[SC.LUG] RE: [SC.LUG]Radio Telescopes = was ISPs

[Martyn Kinder]

see below

----------------------------------------
Martyn Kinder G0CZD

mailto:martyn@czd.org.uk

http://www.czd.org.uk

<:
<: How can that be?

z= sqrt(C sqd x L sqd)

ie Frequency independent - these are passive components. Active devices vary
their L & C components depending on the frequency

<:
<: > Nope - in a perfect mixer you get outputs of f0 + fw and f0 - fw.
<:
<: I'll take your word for it.
<:
<: > See explanation above. Thats the way it is. If you dont believe me,
<: > run the bath cause a fast oscillating wave from one end and a slow
<: > oscillating wave at the other and look at the formation of
<: the waves
<: > when they meet... its an eye opener.
<:
<: Still cant visualise it :-(

You can see it on a SA. But thats the way it is....

<:
<: > The hardest thing about RF is building a stable LO. When
<: we get going
<: > with this I will build you a 1GHz LO or direct you at a a
<: man who will
<: > sell you a kit - then i will let you build it and I will
<: set it up on
<: > my SA.<:
<:
<: Whereabouts do you live?

Haslington

<:
<: And where did you get an SA? I'd love one...
<:
Bought it S/H from Telford Electronic Distribution  about 6 years ago -
totally underused and really needs aligning

<:
<: Well it has to be better than 10 degrees, for example...
<:

Oh much better.

<: > To resolve a signal by drift scanning, you
<: > beamwidth should be 0.5 the size of the source you are
<: measuring.....
<: > if you want to detect a pattern. My gut feel is that it will be
<: > difficult to detect Taurus A (the strongest interstellar
<: Radio Source)
<: >  with back yard equipment.
<:
<: I suspect so.
<:
<: > I am going to start on the Sun and see how that goes.
<:
<: Same here.
<:
<: > You need to be with 1/10 of a wavelength - 10GHz = 3 cm 1/10 = 3mm.
<:

Have a look at the difference between Sin0 and Sin 10, then the difference
between Sin0 and Sin20... Consider multiplying your wanted signal with each
of the two products above.  that is the fiddle factor.

<: unsurprisingly, optical astronomy uses a similar degree of accuracy
<: (relative to the wavelength). I keep meaning to try and make a mirror
<: but never get around to it:(

Funny that - same here..

<:
<: > <: surely if we can get pointed at the same source, then we
<: > <: could find the
<: > <: correct distance by simply moving one antenna until the
<: > <: signal strength
<: > <: peaked?
<: > <:
<: > <: we could practise on the sun or moon ;-)
It may feasible if we were on a perfect E-W direction - however, the element
of N-S and difference in altitude would make it difficult if not impossible,
but I think that this is the lesser of the problems.

<: But would my idea work? simply moving the dishes within a known range
<: until the mixed signal peaks?

If you are looking at two random noise sources, how would you know they have
peaked.?

<:
<:
<: > Still not convinced.
<:
<: I've seen it. Stepped side to side with it and seen the numbers move,
<: and stood still and watched them not dither.

But I think that is a function of the GPS receiver with a damping effect
rather than the knowing your exact position,

<:
<: Even with a crummy GPS I'd imagine an average over a long
<: enough period
<: would get you below 1m accuracy.
<:
<: > Some GPS systems have an RFoutput, you simply divide that
<: down until
<: > you get the basis of you LO - typically 100MHz ish and
<: lock that with
<: > you LO crystal.
<:
<: ok, that (again) only gets us frequencuy lock though. I dont think
<: locking the receivers is worth the effort though, since we'd still be
<: sampling the data onto a computer and the soundcard will have drift,
<: jitter, and whatnot of its own.

Locking the receivers is absolutely essential. Imagine that you are only
10Hz out. That means that every 1/10 second your mixed signals will max and
totally cancel each other out. The absolute frequency doesn't matter, but
they do neeed to be identical.

<:
<: I think that an adaptive lock of some sort on the
<: data-gathering machine
<: would work better. We'd just need to keep the frequency drift between
<: our stations within our bandwidth (20kHz on a decent card). I recon
<: ~10Hz stability shouldnt be impossible.

Assuming that you get your absolute 3dB beamwidth down to 2 degrees. So you
will have to measure over 4 degrees min to watch the source pass through.  -
Celestial objects move at 15 deg per hour - so you will have to sample for
15/4 = approx 4 mins. Now superimpose a frequency shift on that. Don't
forget that an interferometer works by measuring the nulls within the wanted
beamwidth to get the desired resolution. You would not be able to
differentiate.  Locking receivers is essential.. but not the hardest part.

<:
<: > with a dish with an effective
<: > diameter of about 8000 miles.
<:
<: Not between your house and mine ;-)
<:
<: > They call this Aperture Synthesis which
<: > is the next stage on from interferometery... but the
<: stumbling block
<: > is measuring the distance to within 3mm.... or else we all would do
<: > it!
<:
<: :-)
<:
<: > and perhaps subscribe to the SARA mailing list... some good stuff
<: > there.
<:
<: Do you have the address to subscribe to? I tried to before
<: and ended up
<: on a list with about 3 posts a year.
<:
Can't remember - but you will get in on the SARA website.
<:
<: On another note, I have attacked 3 of my LNBs tonight and hopefully
<: should soon have 2 10GHz active aerials and one mixer/amp.
<: (waiting for
<: dremel battery to charge)
<:
<: :-)
<:
<: Oh, and could you do me a favour and quote in a standard way, if
<: possible? I find your replies very hard to read :(

Sure - What is your definition of  standard?

KM