Star'Ex is an extension of the Sol'Ex instrument. This version allows astronomy enthusiasts to perform spectra of stars, nebulae, galaxies ... and thus opens wide the doors of astrophysics. Spectroscopy is the science of light, affordable and easy to access with equipment like Star'Ex. Technically, for a "Solar Explorer" to become a "Stellar Explorer », it suffices to produce a small 3D printing module for pointing the stars, or any other light source whose spectrum you want to take. For the rest, let yourself be guided by the explanations given on this page and the many achievements of amateurs to discover the fascinating world of astronomical spectroscopy. That’s Star’Ex promise.
This section of the site describes how to make the check-in module to add to Sol’Ex, how to set it up and help you get started with using Star’Ex.
Part 1: Make the pointing cube
The image below shows the Star'Ex instrument in a situation for observing the stars. We identify the guidance / pointing module right at the entrance, look like a cube, which makes the link to the telescope, Sol'Ex in itself is unchanged, but we notice an acquisition camera whose detector is cooled for a better performance on weak objects (recommended, but not compulsory for the first weapons).
Complete Star’Ex with its guidance camera and a spectra acquisition camera with a sensor cooling system.
The guide cube is important because it will allow you to aim at the star so we are trying to take the spectrum and position it on a slit as narrow as 10 microns at the focus of the telescope. I show how it is done in a video to follow.
A handful of optical components is required to make the guide cube. They are included in the Sol'Ex basic kit, which you can obtain from Shelyak (see "Contact" section):
- two lenses of 12.7 mm in diameter and 50 mm in focal length, which produce a quality and permanent image of the Sol’Ex slit on the guide camera.
- an aluminized mirror with a side of 15 mm.
The optical diagram of the guidance / pointing system is as follows:
The sharp image of the star field focused on the slit by the telescope is reflected by the slit (except at the location of the slit). This is the reason why the Sol'Ex slit is tilted at 30 ° = after reflection, the light is directed sideways on a small deflector mirror. The latter reflects the light towards the two doublets of 50 mm focal length, positioned head-to-tail, and which form a clear image of the slit on the guide detector. These two lenses form an image carrier, similar to the principle found on periscopes, for example.
The assembly of these optical elements is very simple, it does not require any adjustment.
On the mechanical side, the image below shows the parts to be produced in 3D printing. They are relatively few:
You can download the corresponding STL files by clicking on the following link (ZIP archive): guidage_kit.
You have noticed, the guidance system includes a camera, and also a focusing system, which is undeniable comfort. It can be seen that the production of the guidance module still requires a certain financial investment, even if initially, the guidance camera can be replaced by a guidance eyepiece (guidance is then done by eye, to the "old" in a way!).
A miniature camera model does the trick perfectly. For my part I use an ASI290MM mini camera, but we can easily find the equivalent at QHY or elsewhere. The focusing device is the helical system in the example on the right, but more economical models, or even 3D printed models can be used.
Once these parts have been printed, you can proceed with assembly.
There are 7 RUHEX M3 inserts to be fitted in the parts, 3 in the part "guide tube # 3", and 4 in the part "guide cube #11". The procedure is described in detail in the "Construction" section of Sol’Ex.
The proper assembly requires gluing the 15 mm mirror in the inclined support # 2. This is the only part that needs a little attention. The assembly of the lenses amounts to a simple stacking, taking care in the direction of assembly:
The following photographs show the appearance of the guide tube that integrates the two lenses.
The following photographs show the appearance of the guide tube which integrates the two lenses.
Part 2: Adjust the guiding camera
Before heading towards the telescope or telescope, a good idea is to set the guiding / pointing cameras by day and on the table. The aim is to achieve a sharp image of the slit on the detector and to orient it correctly. Compared to imaging the Sun with Sol'Ex, with Star'Ex the performance in terms of sensor reading speed is not a concern, the slit is oriented vertically to benefit from the maximum width of the sensor according to the spectral axis and thus cover an enlarged spectral domain.
The following video explains how to make this adjustment.
Part 3: First Star’Ex observations
I propose to put you in the situation of using Sol'Ex in a configuration as close as possible to that of the Sol'Ex kit: a slit of 10 microns, a network of 2400 lines / mm and the collimator and camera objectives of 80 mm and 125 mm respectively. Likewise, for the camera, we select an ASI178MM (or equivalent), common in solar observation, but not having a sensor cooling system. This last point is potentially critical when it comes to capturing the spectrum of stars, faint objects that sometimes require exposures of several tens of minutes. Here we are far from the possibilities of a camera with a cooling system: the maximum exposure time is at most that which does not allow exposures of 3 to 5 minutes at most, with the result of very strong thermal noise. Yet this is how I suggest you get started with stellar spectrography, if you are in this situation. There are good surprises at the end of this path and a universe that opens up. You can watch the following film on Observational Spectrography with Star'Ex to learn more.
This video explains how to make these first spectra with a digital camera, how to position a star on the slit, how to calibrate the observations ...
As indicated above, the use of a camera equipped with a sensor cooling system will make it possible to observe stars of shards much weaker than that of Vega./ Among the other options that allow you to go to this direction the choice of the grating (engraving density) as well as the width of the slit (which increases in proportion to the diameter of the telescope).
For example, here is the spectrum of the symbiotic star CH Cyg observed with the video configuration (very high spectral resolution), except that a cooled camera (ASI183MM) is used, which changes a lot of things:
Detail of the very high resolution H-alpha line (R = 40,000) in the spectrum of the star CH Cyg (magnitude close to 5), produced with a telescope only 65 mm in diameter.
The following spectrum, of the star P Cyg (34 Cyg), still taken with a 65 mm diameter telescope, is remarkable in the sense that it allows to detail the heart of the blue wing of the H-alpha line, may be a first in amateur astronomy, which demonstrates the potential of Star'Ex, which is also a research instrument which rises to the level of a much more expensive instrument:
Use of Sol'Ex solar configuration (10 micron slit, 2400 t / mm grating, 125 mm camera lens, small telescope) used to detail H-alpha lines of the star RS Oph during the outburst of 2021 - an observation made possible by the use of a mid-range cooled camera (ASI183MM):
Some of the options that can be used with Star'Ex are described below.
Part 4: The options
Part 4.1: Use a Lhires III slit
Designing a spectrograph like Star'Ex is a matter of compromise. In part 3 (see video) I showed that this instrument can observe stars with a slit of only 10 microns wide. The performance in spectral resolution is then remarkable (close to R = 40,000). But however on one condition: that the telescope is of short focal length (and of small diameter). As soon as the focal length approaches 1 meter, the slit blocks too much of the light rays before they reach the detector due to the enlargement of the image spot induced by turbulence or tracking errors. .
This is where the trade-off comes in: you have to sacrifice a bit of spectral finesse to compensate for the flow of photons into Star'Ex. The solution is simple: you have to use a slit larger than 10 microns. You can't have it all simultaneously.
It is perfectly possible to combine Star'Ex with a wide choice of slots, with widths that even allow our instrument to be mounted on telescopes that can be 1 meter in diameter! I took care of this, because my desire is that the Sol’Ex / Star’Ex project be as flexible as possible.
To do this, I chose to use the set of slots provided for the Lhires III and LISA spectrographs from the Shelyak company. These slots can be ordered from the latter company. They are based on the same principle as the 10 micron slit in your possession, with high precision screen printing in a layer of chrome, all done on a thin glass slide. Your only job is to print a specific support ring that replaces the current support ring .
The STL files (slit support and two flanges) of the parts allowing the use of the Shelyak Liters III slots can be downloaded by clicking on the following link: fente_kit..
How to mount and adjust these slits is explained in the video below.
Part 4.2: The telescope and camera
You can very well use Sol'Ex at the focus of a telescope. Compared to a refractor, the gain will of course be very significant thanks to the larger collecting surface and the absence of chromatism. A Ritchey-Chretien telescope open at f/8, for example, will do very well. If you have a Schmidt-Cassegrain open at f/10, I recommend using it as it is, without adding a focal reducer because of the optical defects that it introduces, which are clearly visible in spectrography. With a Newton open at f / 4.5 or f / 5 (see photograph below), I recommend adding a Barlow lens, however, to bring the aperture between f/8 and f/10.
La largeur de la fente à l’entrée de Sol’Ex détermine le pouvoir de résolution spectral atteint (voir la section « Théorie »). Mais c’est le seeing et la précision de guidage que permet votre monture qui va en fin de compte définir la largeur de la fente. Depuis un observatoire normal pour les amateurs, je recommande que la largeur angulaire de la fente sur le ciel soit comprise entre 3 et 4,5 secondes d’arc.
When using a 2400 lines/mm grating, or even 1200 lines/mm, the internal vignetting of Star'Ex is quite severe if one is using a fast telescope, which causes a loss of signal (see the “Theory” section). The situation is even more critical with a reflectiong telescope, because the image of the secondary mirror projected onto the grating occupies a significant portion of the surface grating. Since the instrument is more open than f/6.5, I recommend for stargazing to add a Barlow lens in the optical path which artificially increases the focal length of the refractor or mirror telescope. A Barlow so the magnification is 1.5 to 1.8 is often a good choice.
The difficulty when adding a Barlow lens is the appearance of chromatic aberration, especially in the blue part of the spectrum:
Observation and optical simulation with an old X2 Barlow lens model (CLAVE) on a 250 mm f / 4.5 Newton telescope. The star observed is Deneb (alpha Cyg). We notice the widening of the 2D trace of the spectrum going towards the ultraviolet. By adjusting the focusing (50 microns, 150 microns, ...) it is possible to refine the trace at certain wavelengths, but unfortunately to the detriment of other parts of the spectrum. This is a typical chromaticism defect.
The APM X2.7 lens model, described as "apochromatic" by the manufacturer, reduces this aberration by using special glass. I was able to see this by theoretical analysis and by observation:
Optical simulation of an APM X2.7 type Barlow lens used on a f/4.5 Newton telescope . The color correction is very good, with uniform image quality from ultraviolet to infrared. An excellent result, especially for spectrography.
Unfortunately, the short focal length of the APM X2.7 lens makes it difficult to use when aiming for a magnification of around 1.6 to 1.8. Recall that the magnification A of a Barlow is given by the formula:
A = 1 - d / f
with d, the distance from the Barlow to the detector and f, the focal length of the Barlow. In the case of the Barlow APN, the focal length is f = -60.3 mm and will easily verify that the lens must be approached very closely to the slit to achieve the desired magnification of X1.6, which generates new optical aberrations (aberration of sphericity, because the component is not calculated to be used in this way). A longer focal length lens will be preferable, like the Tele Vue 1.8x with a focal length of f = -133mm, which I have had the opportunity to test successfully over a large part of the spectrum accessible to Star'Ex. In this case, the magnification of 1.6 is obtained with a lens-detector distance of 80 mm, which is relatively close to the nominal value. Notice in the photograph below that I housed the lens in the 50mm slider after making a small 3D printing adaptation piece (in the interest of having a 3D printer!).
Tele Vue 1.8x Barlow Tele Vue 1,8x at 80 mm distance of entrance slit of Star’Ex.
With a 200mm f/5 Newtonian telescope, this arrangement takes the focal length to 1600mm, i.e. an aperture of f/8, judged to be a good compromise between the (partial) reduction of internal vignetting in Star'Ex and the loss of resolution induced by increasing the width of the slit. A 23 micron wide slit here represents approximately 3 arc seconds on the sky and a 35 micron slit approximately 4.5 arc seconds. One can hesitate between these two values, the configuration with a slit of 35 microns being more comfortable during the observation, because causing less loss of flux, but reducing the spectral resolution by a factor 23/35 = 0.66 .
Taking into account the duration of exposure times in stellar spectrography, which can reach 15 to 30 minutes on individual images, the use of a camera with a detector cooling system is strongly recommended, see opposite. use of an ASI183MM Pro.
A review of some usable cameras can be found in the "Construction" section. Note that it is not in itself the level of the thermal signal that is responsible, because it is eliminated by calibration, but the noise that this thermal signal produces, which cannot be eliminated. However, we have seen that it is perfectly possible to use an uncooled camera in part 3 of this section, at a minimum to get started.
Part 4.3: The grating choice
The design of Sol'Ex / Star'Ex offers the possibility of network choice, which considerably multiplies the possibilities. Any 25 x 25mm x 6mm grating can be used as is (I recommend printing as many grating support and storage boxes - see "Construction" section). We will find in the “Theory” section what to calculate the performance to be expected from the available networks.
Compatible grating will be obtained from ThorLabs or Optometrix. Orders are also possible through Shelyak Instruments, this inquire.
A grating is characterized first of all by the engraving density. The standard values are 150, 300, 600, 1200, 1800, 2400 lines/mm. The higher the etching density, the more dispersive the grating. For some applications, we will look for a low dispersion by force of circumstance, because the object to be seen is very faint. It is also possible to wish to cover a large spectral range in a single image.
Another characteristic is the blaze wavelength. Depending on the manufacturing parameters (shape of the lines), the grating can preferentially concentrate the light in a specific region of the spectrum. For visible part applications, the blaze wavelength is 500 nm. But there are also grating whose peak performance is located at 300, 750 or even 1000 nm, for example. Thus, to observe infrared (which Star'Ex is very good at), we will choose a blazed grating at 750 nm or 1000 nm.
For example, here are observations of the infrared spectrum of stars obtained with Star'Ex at a wavelength greater than 1 micron, which is truly exceptional in the amateur world. A weakly dispersive grating (300 t / mm), blazed at 1000 nm is used here. The optical configuration of Star'Ex is modified for the occasion: the 125 mm lens is replaced by an 80 mm lens, in fact the same one that serves as the collimator (see next section).
Infrared spectrum of the star Vega produced with Star'Ex mounted at the focus of a small Newton 200 mm f/5 telescope. The spectrum shows the Pashen hydrogen line # 8 at 10049 A, and especially the Pashen line # 7 at 10938 A, which is a probable record in amateur astronomy. Note the use of a grating whose blaze angle is optimized for 1 micron wavelength. The camera is equipped with an ordinary CMOS sensor (here we are at the extreme limit of the sensitivity range of silicon technology detectors).
The infrared spectrum of the nova Cas 2021. All the lines of the Pashen series are in strong emission.
For infrared observation, an order filter must be added in front of Star’Ex, here an RG630 (SCHOTT) filter.
Star'Ex can be used perfectly to observe non-stellar objects as long as the spectral dispersion is well adapted. We can even tackle extragalactic objects, as in the following example, with the Messier 51 galaxy as a target, observed with a telescope only 85 mm in diameter and a 300 lines/mm grating (the "redshift" of the galaxy can be measured!):
Note how the core of M51 has been positioned over the slit to capture maximum light. The example to follow also concerns a non-stellar object, the Messier 42 nebula (Orion nebula) using the same equipment (a 85 mm refractor):
The analyzed nebula "slice" can be seen in the guiding camera image on the left. On the right, the spectrum shows many very fine emission lines. The shape of these lines changes depending on the chemical element, which is the origin of the color in this object seen in the photographs.
We can easily see that for stargazing we can be led to use weakly dispersive gratings by sacrificing the spectral resolution when aiming at objects of low brightness. For solar applications (Sol'Ex ), the 2400 lines/mm grating is preferred (but also usable on stars, as we have seen in part 3). For solar again, a 1200 lines/mm grating can also be interesting, in particular for making images of the Sun in the ultraviolet region, at the level of the H and K lines of Ca II (compared to a 2400 lines/mm grating , higher brightness, sufficient spectral resolution, good image quality).
From experience, and this remark is important, if you are new to stellar spectrography, high spectral resolution should be preferred. It's not immediately intuitive, but it's much easier to observe and calibrate a high-resolution spectrum than a low-resolution spectrum. Astrophysically, you will also get results more immediately (spectra "move" faster in high resolution than in low!). So, for your first few weapons with Star'Ex, select a 2400 lines/mm and/or a 1200 lines/mm.
Part 4.4: The « 80 mm / 80 mm » configuration
The Sol'Ex collimator lens, made with a special glass, is also a very good performance camera lens, which can replace the classic 125 mm of the standard configuration. The benefit is an increase in brightness and color correction, which is particularly useful in stellar spectrography.
The "80mm + 80mm" optical combination, here with a 300 lines / mm grating, spectral coverage ranging from 390 nm to 750 nm and an input beam at f/8.
A specific camera interface must be made to focus, typically on a ZWO ASI183MM Pro cooled camera. This interface is common with that of the “APN kit” described in the “Construction” section, part 4 (you can download the STLs).
To appreciate the performance of the "80 mm / 80 mm" option of Star'Ex, here is for example an ultraviolet spectrum of the star RS Oph (in outburst) obtained from a Newton telescope of 200 mm at f / 5 and a 2400 line/mm grating. The performance is such that it is possible to detect an emission in the heart of the H and K lines of Ca II:
Below, with the same configuration, the ultraviolet spectrum of the star Vega (exposure time of only 28 seconds with a Newton 200 mm f/5):
Note: below 3680 A, the absorption of the lenses becomes too severe.
The blue part of the spectrum of the star Deneb with the 80 mm / 80 mm configuration and a of 600 lines/mm grating (44 seconds exposure):
Below, the appearance of the solar spectrum (daylight) and emission line lamps in the form of images (called 2D) produced with the « 80 mm / 80 mm » configuration:
From top to bottom, the daylight spectrum, the high contrast daylight spectrum (notice the H and K lines of Ca II on the left and the infrared spectrum on the right), the spectrum of a neon lamp , the spectrum of a fluorescent lamp (fluo-compact).
The following video (click on the image) explains how to mount the 80mm x 80mm configuration, then how to use it on a telescope, with examples:
Part 5: The spectra processing
Spectra processing consists of classical operations of thermal signal removal, offset and division by the flat-field. This is followed by tasks specific to spectrography, such as geometric correction, extraction of the spectral profile or calibration in wavelength and relative flux.
These relatively numerous operations are facilitated by software such as VisualSpec or ISIS. With ISIS, to find your way through the documentation, consider that Star'Ex spectra are very much equivalent to Lhires III spectra when using the 2400 lines / mm lattice. The operations here are quite highly automated.
The other possibility is to use an application specially written for the Star'Ex project. We are talking about the specINTI software, which works according to the INTI philosophy (see section "Processing"), with great ease of use and extensive automation. The processing of the spectra is here an almost transparent operation.
The specINTI software is under development. It is written in Python, but just like INTI, it is presented as an executable file (specinti.exe), which does not require any programming knowledge. An example of specINTI output at the current stage of development:
Copyright (C) 2020-2021 Christian Buil
Web : www.astrosurf.com/buil