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

Experiments in redshift

  1. Feb 7, 2013 #1
    Anyone interested in going over some crude experiments and having a look to see if anything stands out as incorrect or impracticle?

    I'm not interested in any discussion as to the validity of doing the experiments, kindly keep that nonsense to yourself, I'm only interested in any input to correct, improve on them, or extend them for non visible or polarized light.

  2. jcsd
  3. Feb 7, 2013 #2


    User Avatar

    Staff: Mentor

    Have you done any work in spectroscopy before?
  4. Feb 21, 2013 #3
    Only as an amature, which is why I'm asking for help, that and not being able to find any results for similar experiments.
    I have come across folks that simply dismiss these as invalid because they know for sure what the results would be, yet have never supplied any observational evidence for their claims. I know what the results should be, there's really no harm in making sure.

    I also find alot of folks I talk to don't really understand optics that well and when presented with my diagrams choose the basic understanding they were taught.
    So with that in mind maybe you could view these diagrams :- http://homepage.ntlworld.com/papermodels/reflectors/ , and answer a simple question: Red or Blue?
  5. Feb 22, 2013 #4


    User Avatar
    Gold Member

    I went to that website and looked at everything there. I am unable to connect any image with redshift spectroscopy. You ask "and answer a simple question: Red or Blue?" Is this some kind of joke?

  6. Feb 22, 2013 #5


    User Avatar

    Staff: Mentor

    I don't understand the point of the pictures. They just show parallel light at focus and non-parallel light out of focus. They have nothing to do with spectroscopy.
  7. Feb 24, 2013 #6
    The point of the pictures is an attempt to attain you level of understanding of optics.
    The images in blue are an exageration of the caustic curve produced by effectively (Non)parallel light. The red images are you standard ideal optics taught at a basic level.

    Caustics have everything to do with spectroscopy, maybe you just didn't understand the idea behind the experiments? I thought it was plain to see.
    Ok, consider this, we send a divergent light source through a prism and obtain a spectrum one meter away, now we increase the divergence by 0.001 arcsecond and obtain a second spectrum, what happens to the second spectrum, does it stay the same as the first, or does it broaden and by how much?
    My interest lies not with spectrum broadening within the apperatus, as this has surely been taken into account with telescopic design, my interest lies in what happens to the spectrum when a distant object with irregular shape produces a more complex caustic.

    No-one, as yet, can provide me with an answer, hence the attempt at designing basic experiments to provide an answer.
  8. Feb 24, 2013 #7
    I havent done any spectroscopy so can't input much. I like the premise of your experiment. After your explaination I can relate to what your asking. Unfortunately your knowledge on spectrography far exceeds mine lol
    You have probably thought of this but multiple shapes can be mapped in a lab. Develop how complex objects affects on your caustic regions in the lab then pick close stellar objects see how they compare. Then work at increasingly further distances. Thats about the extent I can suggest lol
    Last edited: Feb 24, 2013
  9. Feb 24, 2013 #8
    One thing that would be interesting is how the caustic regions are affected by cosmic dust on route?
    Another consideration would be gravitational lensing effects.
    Last edited: Feb 24, 2013
  10. Feb 27, 2013 #9


    User Avatar
    Gold Member

    I've looked at all the diagrams you referred to. Will you please explain your request to "answer a simple question: Red or Blue?" I do not get how to answer it because I cannot figure out what you are referring to. Thank you.

    Cheers, Bobbywhy
  11. Feb 27, 2013 #10


    User Avatar

    Staff: Mentor

    Do they? Why have my books on spectroscopy never mentioned them?

    I believe the focal plane is moved backwards, so you would have to move your sensor backwards to attain proper focus. Nothing happens to the spectrum. Unless I'm misunderstanding what you mean by a "divergent light source".

    I have never seen these caustics in anything I've seen or read on spectroscopy, so I couldn't tell you.
  12. Feb 28, 2013 #11


    User Avatar
    Gold Member

    codex, this is from your opening post on 7 February:

    “I'm not interested in any discussion as to the validity of doing the experiments, kindly keep that nonsense to yourself, I'm only interested in any input to correct, improve on them, or extend them for non visible or polarized light.”

    It is rather unusual for a scientist to say this because if another experienced scientist studied your experiments and wanted to offer some constructive suggestions, you have already decided in advance that his comments are “nonsense”. This indicates a lack of openness to evaluation by other experts. We who work in science do not commonly encounter this approach.

    It is especially unusual since you say in post #3 on 22 February: “Only as an amature (sic), which is why I'm asking for help, that and not being able to find any results for similar experiments.” By claiming to be an amateur one would expect a more open-minded request for help. You further state “I know what the results (of these experiments) should be, there's really no harm in making sure.” So, will you please explain how you know what the results should be, what those results are, and what observational evidence you have accumulated? Will you supply some peer-reviewed journal reference article or some textbook that describes the results of these experiments?

    The website you referred readers here to on 22 February did not contain the file: redshiftexperiments...25-Feb2013 04:16 2.4K
    It was added to the site on 25 February.

    I have read the document “Redshift Experiments” carefully and it has become clear why you label yourself as “an amature" (sic). Clearly you are. Clarity and simplicity in writing is essential for accurate and effective communication. Is English your first language? Notwithstanding all the misspelled words there appears to be some fundamental misunderstandings in your ideas about redshift measurements of distant objects. If you had simply posted this document in your opening post it would have simplified the process of communicating your ideas. In the document I found eleven declarative statements and nine specific questions. I’ve taken the liberty to separate them into two lists and to correct most of the spelling errors. This effort is intended to assist us to grasp your proposed experiments and to attempt to formulate useful answers to your questions. If I’ve made any mistakes in transcribing the document, please inform me soon. I’ve commented on a few statements.


    S1. Optically we cannot resolve the light from the edges of distant objects so we use the term effectively parallel.
    S2. Light from distant objects is NOT perfectly parallel.
    S3. Effectively parallel is not true parallel, an angular component exists in ALL incident light.
    S4. The prism, or grating, is based on and set up for parallel light, but light isn't parallel.
    S5. Optically we need to know what happens to the spread of the spectrum of light for parallel, convergent, and divergent light.
    S6. When light is not parallel, which is always the case, we do not obtain a true focal point, we obtain a caustic curve either in front or behind the 'ideal' focal point.
    S7. Since the angular component is based on angular diameter, it is logical to assume the caustic curve is directly proportional to the angular diameter.
    S8. It isn't a secret that redshift values are directly proportional to magnitude, and that magnitude is proportional to angular diameter.
    S9. The angular diameter V's redshift comparisons support a need to investigate the interaction of angular component and the instruments used for observations.
    S10. An investigation is needed to calculate the degree of spread in redshift in the instrumentation compared to the angular component of incident light.
    S11. Until such an investigation is carried out and the findings made public, ALL redshift based theory is merely speculation.

    Q1. So what happens when we use a prism, or diffraction grating, to split light?
    Q2. What degree of angular component is needed, in convergent or divergent light, to affect the result spread of the spectrum?
    Q3. What would happen if we set up our instruments for a nearby star and then compare that spectrum to a spectrum of an object with a different angular component?
    Q4. Do we expect the angular component to somehow magically not alter the spread or position of the spectrum produced?
    Q5. What happens to the spread of the spectrum if the light becomes more, or less, divergent, or convergent?
    Q6. Wouldn't even the smallest angular difference move the spread of the spectrum?
    Q7. If the difference in caustic curvature moves the spread of the spectrum, then isn't it also logical that some portion of the shift in the spectrum comparisons are directly proportional to the angular diameter?
    Q8. What about object shape?
    Q9. How does shape change the observed change in the spectrum?
    my comments on a few statements:

    S1. Only about ten nearby stars can be resolved as more than point sources. Stars more distant do appear as point sources.
    S4. In modern spectrometers the light rays impinging on the prism or grating are parallel because they have passed through the collimator element.
    S8. Measured redshift values are not proportional to magnitude of the object as you claim. Redshifts are proportional to the relative difference between the observed and emitted wavelengths of an object. And magnitude of the distant object is not proportional to angular diameter as you claim. The apparent magnitude of a celestial body is a measure of its brightness as seen by an observer on Earth, adjusted to the value it would have in the absence of the atmosphere. There is no connection between angular diameter and magnitude.
    S11. You claim that your “investigations” should be done and the findings made public. You also claim that “until then, ALL redshift based (sic) theory is merely speculation.” This is a massive, sweeping claim that affects thousands of astrophysicists and astronomers world-wide. Is this a “personal theory”? Is there any supporting literature, any studies, any empirical evidence you can offer to support your claims?

    Cheers, Bobbywhy
    Last edited: Feb 28, 2013
  13. Feb 28, 2013 #12


    User Avatar

    Staff: Mentor

    Quoting from Redshiftexperiments from the link:


    This is incorrect. The prism and grating do not care whether the light is parallel or not. The incoming light is typically collimated prior to the prism/grating anyways.

    The light is parallel, so no we don't.

    That is exactly what we expect. Because it doesn't happen.

    No. The spectral lines may become a little wider or blurrier if your light isn't parallel when it enters the grating/prism, but there is no shift in the lines.

    It doesn't. The spectrum remains exactly the same.
    If by this you mean that redshifted stars/galaxies are dimmer, then overall that's true because they are further away and get redshifted by expansion. But there is no direct relation between redshift and magnitude.

    This is nonsense. Spectroscopy is used for a wide range of subjects, including a great many here on Earth. If the spectrum were affected by the angular component of the incoming light we would notice it immediately.
  14. Mar 5, 2013 #13
    So what is this? or have I mis-interpreted it?
    http://ssg.astro.washington.edu/elsst/images/miRedshift.png [Broken]

    And yet they take every precaution to eliminate light with any angle of incidence as it seriously effects the resulting spectrum. What exactly do you think diffusers are for?
    Try placing 2 bulbs at different distances to a prism, the spectrum of one is broader than another? Why are they different if the prism does not care?

    But would you? Think about it for more than the time it takes you to reply, how the hell would you be able to, try to devise a single experiment that would allow you to notice it.

    Comments on your comments.
    S1 - exactly, we are ignoring angular componant within light because we can't optically resolve it, is it there?

    S4 - collimators are desgined how? Are they part of the X²/4 curve? Is that curve based on parallel light? So if light does not pass through a SINGLE focal point how does it magically become parallel if it wasn't parallel when entering the mirror? Collimators are mostly solid and mathematically based on parallel light ideal models.

    S8 - power per unit area? No Angular Diameter to Redshift Relationships and magnitude to redshift relationships? Google them.

    S11 - My personal opinion is irrelevant, such experiments have never been attempted so you can only spectulate at what the outcome of such experiments would be, we both know what they should be, according to current theory, but should current theory object to such experiments being performed?
    It is not a personal theory, it is curiosity arisen from reading a paper on measuring redshift velocity fields in which an edge on galaxy produced a velocity field when the slit was along the major axis of the galaxy but no velocity field could be produced when the slit was 90° to the major axis - why? You really need to think about that. Why does the light behave differently? It shouldn't matter what orientation the slit has, so long as light enters the slit we can obtain a redshift and therefore a velocity field. If you go back to the experiments, this is the first one suggested. Is the slit orientation to circluar objects and face on galaxies arbitary?
    Is there some angular componant within light, interacting with the curve used for telescopic reflectors, something optic fibres and diffusers can't get rid of?

    The obvious suspect is the non parallel componant which you agree exists but fail to see that collimators can't put right, unless your using a different collimator for every object....

    And that's where you run into real trouble, real big trouble, if an objects shape affects redshift observations and if it is due to an interection between the incoming non parallel light and the x²/4 curve then all redshift observations themselves need looking at, big ifs yes but I repeat myself, try to devise a single experiment that would allow you to notice it. Can we even measure such small distances and angles with our current technology?

    You can even go back to hubbles first observations and question them.
    If a large angular diameter object has a large angular componant which results is a broadened spectrum, and we compare that to objects with smaller angular componants which result in a reduced broadening of the spectrum then we can only conclude that objects with smaller angular componants have a greater redshift, agreed?
    And if we completely ignore the angular componant, thinking collimators somehow magically produce parallel light from non parallel light, we can only conclude what?

    All based on a small curiosity as to why slit orientation has ANY effect on velocity fields.
    (if you have a satisfactory answer let me know)

    Test it, it can either be something or nothing, if something then we learn something new, if nothing then we learn something more about something we thought we knew.

    Even though I am crap at explaining myself and my spell checker is offf, I am talking to a chap from chile via email who thinks this is of some interest, so hopefully he might be able to perform the first experiment at least, if he gets the time, maybe.

    Here is a good question for you, if redshift is velocity related then the BBT fails at around z=1000, if redshift has angular relationships, what z value would be the maximum?
    Last edited by a moderator: May 6, 2017
  15. Mar 5, 2013 #14


    User Avatar

    Staff: Mentor

    You have misunderstood it. Redshift in itself doesn't cause the reduction in magnitude, it is the increasing distance that does so. But since redshift is also related to distance thanks to expansion you have both reduced magnitude and increased redshift as objects get further away. If objects at the same distance where receding at different velocities their redshift and magnitude would not match up like this graph shows.

    Yes, as it broadens the spectral lines. That's it. It has no effect on redshift measurements.

    If you aren't using a slit then of course it will be. But this has nothing to do with the prism. It has to do with the angular size of the light bulbs. The bigger looking bulb produces a broader spectrum unless you use a slit.

    Easy. Measure a near point source and then an extended source in a lab.

    Obviously no collimator is perfect, but the light will be as close to perfectly parallel as it's going to get. I don't see the issue here. It is trivial to build a collimator to work with an optical system to produce parallel light. How did it become parallel? The mirror or lens made it that way. What's the issue? Even a bad collimator will merely cause blurred spectral lines.

    When the slit takes in light from a large slice oriented along the major axis of the edge on galaxy, it captures light from many different stars, all moving at different speeds and directions relative to ourselves. When it is oriented the other way, along the minor axis, and the slice is only over a small section all the stars visible in that small slice are moving with approximately the same velocity relative to us.

    Face on galaxies do not have stars orbiting in a plane parallel to our line of sight, so there are no major differences in their velocity away from or towards us.

    There is no obvious suspect here other than your own understanding of optics. A spectrometer simply doesn't work the way you imagine it to.

    No. A broadened spectrum has absolutely nothing to do with redshift.

    Magically? No. They use plain old reflection and refraction to produce parallel light.
    Last edited by a moderator: May 6, 2017
  16. Mar 7, 2013 #15


    User Avatar
    Gold Member

    codex34, Since you’ve said in post number 13 “It is not a personal theory, it is curiosity arisen from reading a paper on measuring redshift velocity fields...” I ask you: Why have you hidden that from everyone here? Why have you not posted the reference to that paper which seems to be driving this entire thesis of yours?

    You have also written “And if we completely ignore the angular componant, thinking collimators somehow magically produce parallel light from non parallel light, we can only conclude what?”

    There is no magic involved in the operation of an optical collimator’s function in a spectrometer. I’ve found this which seems to describe the exact nature of your thesis. See the website below, especially “Application 2: Collimating Light from a Point Source” and “Application 4: Focusing an Extended Source to a Small Spot”

    Even considering that a slightly imperfect formation of parallel light rays may be sent to the wavelength division element, this would NOT result in an erroneous redshift measurement. It would only cause a slight “broadening” of the spectral lines.

    p.s. Since you say your spell checker is off, here’s a list of misspelled words from your post #13 that I’ve spelled correctly just to help you get them right the next time:


    Cheers, Bobbywhy
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook