are pulsars which emit nanosecond (sub-)pulses, and that these
    are very regular and stable.

    How could one use observations of these to "gauge the expansion of the universe"?

    BTW, here's a link to the Freedman "Final Results" HST paper on the Hubble constant:
    http://arxiv.org/abs/astro-ph/0012376
     
  11. Sep 29, 2003 #10

    wolram

    User Avatar
    Gold Member

    it would depend on how regular the pulsar is, if you
    recorded say 60 seconds of the pulses also record
    the time t1 of first pulse, then wait a week, month
    and do same again, if time of first pulse changes
    ie t1+X seconds or the comparison of first and
    second recording is out of sync then you know the
    source is mooving, after several recordings you
    would build a better picture.


    thankyou for link.
     
  12. Sep 29, 2003 #11

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    If I can summarise:

    We make two measurements of the frequency of the pulses, one 'now' and the other a week (month, year, ...) from 'now'.

    Although we don't know what the pulsar's recession speed is, at either time, we can see if the frequency of the pulses has changed.

    If the pulse frequency is lower at the second measurement, this might be due, in part, to the expansion of the universe; the further away the pulsar is, the higher its recession speed.

    If the pulsar is (observed to be) 1 Mpc from Earth 'now' (so its recession speed is (observed to be) 80 km/s), what will its recession speed be 1 million seconds from 'now' (assume 1 Mpc = 3 x 1019 km)?
    {Assume, for the moment, that the pulsar is not accelerating (in its local environment), and that we can factor out all the changes in the Earth's speed over the 1 million seconds.}
     
  13. Oct 3, 2003 #12

    marcus

    User Avatar
    Science Advisor
    Gold Member
    2015 Award
    Dearly Missed

    Wow! this is the first time I wandered in here (since the format changed and categories like Cel Mech were established) and already I see a problem! This is a sign that Nereid is a nice person. Among astronomers giving each other problems is like grooming among monkeys or giving backrubs, it is a friendly gesture

    She says wait a million seconds so the thing is 80 million km farther and then how much faster is it going?
    She is an astronomer so she makes clear that the distance in the Hubble formula is the distance measured 'now' (so many people do not realize this!)

    The fractional increase in distance is 8E7 km/3E19 km so it is around 2.7E-12. This is a negligible fractional increase so the increase in speed is also negligible, it was 80 km/s and it after you have waited E6 seconds it still is 80 km/s

    How did this come up. Fraid I havent read the thread, just saw Nereids name and scrolled down
     
  14. Oct 5, 2003 #13

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Thank you for the kind words marcus :smile:

    wolfram posted a very good question, about the possibility of using pulsars as (another) means to measure the expansion of the universe. Various people have contributed ideas to how feasible this might be. As you point out, the difference in apparent recession speed would be very small.

    Does anyone know how accurately we could measure the signals from a pulsar? I mean, could we tell if a pulsar's period changed by 1 part in 109? or 1 part in 1015? Assume that the signal is sufficiently strong (or that we can integrate over enough signals) so signal detection is not an issue. Assume also that all local contributions to any apparent change in period can be nulled out.
     
  15. Oct 5, 2003 #14

    marcus

    User Avatar
    Science Advisor
    Gold Member
    2015 Award
    Dearly Missed

    even the simplifying assumptions of this question are fun to consider, like the earth is moving all over the place at one tenthousandth of the speed of light so it is changing the timing of the pulses by one part in 104 all the time by its own
    velocity changes and the rotation of the earth by one part per million by its velocity. So nulling out the local contributions of all the local motion would already be fun.

    this is a question that Phobos or Labguy or chroot maybe or selfAdjoint----cant think of which other people---would love to reply to but might not see because of the FORMAT having so many pigeonholes.

    there is some historical observation of change of timing of pulsar showing radiation away of energy in form of gravitywaves. it is historical because of the accuracy achieved in measuring the change of timing (whenever the accuracy of measurement in astronomy changes this generates historical advances in some area or other) so by looking at that celebrated measurment you could tell what the leading edge accuracy for pulsar timing at the present era----one of those people would remember the numbers

    I dont want to hog the witnessbox so I will get out and let someone else reply. Jeez why isnt everybody all over this question?
     
  16. Oct 7, 2003 #15

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    accuracy of measuring time

    IIRC, the NIST time standard claims an accuracy of 10-15. Within a decade this may be improved by up to three orders of magnitude. However, this degree of accuracy is, for all practical purposes, useless. Why?

    Assuming a pulsar's distance from the Earth was determined solely by the expansion of the universe (and all local - to the Earth - motions could be accurately determined), at what distance (and over what time period) would a pulsar be for observations of a change in its period to be able to show the Hubble expansion? Assume the arrival time of the pulsar's pulses could be recorded accurately to 1 part in 1015.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Nanosecond pulsars
  1. Binary pulsars (Replies: 1)

  2. Millisecond pulsar (Replies: 9)

  3. Pulsar's spectrum (Replies: 2)

Loading...