Still lower noise radio astronomy (was: low-noise amplifiers for radio astronomy )

John (Liberty) Bell wrote:
wrote:
John (Liberty) Bell wrote:
wrote:



True, but if we treat the amplifier as a black box, inside of which is
a transmission line impedance matching resistor at ambient temperature
T, the thermal noise formula would lead us to believe that the noise T
of the black box cannot be less than ambient T, contrary to
observation.

It might lead you to believe that, it might lead
me to think that the internal resistor would
shunt the noise from the external resistor
thus giving a negative noise figure, and the
noise temperature is the noise _added_ by
the amp, hence would be negative!

I don't think there is any point in continuing
to argue about this John, all we need to know
is that a perfect amp has a noise temperature
of 0K and a noise figure of 0dB.

snip

OK We are in agreement here too. ...

snip

Yes, probably. ...

snip

What I read was that the electrons involved in conduction
have a substantially lower temperature then the atoms.

OK, that sounds fine by me too. However, that would also mean that the
thermal noise formulae for conductors are wrong. ( ie at
http://en.wikipedia.org/wiki/Thermal_noise ).

I find it very difficult to believe, for example, that the noise energy
injected into the input of an amplifier using an ambient T resistor, is
the same as the energy supplied when the input is shorted to earth.

The power in the _wire_ is the same but the voltage
would be lower while the current is higher. That would
mean the power injected is lower unless you compare
cases where the input impedance is matched.

(I intend testing that too, during the next breadboarding.)

In that case be sure to measure the noise current.

snip

Replace "atoms" by "electrons" and you have what I
said.

Excellent. Then we are in agreement again.

The coupling of the electrons to the dish is as
good (assuming the impedances are matched) while
that to the atoms is poor hence the temperature of
the electrons is closer to that of space plus
atmospheric noise etc.) than that of the the atoms.

Now, whether that behaviour also pumps thermal energy out of a final
impedance matching resistor at such frequencies, is an interesting
question over which I will let you contemplate the answer.

I believe it would but the question then is the thermal
conductivity between the fixed atoms and the electrons
in the resistor responsible for conduction. If it is very
good then the electrons will remain at nearly the same
temperature a the body of the resistor and the thermal
power will be radiated through the dish.

Right. I would agree again. I don't know how good or how bad that atom
to electron thermal conductivity is in off-the-shelf resistors, or
whether it varies from type to type. I was anticipating using metal
film as opposed to metal oxide, because they have closer tolerances
(and smaller size). I also seem to remember that they are supposed to
be less noisy, which may, in part, relate to this thermal coupling
question. Any ideas?

Certainly metal film is a better choice but note that
higher values are obtained by cutting a spiral groove
in the film with a laser hence forming a coil. The
inductance my be a problem depending on your
test frequencies.

George

John (Liberty) Bell wrote:
wrote:

http://www.arxiv.org/abs/astro-ph/0503116
http://www.arxiv.org/abs/astro-ph/0510697

These abstracts do not necessarily confirm what you think they do. (As
your third reference has already been deleted by Yahoo, I cannot
actually comment on that here.)

In the later abstract, Bouwens and Illingworth state: "we have been
able to demonstrate that the bright end of the LF (0.3L*) is at least
5 times lower at z~10 than at z~4".

From this I infer that what they are actually saying is that when
viewed at z = 10, there is at most 20% of the luminous material that is
visible at z = 4, after taking account of Doppler shifts in energy.

I believe they take acount of lots more than that,
they have a special 'cloning' program written for the
purpose. In particular they would need to take
account of the Doppler shift moving energy between
filter bands, the reduction in flux due to inverse square
loss and the headlight effect. Perhaps other optical
factors come in too including things like Lyman alpha
absorbtion or dust reddening. I have no idea of the
details. Gravitational lensing matters too at these
high z values.

They and you appear to be concluding that this is strong evidence for
substantial galaxy evolution over that distance range. However, this is
not necessarily the case.

Using the UCLA calculator you recommended,
at z=4, the distance to the big bang is 1.571 Glyr.
This squared is 2.468 SqGlyr.
at z=10, the distance to the big bang is 0.482 Glyr.
This squared is 0.232 SqGlyr.

That's the distance from the bang, not from us.
That increases from 13.7-1.571 = 12.129 (which
squared gives 147) to 13.7- 0.482=13.218, squared
gives 175) hence the area increases by 28 or 19%.

Now we know that, if we trace back light paths descending vertically
towards the Earth's south pole, they will all eventually meet the
traced-back light paths descending vertically towards the Earth's north
pole, at the point we all affectionately call the big bang.This is
caused by the gravitational curving of light by the universe, or, what
amounts to the same thing, by the curvatue of the spacetime continuum.

Yep, but the number of lines 30 longitude apart
remains 12 as far as you go. Galaxies at earlier
times had to be closer together.

Now, what would that mean, to take an extreme case, for a steady state
universe, such as Fred Hoyle's? It would mean that, if we follow the
light cone emanating from the big bang,

Hold on a moment, there was no bang in the
steady state model! If you are thinking of
quasi-steady then the conventional analysis
applies.

this will encompass
progressively more of the universe within our field of vision, as we
proceed to lower z. For very high z, the sides of that light cone would
have negligible curvature, and the number of galaxies potentially
visible at a given distance d from the big bang, would then be
proportional to d squared.

On the assumption that the curvature of light from the origin to 2.5
Glyr can be ignored to a first approximation, this model then gives
only 9.5% of the galaxies visible at z= 4, still visible at z = 10.
This is confirmed by Bouwens and Illingworth's conclusion that the
number at z = 10 is 20% of the number at z = 4.

Consequently, far from confirming (recently revised) galaxy evolution
models, their findings could equally well have confirmed that the
universe is, in fact, steady state.

I am not actually claiming here that the universe _is_ steady state, I
am merely pointing out that it is far too easy to think that
astronomical data confirms a theory (if the data is interpreted within
the context of that theory), when, in practice, it does no such thing,
objectively.

This isn't confirming anything. Recent theory
said galaxies would be rare at these times. The
observations suggest lots of large galaxies at
900Ma but very few at 700Ma hence a period
of rapid merger of smaller undetectable objects
in the intervening period. Next we need a theory
to explain that.

George

John (Liberty) Bell wrote:
wrote:

snip

"They found hundreds of galaxies at redshifts around
900 million years after the Big Bang.

But when they looked at higher redshifts, at about
700 million years after the Big Bang, they found
unconfirmed evidence for only one galaxy, when
they had expected to find many more.

Yes, that is particularly interesting. Nota bene: "they expected to
find many more."

This backs theories about a "hierarchical" formation
of big galaxies -- that these huge clusters were built
up over time as smaller galaxies collided and merged,
they believe.

Not necessarily. See posting of Thurs, Sep 21 2006 8:26 pm, and below.

See my reply to that.

snip

I think you may be missing something here. The EM spectrum is a
continuum.
Optical - infrared - microwave - lower RF, so, in principle there is
no problem going beyond, say, Z = 1000, if there was anything there
to see.

Agreed but there may be little to see by then. This
is the record holder at z = 6.96 and the chances are
the first stars formed at less than z=14.

http://arxiv.org/abs/astro-ph/0609393

Developing on that argument proposed in my response of Thurs, Sep 21
2006 8:26 pm, this final paper too is far from conclusive, despite the
statement "The number density of galaxies at z = 7 seems to be only
18-36 per cent of the density at z = 6.6."

The operative phrase here is "seems to be". In order to change that to
"is", we would need to have comparable data over a solid angle of 4 pi
steradians to be conclusive.

No, that's not the problem. At z~6 they found over
500 galaxies so the statistics are fairly robust. The
problem is that the limitations of the Hubble mean
the same density at z~7.5 wouldn't predict 500,
it only predicts less than 20. We are seeing only
the tail of the distribution because they are so much
fainter. What is needed is a more suited telescope.

Even then, the data could be less
conclusive than one might imagine it to be.

Assuming for approximate order of magnitude comparison purposes that
the maximum surface area of the universe visible to us is pi r squared,
where r is half the distance to the big bang, that gives a surface area
of ~ 600 sqGlyr. Contrast that with 0.6 sqGlyr at z = 7.

You are going the wrong way, from us, the cone that
the HUDF covers would get slightly larger and at the
same time the density is increasing so the number
should go up. The explanation why neither approach
gives the right answer is that light is curved significantly
at these distances.


Thus, in order for the conclusion to be rigorously applicable for
galaxy evolution, as opposed to steady state conclusions, we need to
assume that the spatial distribution of galaxies is isotropic at least
on a scale 3 orders of magnitude smaller than the whole observable
universe. However, we know that, in detail, the universe contains large
clumps and filaments of galaxies with large volumes of vacuum in
between.

IIRC it becomes homogenous and isotropic at scales
around 100MPc so about 2.5 orders of magnitude.
That's perfectly acceptable given the other problems.

snip

You won't see stars or galaxies redshifted to RF
frequencies.

Well, that presuposes:
a) that nothing I have said above could be true.
b) that stars and galaxies do not already emit at higher radio
frequencies.

If (b) is true, what are all those radio astronomers supposed to be
doing, with their time and our money?

Hah! Sorry, I should have said "You won't see
visible light from stars or galaxies redshifted ..".
You got me.

Getting from t=700 million years to 378000 years goes
from z=7.6 to z=1090 and the Tolman test says the
brightness goes as (1+z)^4 so requires an increase in
sensitivity of about 260 million times. That takes you
from 13 billion years ago to 13.69622 billion assuming
an age of exactly 13.7 billion. Since a photon's energy
is proportional to its frequency, redshift has a very
significant impact.

The CMBR peaks at around 160 GHz and stars at z~11
would be redshifted by a factor of 100 less,

OK, so you are saying the CMB has a z of ~1100

Yes, 1079 from WMAP.

and their
surfaces are perhaps another order of magnitude hotter.

OK so you seem to be saying, from Wien's displacement law (see
http://en.wikipedia.org/wiki/Blackbody_radiation ) that a "Fred Hoyle
steady state universe" star or galaxy at z = 1100 would have a black
body radiation peak at ~ 1.6 THz.

I'm saying that if we could see the CMBR from
a location where z~11 then it would have a peak
around 16THz. Hot blue stars in the early universe
would have a temperature of the order of 10^4 K
or say a factor of 10 hotter than the CMBR (2950K)
so as a back-of-envelope guess, stars at z~11
would peak around 160THz. You will have trouble
finding transistors to work at that and you might
need a better scope.

In that case, it looks like I am going to have to take this "Devil's
Advocate" argument the whole 9 yards now, by going to extremes.

I think you might have misunderstood my point,
I'm going to skip this bit:

Within the context of that argument, the CMB would now be interpreted
as a a black body radiation peak at ~ 1.6 THz, thus having a z shift of
~ 11,000. According to http://www.astro.ucla.edu/~wright/CosmoCalc.html
, that gives a distance from the big bang of 6,000 lyr (and would
give a corresponding surface radius).

snip steady state

[Mod. note: this is straying into the sort of speculation that is
forbidden by the charter of this newsgroup, particularly when
arguments start to be based on web sites that aren't accessible to the
general public rather than peer-reviewed journals and textbooks.
Please try either to stick to known physics in this discussion, or, if
you really need to claim that known physics is wrong, start a new
thread and explain the `proofs' you're relying on. -- mjh]

I think my quick figures misled John, hoopefully what I
was estimating is clearer now. Sorry for any confusion.

George

John (Liberty) Bell wrote:

.....

They and you appear to be concluding that this is strong evidence for
substantial galaxy evolution over that distance range. However, this is
not necessarily the case.

Using the UCLA calculator you recommended,
at z=4, the distance to the big bang is 1.571 Glyr.
This squared is 2.468 SqGlyr.
at z=10, the distance to the big bang is 0.482 Glyr.
This squared is 0.232 SqGlyr.

Now we know that, if we trace back light paths descending vertically
towards the Earth's south pole, they will all eventually meet the
traced-back light paths descending vertically towards the Earth's north
pole, at the point we all affectionately call the big bang.This is
caused by the gravitational curving of light by the universe, or, what
amounts to the same thing, by the curvatue of the spacetime continuum.

John, have a look at the spacetime diagram he

http://www.astro.ucla.edu/~wright/cosmo_02.htm#dh

The straight black lines are selected galaxies while
the red lines are light from the early universe reaching
us. You need to consider the number of galaxies
included within some set of lines bounding our past
light cone and I expect there predicted the number
based on the LCDM model (see a little farther down
that page).

George

The moderator commented at
http://groups.google.com/group/sci.a...546d1e273413ff
:

[Mod. note: this is straying into the sort of speculation that is
forbidden by the charter of this newsgroup, particularly when
arguments start to be based on web sites that aren't accessible to the
general public rather than peer-reviewed journals and textbooks.
Please try either to stick to known physics in this discussion, or, if
you really need to claim that known physics is wrong, start a new
thread and explain the `proofs' you're relying on. -- mjh]

Thank you for this comment. In that case, can I simply change that last
paragraph to read:

Again, I am not actually claiming here that the universe _is_ a Hoyle
steady state one (or anything like it). However, given that
observational astrophysics so frequently produces results that were not
predicted by established theory, what I am actually saying is that it
would be wise to exercise some caution in assuming that those results
must necessarily then be interpreted within that established theory
context.

Several examples of such unpredicted, or at least unexpected,
observational results (in the context of then established theory) occur
to me immediately.

The first is the accelerating expansion of the universe.
http://www.aip.org/enews/physnews/19...t/pnu361-1.htm (1998)
cf http://www.aip.org/pnu/1997/split/pnu345-2.htm (1997)

The second is the observation of galaxies far earlier in time than
originally predicted, and the associated observation that 'young
galaxies' still appeared to contain old stars, and subsequent
observations which pushed those limits back still further.

http://oposite.stsci.edu/pubinfo/bac...t/galaxpdx.txt (1994)
Quote: "Difficult to explain"

http://star-www.dur.ac.uk/cosmology/pressrelease.html (2000)
Quote: "Beyond the final frontier"

http://www.aip.org/enews/physnews/2004/split/668-1.html (2004)
Quote: "Surprisingly early"

Following theoretical adjustments to accommodate the above facts
(within established theory), the third example might now seem to be the
then 'unexpected' observed paucity of galaxies at z =7 (if George
Dishman's comments based on the now deleted Yahoo article are anything
to go by).

Theories stand or fall on their ability to make unique predictions that
are subsequently confirmed by laboratory and/or astronomical
observations. Discoveries such as the above would seem to confirm that
established theory is no longer doing that.

I have copied the above references (and quotations from those
references), with permission, from the carousel of verified predictions
at the previously mentioned site.
Incidentally, I understand that the only reason why that site has not
yet been re-opened to the general public, is because there is not yet
sufficient declared interest to justify the additional cost overheads
of doing so.

Regards

John Bell

wrote:
John (Liberty) Bell wrote:

Using the UCLA calculator you recommended,
at z=4, the distance to the big bang is 1.571 Glyr.
This squared is 2.468 SqGlyr.
at z=10, the distance to the big bang is 0.482 Glyr.
This squared is 0.232 SqGlyr.

That's the distance from the bang, not from us.

Of course. That is why I used these figures.

That increases from 13.7-1.571 = 12.129 (which
squared gives 147) to 13.7- 0.482=13.218, squared
gives 175) hence the area increases by 28 or 19%.

No. That is clearly wrong. By your argument, the big bang has a radius
of 13.7 Glyr.
In reality it has Planck dimensions using classical GR analysis.
Admittedly that radius is many orders of magnitude larger using the
newer field equation, but that 'many orders of magnitude larger' is
still only the size of a strong nuclear force carrying meson.

Now we know that, if we trace back light paths descending vertically
towards the Earth's south pole, they will all eventually meet the
traced-back light paths descending vertically towards the Earth's north
pole, at the point we all affectionately call the big bang.This is
caused by the gravitational curving of light by the universe, or, what
amounts to the same thing, by the curvatue of the spacetime continuum.

Yep, but the number of lines 30 longitude apart
remains 12 as far as you go. Galaxies at earlier
times had to be closer together.

This is true (except, of course, in a steady state model). The question
is, how much closer together. It is now generally accepted, in large
part due to Guth's inflationary model, that as you move further from
the big bang, the 'observable universe' becomes a larger proportion of
the totality generated out there.

Now, what would that mean, to take an extreme case, for a steady state
universe, such as Fred Hoyle's? It would mean that, if we follow the
light cone emanating from the big bang,

Hold on a moment, there was no bang in the
steady state model!

I earlier defined the big bang as the point at which traced back
divergent light rays converge again to a point. There is, therefore, no
contradiction involved, provided we accept the tenet that gravity bends
light.

this will encompass
progressively more of the universe within our field of vision, as we
proceed to lower z. For very high z, the sides of that light cone would
have negligible curvature, and the number of galaxies potentially
visible at a given distance d from the big bang, would then be
proportional to d squared.

On the assumption that the curvature of light from the origin to 2.5
Glyr can be ignored to a first approximation, this model then gives
only 9.5% of the galaxies visible at z= 4, still visible at z = 10.
This is confirmed by Bouwens and Illingworth's conclusion that the
number at z = 10 is 20% of the number at z = 4.

Consequently, far from confirming (recently revised) galaxy evolution
models, their findings could equally well have confirmed that the
universe is, in fact, steady state.

I am not actually claiming here that the universe _is_ steady state, I
am merely pointing out that it is far too easy to think that
astronomical data confirms a theory (if the data is interpreted within
the context of that theory), when, in practice, it does no such thing,
objectively.

This isn't confirming anything. Recent theory
said galaxies would be rare at these times. The
observations suggest lots of large galaxies at
900Ma but very few at 700Ma hence a period
of rapid merger of smaller undetectable objects
in the intervening period.

Next we need a theory to explain that.

That was my intended point really. Established theory failed to predict
observation. Since I can't really discuss the newer theory in detail
here, I chose to demonstrate instead that even a steady state model
would have done a better job of predicting observed population
densities.

Regards

John

wrote:
John, I've been falling behind with the posts
so some of this may be out of date.

That's ok. But since responding to all your points would now get me
behind in other matters, I am going to keep this as brief as possible.

snip

I believe we have arrived at a satisfactory concensus of understanding
on all the above points by now. Thanks for your help. (I think any
remaining apparent differences are due to linguistics and different
viewpoints as opposed to firm technical facts)

Dispense with the line so the amp is directly connected to
the horn and you should still match but the reflection time
will be smaller so problems moved to higher frequencies.

Yes, I was also wondering about this (still mysterious to me) aspect of
things in relation to impedance matching at the amp input. Clearly we
can neutralise the input capacitance with an inductor at the centre
design frequency, and broaden the bandwidth somewhat, but at
frequencies far from that bandwidth, any transmission line will be far
from matched. There should thus be horrendous effects of line
mismatching within the transmission line at those higher and lower
frequencies. Presumably, this does not concern us provided the range we
want is reasonably clean?

I guess also that this will preclude one from amplifying two different
passbands simultaneously from the same antenna? Consequently, it looks
to me like, whenever an astronomer wants to examine a different
passband of the VLA spectrum, somebody has to go out and physically
unplug 10,000 preamps from 10,000 antennae, and then physically plug in
10,000 differently tuned ones?

snip

This
consideration is academic here, however, because the analysis of that
paper was theoretical. The amplifiers were not actually built or tested
on a bench.

I think they were since they had images of the chips. I
understand the graphs to be actuals measured, they don't
look like simulator outputs.

No. You are remembering the preceding ref. given by JT.
http://www.skatelescope.org/document...8MTTSPaper.pdf has no
photos.
The title is "noise temperature estimates..." not "measurements".
Furthermore, the model allowed the transistor width to be varied for
optimum matching at different frequencies, which is clearly only really
feasible within a theoretical model.

However, this is not a particularly important point.

Regards

John Bell
(Change John to Liberty to bypass anti-spam email filter)

John (Liberty) Bell wrote:
wrote:

snip

I believe we have arrived at a satisfactory concensus of understanding
on all the above points by now. Thanks for your help. (I think any
remaining apparent differences are due to linguistics and different
viewpoints as opposed to firm technical facts)

OK, on the amplifier subject, I think any lingering
differences of opinion will be resolved when you
actually run some measurements in the lab.

Dispense with the line so the amp is directly connected to
the horn and you should still match but the reflection time
will be smaller so problems moved to higher frequencies.

Yes, I was also wondering about this (still mysterious to me) aspect of
things in relation to impedance matching at the amp input. Clearly we
can neutralise the input capacitance with an inductor at the centre
design frequency, and broaden the bandwidth somewhat, ...

I don't think they would be tuned in the normal
sense unless it was unavoidable. The passband
I would expect would look like this:

/\ /\____________/\ /\
/ \/ \/ \
| |
| |
|----------------------|
| |
| |
____ / \ ___
\/ \/

The passband would be measured at -3dB relative
to the centre, the ripple within the passband might
be perhaps +/- 1dB, the edges might be 12dB per
octave or hopefully better to cut out adjacent band
traffic and the attenuation some way from the band
might be of the order of 60dB.

Certainly that's the sort of response the gear we
produce must have although that's a different field.
To get that, the input and output ports usually look
purely resistive over the band.

... but at
frequencies far from that bandwidth, any transmission line will be far
from matched. There should thus be horrendous effects of line
mismatching within the transmission line at those higher and lower
frequencies. Presumably, this does not concern us provided the range we
want is reasonably clean?

To be exact we have not been talking about
transmission lines but mostly waveguides with
some mention of coax though the principles are
closely related. I think the best would be for you
to research those on the web yourself. The key
point is that they are not tuned but have cutoffs.
Coax will go to DC and has a practical upper
limit while a waveguide has a lower limit. Within
the useable range, there is no tuning effect in
a waveguide or coax as I'm sure you know.

I guess also that this will preclude one from amplifying two different
passbands simultaneously from the same antenna?

That depends on the width of the bands.
Remember in the DSN which we were talking
about the LNA has a bandwidth of 30MHz
which is I believe the entire width allocated
for their use at S-band, from which four
narrower operational bands are extracted. It
is perhaps of limited use in that they will seldom
have more than one craft in the beam but for
radio telescope work it's another matter.

Consequently, it looks
to me like, whenever an astronomer wants to examine a different
passband of the VLA spectrum, somebody has to go out and physically
unplug 10,000 preamps from 10,000 antennae, and then physically plug in
10,000 differently tuned ones?

No, I would expect they would put passive
diplexors in front to split the bands to the
amps but keep all connected.

snip

.. You are remembering the preceding ref. given by JT.
http://www.skatelescope.org/document...8MTTSPaper.pdf has no
photos.

OK I was talking about these

http://www.imec.be/esscirc/esscirc20...gs/data/76.pdf
http://amsacta.cib.unibo.it/archive/...1/GA043200.PDF

George

John (Liberty) Bell wrote:
wrote:
John (Liberty) Bell wrote:

Using the UCLA calculator you recommended,
at z=4, the distance to the big bang is 1.571 Glyr.
This squared is 2.468 SqGlyr.
at z=10, the distance to the big bang is 0.482 Glyr.
This squared is 0.232 SqGlyr.

That's the distance from the bang, not from us.

Of course. That is why I used these figures.

That increases from 13.7-1.571 = 12.129 (which
squared gives 147) to 13.7- 0.482=13.218, squared
gives 175) hence the area increases by 28 or 19%.

No. That is clearly wrong. By your argument, the big bang has a radius
of 13.7 Glyr.

Well in one sense that is correct. The bang itself
was everywhere so the concept of a radius in not
applicable but the "surface of last scatterring" which
produced the CMBR is a sphere whose radius is a
"lookback time" of 13.7 billion years.

In reality it has Planck dimensions using classical GR analysis.

However, all the matter in the galaxies we see later
spread over the whole sky was concentrated onto
that surface at that time.

Admittedly that radius is many orders of magnitude larger using the
newer field equation, but that 'many orders of magnitude larger' is
still only the size of a strong nuclear force carrying meson.

What "newer field equation" are you talking about?

Now we know that, if we trace back light paths descending vertically
towards the Earth's south pole, they will all eventually meet the
traced-back light paths descending vertically towards the Earth's north
pole, at the point we all affectionately call the big bang.This is
caused by the gravitational curving of light by the universe, or, what
amounts to the same thing, by the curvatue of the spacetime continuum.

Yep, but the number of lines 30 longitude apart
remains 12 as far as you go. Galaxies at earlier
times had to be closer together.

This is true (except, of course, in a steady state model). The question
is, how much closer together.

Think of the matter from the bang moving up lines of
longitude from the south pole as cosmic age increases.
We are looking back along light lines from somewhere
on the equator (not strictly accurate but enough to give
the idea) and those lines are always progressing exactly
south-west and south-east. We see everything between
that lies on some line of latitude. The farther back we
look, the more we see.

It is now generally accepted, in large
part due to Guth's inflationary model, that as you move further from
the big bang, the 'observable universe' becomes a larger proportion of
the totality generated out there.

No, inflation in the conventional model is thought to have
occurred between 10^-34 and 10^-32s after the bang.

Now, what would that mean, to take an extreme case, for a steady state
universe, such as Fred Hoyle's? It would mean that, if we follow the
light cone emanating from the big bang,

Hold on a moment, there was no bang in the
steady state model!

I earlier defined the big bang as the point at which traced back
divergent light rays converge again to a point. There is, therefore, no
contradiction involved, provided we accept the tenet that gravity bends
light.

You defined it as follows:

Now we know that, if we trace back light paths descending vertically
towards the Earth's south pole, they will all eventually meet the
traced-back light paths descending vertically towards the Earth's north
pole, at the point we all affectionately call the big bang.This is
caused by the gravitational curving of light by the universe, or, what
amounts to the same thing, by the curvatue of the spacetime continuum.

In a steady state, the lines don't converge and
a suitable similar analogy might but radial lines
on a disc. They will be randomly bent passing
individual galaxies but there is no way they can
all converge.

snip

This isn't confirming anything. Recent theory
said galaxies would be rare at these times. The
observations suggest lots of large galaxies at
900Ma but very few at 700Ma hence a period
of rapid merger of smaller undetectable objects
in the intervening period.

Next we need a theory to explain that.

That was my intended point really. Established theory failed to predict
observation.

That's not really true, we have a number of models
of galaxy formation but none are what I would call
'established', they are all tentative. The roles of dark
matter and supermassive black holes for example
are poorly understood. They are really work in
progress and the current observations suggest that
the mergers expected in one of those models might
be happening between 700 to 900 Ma rather than a
little later. In general what is seen is qualitatively in
line with expectation.

Since I can't really discuss the newer theory in detail
here, I chose to demonstrate instead that even a steady state model
would have done a better job of predicting observed population
densities.

Again I'm not sure what "newer theory" you mean
but your steady state analysis is not correct. What
it would predict is not easy to work out since you
need a model for cosmological redshift. Using tired
light for example would give one value while dust
reddening might give another.

Anyway, the bottom line here is that we are on the
verge of getting the first hard data on early galactic
mergers and morphology in that era with which the
theories can be refined, this is far from a finished
story. Watch this space ....

George

wrote:
John (Liberty) Bell wrote:
wrote:
John (Liberty) Bell wrote:

Using the UCLA calculator you recommended,
at z=4, the distance to the big bang is 1.571 Glyr.
This squared is 2.468 SqGlyr.
at z=10, the distance to the big bang is 0.482 Glyr.
This squared is 0.232 SqGlyr.

That's the distance from the bang, not from us.

Of course. That is why I used these figures.

That increases from 13.7-1.571 = 12.129 (which
squared gives 147) to 13.7- 0.482=13.218, squared
gives 175) hence the area increases by 28 or 19%.

No. That is clearly wrong. By your argument, the big bang has a radius
of 13.7 Glyr.

Well in one sense that is correct. The bang itself
was everywhere

But that 'everywhere' had a diameter of 10^-20 cms

so the concept of a radius in not
applicable but the "surface of last scatterring" which
produced the CMBR is a sphere whose radius is a
"lookback time" of 13.7 billion years.

That sphere, nevertheless, has a radius of 6,000 lyr if all its
energy travelled at the speed of light from the Big Bang (and the
default UCLA calculator model is reasonably accurate, at 'flat'
setting).

In reality it has Planck dimensions using classical GR analysis.

However, all the matter in the galaxies we see later
spread over the whole sky was concentrated onto
that surface at that time.

That may well the case for the _energy_ of that matter, in classical
big bang theory. That does not necessarily mean that the assertion is
automatically true.

Admittedly that radius is many orders of magnitude larger using the
newer field equation, but that 'many orders of magnitude larger' is
still only the size of a strong nuclear force carrying meson.

What "newer field equation" are you talking about?

The one described at http://www.1stlight.org, and mirrored at
http://global.accelerators.co.uk. (NYA without confidentiality
agreement and, therefore, not yet appropriate for detailed discussion
here)

Now we know that, if we trace back light paths descending vertically
towards the Earth's south pole, they will all eventually meet the
traced-back light paths descending vertically towards the Earth's north
pole, at the point we all affectionately call the big bang.This is
caused by the gravitational curving of light by the universe, or, what
amounts to the same thing, by the curvatue of the spacetime continuum.

Yep, but the number of lines 30 longitude apart
remains 12 as far as you go. Galaxies at earlier
times had to be closer together.

This is true (except, of course, in a steady state model). The question
is, how much closer together.

Think of the matter from the bang moving up lines of
longitude from the south pole as cosmic age increases.
We are looking back along light lines from somewhere
on the equator (not strictly accurate but enough to give
the idea) and those lines are always progressing exactly
south-west and south-east. We see everything between
that lies on some line of latitude. The farther back we
look, the more we see.

Observational evidence would seem to indicate that is only true up to a
certain point.
In theory, that is also only true up to a certain point, at least in
all inflationary models.

It is now generally accepted, in large
part due to Guth's inflationary model, that as you move further from
the big bang, the 'observable universe' becomes a larger proportion of
the totality generated out there.

No, inflation in the conventional model is thought to have
occurred between 10^-34 and 10^-32s after the bang.

But the consequences of that inflation can only gradually come into
view later, because of the limiting velocity of light!

snip

In a steady state, the lines don't converge and
a suitable similar analogy might but radial lines
on a disc. They will be randomly bent passing
individual galaxies but there is no way they can
all converge.

That would only be true in an infinite universe. In a finite universe,
gravity would always bend light towards the centre of gravity of
matter.

However, you are still taking this 'steady state' thing far too
literally.
Please see my response to Steve Willner.


John