A modification for improve eShel spectrograph

Design, construction, tuning of spectroscopes
Information and discussion about softwares (telescope remote, autoguiding, acquisition, spectral processing ...)
Joan Guarro Flo
Posts: 611
Joined: Thu Nov 03, 2011 8:50 pm

Re: A modification for improve eShel spectrograph

Post by Joan Guarro Flo »

Hi James,

Sometimes when you are alone, the main problem isn't lack of self-ability but yes lack of information, I passed for this situation a lot of times.

As a complementary information I attached this picture, I've indicated two possible numerations, I don't know which of them will be better for you.

Really the Thor and Relco lamps spectra have many lines in common.

Let me know if you need anything.

Cheers, Joan.
Attachments
First step Camera position 4.png
First step Camera position 4.png (334.19 KiB) Viewed 4108 times
James Clem
Posts: 9
Joined: Thu Jun 28, 2018 4:48 pm

Re: A modification for improve eShel spectrograph

Post by James Clem »

I just completed the UV upgrade of my eShel and purchased an ASI294MM camera. My first images are showing that the orders are not "centered" in the camera anymore. See in the image below how the upper orders are lying off the left edge? Is there any way to fix this?
orders.png
James Foster
Posts: 439
Joined: Sun Jan 24, 2016 7:14 am

Re: A modification for improve eShel spectrograph

Post by James Foster »

To: James,

Please make sure your chip can handle orders 30-60, just enclosing the "duplicate" orders on the right hand side.
If your using the Canon 135mm F/2 lens, you will probably need at least the ASI 2600MM.
Here is the chip size difference btw yours and the ASI6200MM:
Image

Christian uses the ASI 6200MM for the eShel2 and his article is based on this hardware in conjunction with Isis V6 and up; see:
http://www.astrosurf.com/buil/asi6200mm/

I'm still half way processing my eShel2 data, but I try to adhere to his hardware/software parameters including:

1. Canon 135mm F/2 L lens as eShel condenser lens
2. Thor labs 200 (calibration) and 50 (science) micron (15m) optical fibers
3. Determining the ideal fiber direction of max efficiency with science (stellar spectra) 50micron fiber; Shelyak did this for me.
4. Two fans for optical fiber agitation (lowers modal noise)
5. Wine cooler for eShel enclosure to minimize thermal flexure.
6. Use of 4700K halogen lamp for tung flat; note in the new Isis6 eShel processing procedures, LED and Tung. in the Shelyak calibration box are not used.
7. ASI6200MM Pro to capture full eShel spectral order 26-60; ASI2600MM might work as well, but this is bigger that your ASI294.
8. Acquire camera data using PRISM software with a CMOS set at Gain=200, & Offset=60
9. Process eShel data using Isis V6.1.1

Let me know if you think I've forgotten anything.....as documentation for set-up of the eShel2 with new camera/condenser is fragmented between Isis 3 to Isis 6.1.1, help from this forum is a MUST to succeed on processing data with your eShel2!

James Foster
James Foster
eShel2-Zwo ASI6200MM Pro
Lhires III (2400/1800/600 ln/mm Grat) Spectroscope
LISA IR/Visual Spectroscope (IR Configured)
Alpy 200/600 with Guide/Calibration modules and Photometric slit
Star Analyzer 200
James Foster
Posts: 439
Joined: Sun Jan 24, 2016 7:14 am

Re: A modification for improve eShel spectrograph

Post by James Foster »

Using Francois Teyssier's 2018 post on eShel+ configuration, I was finally able to "line-up" the 3 wavelengths (6584.9A 6677.3A & 4510.7A)
to the three "circles" in check2&3 of the CamRot, ThAr, & Zone tests; see:
Image

To get this far with my set-up (ASI6200MM Pro cmos camera and Canon 135mm F/2 lens) I had to rotate my camera so that the line connecting
the ThAr wavelength emission at 6584.9A (part of the 4-lines in Francois reference) by -2 degrees and shim the right side of the eShel with
some soda can Al skin. Note, that I manually rotated a post processed (Cmos CMED aglorithm) thor_vega image in Prism to arrive at my ideal angle;
I will have to manually rotate the ASI6200MM on the eShel to get my "ideal" -2 degrees inclination.

Here are the setting I used in the eShel set-up box:
Image

Finally, here is the (bumpy) resolutions IsisV6.1.1 gave me for orders 29 to 58:
Image
I probably have to improve my "slant" angles better in the def_order.LST to get more consistent resolutions.

Still have to process data to get 1D spectrum of Vega, but now I have some good parameters to finalize the mechanical set-up of the eShel.

James
James Foster
eShel2-Zwo ASI6200MM Pro
Lhires III (2400/1800/600 ln/mm Grat) Spectroscope
LISA IR/Visual Spectroscope (IR Configured)
Alpy 200/600 with Guide/Calibration modules and Photometric slit
Star Analyzer 200
James Foster
Posts: 439
Joined: Sun Jan 24, 2016 7:14 am

Re: A modification for improve eShel spectrograph

Post by James Foster »

Here is my 1st processed eShel spectra of Vega (note, no image response curve made yet):
Image
I was able to process 3x100 sec 1x1bin images of Vega taken with my CDK17 at F/6.8 with my F/7 eShel Injector and 50micron thick, 15m long Thor optical
fiber being agitated by two (near the scope and near the "wine cooler") oscillating fans.

I was able to process Vega from Order 59 to order 29 for a given spectral response of 3765A - 7860A. I need to improve my def_order.lst slants and process
my response curve per order since there is some non-linearity when adjusting on H-Alpha line with relations to H-Beta and so on.
Hope to get the response curve tonight and finally start to process real data!

James
James Foster
eShel2-Zwo ASI6200MM Pro
Lhires III (2400/1800/600 ln/mm Grat) Spectroscope
LISA IR/Visual Spectroscope (IR Configured)
Alpy 200/600 with Guide/Calibration modules and Photometric slit
Star Analyzer 200
Joan Guarro Flo
Posts: 611
Joined: Thu Nov 03, 2011 8:50 pm

Re: A modification for improve eShel spectrograph

Post by Joan Guarro Flo »

Hi James,

If it is really true that you need one hundred seconds to take a Vega integration, it means that you have a serious problem in your equipment..

Cheers, Joan.
James Foster
Posts: 439
Joined: Sun Jan 24, 2016 7:14 am

Re: A modification for improve eShel spectrograph

Post by James Foster »

To: Joan,

RE:"...to take a Vega integration, it means that you have a serious problem in your equipment.."

I don't know if this is the case but I needed a longer exposure to get more than s/n on orders 55 and over with my eShel, ASI6200MM Cmos camera and
Canon 135 Fl/2 dispersor lens. Here is what I got on a post processed (CMED) image of Vega with bin 1x1 exposure of 100 seconds and PRSIM
acquisition setting of offset 60 and gain of 200:

Order Max Adu
55 8800
56 7600
57 4300
58 2130
59 935
60 195

I barely get any signal at orders below 58. Also keep in mind that my eShel injector is rather slow, F/7, unlike Christian's Newtonian injector which is optimized for F/5 or 4x faster than my system. Let me know if you want and of my raw images to examine. Thanks for the feedback!

James

p.s. On my Alpy600, my exposures on vega are very very short (0.2 sec), even using the Atik414 CCD with 1x1 bin. But the R<600.
James Foster
eShel2-Zwo ASI6200MM Pro
Lhires III (2400/1800/600 ln/mm Grat) Spectroscope
LISA IR/Visual Spectroscope (IR Configured)
Alpy 200/600 with Guide/Calibration modules and Photometric slit
Star Analyzer 200
Olivier GARDE
Posts: 1243
Joined: Thu Sep 29, 2011 6:35 am
Location: Rhône Alpes FRANCE
Contact:

Re: A modification for improve eShel spectrograph

Post by Olivier GARDE »

Hello James,

I think the problem comes from the CDK which is not a 100% mirror optical formula but also contains 2 lenses which undoubtedly limits the flux in the blue below 400nm. These lenses have been optimized for imaging but not for spectroscopy. The number of ADUs on extreme orders that you get shows this.

The best telescope to get wavelength below 400nm are Newton and RC.

is it possible to remove these two lens from the CDK ? This lens are useful for imagery to have a flat field of view on a large surface but in spectroscopie, we use only the beam at the center of the optical.
LHIRES III #5, LISA, e-Shel, C14, RC400 Astrosib, AP1600
http://o.garde.free.fr/astro/Spectro1/Bienvenue.html
James Foster
Posts: 439
Joined: Sun Jan 24, 2016 7:14 am

Re: A modification for improve eShel spectrograph

Post by James Foster »

To: Joan,

Thanks for the feedback!

RE:"I think the problem comes from the CDK which is not a 100% mirror optical formula but also contains 2 lenses which undoubtedly limits the flux in the blue below 400nm. These lenses have been optimized for imaging but not for spectroscopy. The number of ADUs on extreme orders that you get shows this."

Its funny, but I was get fairly decent results btw 3750-4000A with the LhiresIII, but of course this is direct imaging and not through optical fibers.

RE:"is it possible to remove these two lens from the CDK ? This lens are useful for imagery to have a flat field of view on a large surface but in spectroscopie, we use only the beam at the center of the optical."

I tried this and the optics were too poor to image even spectroscopically. I do have a 13" classical cassgrain, without any refractive elements, but it is slower at F/7.5, but may work. Of course I lose 4" of aperture and 0.7 F-stops. If most of the ADU loss is only in the 55-60 order range I'd probably stick with the CDK17 at F/6.8.

James
James Foster
eShel2-Zwo ASI6200MM Pro
Lhires III (2400/1800/600 ln/mm Grat) Spectroscope
LISA IR/Visual Spectroscope (IR Configured)
Alpy 200/600 with Guide/Calibration modules and Photometric slit
Star Analyzer 200
Joan Guarro Flo
Posts: 611
Joined: Thu Nov 03, 2011 8:50 pm

Re: A modification for improve eShel spectrograph

Post by Joan Guarro Flo »

Hi James,

I'm not a friend of making parallelisms, so I want to only tell you some performances of SMM remote equipment,  then you can draw your own conclusions.

Well, SMM telescope is a Meade16 ACF reduced to F6.5, it wants to say that its Schmidt plate is a very strong and thick blue filter, the same that the Alan Gee Mark II Telecompressor lens,  and sure that the two mirrors modern metallic coating don't help anything for the blue light transmission. You can deduce that this 16 ACF is the most mediocre telescope for making spectroscopy.

Some weeks ago, before this cloudy season we have here, I observed the A0V HD39357 4.55 magnitude, I took some 800 seconds integrations of this star, you know that is more than 100 times minus brilliant as Vega is.

I've reached a maximum ADU of 59500 levels, and in my numeration Orders the ADU was :( Add +1 for your order numeration )

#54 = 40000, #55 = 37000, #56 = 30000, #57 = 25000, #58 = 15000, #59 = 5600, #60 = 1800.

The dynamic range is from 3706.500 to 9980.050 Angstroms, the R is around 9000. You know that I use a Nikon 85mm. F1.8 objective instead of a 135mm. one.

On the other hand, I'm taking advantage of these cloudy days and I'm trying a modest, old and simple Nikon 135mm. F2.8 in one NOU_T with very encouraging results, but that is another matter.

Regards, Joan.
Post Reply