Alpy 600
User Guide
(olivier.thizy@shelyak.com)
(francois.cochard@shelyak.com)
Olivier Thizy
François Cochard
DC0016B : feb.-2014
Alpy 600 spectroscope
User Guide
Olivier Thizy
(olivier.thizy@shelyak.com)
François Cochard
(francois.cochard@shelyak.com)
February 2014
Ref. DC0016 - rev B
Table of Contents
Introduction 4
1 Discover your Alpy 600 6
1.1 Out of the box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Technical specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Alpy 600 elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Mechanical interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 Visual observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.6 Alpy 600 principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.7 Different slits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 Alpy 600 - Spectroscopy in slit mode 14
2.1 Configuration for imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2 Color or black & white ? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3 Preview of your spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4 Reference images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.4.1 Offset image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4.2 Dark frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4.3 Flat field spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.4.4 calibration spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.4.5 Typical observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5 Data reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.1 Prepare master files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.2 First pass : raw profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.5.3 Second pass : wavelength
calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.5.4 Third pass : Response curve
correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3 Alpy 600 - Stellar spectroscopy in slitless mode 26
3.1 Changing the slit position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.2 Visual observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.3 Recording a star spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.3.1 Tuning the instrument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.3.2 Stars observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.4 data reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4.1 First pass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4.2 Second pass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4.3 Third pass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4 Star spectra in slit mode 33
5 Next steps 35
5.1 Improve data quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.1.1 Signal / Noise Ratio (SNR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.1.2 Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.1.3 Wavelength calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.1.4 Autoguiding (guiding module) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.1.5 Cosmetic file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2
TABLE OF CONTENTS
5.1.6 Instrumental Response curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.1.7 Write a log file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.2 Improve productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.2.1 Keep the same setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.2.2 Dark frames library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.2.3 Exposure time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.2.4 Prepare for your observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.2.5 One directory per observing session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.2.6 Reduce your data quickly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.3 Share your results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.4 Get pro-like quality spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6 Typical observing session 38
7 Appendix 39
7.1 Alpy Modules & Accessories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.1.1 Alpy Guiding module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.1.2 Alpy Calibration module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.1.3 DSLR camera barlow adapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
7.1.4 C-mount camera adapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
7.1.5 Optical fiber adapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
7.2 Alpy 600 with a DSLR camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3
Introduction
Welcome to the world
of spectroscopy !
Alpy is a modular product line, allowing you to
start in spectroscopy in a very easy way, and progress
up to producing high quality spectra on astronomical
objects. Alpy 600 is the main module in this family.
It has several operating modes. You can use it visually, on different light sources as well as on stars with
a telescope. You can also take pictures to share what
you see, and make scientific measurements : you’ll
be surprised how far you can go in astrophysics with
such a basic instrument. For instance, you’ll be able
to measure the temperature of a star, define its spectral class, measure the expansion velocity of a nova,
observe the emission line of a Be star1, measure the
expansion of the universe, and so on.
The Alpy 600 design meets several objectives:
– To offer an instrument which is easy to start with
and fun to use, but which is also powerful enough
for people who want to go deeper into the science.
There is no compromise on the spectrum quality:
the optics are specially designed for this spectroscope.
– To provide an affordable and scalable instrument.
Several additional modules are available. You can
use the Alpy 600 module alone to produce high
quality spectra, but adding the guiding and/or calibration modules will make your astronomical observations much easier and more productive.
– To be easily adaptable to your own setup (tele-
scope, camera...). the mechanical interfaces are
standard, and allow many different usages.
– In spectroscopy, there are different kinds of in-
strument : slitless, with slit, or with fiber optics.
All of them have advantages and drawbacks: It is
the type of observation which defines the optimal
configuration. The Alpy 600 allows all of them!
Then, you can decide for yourself what is best in
your case.
to spectroscopy, but that you’re already familiar with
the basics of astrophotography. Specifically, we’ll not
cover here the following points:
– Setting up and tuning your telescope. You must be
able to point and track a star. Ideally, you should
have autoguiding (with external parallel optics,
off-axis or with the Alpy guiding module).
– Installing, starting and using most common soft-
ware. You should have a PC available, running under Windows (we use MS Windows 7 in this document).
– Astronomical imaging. You should know what is
a dark, an offset, a flat field and classical image
processing.
If you’ve already made some deep sky images, with long exposure and good process-
ing, moving to spectroscopy will be easy for you
- a new adventure which gives access to a new
dimension.
Recommended learning path
Here is a proposed learning path to guide you step
by step through the discovery of your Alpy 600:
– First, use the Alpy 600 in slit mode (as it is con-
figured in our factory), during daylight.
– start by looking at some lamps and daylight spec-
tra - the Sun spectrum, in fact.
– record images and reduce data.
– Secondly, change to slitless configuration, and in-
stall the Alpy 600 on your telescope.
– start first with visual observation of some bright
stars,
– then record & process star spectra images.
– Lastly, take star and extended object spectra in slit
mode.
This document follows this path.
Prerequisites
Spectroscopy is a very large subject. Some parts are
very easy. For instance, looking visually at a spectrum,
during daylight or with a star, requires no experience.
However, if you want to take images or make scientific measurements, it is highly recommended to have
some experience with your instrument - telescope and
camera. In this document, we assume that you’re new
4
Software
The Alpy 600 is shipped with a CD-ROM containing
all the software tools you need to observe and process
your data. These software are free - you can find their
latest versions on the web. Of course, you can use
your own software if you prefer. In this document,
1. http://arasbeam.free.fr/
we use AudeLA2for acquisition and ISIS3for data
reduction.
Equipment configuration
The Alpy 600 can be used in many configurations,
but we cannot document all of them ! This documentation is based on a specific configuration (which fits
Alpy requirements perfectly). It will be easy for you
to adapt the instructions to your own equipment &
software.
– A Goto telescope, able to point and track a star in
autoguiding mode.
– A C8 telescope with F/6.3 reducer.
– A CCD camera Atik 314L+. It has a 8.9 x 6.7mm
CCD chip, with 6.45µm pixels. It covers the full
Alpy 600 spectrum.
– A PC running under Windows 7.
A CCD camera is the optimal choice for as-
tronomical observations. However, Alpy 600
can also be used with a DSLR camera. Because
the backfocus (distance between camera mount
and sensor chip) of a DSLR camera is much
higher than in a CCD camera, we’ve designed
an optical adapter for DSLR cameras. Refer to
Appendix for accessory descriptions.
However, we know that you’ll invent new ones. We’ll
be very happy to take your experiences into account
to continuously improve this product : do not hesitate
to contact us if you have any comments !
We invite you to join the growing Shelyak Instruments users and amateur spectroscopist community
on the Spectro-L Yahoo group5and the Aras forum
to share your own experiences and ask questions to
the community. We are really interested to see your
results there.
It is now time to enter in this wonderful world...
Are you ready ? Enjoy Spectroscopy !
Olivier Thizy
7
François Cochard
6
8
Alpy modules
Alpy is a scalable instrument. This a family of modules, to best fit your own needs & observations. These
are the main modules :
– The Alpy 600 is the main module. It is the spec-
troscope itself. It can be used in different configurations, to cover most applications in astronomy
& the lab.
– The Guiding module : For astronomical applica-
tions, it is difficult to ensure that all the light from
the object you’re observing goes into the spectroscope. This module will help you to observe & control the spectroscope entrance.
– The Calibration module: For accurate results, a
spectrum must be properly calibrated, in wavelength and intensity. The calibration module simplifies this operation by providing both a calibration and white light. It can be used manually or
remotely.
– Accessories : DSLR camera adapter, C-mount cam-
era adapter, fiber optics connector.
Astronomical sectroscopy is a never ending story. We
have thought about a lot of applications (you can find
some ideas on the Shelyak Instruments website4).
2. http://www.audela.org/
3. http://www.astrosurf.com/buil/isis/isis.htm
4. http://www.shelyak.com
5. http://groups.yahoo.com/group/spectro-l/
6. http://www.spectro-aras.com/forum/
7. olivier.thizy@shelyak.com
8. francois.cochard@shelyak.com
5
1
Discover your Alpy 600
1.1 Out of the box
When you receive your Alpy 600, it is assembled
in slit configuration, ready for mounting on a telescope and for imaging with a CCD camera. The package contains the Alpy 600, as well as a tool set and a
CD.
1.2 Technical specifications
The Alpy 600 specifications are given in table 1.1,
and dimensions are given in figure 1.1. In addition,
the table 1.2 gives the wavelength covered depending
on the CCD length.
CCD Sensor CCD length λ min λ max
4 mm 4560Å 6055 Å
5 mm 4293Å 6322 Å
1/2” 6.4 mm 3920 Å 6695 Å
KAF400 6.9 mm 3790 Å 6825 Å
9 mm 3650 Å 7380 Å
10 mm 3650 Å 7656 Å
11 mm 3650 Å 7923 Å
12 mm 3650 Å 8190 Å
ICX694 12.5 mm 3650 Å 8319 Å
13 mm 3650 Å 8457 Å
KAF1600 13.8 mm 3650 Å 8500 Å
15 mm 3650 Å 8500 Å
16 mm 3650 Å 8500 Å
Table 1.2: Covered bandwidth vs CCD size. Note that
above 7500 Å, the 2ndorder may recover the 1storder.
1.3 Alpy 600 elements
Remove the external body from the big ring:
6
Look at the slit, at the end of the spectroscope:
CHAPTER 1. DISCOVER YOUR ALPY 600
Description Value Unit Comment
Dimension Alpy 600 mm See drawing below
Weight 200 g with 1,25” nosepiece & without CCD camera
Resolution Power (R) @ 650nm ~600 - with 25µm slit
Resolution Power (R) @ 450nm ~400 - with 25µm slit
Spectral domain 370 - 750 nm
CCD min length 8.5 mm if smaller, only a partial spectrum is visible
Optical dispersing element Grism - grating + prism
Grating density 600 l/mm
Typical spectral dispersion 480 Å/mm
Input beam F-ratio F/4 - Telescope beam can be from F/4 to F/10
(optimal luminosity performance is at F/5)
Slit width (standard) 25 µm also included : hole 3mm, slit 300µm,
100µm 50µm, hole 25µm
Slit length 3 mm except for hole 25µm
Mechanical interface M42 x 0.75
thread
Table 1.1: Alpy 600 specifications
- Standard T-mount (or T2)
Figure 1.1: Alpy 600 dimensions (in mm)
7
Alpy 600 - user guide
Loosen the three small screws around the spectro-
scope body, and remove the Alpy 600 core element :
As far as possible, standard threads have
been used on the Alpy 600 parts, to allow
you maximum flexibility with your own equipment. For instance, the external body is made
with standard M42x0,75mm threads : you can
use a 2” eyepiece holder instead of the 1,25”
one.
Several slits are available. We’ll see in the next chapters how to change it and under which conditions to
use each of them.
You have now a view of all the elements of the Alpy
600: Core element, Objective lens, external body and
nosepiece 1,25”.
1.4 Mechanical interfaces
The Alpy 600 uses several standard threads for mechanical interfaces, to adapt the instrument to your
own setup.
The CCD camera interface is a standard T-mount
thread (M42 x 0.75mm)
The same thread is used on both ends of the external body. For instance, you can replace the 1,25” nosepiece by a 2”, using a common 2” to T-mount adapter.
8
The thread around the slit is of SM1 type: external
diameter 1,035" (26,29mm), 40 threads/inch (pitch
= 0.635mm).
CHAPTER 1. DISCOVER YOUR ALPY 600
You’ll see your first spectrum. Look at different sources,
to see the different spectra types :
A tungsten lamp has a continuous spectrum:
A power saving lamp has several bright emission
lines and some bands:
The standard 1,25” nosepiece includes a standard
filter thread (M28.5 x 0.6mm): you can easily measure the response curve of any filter mounted in a
standard ring.
1.5 Visual observation
Now, take the core element and look at any light
source: classical lamp, or spectral lamp, or even the
sky, cloudy or not (do not look directly at the sun, it
is far too bright).
The Sun shows a continuous spectrum, with a huge
number of Fraunhofer absorption lines:
This shows that each source of light has it own
spectrum, and the spectrum reveals a lot of scentific
information about its origin.
This will be the same for stars and any astronomical
object : each one has its own spectrum, which reveals
many physical parameters of the star.
Do not look directly at the Sun, with or even
without a telescope !
9
Alpy 600 - user guide
This document follows the standard con-
vention for presenting astronomical spectra.
The blue end of the spectrum corresponds to
the highest frequencies, or lower wavelength.
The red end is for lower frequencies or higher
wavelength. Usually, in Astronomy, we show the
spectra from lower to higher wavelength. From
now on, get into the habit of putting blue on
the left, and red on the right.
1.6 Alpy 600 principle
The principle of the Alpy 600 is as follows. A point
light source is sent to a collimator lens which makes
the beam parallel. This parallel beam goes through
a dispersing element (usually a prism or a grating):
all outcoming rays of the same wavelength (or color)
go in the same direction, but the output angle differs
for other wavelengths. Then, an objective lens collects
all these beams and focuses them on the focal plane,
where we put the sensor.
In Astronomy, we’re lucky : stars themselve are pointlike. So, when looking at stars, we can afford to remove the slit:
The Alpy 600 is designed in a modular way. The
core element consists of the the slit, the collimator
lens, and a grism. A grism is a combination of a grating and a prism. The grating spreads out the light, but
also deviates the beam by a significant angle. Using a
prism brings the beam back onto the axis of the input
light. In this way we have an on-axis instrument.
To make the source point-like, we usually use a narrow slit: only the light which goes through the slit is
spread out. The narrower the slit, the higher resolution the spectrum is, but the less light goes through:
there is a compromise to find for each case.
10
The light coming out the core element is then a set
of parallel beams. This is exactly what the eye needs :
a parallel beam is comparable to a beam coming from
infinity. Then, if you put your eye behind the core element, you’ll see the spectrum.
If you want to take an image with a camera, you
need to add the objective lens, to focus the parallel
beams onto the CCD sensor. This is built into the Alpy
600 body:
On the next picture, you can see the objective lens,
part of the Alpy 600 body:
CHAPTER 1. DISCOVER YOUR ALPY 600
The eye contains its own objective lens - the
crystalline lens. Then, it accomodates any
beam coming from the infinite !
1.7 Different slits
The Alpy 600 is a versatile instrument. As described
in section1.6, a spectroscope requires a point light
source to properly work - the smaller the source, the
better is the resolution. Alpy 600 contains several slits,
to fit to several usages. Table 1.3 shows some of them.
To work with an extended source (which is the case
in most everyday situations), use the standard slit 25µm wide. It is optimized for visual usage, and for
bench measurements. The resulting spectrum is a wide
band (top line of table 1.3).
In some conditions, it is better to only have a small
hole : this way, the spectrum is always at the same
position in the image - but it is more difficult to see
visually. To a certain extent, this is a way to reproduce
in the lab what will be seen with stars (second line).
There is another useful case : working in slitess
mode. This is applicable to stars, which are point-like.
The advantage is that you’ll see a bigger star field, and
you’ll be able to find easily the star you’re looking for.
To do that, you can totally remove the slit from the
Alpy 600, but it is better to keep in position the “3mm
hole” : this way, you’ll limit the optical aberrations at
the field edges - and you’ll not lose the slit (third line).
Sometimes a wide slit is useful. For instance, when
you have no guiding module but want to eliminate
most of the sky background, or when you want to
collect the full star light if the star image is bigger
than 25µm. In this case, you can use the 300µm slit
(last line).
The Alpy 600 also includes 50µm and 100µm slits.
Changing the slit position
The slit position is easy to change. It is mounted
with two screws. Remove the outer one, and just loosen
the central one. Choose the slit you want, and put its
notch in front of the outer screw’s hole. Put back the
11
Alpy 600 - user guide
Standard slit : 25µm
Hole 25µm
Hole 3mm
12
Wide slit 300µm
Table 1.3: Alpy 600 slits
outer screw, and tighten carefully the two screws.
CHAPTER 1. DISCOVER YOUR ALPY 600
The slit is a fragile component - take extra
care when changing its position, and make
sure that you can move it freely, before turning
it.
Cleaning the slit
Over the time, there can be some dust on the slit. A
dust particle in the slit is seen in the image as a dark
line along the dispersion axis. You can clean-up the
slit with a soft, dry cloth, or gently blow away loose
particles.
13
Alpy 600 - Spectroscopy in slit mode
In the previous chapter, you’ve looked at lamp spectra using the Alpy 600. Now, you probably want to
take a picture of what you’ve seen, either to share
it with others, or to make measurements. This is the
purpose of this chapter. Here you will work in slit
mode, with everyday light, working on a bench.
2.1 Configuration for imaging
As explained in section 1.6, compared to visual observation, we need to add more optics behind the
grating, to create an image at the focal plane (the
CCD surface). These optics are contained in the Alpy
600 body:
2
Assemble the CCD camera adapter to your camera:
The body is threaded, for tuning the distance between the optics and CCD when it is mounted in the
CCD camera adapter. (The nominal distance is about
19,5mm) . A threaded ring is used to lock the body in
the focused position.
Watch out that the optics do not touch or
damage your camera sensor.
14
The camera used here has a T-mount
(M42x0,75mm thread). If you are using a
C-mount camera, an optional C-mount adapter
is available (see appendix 7.1.4).
CHAPTER 2. ALPY 600 - SPECTROSCOPY IN SLIT MODE
Put the Alpy 600 core element in the body. It can turn
freely : this will be used to put the spectrum horizontally.
There are three small screws around the body, to
lock it in the right position:
The first point to check is that the image is not saturated. The CCD camera has a capability of recording
a certain level of light (this is the dynamic range specification). If the source exceeds this level, the camera
considers it is like a “maximum light”, and cannot see
any details in it. The Atik 314L+ camera has a dynamic range of 65535 ADU. If a pixel in the image
is above this value, it means that image is saturated.
To solve this problem, you must reduce the exposure
time, or reduce the source intensity if it is possible.
At the opposite end, if the maximum level of the
image is at the bottom end of the dynamic range of
the camera, you lose the capability to see details in
your images. You can then increase the exposure time.
The optimal level is about 80% of the camera max
level.
If the image is totally black or white, it can
be due to the image visualization threshold
settings in the software. Move the cursor in the
image and check the pixel level (bottom right).
Or change the thresholds by clicking on “Auto”
button (bottom left).
Now, put the camera in a stable position, and point
the slit towards a daylight source (sky or window).
Start your acquisition software. Setup the CCD camera cooling at -5°C for instance. Take a first image of
the spectrum. You may have something like this:
Even if the weather is cloudy, the daylight is
always the Sun light. You don’t need to wait
for good weather when taking a solar spectrum.
The spectrum is probably unfocused, and in any direction. To put it horizontally, simply rotate the Alpy
600 core element in the body:
15
Alpy 600 - user guide
To focus the spectrum, turn the body in the CCD
camera adapter. Proceed step by step, taking continuous images.
When you turn the body in the CCD camera
adapter, you change the spectrum focus, but
also its orientation. To see the focus improvement, it is better to keep the spectrum with the
same orientation. To do that, fix the the CCD
camera firmly (we often use a small hand-vice),
take the Alpy 600 core element in one hand
to keep it in place, and turn the body with the
other hand.
The spectrum must be orientated with blue on the
left, and red on the right... It is not obvious to see in
a black & white image ! With experience, you will immediately recognize the Sun spectrum. Refer to this
image to put it in the right position:
When the spectrum image is correct, tighten the threaded
ring around the body, and the three screws to lock the
Alpy 600 core element.
You should have something like this:
You’ve completed the most complex part of spectroscopy with Alpy !
At this stage, your camera is equiped with the spectroscope, and any image you take will be a focused
and horizontal spectrum of the light coming into the
slit.
When The Alpy 600 is focused, you can
remove the core element and put it back
quickly : the focus is not lost. You’ll have only
to tune the rotation angle, to put the spectrum
horizontal in the image.
Test now different kinds of lamps: tungsten, power
saving lamps, neon lamp, Alpy calibration lamp... each
source has its own spectrum. The table 2.1 shows
some examples.
For the next step, record at least one image of the
Sun spectrum in FITS format.
FITS format is the image file format for As-
tronomy. This is a format which keeps the
whole information in the image (no compression), and adds some key data in a file header,
such as date, image size, observer, observing
site, object coordinates, and so on.
16
Description Images
Sun light
Halogen lamp (flat field lamp of the Alpy
calibration module)
Candle light
Argon / Neon (Alpy calibration module)
CHAPTER 2. ALPY 600 - SPECTROSCOPY IN SLIT MODE
Neon lamp
Argon (Habitat lamp)
White LED
Power saving lamp
Blue paper, illuminated by a halogen lamp
Red paper, illuminated by a halogen lamp
Table 2.1: Some examples of common light sources. The color version of each spectrum is colorized by software.
17
Alpy 600 - user guide
2.2 Color or black & white ?
When you look visually at the spectrum, you see a
rainbow, with beautiful colors. Using most CCD cameras, you get only a black & white image. This is normal, because the CCD camera is not color-sensitive.
If you had used a color camera (or a DSLR camera),
you would have a colored spectrum like this one:
Go in the tab “Image”. Select your Sun image (Sun-
1.fit), and display it :
You might think it is better to have the color... it
depends ! If this is only for sharing what a spectrum
is, this is true. But if you want to make measurements
on your spectrum, a black & white camera is preferable : its surface is uniformly sensitive, and there is
no effect of the colored pixels (Bayer matrix effect).
Anyway, keep in mind that color or black & white
only depends on the camera : the spectroscope itself
works in color.
2.3 Preview of your spectrum
You’ve recorded your first spectrum image. That’s
great, but not very useful in this format. Usually, a
spectrum is presented as a profile, giving the light
intensity vs the wavelength. To make the conversion
from image to profile, you must use spectral data reduction software. Alpy 600 is provided with ISIS software, which is one recommended tool. This document isn’t a tutorial of ISIS (refer to the ISIS documentation1, and specially the Alpy tutorial2), but
shows very quickly the basics for this first experience.
Here is how to proceed with your Sun spectrum.
Run ISIS, ang go to the "Settings" tab. Select "ALPY
600 (without calibration module)" in the spectrograph
model, and specify the working directory where you’ve
storedyour images (FITS files):
If you don’t see the spectrum, it is probably because
it is outside of the displayed zone. You can move in
the image using the scrollbars
Select the checkbutton “Reticule”.
It will display the area of the image which will be
selected for profile extraction. If the area is not in the
spectrum, double click in the middle of the spectrum
: it will put the reticule there.
Then, click the button “Next”; ISIS switches to the
“2.General” tab, and updates automatically several fields.
Put “Sun” in the object name. In the options area, select the “don’t remove background” checkbutton. This
is to tell ISIS that your spectrum is from an extended
object (the spectrum is not a single line in the image). Unselect “wavelength calibration”, because at
this stage, we’ve no data for calibration.
18
1. http://www.astrosurf.com/buil/isis/isis.htm
2. http://www.astrosurf.com/buil/isis/guide_alpy/tuto_en.htm
CHAPTER 2. ALPY 600 - SPECTROSCOPY IN SLIT MODE
Click on the "Next" button. Isis switches to the "3.
Calibration" tab. Check that “Tilt angle”, “Smile Y”, “X
coordinate of line at wavelength” and “A =” have the
value zero. Put 10.000 in the “Radius” value. Uncheck
the “Auto” after “vertical coordinate”:
Click on the "Next" button. Isis switches to the "4.
Go” tab. Click on the GO button; ISIS runs the data
reduction:
This is your first Sun spectrum profile ! There are a
lot of absorption lines - each of them is the signature
of a chemical element in the external layers of the
Sun’s atmosphere (photosphere).
However, at this stage, your profile is still not of a
high scientific interest. For three reasons, at least :
–The 2D image has not been pre-processed,
–The spectrum is not calibrated in wavelength,
–The profile is not corrected for the Instrumental
Response.
In the next sections, you’ll learn how to proceed to
make these corrections and calibrations.
2.4 Reference images
Digital imaging is affected by several instrumental
effects, such as offset, thermal signal, read-out noise,
hot pixels, and so on. These effects are quite easy
to correct, using reference images : offset, dark, flat
fields.
The pre-processing can be highly automated, once
you have made the proper reference images. These
images are “offset”, “dark frames”, “flat field” and “calibration”.
It is always better to take a series of images instead
of a single image. From the series, you can make either an average or a median. This way, you’ll be able
to eliminate any accidental events (such as cosmic
rays in images), and improve the image quality. ISIS
software can manage series of images for all type of
reference images.
After a few seconds, the process is done. Click on
“Display profile”. You should have this kind of image:
19
Alpy 600 - user guide
2.4.1 Offset image
To avoid negative signal digital conversion, CCD
cameras have a fixed offset on the analog signal which
is called Offset. Any data readout from the CCD chip
includes this offset.
Offset images are also called Bias in some
documents. In this User Guide, we made the
choice to use Offset, but consider it is equivalent
to Bias.
This offset is fairly stable and once you have a master offset frame, you can keep it for long time. You
can make one again once per year approximately. It is
also not too much temperature dependant. Because it
is very stable, you can improve your data by taking a
long series of offset frame (around 100 exposures)
and perform a median average to create a smooth
master offset frame for your camera. A median average is better than a simple average as it will totally
remove out of the ordinary defects on some of your
images, such as cosmic rays for example.
To get an offset image, put a cap on the Alpy 600
light entrance to ensure that no light comes in. Take
an exposure of 0 second, binning 1x1.
Offset images are sometime called Bias im-
ages.
You can measure the average level in the image (draw
a rectangle, and right-click to get the statistics in the
window):
2.4.2 Dark frames
CCD chips have a tendency to build up electrons
over time, particularly at high temperature. This temperature dependant effect is called « thermal noise ».
The thermal noise is included in dark frames which
are exposures taken with the camera shutter closed;
the offset is also included in dark frames.
Hot pixels are individual cells that are particularly
sensitive to this thermal noise and tend to saturate
quickly during long exposure. These hot pixels have
to be processed separately (usually replaced by the
average of the surrounding pixels). An interesting property of thermal noise is that it is linear with exposure
time. For example, the thermal noise in a 2sec exposure is twice the thermal noise in a 1 sec exposure.
This means that you can extrapolate any shorter thermal noise frame from a master thermal noise frame
with long exposure – assuming you have mapped the
hot pixels properly. This is done by doing a median average of a long exposure dark frame series (at least 7
frames, if possible more). Most astronomical software
require both offset and dark frames to pre-process
your image. The thermal noise is of course embeded
in the dark frame. Be very careful as the thermal noise
(thus the dark frame) is very sensitive to temperature.
Make sure to use a dark frame taken at the same temperature as your spectra. If your camera doesn’t have
temperature regulation, you should take dark frames
close to the time you expose your spectra. Some software can still extrapolate a proper thermal noise map
based on the area of your chip with no signal, which
is very easy in the case of Alpy 600 spectra. You do not
have to rebuild your dark frame every night; you can
create your library of dark frames taken at different
temperatures and use it for several months.
A dark frame is taken with the same conditions
as the target spectrum (exposure time, temperature),
but with no light coming into the spectroscope.
For instance, if your star spectra are made with
300s exposure, dark frames must be taken with the
same duration.
Here is a typical dark frame of 300s (binnng 1x1,
of course):
20
2.4.3 Flat field spectrum
A flat field spectrum is obtained by illuminating
the spectroscope with a continuous spectrum of white
light. Ideally, it should be a lamp that has the same
energy level in each wavelength. It doesn’t exist in
real life... As a compromise, we use tungsten or halogen lamps (but not power saving, calibration lamps,
or sunlight !). The exposure time must be choosen to
give 80% ot the max level of the camera.
This image is mainly used to determine low frequency non uniformities, caused by dust on the CCD
front glass for example or overall instrumentation vigneting. Make sure to take a series (at least 5 exposures) of flats. The signal has to be as high as possible
without any saturation.
CHAPTER 2. ALPY 600 - SPECTROSCOPY IN SLIT MODE
2.4.5 Typical observation
To get all the data you need for a high quality observation, here is a suggested list of all the images
that you should have before switching off the equipment. This is applicable for your Sun observation, but
is also valid for night observations:
– 7 offset images (or more)
– 7 dark images (or more)
– 7 flat field spectrum images (or more)
– 1 calibration spectrum image
– 5 reference star spectrum images (or more)
– and of course all the images of your object !
For each image, you must check that there is no saturation.
The Alpy calibration module contains a halogen lamp which gives a perfect flat field spec-
trum.
2.4.4 calibration spectrum
As explained in section 2.5.3, you will process your
Sun spectrum using the calibration information contained within itself. But you can also use a calibration
lamp - if you are working in slit mode. For instance,
below is a neon spectrum
The Alpy calibration module contains a cal-
ibration lamp made of neon, argon and hydrogen which makes it very easy to calibrate in
slit mode.
A scientific measurement on an astronomi-
cal object spectrum requires not only a single spectrum of this object, but a whole set of
images. If you make sure you have all these images during an observing session, data reduction
will be easy and fast.
2.5 Data reduction
You’ve now acquired all images for a full data reduction of the Sun observation. In this section, you
will see how to proceed with ISIS software. You will
proceed with these images as if they were from any
star.
2.5.1 Prepare master files
This step is to convert the series of reference images into single images, called master files. It is done
very quickly in ISIS. Go to the “Masters” tab, and select each series of images (offset, dark, flat), to create
a master. Click on “Go” for each series:
21
Alpy 600 - user guide
Also create a cosmetic file, which records all the hot
pixels in your camera. Hot pixels are pixels that react
differently from the others; we consider they are defective and the processing will replace their value by
the average of the surroundig pixels. The cosmetic file
is made from the dark image (set up the threshold in
such a way that you get around 200-300 hot pixels this is a reasonable value):
The profile is significantly different from the initial
one made with no reference images at all. This difference is mainly due to the flat field correction.
2.5.3 Second pass : wavelength
calibration
You now have to establish the calibration law (the
relationship between pixel position and wavelength),
to display the spectrum in wavelength units. The more
lines you have, the more precise is the calibration.
The dispersion law is usually a 3rd order polynomial.
To reduce the Sun spectrum, you will proceed in a
special mode : in the Sun spectrum itself, there is a list
of known absorption lines that can be used for the calibration. It is not the simplest way (some lines can be
quite difficult to identify), but this is a reference spectrum that you can use without any additional measurements. Of course, if you have access to a calibration lamp, you can use it in preference.
2.5.2 First pass : raw profile
The first pass is very similar to your first spectrum
preview (section 2.3), but you can now include the
master files for pre-processing. In the “2. General”
tab, select offset, dark, flat and cosmetic master files:
Then, switch to the “4. Go” tab, and click on Go.
Display the resulting profile:
wavelength Line
3933 Å Calcium K line
3968 Å Calcium H line
4307 Å G band (molecular)
4861 Å Hβ (balmer line)
5167 Å Magnesium triplet
5270 Å Iron (Fe) line
5893 Å Sodium doublet
6563 Å Hα (balmer line)
Click on the button “Dispersion”, in the right side
of the screen. The dispersion panel appears. Fill in all
22
the wavelengths above, select “absorption line” and
click on the first radio button (matching 3933 Å):
Now, double click in the profile, on each side of the
absorption line at 6563A. This will measure precisely
the center of the line (in pixels). The next line will be
automatically selected.
CHAPTER 2. ALPY 600 - SPECTROSCOPY IN SLIT MODE
In the bottom area, you have the RMS error resulting from the calibration. It should be around few
Å(here, the error is of 3,7 Å)
Now re-run the reduction process, but selecting the
“Spectral calibration” button in the “2. General” tab.
You also have to select “Predefined dispersion equation” in the “spectral calibration” panel:
Repeat the same operation for each recorded line.
When it is done, select “3rd order” option, and click
on button “Compute polynomial”. Based on the lines
you’ve selected, it calculates the dispersion polynomial law.
Display the spectrum: you can see that it is now cal-
ibrated in wavelength:
The calibration law can be established from
any spectrum containing several known features. It is highly recommended to use a calibration light to do this when working in slit mode:
the more lines you can identify, the more precise
the calibration law will be.
23
Alpy 600 - user guide
2.5.4 Third pass : Response curve
correction
The whole instrument, including telescope (if any),
spectroscope, and CCD camera, has its own response
to the light. It is called Instrumental Resonse Curve.
For instance, a CCD is much more sensitive to red
than to blue light. If you want to get the true spectral profile of the source (the Sun in your case), you
must correct for this very important effect. The best
way to do this is to record the spectrum of a known
star (a reference star), and compare it to its “theoretical” spectrum found in the literature.
In the current case, you’re observing the Sun... and
the Sun is a very well known G2V type star ! So, we’ll
make the correction from a G2V spectrum.
When you observe astronomical targets, it is necessary to observe a reference star, generally a hot star
(A-type, with few absorption lines), close to your own
target to prevent atmospheric effects.
A reference star is required for both calibration and
response curve correction. Of course, the same star
can be used for both needs.
Click on the "Display" button for reloading the sun
profile. Then click on the “Response” button. A response panel appears : select the G2V profile as reference. Click on the “Response” button (red).
In the main window, you now have 3 profiles: your
Sun spectrum, the G2V theoretical profile, and the
division of both - which is indeed the raw response
curve.
Each observing session must start by observing a known hot reference star, to allow the
correction of any instrumental effect.
ISIS includes a database of spectra recorded at different resolutions. Here is how to get a G2V star reference spectrum. In the “4. Profile” tab, click on the
database button. The database panel appears:
At the top left, you can select an G2V star from the
Pickles list:
Click on the OK button: only the response curve remains. As you can see, this profile is very noisy, and
there are some “emission lines”. This is a side effect
of the division between the profiles, because the Alpy
600 and reference profiles do not have the same resolution.
To get the effective response curve, you need to remove these lines, and smooth the profile. This can be
done with the “continuum button”:
This profile is different from yours. This is due to
the Instrumental Response curve and the wavelength
range covered.
24
To remove the emission lines, double click on both
sides of each line. The line is replaced by a straight
line:
CHAPTER 2. ALPY 600 - SPECTROSCOPY IN SLIT MODE
Now, move the cursor in the continuum panel to
smooth the profile, and remove all the noise in the
profile:
Click on OK, and save this response curve.
Now go back to the “2. General” tab, and fill in the
“response curve” field.
You can see that the result is excellent - much of the
detail that we could suspect to be noise is in fact the
sunlight signal.
Summary
The spectrum of the Sun is now fully processed.
To do it, you’ve run the data reduction process three
times: the first time to get the raw profile, the second one to include the wavelength calibration, and
the last one to correct for the instrumental response
curve. This process needs to be done only once per observing session. When you observe several objects in
the same night, you can reuse the calibration and the
response curve - to the first order, you can consider
that it only depends on the instrument, and not on
the observing conditions (this may be a little bit more
complex if you need a very accurate measurement).
We invite you to observe many spectra with this
setup and run the data reduction process for each of
them. For a new object, the only point to change is the
input file and the source name in the tab “2. General”
- and run the process in the “4. Go” tab. It takes only
a few seconds.
You can then re-run, for the third and last time:
This is the actual spectrum of the Sun !
You can compare it with the G2V theoretical spec-
trum (click on the “compare” button):
25
3
Alpy 600 - Stellar spectroscopy in slitless mode
In the previous chapter you learned how to produce
a basic spectrum. You know what images are needed
and how to process them to get the actual signal contained in the light you observed. You are now ready
to go to the telescope and to apply the same method
on astronomical objects.
So far, you’ve worked with the Alpy 600 in slit mode.
Ideally, you should do the same with the stars. But
there is a problem: when using a slit spectroscope,
the only field you get is the slit. With an extended
source (daylight, calibration lamp...) it’s easy to aim
for the source. But stars are point-like and it’s not obvious how to put it in the slit (If the star is in the slit,
you get a spectrum, but if it is just beside the slit, you
see... nothing and when you see nothing, there is no
way to define in which direction you must move the
telescope).
>From a practical standpoint, this is not feasible,
at least when starting in spectroscopy. You’ll see in
chapter 4 how to deal with that. For now, switch to
slitless mode, applicable for point-like sources..
Slitless mode has advantages. For instance, it collects the full light of the star, whereas slit mode has
some loss at slit entry. This is useful for very faint objects, but it has also some drawbacks : there is no
way to use a calibration lamp (which is an extended
source), and the position of the spectrum will move
in the field, depending on your telescope tracking.
In a few words, using the Alpy 600 in slitless mode
is the way to work when you have no guiding device,
but this is not the simplest way to get accurate measurements. The good news is that when you become
familiar with this process, switching to a guided mode
will reveal how easy it is to do spectroscopy !
The observation itself must be done during
the night, but the preparation of the telescope and the tuning of the Alpy 600 with the
CCD camera has to be done during daylight.
3.1 Changing the slit position
You’ve seen in section 1.7 how to change the slit.
Here, you should select the 3mm hole (the “big” one):
Now, if you take an image of Sun light, you’ll see
a very rough spectrum, as if if it were strongly defocused. It is in fact focused, but the light source is so
big that the spectrum is unusable.
Why not totally remove the slit ? You could
do this but the size of the “high quality” field
of the Alpy 600 is limited (it is very good over
the whole 3mm hole, but decreases outside this
area), and... keeping the slit in place is the best
way to not damage or lose it.
3.2 Visual observation
To start with astronomical objects, we invite you to
look visually at a star spectrum. Remove the Alpy 600
core element from the body (loosen the three small
screws around the body):
26
CHAPTER 3. ALPY 600 - STELLAR SPECTROSCOPY IN SLITLESS MODE
This nosepiece can be put in a standard 1,25” eye-
piece holder:
Attach the Alpy 600 core element to the nosepiece
(remove it from the external body):
Select a bright star in the sky (preferably B or A
type), and point the telescope at it. Now look through
the Alpy 600 core element, and focus the telescope to
give a very thin spectrum. Turn the Alpy 600 in such a
way that the spectrum is horizontal, with blue on the
left and red on the right.
You should see something like this:
You can now look at different stars and bright objects.
3.3 Recording a star spectrum
After observing visually, you will certainly want to
record images. You’ve learned how to proceed with
spectrum acquisition and reduction, you’ve put the
spectroscope on your telescope: now, it’s time to do
both together !
3.3.1 Tuning the instrument
In section 2, you’ve seen how to tune your Alpy
600 (to produce a focused and horizontal spectrum
with the CCD camera). You must start from there and
switch to the slitless mode (change the slit to the
27
Alpy 600 - user guide
“3mm hole” position).
Ensure that spectrum is horizontal, with the right
orientation (blue on the left) :
Put the whole assembly into the eyepiece holder of
the telescope:
Assemble the external body and the eyepiece ring
on the Alpy 600:
The external body includes a standard
T-mount thread (M42x0,75mm). If your
telescope includes a 2” eyepiece holder, you can
remove the Alpy 600 eyepiece ring and replace
it by a T-mount to 2” adapter (optional).
Plug in the cables (power supply and USB), and balance the instrument.
Do not forget to set your CCD temperature
control.
We recommend setting up an empty direc-
tory on your PC to record all the images of
the night.
28
CHAPTER 3. ALPY 600 - STELLAR SPECTROSCOPY IN SLITLESS MODE
Select a star to observe. It must be bright, and you
should preferably look for a hot star (A or B type), because its spectrum has a clear continuum, with some
deep absorption lines. In the images below, we observed Regulus, which is a B7 type star.
When night falls, point at the star using the telescope finder. Take an image of the spectrum. Compared to the Sun spectrum you made previously, the
exposure time must be much higher for stars! If only
tenths of a second were enough for Sun, seconds or
minutes can be required for stars. If you’ve chosen a
bright star (magnitude 0 to 3), then only few seconds
is enough - with a small telescope. Anyway, always
control the exposure time to prevent saturation or low
levels.
You should see something like this:
Look at the left end of the spectrum : there are clear
absorption lines (only for hot stars):
If your image is black, it can be due to the im-
age visualization thresholds. In AudeLA, click
on the “auto” button (bottom left) to adapt the
thresholds to the image.
You can rotate the Alpy 600 in the eypiece holder : the
spectrum remains horizontal in the image. This is normal: the orientation of the spectrum only depends on
spectroscope position vs the CCD camera. However,
it does not mean that you should put the Alpy 600
in any position. If you want to make it easy to find
and center the spectrum in the image, it is better to
align the spectroscope axis to the telescope axis. Rotate Alpy 600 (the whole instrument, including CCD
camera and external body) in such a way that when
you move the telescope in Declination, the spectrum
moves along the vertical axis.
Your instrument is now ready for observation.
Move the telescope to position the spectrum in the
center of the image.
Your telescope is probably unfocused. Take spectra
continuously while focusing the telescope. Little by
little, you’ll see the spectrum get thiner and thiner.
When focused, you should have this (it must be only
few pixels wide):
In slitless mode, the width of the spectrum is
directly linked to the star image size in the
focus plane of the spectroscope (slit plane).
3.3.2 Stars observation
Take some time to acquire several images of the
star. Of course, if you make exposures of several seconds (which is more than probable), you must rely
on the telescope tracking. If the tracking is bad, the
spectrum will move in the image during the acquisition, and the result will be a band instead of a line:
As you did with the Sun, you can immediately process your first spectra with ISIS. The only difference
is that you now have a star spectrum (Regulus in
our case), and you should ask ISIS to remove the sky
background (in the “2. General” tab). You should get
29
Alpy 600 - user guide
a spectrum like this:
In astronomical imaging, the signal level (light level)
is very low and all the instrumental defects are proportionally more important. It is even more important therefore to acquire the whole set of reference
images, to ensure good data reduction.
In slitless mode, the flat field spectrum can be produced by putting a diffuser in front of the telescope
(see an example below), and illuminating it with a
tungsten or halogen lamp (during the night, to prevent any sunlight polluting the image).
Here is the list of the reference images you need to
take in slitless mode, to ensure a proper data reduction:
– 7 offset image or more (same as before)
– 7 dark images or more (same as before)
– 5 flat field images or more (take care of the right
exposure time to get 80% of the maximum level)
– 5 spectra of the reference star (A or B type) or
more
– Point at several stars (adapt the exposure time
for each star), and take 3 to 15 spectra of each of
them.
3.4 data reduction
Create master files for offset, dark, flat fields, and a
cosmetic file from the master dark,
As you learnt in chapter 2, reduce your data in
three steps to obtain the wavelength calibration and
instrumental response curve.
3.4.1 First pass
First run the data reduction without wavelength
calibration and response curve correction. You should
get this (note that ISIS can export your profiles as
PNG, using the GnuPlot graphics tool):
The result will be like this (in flats taken without a
slit, there is no evidence of the response curve. All the
details are smoothed out):
The wavelength calibration and response curve correction will both be extracted from a reference star
spectrum.
This profile is different from the first Regulus one
because here we used the flat field. It is closer to the
actual star spectrum, but is still strongly affected by
the instrumental response curve.
3.4.2 Second pass
The wavelength calibration law can be established
using the Balmer lines (hydrogen lines), which are
the most visible in the spectrum. They are spread out
all along the visible spectrum, like a geometrical series. Here is the list of Balmer lines:
30
wavelength Line
3835 Å Hη (eta)
3889 Å Hζ (dzeta)
3970 Å Hǫ (epsilon)
4102 Å Hδ (delta)
4340 Å Hγ (gamma)
CHAPTER 3. ALPY 600 - STELLAR SPECTROSCOPY IN SLITLESS MODE
4861 Å Hβ (beta)
6563 Å Hα (alpha)
These lines are located as below:
With this information, you can establish the calibration law - select a 4th degree polynomial (because
there are several available lines).
3.4.3 Third pass
The last operation is to establish the instrumental
response curve. In this document, we used Regulus
as reference star. This is a B7V. In the ISIS database
(Pickles), there is no B7V, but we can select a B8V the difference is not significant at our level. Display
this profile, and save it in the current directory:
The spectrum has a much wider range than yours:
it goes from 1.000 to 10.000 Å. Reload your spectrum,
and click on “Response”:
When the polynomial is available, re-run the data
reduction process, including the wavelength calibration. You will get the same profile, but now calibrated
in Å(see the scale in the bottom axis - note that the
spectrum has been cropped between 3650 and 7300
Å):
Click on OK, to get just the raw response curve. Remove the lines and smooth the curve to reduce the
noise:
31
Alpy 600 - user guide
This response curve is very embossed.
This is different from what you saw on the Sun,
without the flat correction, all the flat field details
were smoothed.
Save the profile, and re-run the data reduction process, including the response file name (in “2. General”
tab).
Now the result is only the star profile:
You have now finished the data reduction for your
reference star, and you now have the dispersion law
and the response curve - these data will be reused for
all the stars you observed during the night. For the
other stars, you need run the data reduction process
only once.
However, you must pay special attention to an important point. You’re working in slitless mode, and
there is no way to put the next star exactly in the
same position as your reference star spectrum. There
will necessarily be a shift between the reference star
and the new target. The dispersion law remains the
same, but it must be shifted to match the new star
position. To do it, you need to identify one line in the
new spectrum, and give its wavelength to ISIS.
For instance, here is a spectrum of 38 Lyn, another
hot star. In the 2D image, you can easily identify the
Balmer line at 4861 Å. Measure its position, and enter
it in the ISIS field at the “3. Calibration” tab:
Continue with the same process for all the stars you
observed during the night. Here is an other example,
mu Leo which is K2III:
and a last one, Betelgeuse (M2I type):
You may observe that there are some shifts
in the calibration law from one star to the
other, despite the realignement you made using
a known line. This is due to the fact that the dispersion law can be dependent on the spectrum
position in the image. To minimize this effect,
always put the spectrum exactly at the same position in the image.
Run the process, and get the fully processed spec-
trum:
32
You now know how to proceed with a complete and
precise data reduction. With experience, you’ll do this
process in a few minutes for the reference star, and in
a few seconds for all the stars of the night.
4
Star spectra in slit mode
In previous chapters, you’ve see how to use the Alpy
600 in slit mode (see chapter 2) and slitless mode
(see chapter3). Slitless mode allows you to find your
target more easily, and capture all the light, which is
important for faint objects.
But slit mode has several advantages:
– A better wavelength calibration (using calibration
lamps),
– Observation of extended objects (nebulae, comets,
galaxies...),
– Allows the sky background to be removed (useful
for faint objects or urban observing conditions),
– The resolution depends only on the slit, not on
observing conditions (tracking, seeing, telescope
focus...).
For these reasons, most observations are made in slit
mode. In this chapter, you’ll see how to work around
the problem of pointing and tracking a star in slit
mode, though
the best solution is definitely to add the Alpy guiding module to your Alpy 600. This is a device which
shows you the entrance of the slit: you can see the star
you are observing. You can use this image to move
the star onto the slit and track it. If you plan to make
regular observations of astronomical objects, this is a
required device - for comfort and efficiency, but also
for the quality of your measurements. It is described
in 7.1.1
However, there are some ways to observe in slit
mode, even without the guiding module. It requires
some experience - most astronomers like challenges !
There are two approaches:
1. Observing with the wide slit and a flip mirror,
2. Observing with the narrow 25µ slit.
The first one is really accessible. The second one is
clearly a challenge - but we did it ourseleves.
Observing with the wide slit
The Alpy 600 includes several slits (refer to section
1.7). One of them is the wide slit, 300µm wide. This is
a trade-off between the thin slit and the slitless mode.
Compared to slitless mode, the size of the slit means
that your spectrum is always at roughly the same position horizontally in the image, and reduces the uncertainty in wavelength calibration. Also, it reduces
the sky background effect, and gives more detail in
the flat field.
The difference with the thin slit is that the spectrum
resolution is still given by the star image size, whereas
the thin slit ensures that the resolution comes from
the instrument itself. Of course, because of the field
of view, it is much easier to center the star.
To help you in star pointing, we recommend using a
flip mirror (with a reticle eyepiece) between the telescope and the Alpy 600:
This optional device will help you to put the star in
a very precise position, enough to put it in the wide
slit at least.
Centering the star
Use the telescope finder to locate the star (start always with a bright one), and flip the mirror to see
the telescope field in the eyepiece. Put the star well
centered in the Field of View (FOV). Then, flip back
the mirror to let the starlight enter the spectroscope.
Take continuous images, until you can see a spectrum
in the image. If there is nothing, put all the thresholds at max sensitivity, and change the focus of the
telescope - the first time you install the flip mirror,
the telescope is probably very unfocused. When star
becomes focused, you should see a large spectrum
(band), which will become brighter and brighter. The
band is due to the fact that an unfocused star is like
33
Alpy 600 - user guide
an extended object. When you’re close to the optimal
focus, the spectrum may disappear : This is because
the star image is now small, and may be outside the
slit. In this case, slowly move the telescope across the
slit: you should soon recover the star. If not, turn back
the focus, to extend again the object size, until you recover its spectrum. Step by step, you will be able to
focus the telescope perfectly and put the star right in
the middle of the slit.
At this point, look through the flip-mirror’s eyepice,
and note the actual position of the star. This position
will have to be the same for the next observations. If
you put any star exactly at this position, you’ll have
its spectrum in the Alpy 600 camera each time.
Tracking the star
You’ve been able to put the star in the Alpy 600
slit... now, you have to keep it there during the whole
observation. This depends of course on the quality of
your mount, and if it is precisely aligned or not. You
need to adapt the exposure time to ensure that spectrum remains very thin, with no visual effects of tracking issues. The spectrum must be thin, and absorption
lines must be sharp.
If after some exposures, the spectrum disappears,
this is because the star moved out of the slit. You can
of course flip the mirror and look for it again, but in
practice, it is probable that the star always moves in
the same direction, due to a mount alignement error.
So to return the star to its original position, you can
just move the telescope in the right direction. You will
quickly find this direction and the level of required
corrections after some trials.
We may even suggest, in some conditions, to slightly
misalign the mount. This way, the star will slowly
move in the FOV, but always in the same direction.
Then, with experience, you will know in which direction you need to move the telescope when spectrum
level is decreasing.
compared with the wide slit. The process is exactly
the same as for the previous section, but the tolerance
for star positioning is more than ten times smaller
(300µm vs 25µm). However, this is really feasible we did it... even without a flip mirror (only a finder).
The main tip is to slightly unfocus the star. It becomes bigger in the focal plane (slit plane), and easier
to find. We recommend first working in slitless mode,
to be sure that telescope is close to focus. If the focus
plane is too far out, the star can be huge (even bigger
than FOV) and the light level very low.
Be careful : if you unfocus the telescope, only
a samll fraction of the light will enter the
spectroscope. This is applicable only for bright
stars.
Also, with some experience, and if your telescope is
precise enough, you can use your finder (or even better, an electronic finder) to put the star very close to
the slit .
Then, depending on the spectrum level, you can decide to refocus the telescope step by step. By doing
that, the star may go out of the slit, because it becomes smaller and smaller. Then, move the telescope
again until you recover it. You can make several iterations, until you juge that the star size and spectrum
level are compatible with your observing conditions.
Again, proceeding this way is more of a a fun challenge than a recommended observing method. But it
can show you the significant difference between slitless and slit spectra: you’ll quickly see that in most
cases, there is a much higher potential in slit spectroscopy.
Then, keep in mind that by using the Alpy guiding
unit, you’ll get the high quality of a slit spectroscope,
in a very easy way, because you can permanently see
the star you are observing in your guiding image.
Of course, if you have spent hours to align
your mount, this could be bad advice ! Only
misalign your mount if you consider there is no
such risk.
Pointing and tracking in wide slit mode with a flip
mirror requires some experience, but this is a real improvement compared to the slitless mode. You will
not be wasting your time if you do it !
Observing with the narrow (25µm) slit
This section is for people who like challenges ! It is
possible to observe in ”full slit mode” (using the 25µm
slit), but it requires still more accuracy and patience
34
5
Next steps
You have now passed all the steps to produce your
first reduced spectra. With experience, you will perform the acquisition and data reduction more and
more efficiently. In this section, we’ve put some tips
and tricks to improve your observations. Improved
observations means better data quality (up to professional quality level data), and better efficiency (or
productivity) in your observing procedures. Most of
the advice below assumes that you’re working in slit
mode, ie you have a guiding module.
5.1 Improve data quality
5.1.1 Signal / Noise Ratio (SNR)
Star light is faint... and you spread it out with your
spectroscope. At the end of the chain, only a few photons reach each pixel of your CCD camera. The whole
instrument (telescope, spectroscope, camera) limits
your ability to detect faint signals. The quality of your
spectrum is directly linked to the level of signal you
get on your CCD compared to the “noise” of the instrument, which is its limit of detection. This is what
we call the Signal / Noise Ratio (SNR).
Improving SNR can be done in two ways: reduce
the noise or increase the signal.
Reducing the noise is hard: it is limited by physics.
But we can do it to a certain extend. For instance,
by cooling down the camera, you reduce the internal
activity of the CCD, and reduce the noise.
The best way to improve SNR is then to increase
the Signal. It can be done by several ways:
– Increase exposure time (the SNR will be increased
as the square root of the time multiplication factor),
– Increase the telescope size (this is why all as-
tronomers are always looking for bigger telescopes),
– Ensure that most of the light is going through the
spectroscope. Star images and slits are very small
(a few µm), and it is very easy to lose a high pro-
portion of the light at the instrument.
The two main reasons for losing light at the spectroscope entrance are usually:
– Telescope focusing. If the focus is not optimal,
the object size will be bigger than the slit width,
and only some percentage of the object’s light will
enter in the instrument.
– Position of the star in the slit. When working in
slit mode, it is very easy to put the the star partly
outside the slit. The result is the same: you lose a
high percentage of your source.
The best way to improve this point is to check that
when the star crosses the slit (if any), it almost disappears from your guiding image. Carefully tune your
guiding camera exposure time to prevent any saturation.
If the star disappears from the guiding image
when crossing the slit, it means that most of
light is going through the spectroscope.
With the experience, you will know how much signal
your instrument will get for a given magnitude. You
should also compare your data with others with similar equipment, as an external reference (see section
5.3). If you see that this level is lower than is usual,
take some time to find the source of this loss.
To go further into SNR improvement, you can measure the total power of your spectrum, and tune the
telescope focusing to optimize this measurement.
5.1.2 Resolution
If you work in slit mode, the resolution of your
spectrum comes mainly from the spectroscope itself.
The resolving power of the Alpy 6001is approximately
R=600 around Hα. The ISIS software calculates the
actual resolving power as part of the data reduction.
Check that your actual result is close to this value.
The resolution is calculated from the width
of the calibration lines (∆λ = FWHM of calibration lines. The better the focus of the calibration lines, the better is the resolution.
1. Resolution power gives the ability of the spectroscope to see
details. It is calculated as λ/∆λ, where∆λ is the smallest visible
detail in the spectrum.
35
Alpy 600 - user guide
5.1.3 Wavelength calibration
Take special care when performing the wavelength
calibration. Normally, you should have an RMS smaller
than 0.5 Å. Take the time to record a calibration quite
often, and use this image for the data reduction. Always double check that there is no major error in the
calibration. The simplest way is to look at the positons
of the balmer lines, visible in most types of stars.
Also, remember that the calibration law can change
across the FOV. The key is to always put the spectrum
at the same position in the image. If you work in slit
mode, the horizontal position is constrained by the
slit but also ensure that the spectrum is always at the
same position vertically.
5.1.4 Autoguiding (guiding module)
Autoguiding is very dependent on your equipment.
Autoguiding means that you send the guiding image
to the computer, which controls the telescope based
on star movement in the image. AudeLA software can
do that (like many other software).
Autoguiding is not only for comfort (when you activate the autoguiding, you don’t need to permanently
check the guiding image any more), but it also improves the spectrum quality, because it always reproduces the same conditions. Consider autoguiding a
high priority for obtaining good results.
5.1.6 Instrumental Response curve
We have already described a simple way to get the
Instrumental Response curve with ISIS. In fact, this
is a tricky operation. The response curve can evolve
with some parameters, like sky conditions, target position in the sky (air mass), reddening, and so on. To
be perfect, you need a reference star close to your
target, of the same spectral type, with similar magnitude... and this never happens. Getting a bad response curve will have an immediate effect: the general shape of your spectrum will be distorted. It is a
too long story for this document, but keep in mind
that getting the proper Instrumental Response curve
requires some experience.
5.1.7 Write a log file
It can take various formats, but we strongly encourage you to write a detailed log file when observing.
Note all the conditions of your observations - setup,
weather, goal, target list, problems encountered during the night. It will help you, in the future, to make
the best use of your data.
5.2 Improve productivity
5.2.1 Keep the same setup
5.1.5 Cosmetic file
The CCD chip has some “hot pixels”. These pixels
are much more sensitive than others. These pixels
will give wrong measurements. It is better to remove
them from your image. This can be managed by ISIS
software, by creating a map of hot pixels. All pixels
considered as hot are replaced in the image by an average of the surrounding pixels. To generate the hot
pixel map, you use a dark frame and record all pixels above a certain value. In ISIS, go to the Master
images tab, and focus on the “Computing a cosmetic
file” area. Enter the threshold value - for instance 100
ADU (in the dark we use, most of the pixels are close
to 0 ADU). Click on the Go button, and look at the
console: you’ll see the number of hot pixels detected.
Around 100-500 hot pixels is considered normal.
The master dark is not really a dark frame,
but the thermal map of the camera after the
offset has been removed. This is why the average level of the image is close to zero.
You can use the hot pixel map in the “2. General” tab.
Simply enter the name of your map file in the field
“Cosmetic file”, and re-run the data reduction. ISIS
will now include this hot pixel list in its calculation.
The best way to improve your productivity is to always keep the same setup, and always process the
same way. Of course, observers often want to improve
their setup (this is a good behavior), but as soon as
you change parameters in your setup, you have to
adapt (to change...) your process. Sometime, it is preferable to keep the same conditions, for both quality and
productivity.
5.2.2 Dark frames library
Dark frames are long exposure time images (at least
as long as the exposure time of your objects). And the
more darks you take, the better (you will improve the
SNR of the dark frame). This is a very time consuming operation - something we should do at the end of
the night, when we know the longest exposure time.
But dark frames are an intrinsic charateristic of your
equipment. They only depend on the temperature and
exposure time. They don’t depend on the observing
conditions. So there is no need to do them during the
observing session itself: you can do them once, for instance during a cloudy night (or even during the day
- just take care of light leaks).
We suggest that you make your own dark frame library, with different exposure times and camera temperatures (you could even put the camera in the fridge
36
CHAPTER 5. NEXT STEPS
to simulate a cold night). Take a large number of images, and create a set of master darks from them just
once.
5.2.3 Exposure time
We suggest you always use the same exposure time
for your observations (typically 5 minutes - 300 seconds - for faint objects). Of course, the total exposure
time will depend on the object magnitude, but you
can adapt the total exposure time by changing the
number of exposure. They are several avantages to
proceeding this way:
– You can always use the same dark frames,
– Data reduction is very fast (no parameter is changed),
– No risk of error during acquisition (you never
change the time).
There are some limits with this process, for instance
for bright or extremely faint objects, but in general it
will make your life easier.
5.2.4 Prepare for your observation
Amateur astronomers often wait for the last minute
to decide what to observe. We recommend to prepare
for your observation well in advance. If you properly
organize your observing session, you will optimize the
target order, their position in the sky (higher is better), and so on. You can even define the exposure time
for each object, depending on their brightness.
If you are looking for target and observing programs, you can look at the Shelyak Instruments website2: we’ve listed many ideas, ranging from from easy
to ambitious.
5.2.5 One directory per observing session
This is a basic advice, but doing spectroscopy will
quickly produce a high volume of files. To be sure that
you will be able to proceed with a reliable data reduction, it is important to not mix files from different
observing sessions. We recommend that you to create
a new empty directory at the beginning of each observing session (an observing session is generally an
observing night). Put your log file (see section 5.1.7)
in this directory.
5.2.6 Reduce your data quickly
It is very common that astronomers use the time
during the night to observe and wait until the day after to reduce the data. Be careful: if you wait too long,
you’ll probably never reduce your data. And having
just raw data on your hard disk is the same as if you
did nothing. We strongly encourage you to reduce
your data immediately after acquisition. This way it
is all fresh in your mind, and there is much less risk
of mistakes during the data reduction.
5.3 Share your results
Our experience shows that the best way to improve
your data is to compare your results with others. When
you compare your spectra with the results from experienced people - with comparable instruments, of
course - you can get an idea of the potential for improvement you have in front of you. We encourage
you, for instance, to observe Be stars, and look at the
BeSS database3for comparable observations. You can
even send your spectra to the database - they will be
checked by administrators who give feedback on potential improvements.
In addition, spectroscopy is certainly still at a pioneering stage. All the data you collect are of the highest interest for the scientific community: don’t keep
them to yourself.
5.4 Get pro-like quality spectra
Very often, amateur astronomers think they are not
able to provide high quality data, comparable to those
produced by professionals. Of course, amateurs usually don’t have professional quality instruments - but
the quality of the data is more a question of methodology than instrumentation and you will quickly see
that you can reach professional quality, with just as
much rigour. You can even contribute to research in
several Pro-Amateur collaborations. In a few words,
to get good quality data, you need to pay attention
to:
– Accurate auto-guiding
– Repeatable configuration and processing
– Correct data reduction: SNR, Wavelength calibra-
tion, Instrumental response curve correction.
– Standard file format (see for instance BeSS file
format)
– Archiving and backing up the raw data
Never forget that when you’re observing the
spectra of an object, you’re probably the only
one doing it at this time on earth - this is incredibly valuable data... and you must absolutely
archive it very carefully (raw images as well as
processed data).
2. www.shelyak.com - menu spectroscopy / educationnal &
projects
3. http://basebe.obspm.fr/basebe/Accueil.php?flag_lang=en
37
Here is a summary of a typical observing session,
to ensure data quality and productivity. It is a basic
checklist that you can use every time you observe.
Preparation
– Start-up the equipment, and cool down the acqui-
sition camera (max 80% of cooling power),
– Check your session parameters in AudeLA and
ISIS: observer’s names, observing site, equipment,
working directory...
– Prepare your session directory,
– Check your computer time, to better than 1 sec-
ond (the exposure date & time will be based on
it),
– Open a log file,
– Run a test for all images: sky (Sun’s spectrum),
calibration, flat, and check that all is Ok (images
correct, file headers with proper data...),
– Prepare your target list,
– Acquire dark and offset images and prepare mas-
ter images (or copy from the master directory),
– Get flat field images.
6
Typical observing session
During observation
For each target observation, you will need to:
– Locate the target (start with a reference star),
– Run the acquisition,
– Get the calibration images,
– Fill in the log file with any significant informa-
tion that will help in the future to remember the
observing conditions,
– Check that the images are properly recorded in
the session directory,
– (Reduce the data as soon as possible),
At the end of the night
– Double check that you have all reference images
(dark frames, offset, flat fields, neon, reference
star, objects)
– Archive your raw data securely.
38
7.1 Alpy Modules & Accessories
7.1.1 Alpy Guiding module
As explained in this document, working in slit mode
gives some key improvements to your observations.
But in this mode, you cannot see the telescope field,
and it is difficult to position the star on the slit.
The Alpy guiding module makes it possible to observe the slit entrance with a second camera (usually
called the guiding camera). This allows you to position the star precisely on the slit, and to track it continuously. This gives comfort, of course, but also improves the quality of data, and allows long exposure
times. If you’re using your Alpy 600 for astronomical observations, the Alpy guding module is almost
mandatory.
7
Appendix
During the night, you can observe and control the
star position vs the slit:
In the image below, you can see the telescope field
and the Alpy 600 slit:
7.1.2 Alpy Calibration module
As described in this document, getting high quality spectra requires reference images. These images
can be made “manually”, by putting a calibration or
flat lamp in front of the telescope. But this operation
is dramatically simplified by using the Alpy calibration module. This module includes a halogen lamp
for flat field images and an Argon-Neon lamp for calibration. It can even be used remotely, to avoid any
operations at the telescope during observations. The
Alpy calibration module is not required to produce
39
Alpy 600 - user guide
quality data, but it improves the repeatability and efficiency of your observations.
7.1.3 DSLR camera barlow adapter
The Alpy 600 has a short backfocus, to comply with
compact CCD cameras. However, in some conditions,
a longer backfocus is required. This is the case for
DSLR camera usage. A DSLR camera with a T-mount
adapter has a long backfocus (54,85mm) compared
to most CCD cameras (15-25mm range). The Alpy
DSLR camera barlow adapter extends the Alpy 600
backfocus to match that required by DSLR cameras.
Calibration and flat field images made with the Alpy
calibration module:
Here, the DSLR camera barlow adapter is mounted
on the Alpy 600:
40
Even though a CCD camera is generally better for
astronomical applications (cooling, sensitivity, no color
bayer matrix effect...), using a DSLR camera makes
astronomical spectroscopy accessible with a low budget. How to use a DSLR camera with the Alpy 600 is
described in section 7.2 below.
7.1.4 C-mount camera adapter
Most CCD cameras are today using T-mount thread
(M42 x 0.75mm). But there is also a large range of
compact CCD cameras which use a C-mount thread
instead (1” diameter). Some of them can be used with
Alpy 600 thanks to the optional C-mount camera adapter
- just keep in mind the CCD size, which must be at
least 8.5mm to get the full spectrum.
The C-mount adapter replaces the main T-mount
adapter of the Alpy 600:
CHAPTER 7. APPENDIX
7.1.5 Optical fiber adapter
Alpy 600 can be used with a fiber optics. The optical fiber adapter has a Type-FC connector:
7.2 Alpy 600 with a DSLR camera
Your Alpy 600 can be used with a DSLR camera.
A lot of people today have such a camera at home,
and it allows you to start in spectroscopy with a low
investment. Because of the long backfocus of DSLR
cameras, you must add the barlow adapter. In the
nominal configuration, the spectrum length is about
18mm (it is 6mm without the barlow). Since pixels
are usually small in a DSLR camera (5-6µm range),
the spectrum is over-sampled.
Installing the DSLR camera barlow adapter
The DSLR camera barlow adapter is mounted on
the back of the Alpy 600. If necessary, remove the
main ring for better access to the thread. Make sure
to tighten it firmly:
41
Alpy 600 - user guide
The Alpy 600 is mounted on the DSLR camera with
a standard T-mount adapter:
a CCD camera (refer to section 2.1). You can use sunlight to make this operation.
But you may also use any lamp with emission lines,
such as Neon or Argon.
Focusing the Alpy 600
We suggest installing your DSLR camera on a tri-
pod. Focus the Alpy 600 with the same method as for
42
DSLR camera vs CCD camera
You can use your Alpy 600 and DSLR camera with
all Alpy modules and accessories (guiding, calibration, fiber optics...), for lab activities as well as for Astronomy. However, keep in mind that a DSLR camera
doesn’t have the same performance as a CCD camera.
The main differences are:
– Sensitivity. A CCD camera is roughly ten times
more sensitive than a DSLR camera. It means that
for the same result, you will need ten times longer
exposures. This is not a problem for bright objects,
but it limits the observation of faint targets.
– Spectral bandwith is smaller for a DSLR camera.
A CCD camera can go up to near UV and near IR,
where DSLR camera is strictly limited to the visible domain. Note that it is better to remove the
IR-cut filter from the DSLR camera, to cover Hα
(this is a usual operation in astronomy).
– The DSLR camera has a color sensor. This makes
raw images more complex to work with (there is
the Bayer matrix to manage). Note that the oversampling in the DSLR camera configuration means
that the Bayer matrix has no effect in practice on
the Alpy 600 spectra resolution.
CHAPTER 7. APPENDIX
43
Loading...