DIMM seeing test, seeking prism source

A question came up in the STV group (SBIGSTV) at Yahoo regarding
the DIMM measurement. DIMM = Differential Image Motion Monitor,
and it appears to be the present de facto method to evaluate sites
for astronomical use as I've found in 100s of papers and journals
over the past 5 days.

The SBIG STV device has a DIMM operational mode, but it seems
everyone who has attempted it has run into some difficulties, and
it doesn't appear to be due to the algorithms but, instead, to
the aperture mask and whether one should have a prism over one of
the sub-apertures or not (with "not" being a slightly defocused
image). In brief, the mask is a Hartmann mask and the goal is to
have 2 images of the same star appear on the CCD chip after passing
through the mask whose baseline is typically 6 to 10 inches
depending on the scope used.

After a lot of web surfing and document search, the consensus is
that a prism "should" be used in one of the sub-apertures of the
DIMM mask for optimal results. SBIG's docs make no mention of a
prism being used or needed for this application.

Everywhere I've read about the prism, a 30 arcsecond deviation is
what's being used whether with the SBIG STV or a custom, home-brew
imaging station at some of the world's major observatories.

The prism must be custom made since I've found nothing like it in
the Edmund Industrial Optics 2004 catalog 41B or elsewhere.

Some have reported minor success with slight defocusing and with
both sub-apertures "clear" (i.e., no prism or beam-splitter).

I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


If anyone's interested in further reading, some of the better
references site testing and DIMM usage are in the following list;
most of the scopes used are SCTs from 8" to 12".


Prisms used for DIMM in mask's sub-apertu

http://www.ctio.noao.edu/telescopes/dimm/dimm.html

http://www.ing.iac.es/Astronomy/development/hap/dimm.html

http://www.astro.virginia.edu/~cp8h/dimm/

http://www.astro.washington.edu/rest/dimm/dimmtext.htm

http://www.astro.caltech.edu/~george/celt/schoeck.ps

http://www-astro.unice.fr/Concordia/sitetesting.html
http://www-astro.unice.fr/Concordia/Documents/dimm.pdf

The above PDF document (finally!) describes the prism:
7cm x 7cm with 30 arcsec deviation angle on one of the
sub-apertures. Note this is being done in Antarctica
and also referenced in the paper below:

http://sait.oat.ts.astro.it/MSAIS/2/PDF/146.pdf

http://galaxy.ps.uci.edu/~celtsite/equipment.html

Also at the above site ("uci.edu"), click on the link for
"Sarazin & Roddier 1990" for the original definitive
article about DIMM.



Defocused DIMM:

http://www.iac.es/proyect/sitesting/PDF/defocus.pdf


Defocused and with prism:

http://www.tug.tubitak.gov.tr/gozlemler/SDIMM/



Hartmann-DIMM (without prism):

http://astro.uchicago.edu/cara/research/papers/bob/

A question came up in the STV group (SBIGSTV) at Yahoo regarding
the DIMM measurement. DIMM = Differential Image Motion Monitor,
and it appears to be the present de facto method to evaluate sites
for astronomical use as I've found in 100s of papers and journals
over the past 5 days.

SNIP

The prism must be custom made since I've found nothing like it in
the Edmund Industrial Optics 2004 catalog 41B or elsewhere.

Some have reported minor success with slight defocusing and with
both sub-apertures "clear" (i.e., no prism or beam-splitter).

I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


I have always used Technical Optics (now CVI Technical Optics) for my
"specials" as they have always been very competetive on pricing. I doubt
if they can match your price target for a window as large as 7x7cm but
they would still be my first port of call.

CVI Technical Optics Ltd
Second Avenue,
Onchan,
IOM
IM3 4PA
United Kingdom
Europe

http://www.cvi-tol.co.uk/

Tel.: +44 - (0)1624 - 647 000
Fax.: +44- (0)1624- 676 859
e-mail:


Hope this helps you.
************************************************** ******

Dr Peter R Hobson
Reader, Department of Electronic and Computer Engineering
Brunel University, Uxbridge UB8 3PH G.B.

A question came up in the STV group (SBIGSTV) at Yahoo regarding
the DIMM measurement. DIMM = Differential Image Motion Monitor,
and it appears to be the present de facto method to evaluate sites
for astronomical use as I've found in 100s of papers and journals
over the past 5 days.

SNIP

The prism must be custom made since I've found nothing like it in
the Edmund Industrial Optics 2004 catalog 41B or elsewhere.

Some have reported minor success with slight defocusing and with
both sub-apertures "clear" (i.e., no prism or beam-splitter).

I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


I have always used Technical Optics (now CVI Technical Optics) for my
"specials" as they have always been very competetive on pricing. I doubt
if they can match your price target for a window as large as 7x7cm but
they would still be my first port of call.

CVI Technical Optics Ltd
Second Avenue,
Onchan,
IOM
IM3 4PA
United Kingdom
Europe

http://www.cvi-tol.co.uk/

Tel.: +44 - (0)1624 - 647 000
Fax.: +44- (0)1624- 676 859
e-mail:


Hope this helps you.
************************************************** ******

Dr Peter R Hobson
Reader, Department of Electronic and Computer Engineering
Brunel University, Uxbridge UB8 3PH G.B.

"Thad Floryan" wrote in message
om...
A question came up in the STV group (SBIGSTV) at Yahoo regarding
the DIMM measurement. DIMM = Differential Image Motion Monitor,
and it appears to be the present de facto method to evaluate sites
for astronomical use as I've found in 100s of papers and journals
over the past 5 days.

The SBIG STV device has a DIMM operational mode, but it seems
everyone who has attempted it has run into some difficulties, and
it doesn't appear to be due to the algorithms but, instead, to
the aperture mask and whether one should have a prism over one of
the sub-apertures or not (with "not" being a slightly defocused
image). In brief, the mask is a Hartmann mask and the goal is to
have 2 images of the same star appear on the CCD chip after passing
through the mask whose baseline is typically 6 to 10 inches
depending on the scope used.

After a lot of web surfing and document search, the consensus is
that a prism "should" be used in one of the sub-apertures of the
DIMM mask for optimal results. SBIG's docs make no mention of a
prism being used or needed for this application.

Everywhere I've read about the prism, a 30 arcsecond deviation is
what's being used whether with the SBIG STV or a custom, home-brew
imaging station at some of the world's major observatories.

The prism must be custom made since I've found nothing like it in
the Edmund Industrial Optics 2004 catalog 41B or elsewhere.

Some have reported minor success with slight defocusing and with
both sub-apertures "clear" (i.e., no prism or beam-splitter).

I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


If anyone's interested in further reading, some of the better
references site testing and DIMM usage are in the following list;
most of the scopes used are SCTs from 8" to 12".


Prisms used for DIMM in mask's sub-apertu

http://www.ctio.noao.edu/telescopes/dimm/dimm.html

http://www.ing.iac.es/Astronomy/development/hap/dimm.html

http://www.astro.virginia.edu/~cp8h/dimm/

http://www.astro.washington.edu/rest/dimm/dimmtext.htm

http://www.astro.caltech.edu/~george/celt/schoeck.ps

http://www-astro.unice.fr/Concordia/sitetesting.html
http://www-astro.unice.fr/Concordia/Documents/dimm.pdf

The above PDF document (finally!) describes the prism:
7cm x 7cm with 30 arcsec deviation angle on one of the
sub-apertures. Note this is being done in Antarctica
and also referenced in the paper below:

http://sait.oat.ts.astro.it/MSAIS/2/PDF/146.pdf

http://galaxy.ps.uci.edu/~celtsite/equipment.html

Also at the above site ("uci.edu"), click on the link for
"Sarazin & Roddier 1990" for the original definitive
article about DIMM.



Defocused DIMM:

http://www.iac.es/proyect/sitesting/PDF/defocus.pdf


Defocused and with prism:

http://www.tug.tubitak.gov.tr/gozlemler/SDIMM/



Hartmann-DIMM (without prism):

http://astro.uchicago.edu/cara/research/papers/bob/

it appears that this is simply a 2 hole Hartmann mask . The new twist is
having a 30 arcsecond permanent image shift by using the wedge in one of the
2 holes , which allows the 2 images to be permanently separated at the focal
plane . The same effect of having the 2 images separated is accomplished by
a slight defocus . I'd try first without any prism/wedge . I'd just make a
cardboard Hartmann mask and see if the defocus that's required for a few
arcseconds of star image separation makes the 2 star images too fuzzy/dim .
In my C-11 scope for example I know this is ok, as I usually use a similar
mask for focusing . If you get an optical window, most have non parallel
back to front , the amount of wedge being more than enough for your purposes
.. Try it first with any filter you may already own by simply placing it in
front of one of your Harmann mask holes after focusing and see if you get
enough separation. Many beamsplitters and cold or hot mirrors have a built
in wedge to eliminate internal ghosting so getting one of them might work
too . Check Thorlabs for broadband beamsplitters with a built in wedge angle
of 30 arcseconds.

best regards,
matt tudor

"Thad Floryan" wrote in message
om...
A question came up in the STV group (SBIGSTV) at Yahoo regarding
the DIMM measurement. DIMM = Differential Image Motion Monitor,
and it appears to be the present de facto method to evaluate sites
for astronomical use as I've found in 100s of papers and journals
over the past 5 days.

The SBIG STV device has a DIMM operational mode, but it seems
everyone who has attempted it has run into some difficulties, and
it doesn't appear to be due to the algorithms but, instead, to
the aperture mask and whether one should have a prism over one of
the sub-apertures or not (with "not" being a slightly defocused
image). In brief, the mask is a Hartmann mask and the goal is to
have 2 images of the same star appear on the CCD chip after passing
through the mask whose baseline is typically 6 to 10 inches
depending on the scope used.

After a lot of web surfing and document search, the consensus is
that a prism "should" be used in one of the sub-apertures of the
DIMM mask for optimal results. SBIG's docs make no mention of a
prism being used or needed for this application.

Everywhere I've read about the prism, a 30 arcsecond deviation is
what's being used whether with the SBIG STV or a custom, home-brew
imaging station at some of the world's major observatories.

The prism must be custom made since I've found nothing like it in
the Edmund Industrial Optics 2004 catalog 41B or elsewhere.

Some have reported minor success with slight defocusing and with
both sub-apertures "clear" (i.e., no prism or beam-splitter).

I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


If anyone's interested in further reading, some of the better
references site testing and DIMM usage are in the following list;
most of the scopes used are SCTs from 8" to 12".


Prisms used for DIMM in mask's sub-apertu

http://www.ctio.noao.edu/telescopes/dimm/dimm.html

http://www.ing.iac.es/Astronomy/development/hap/dimm.html

http://www.astro.virginia.edu/~cp8h/dimm/

http://www.astro.washington.edu/rest/dimm/dimmtext.htm

http://www.astro.caltech.edu/~george/celt/schoeck.ps

http://www-astro.unice.fr/Concordia/sitetesting.html
http://www-astro.unice.fr/Concordia/Documents/dimm.pdf

The above PDF document (finally!) describes the prism:
7cm x 7cm with 30 arcsec deviation angle on one of the
sub-apertures. Note this is being done in Antarctica
and also referenced in the paper below:

http://sait.oat.ts.astro.it/MSAIS/2/PDF/146.pdf

http://galaxy.ps.uci.edu/~celtsite/equipment.html

Also at the above site ("uci.edu"), click on the link for
"Sarazin & Roddier 1990" for the original definitive
article about DIMM.



Defocused DIMM:

http://www.iac.es/proyect/sitesting/PDF/defocus.pdf


Defocused and with prism:

http://www.tug.tubitak.gov.tr/gozlemler/SDIMM/



Hartmann-DIMM (without prism):

http://astro.uchicago.edu/cara/research/papers/bob/

it appears that this is simply a 2 hole Hartmann mask . The new twist is
having a 30 arcsecond permanent image shift by using the wedge in one of the
2 holes , which allows the 2 images to be permanently separated at the focal
plane . The same effect of having the 2 images separated is accomplished by
a slight defocus . I'd try first without any prism/wedge . I'd just make a
cardboard Hartmann mask and see if the defocus that's required for a few
arcseconds of star image separation makes the 2 star images too fuzzy/dim .
In my C-11 scope for example I know this is ok, as I usually use a similar
mask for focusing . If you get an optical window, most have non parallel
back to front , the amount of wedge being more than enough for your purposes
.. Try it first with any filter you may already own by simply placing it in
front of one of your Harmann mask holes after focusing and see if you get
enough separation. Many beamsplitters and cold or hot mirrors have a built
in wedge to eliminate internal ghosting so getting one of them might work
too . Check Thorlabs for broadband beamsplitters with a built in wedge angle
of 30 arcseconds.

best regards,
matt tudor

Peter R Hobson wrote in message ...
[...]
I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


I have always used Technical Optics (now CVI Technical Optics) for my
"specials" as they have always been very competetive on pricing. I doubt
if they can match your price target for a window as large as 7x7cm but
they would still be my first port of call.

CVI Technical Optics Ltd
Second Avenue,
Onchan,
IOM
IM3 4PA
United Kingdom
Europe

http://www.cvi-tol.co.uk/

Tel.: +44 - (0)1624 - 647 000
Fax.: +44- (0)1624- 676 859
e-mail:


Thank you VERY much for that reference! I will contact them to see what can
be arranged. I'm also sending email to several authors of the various DIMM
papers since one person was thanked (in a postscript) for having secured the
prism for use in the tests, and with the increasing usage of DIMM there should
(soon?) be a larger market for such prisms.

Peter R Hobson wrote in message ...
[...]
I'm still experimenting with DIMM and would like to continue the
tests this time using a prism. Does anyone have any idea where
such a prism (roughly 2 to 3 inches square or circular) with a
30 arcsec "bending" can be found without paying an arm and a leg?
For these purposes "an arm and a leg" can be defined to be no more
than US$200. :-)


I have always used Technical Optics (now CVI Technical Optics) for my
"specials" as they have always been very competetive on pricing. I doubt
if they can match your price target for a window as large as 7x7cm but
they would still be my first port of call.

CVI Technical Optics Ltd
Second Avenue,
Onchan,
IOM
IM3 4PA
United Kingdom
Europe

http://www.cvi-tol.co.uk/

Tel.: +44 - (0)1624 - 647 000
Fax.: +44- (0)1624- 676 859
e-mail:


Thank you VERY much for that reference! I will contact them to see what can
be arranged. I'm also sending email to several authors of the various DIMM
papers since one person was thanked (in a postscript) for having secured the
prism for use in the tests, and with the increasing usage of DIMM there should
(soon?) be a larger market for such prisms.

"matt" wrote in message ...
[...]
it appears that this is simply a 2 hole Hartmann mask . The new twist is
having a 30 arcsecond permanent image shift by using the wedge in one of the
2 holes , which allows the 2 images to be permanently separated at the focal
plane . The same effect of having the 2 images separated is accomplished by
a slight defocus . I'd try first without any prism/wedge . I'd just make a
cardboard Hartmann mask and see if the defocus that's required for a few
arcseconds of star image separation makes the 2 star images too fuzzy/dim .

That's what several of us have already attempted, and that's where the problem
seems to lie. The slightly-defocused images aren't "good enough" for the DIMM
and that's why the wedge prisms have been used as stated in most of the reports
and papers I've found the past week.

A second issue arises with the sub-apertures' baseline and diameters; There is
disagreement as to the correct baseline percentage of clear aperture, with
values ranging from 70% to 80% (and other) and sub-aperture diameter percentage
of 30% to 20% (and other). I'm tabulating all the dimensions from the several
100s of reports I've found and need to re-read the 1990 paper for some more
insight.

In my C-11 scope for example I know this is ok, as I usually use a similar
mask for focusing . If you get an optical window, most have non parallel
back to front , the amount of wedge being more than enough for your purposes
. Try it first with any filter you may already own by simply placing it in
front of one of your Harmann mask holes after focusing and see if you get
enough separation. Many beamsplitters and cold or hot mirrors have a built
in wedge to eliminate internal ghosting so getting one of them might work
too . Check Thorlabs for broadband beamsplitters with a built in wedge angle
of 30 arcseconds.

Thank you very much for the reference to Thorlabs (www.thorlabs.com); I've
already noted they have 2 degree wedge prisms at a VERY reasonable price, so
it should be possible to get some made with 30 arcsecond deviation within my
original budget constraint.

"matt" wrote in message ...
[...]
it appears that this is simply a 2 hole Hartmann mask . The new twist is
having a 30 arcsecond permanent image shift by using the wedge in one of the
2 holes , which allows the 2 images to be permanently separated at the focal
plane . The same effect of having the 2 images separated is accomplished by
a slight defocus . I'd try first without any prism/wedge . I'd just make a
cardboard Hartmann mask and see if the defocus that's required for a few
arcseconds of star image separation makes the 2 star images too fuzzy/dim .

That's what several of us have already attempted, and that's where the problem
seems to lie. The slightly-defocused images aren't "good enough" for the DIMM
and that's why the wedge prisms have been used as stated in most of the reports
and papers I've found the past week.

A second issue arises with the sub-apertures' baseline and diameters; There is
disagreement as to the correct baseline percentage of clear aperture, with
values ranging from 70% to 80% (and other) and sub-aperture diameter percentage
of 30% to 20% (and other). I'm tabulating all the dimensions from the several
100s of reports I've found and need to re-read the 1990 paper for some more
insight.

In my C-11 scope for example I know this is ok, as I usually use a similar
mask for focusing . If you get an optical window, most have non parallel
back to front , the amount of wedge being more than enough for your purposes
. Try it first with any filter you may already own by simply placing it in
front of one of your Harmann mask holes after focusing and see if you get
enough separation. Many beamsplitters and cold or hot mirrors have a built
in wedge to eliminate internal ghosting so getting one of them might work
too . Check Thorlabs for broadband beamsplitters with a built in wedge angle
of 30 arcseconds.

Thank you very much for the reference to Thorlabs (www.thorlabs.com); I've
already noted they have 2 degree wedge prisms at a VERY reasonable price, so
it should be possible to get some made with 30 arcsecond deviation within my
original budget constraint.

"Thad Floryan" wrote in message
om...
"matt" wrote in message
...
[...]
it appears that this is simply a 2 hole Hartmann mask . The new twist is
having a 30 arcsecond permanent image shift by using the wedge in one of
the
2 holes , which allows the 2 images to be permanently separated at the
focal
plane . The same effect of having the 2 images separated is accomplished
by
a slight defocus . I'd try first without any prism/wedge . I'd just make
a
cardboard Hartmann mask and see if the defocus that's required for a
few
arcseconds of star image separation makes the 2 star images too
fuzzy/dim .

That's what several of us have already attempted, and that's where the
problem
seems to lie. The slightly-defocused images aren't "good enough" for the
DIMM
and that's why the wedge prisms have been used as stated in most of the
reports
and papers I've found the past week.

A second issue arises with the sub-apertures' baseline and diameters;
There is
disagreement as to the correct baseline percentage of clear aperture, with
values ranging from 70% to 80% (and other) and sub-aperture diameter
percentage
of 30% to 20% (and other). I'm tabulating all the dimensions from the
several
100s of reports I've found and need to re-read the 1990 paper for some
more
insight.

In my C-11 scope for example I know this is ok, as I usually use a
similar
mask for focusing . If you get an optical window, most have non parallel
back to front , the amount of wedge being more than enough for your
purposes
. Try it first with any filter you may already own by simply placing it
in
front of one of your Harmann mask holes after focusing and see if you
get
enough separation. Many beamsplitters and cold or hot mirrors have a
built
in wedge to eliminate internal ghosting so getting one of them might
work
too . Check Thorlabs for broadband beamsplitters with a built in wedge
angle
of 30 arcseconds.

Thank you very much for the reference to Thorlabs (www.thorlabs.com); I've
already noted they have 2 degree wedge prisms at a VERY reasonable price,
so
it should be possible to get some made with 30 arcsecond deviation within
my
original budget constraint.

They also have some plate beamsplitters which are broadband antireflection
coated on one side and have a built in wedge between front and back of 30
arcmin to eliminate reflections . They come in a BK7 and a quartz version,
both 1/10lambda surface . They go up to 2" in diameter . Part numbers are
BSW13 to BSW18 depending on the beamsplitter coatings. Probably selecting
the "wrong" coating would yield very little reflectivity at 0 deg .

An even better fit would be the Thorlabs beamsamplers, which are optical
windows made of BK7 or quartz, 1/10lambda , with an uncoated front and a
broadband antireflection coated back . They too come in 2" diameter and have
the built in 30 arcmin wedge between front and back surfaces . They're
intended to reflect between 1% - 10% at 45 deg angle of incidence . Part
numbers are BSP20-A1 and BSF20-A1 .

If the 30 arcmin wedge is too much, although this is 4 times smaller than
the wedge prisms, you may try to combine 2 of these beam sampler wedges .
Depending on their orientation you can steer incident light and by rotating
one wedge with respect to the other you can change the steering angle
continuously from 0 . Simply rotating the 2 wedges will allow you to
experiment with the angle value .

best regards,
matt tudor