Forum der Vereinigung der Sternfreunde

Hello

Is anyone going to take IR spectra of the targets?
Has anyone tried it so far?

Me and Aristidis Voulgaris are interested in IR.
Aristidis has made an IR spectrograph and with a CANON 300D(IR filter removed) has gone a bit further than 10000A.

We still have to check the CCDs we have access to.
So, has anyone tried such an observation and with which one?

Cheers

_________________
Everything is Mathematics !

We have been talking about getting some NIR spectra (JHK bands) next summer during the campaign, assuming telescope allocations work and instruments are available, etc. It would be great to see us get some additional IR data. However, with CCDs this may be very difficult. I'd like to see some test spectra of WR 134 around He I 10830 - that line would be fabulous to monitor, but is out of range of most instruments.

Hi Noel,

What sort of resolution would be needed for the NIR work you are suggesting? I know that Buil, using a low resolution a LISA (IR) spectrograph with his 11" scope, has obtained reasonable spectra around 10,000A with a resolution around 10A ... see http://www.astrosurf.com/buil/lisa5/lisa.htm . A Lhires III with a 600 l/mm grating blazed at 7500A or 10000A (and an order filter) should be able to do as well, or better with 14" or 16" scopes.

Cheers, Bernard

_________________
Quicquid Nitet Notandum

Hi Bernard!

Watch out, it's not the grating. Noel already mentioned it. A typical CCD has an efficiency well below 5% at 1 micron. So, you loose at least about 3 magnitudes.

Cheers, Thomas

Hi Thomas,

Exactly, the CCD is the main weakness, which is why I'm implicitly suggesting that the only hope with amateur equipment is if lower resolution is acceptable. I believe a Lhires III with a 600 l/mm grating is roughly equivalent to a LISA ... which my link shows is able to record a possibly usable spectra at 1um ... and the 1um blazed grating has an efficiency around 70% vs 30% for the standard visual grating. The reason I'm suggesting the Lhires, rather than the LISA, is that I believe most CONVENTO amateurs have the former instrument.

I've been wanting to experiment in the NIR region but hadn't decided if the 0.75um or 1um blazed 600 l/mm grating was the best to purchase, now this discussion has pushed me towards the 1um grating.

Cheers, Bernard

_________________
Quicquid Nitet Notandum

Hello everyone

With a custom made IR spectrograph and a modded CANON EOS 350D(filter removed),we managed to go beyond 10000 A.
Next try will have to be with an FLI CCD,but it will have to wait until the Australia eclipse.

What resolution,SNR and dispersion should we aim at? The same as in the optical domain?

_________________
Everything is Mathematics !

Can you tell us more about the 'custom made IR spectrograph'?

Bernard

_________________
Quicquid Nitet Notandum

Hello

unfortunately I don't have a lot of info.
It is made by Aristeidis Voulgaris in Greece.The diffraction grating has a gold substrate.
It has 600l/mm blazes at 750nm.

I don't have a lot more information.
I saw very quickly that it has a dipsersion of 0,995 A/px at the Sodium double line.

_________________
Everything is Mathematics !

If the achromat colimator/objective at F/8 still puts all the spectral range at the same focus, the lhires should still work with a gold coated IR blazed grating as well, I suppose..

_________________
Fil.

Using a gold overcoat would not be that beneficial in the 1.0-1.1 micron range (see graph), which is where Noel is interested, especially given the extra $60 cost, but I don't see any problem using a gold version in the Lhires if that is deemed worthwhile. If we are interested in the NIR above 1 micron then I think the 1 micron blazed 600 line grating is the best choice, or the 750nm blazed (also Buil's choice) below 1 micron as that also gives a good H alpha response.

Cheers, Bernard

_________________
Quicquid Nitet Notandum

Hi Bernard,Well, the gold overcoat could perhaps eliminate the order filter, no? I mean, Gold is really bad reflector for anything under 5500 A, so I figure we will be safe of any second order diffraction of wavelengths bellow 11000A of the first order.. The order filter will take away some signal as well...  But this this was my machiavelic thoughts suggesting...
But in that plot, I see that gold perhaps still reflects considerably

_________________
Fil.

Good thinking!
While the gold overcoat would certainly reduce the 2nd order overlap problem I don't think the cut-off is sharp enough ... still around 50% reflectivity at 5450A. With my Atik314L CCD giving a window of 10150A to 10850A (600 l/mm grating) at the 10500A target wavelength I would need near zero reflectivity below 5435A.

Cheers, Bernard

_________________
Quicquid Nitet Notandum

Another issue is the CCD sensitivity!

Most of the normal cameras are almost dead at these wavelegnths.

_________________
Everything is Mathematics !

Not another issue ... the main issue ... which is what Thomas and I were discussing a little earlier.

Bernard

_________________
Quicquid Nitet Notandum

What about photonmultipliers?

Could they be of any use?
I started some reading about them and their spectral response.

We have available everything needed to start with them.Maybe we could use them with the IR spectrograph.

Any ideas on that?

_________________
Everything is Mathematics !

Dear All - some more notes on the 10830 Angstrom line.

For more information on what you can do with the HeI 10830 line in WR
stars see the observations by Eenens and Williams

http://adsabs.harvard.edu/abs/1994MNRAS.269.1082E

who observed a sample of WR stars (including some of the current
campaign) and work I did with Ian Howarth

http://adsabs.harvard.edu/abs/1999MNRAS.302..549S

where we observed the variability of the HeI 10830 for a smaller sample of
WR stars.

The observations by Philippe and Perry were done with UKIRT (3.8m) and
the CGS2 instrument, while the later were done with the INT on La Palma
(2.5m) and a more standard spectrograph and CCD (and where the CCD
efficiency was dropping off very sharply with wavelength).

For the INT observations we did 6 WR stars. The exposures were typically
900-1800 secs and as you will see the SNR was not that great, but good
enough to see large scale changes in the line profiles. With a 0.5m
telescope the exposures will need to be VERY long to be useful.

Of the 6 stars, 5 were known WR+O binaries and showed major line profile
changes, associated with the wind-wind collision. The one that was not an
obvious binary (WR136) showed much less variability. Higher SNR would
probably help and enable us to see intrinsic wind variability, but the
exposure times will be extremely long (without a detector with a much
higher NIR efficiency). As has been mentioned before - that is the real
killer for this emission line. InSb arrays (or similar) may well become
more available for amateurs - indeed with a bit of digging you can have
an NIR photometer for 3000USD.

http://www.optecinc.com/astronomy/catalog/ssp/ssp4.htm

and maybe there is already something out there for NIR spectroscopy (but
it probably involves liquid nitrogen, which starts getting serious).

So, it would be interesting to see whether you can detect the line at
all with a standard CCD, but they will be challenging observations to be
useful (compared to 5696Angstrom, which is so much easier all round).

Cheers

ian





-------------------------------------------------------------------------
Dr. Ian Stevens
School of Physics & Astronomy Tel. +44 121 414 6450
University of Birmingham, Edgbaston Fax +44 121 414 3722
Birmingham B15 2TT, UK E-mail irs@star.sr.bham.ac.uk
-------------------------------------------------------------------------


On Sat, 13 Oct 2012, Alexandros Filothodoros wrote:

All good points, Ian ? thanks! But of course this WR campaign will concentrate on non-binary phenomena: clumps and CIR, so we do indeed need quite high S/N in a relatively short time (< an hour). If that is still possible in HeI 10830A, then that would be a great add-on. But I agree that normal optical CCDs are not well suited for this. We should probably still concentrate on the optical lines, as noted in the proposal.

Best, Tony

From: Ian Stevens (fg-spek-convento@vds-astro.de)
Sent: Monday, October 15, 2012 10:39 AM
To: fg-spek-convento@vds-astro.de (fg-spek-convento@vds-astro.de)
Subject: [fg-spek] Re: IR observations of WR 134+135




Dear All - some more notes on the 10830 Angstrom line.

For more information on what you can do with the HeI 10830 line in WR
stars see the observations by Eenens and Williams

http://adsabs.harvard.edu/abs/1994MNRAS.269.1082E

who observed a sample of WR stars (including some of the current
campaign) and work I did with Ian Howarth

http://adsabs.harvard.edu/abs/1999MNRAS.302..549S

where we observed the variability of the HeI 10830 for a smaller sample of
WR stars.

The observations by Philippe and Perry were done with UKIRT (3.8m) and
the CGS2 instrument, while the later were done with the INT on La Palma
(2.5m) and a more standard spectrograph and CCD (and where the CCD
efficiency was dropping off very sharply with wavelength).

For the INT observations we did 6 WR stars. The exposures were typically
900-1800 secs and as you will see the SNR was not that great, but good
enough to see large scale changes in the line profiles. With a 0.5m
telescope the exposures will need to be VERY long to be useful.

Of the 6 stars, 5 were known WR+O binaries and showed major line profile
changes, associated with the wind-wind collision. The one that was not an
obvious binary (WR136) showed much less variability. Higher SNR would
probably help and enable us to see intrinsic wind variability, but the
exposure times will be extremely long (without a detector with a much
higher NIR efficiency). As has been mentioned before - that is the real
killer for this emission line. InSb arrays (or similar) may well become
more available for amateurs - indeed with a bit of digging you can have
an NIR photometer for 3000USD.

http://www.optecinc.com/astronomy/catalog/ssp/ssp4.htm

and maybe there is already something out there for NIR spectroscopy (but
it probably involves liquid nitrogen, which starts getting serious).

So, it would be interesting to see whether you can detect the line at
all with a standard CCD, but they will be challenging observations to be
useful (compared to 5696Angstrom, which is so much easier all round).

Cheers

ian





-------------------------------------------------------------------------
Dr. Ian Stevens
School of Physics & Astronomy Tel. +44 121 414 6450
University of Birmingham, Edgbaston Fax +44 121 414 3722
Birmingham B15 2TT, UK E-mail irs(==>)star.sr.bham.ac.uk ([email]irs(==>)star.sr.bham.ac.uk[/email])
-------------------------------------------------------------------------


On Sat, 13 Oct 2012, Alexandros Filothodoros wrote:

: Hello everyone

With a custom made IR spectrograph and a modded CANON EOS 350D(filter removed),we managed to go beyond 10000 A.
Next try will have to be with an FLI CCD,but it will have to wait until the Australia eclipse.

What resolution,SNR and dispersion should we aim at? The same as in the optical domain?

------------------------
Everything is Mathematics !

_________________
Tony Moffat

I agree with Tony, and for the campaign we should concentrate on the other lines. It seems that the best we could do with a small telescope (<0.5 m) is to observe it with the 600 g/mm grating, and the resolution would not find the features we are most interested in for the campaign. There is some merit to the low resolution long-term monitoring of the line in stars such as P Cygni (LBV) or in other WR stars, but we should concentrate on the clumping lines next summer!

I haven't worked with PM tubes for at least thirty years so no doubt the technology has moved on, but in 'my day' they were single channel devices ... like a single pixel CCD ... somewhat limiting.

Bernard

_________________
Quicquid Nitet Notandum

Dear Bernard, Ian, Tony and all

I've just got back from travelling to see the spectra reaching out
to 1 micron, which is very exciting. It would be great to have
spectroscopy of the 10830 line but I agree very long integration
times would be needed. There are also atmospheric features in this
wavelength region so it would be necessary to observe a comparison
star at the same elevation, making it more difficult to pick up
subtle changes. WR134 and WR137 (but not WR138) have very strong
sub-peaks and deserve to be observed over time to look for changes
which may be caused by wind-collision effects in a binary but I
don't think this fits the current campaign.

Regards to all
Peredur

-------------------------------------------------------------------
Peredur Williams, Institute for Astronomy, University of Edinburgh

Royal Observatory, e-mail: pmw@roe.ac.uk
Blackford Hill, phone: (+44) (0) 131 668 8300
Edinburgh fax: (+44) (0) 131 668 8416
EH9 3HJ www: http://www.roe.ac.uk/~pmw

-------------------------------------------------------------------

-----Original Message-----
From: Ian Stevens [mailto:fg-spek-convento@vds-astro.de]
Sent: Monday, October 15, 2012 15:40
To: fg-spek-convento@vds-astro.de
Subject: [fg-spek] Re: IR observations of WR 134+135


Dear All - some more notes on the 10830 Angstrom line.

For more information on what you can do with the HeI 10830 line in WR
stars see the observations by Eenens and Williams

http://adsabs.harvard.edu/abs/1994MNRAS.269.1082E

who observed a sample of WR stars (including some of the current
campaign) and work I did with Ian Howarth

http://adsabs.harvard.edu/abs/1999MNRAS.302..549S

where we observed the variability of the HeI 10830 for a smaller sample of
WR stars.

The observations by Philippe and Perry were done with UKIRT (3.8m) and
the CGS2 instrument, while the later were done with the INT on La Palma
(2.5m) and a more standard spectrograph and CCD (and where the CCD
efficiency was dropping off very sharply with wavelength).

For the INT observations we did 6 WR stars. The exposures were typically
900-1800 secs and as you will see the SNR was not that great, but good
enough to see large scale changes in the line profiles. With a 0.5m
telescope the exposures will need to be VERY long to be useful.

Of the 6 stars, 5 were known WR+O binaries and showed major line profile
changes, associated with the wind-wind collision. The one that was not an
obvious binary (WR136) showed much less variability. Higher SNR would
probably help and enable us to see intrinsic wind variability, but the
exposure times will be extremely long (without a detector with a much
higher NIR efficiency). As has been mentioned before - that is the real
killer for this emission line. InSb arrays (or similar) may well become
more available for amateurs - indeed with a bit of digging you can have
an NIR photometer for 3000USD.

http://www.optecinc.com/astronomy/catalog/ssp/ssp4.htm

and maybe there is already something out there for NIR spectroscopy (but
it probably involves liquid nitrogen, which starts getting serious).

So, it would be interesting to see whether you can detect the line at
all with a standard CCD, but they will be challenging observations to be
useful (compared to 5696Angstrom, which is so much easier all round).

Cheers

ian





-------------------------------------------------------------------------
Dr. Ian Stevens
School of Physics & Astronomy Tel. +44 121 414 6450
University of Birmingham, Edgbaston Fax +44 121 414 3722
Birmingham B15 2TT, UK E-mail )star.sr.bham.ac.uk">irs(==>)star.sr.bham.ac.uk <mailto:irs(==>
-------------------------------------------------------------------------


On Sat, 13 Oct 2012, Alexandros Filothodoros wrote:


:
Hello everyone

With a custom made IR spectrograph and a modded CANON EOS 350D(filter removed),we managed to go beyond 10000 A.
Next try will have to be with an FLI CCD,but it will have to wait until the Australia eclipse.

What resolution,SNR and dispersion should we aim at? The same as in the optical domain?

------------------------
Everything is Mathematics !













--
Scanned by iCritical.

Hi Peredur:

Thanks for your input! WR134 is not a binary, though, as far as we know. It does show strong CIR effects, though, with P = 2.25d, presumably the rotation period (Morel et al. c. 2000). We won?t be monitoring WR138 this time, but do include WR135, a definitely (well, as far as we can tell!) single WC8 star with lots of wind clumps. I.e. we?ll be doing WR134, 135, 137. To monitor WR137 and 138 over a complete binary orbit (13 and about 5 yrs ? I don?t have its period at my finger-tips ? resp.) would take much longer than the 4 months? monitoring we propose here to examine clumping and CIRs, the main motivation.

Cheers, Tony

From: Peredur Williams (fg-spek-convento@vds-astro.de)
Sent: Thursday, October 18, 2012 8:07 AM
To: fg-spek-convento@vds-astro.de (fg-spek-convento@vds-astro.de)
Subject: [fg-spek] Re: IR observations of WR 134+135




Dear Bernard, Ian, Tony and all

I've just got back from travelling to see the spectra reaching out
to 1 micron, which is very exciting. It would be great to have
spectroscopy of the 10830 line but I agree very long integration
times would be needed. There are also atmospheric features in this
wavelength region so it would be necessary to observe a comparison
star at the same elevation, making it more difficult to pick up
subtle changes. WR134 and WR137 (but not WR138) have very strong
sub-peaks and deserve to be observed over time to look for changes
which may be caused by wind-collision effects in a binary but I
don't think this fits the current campaign.

Regards to all
Peredur

-------------------------------------------------------------------
Peredur Williams, Institute for Astronomy, University of Edinburgh

Royal Observatory, e-mail: pmw(==>)roe.ac.uk ([email]pmw(==>)roe.ac.uk[/email])
Blackford Hill, phone: (+44) (0) 131 668 8300
Edinburgh fax: (+44) (0) 131 668 8416
EH9 3HJ www: http://www.roe.ac.uk/~pmw

-------------------------------------------------------------------

-----Original Message-----
From: Ian Stevens [mailto:fg-spek-convento(==>)vds-astro.de]
Sent: Monday, October 15, 2012 15:40
To: fg-spek-convento(==>)vds-astro.de ([email]fg-spek-convento(==>)vds-astro.de[/email])
Subject: [fg-spek] Re: IR observations of WR 134+135


Dear All - some more notes on the 10830 Angstrom line.

For more information on what you can do with the HeI 10830 line in WR
stars see the observations by Eenens and Williams

http://adsabs.harvard.edu/abs/1994MNRAS.269.1082E

who observed a sample of WR stars (including some of the current
campaign) and work I did with Ian Howarth

http://adsabs.harvard.edu/abs/1999MNRAS.302..549S

where we observed the variability of the HeI 10830 for a smaller sample of
WR stars.

The observations by Philippe and Perry were done with UKIRT (3.8m) and
the CGS2 instrument, while the later were done with the INT on La Palma
(2.5m) and a more standard spectrograph and CCD (and where the CCD
efficiency was dropping off very sharply with wavelength).

For the INT observations we did 6 WR stars. The exposures were typically
900-1800 secs and as you will see the SNR was not that great, but good
enough to see large scale changes in the line profiles. With a 0.5m
telescope the exposures will need to be VERY long to be useful.

Of the 6 stars, 5 were known WR+O binaries and showed major line profile
changes, associated with the wind-wind collision. The one that was not an
obvious binary (WR136) showed much less variability. Higher SNR would
probably help and enable us to see intrinsic wind variability, but the
exposure times will be extremely long (without a detector with a much
higher NIR efficiency). As has been mentioned before - that is the real
killer for this emission line. InSb arrays (or similar) may well become
more available for amateurs - indeed with a bit of digging you can have
an NIR photometer for 3000USD.

http://www.optecinc.com/astronomy/catalog/ssp/ssp4.htm

and maybe there is already something out there for NIR spectroscopy (but
it probably involves liquid nitrogen, which starts getting serious).

So, it would be interesting to see whether you can detect the line at
all with a standard CCD, but they will be challenging observations to be
useful (compared to 5696Angstrom, which is so much easier all round).

Cheers

ian





-------------------------------------------------------------------------
Dr. Ian Stevens
School of Physics & Astronomy Tel. +44 121 414 6450
University of Birmingham, Edgbaston Fax +44 121 414 3722
Birmingham B15 2TT, UK E-mail )star.sr.bham.ac.uk">irs(==>)star.sr.bham.ac.uk
-------------------------------------------------------------------------


On Sat, 13 Oct 2012, Alexandros Filothodoros wrote:


:
Hello everyone

With a custom made IR spectrograph and a modded CANON EOS 350D(filter removed),we managed to go beyond 10000 A.
Next try will have to be with an FLI CCD,but it will have to wait until the Australia eclipse.

What resolution,SNR and dispersion should we aim at? The same as in the optical domain?

------------------------
Everything is Mathematics !













--
Scanned by iCritical.

_________________
Tony Moffat

The effectiveness of the order filter will be critical too. With the CCD QE being perhaps as much as 50x lower at 10830A compared with at 5415A Any leakage will potentially cause big problems.

Robin

Hi Robin,

A quick review of commonly available (i.e. amateur) front illuminated CCDs suggests that, due to the increasing transparency of the silicon substrate at NIR wavelengths, the best option is still a camera based on the KAF-3200ME, such as the ST-10, STT-3200 or QSI-532, with a QE around 5% at 10,000A and not apparently dropping very fast above that. It looks like the SONY 'flavour of the month' ICX694ALG is only around 3%. These are certainly not very encouraging figures for NIR work.

I agree with your comments on the need for excellent 2nd order filtering.

Cheers, Bernard

_________________
Quicquid Nitet Notandum

Just a brief remark. Don't forget that we talk about the HeI 10830 line. This is not 10000. In this region the QE curve is dramatically dropping.

Thomas