New Spectroscope.... So far

[theodog]

Hi All,
I had the first go at taking a test spectra through the new spectroscope. It has the split mirror fitted, adjustable lens in, 200mm collimator lens in place and the 600l/mm grating housed and fitted. About to mount the camera lens. However I could attach the C. lens to a 300d and get an image of an ES light globe.
Below;
1. Old spectroscope image of ES globe using DSI.
2. Raw 300d full frame.
3. Spec processed through VSpec.
:)

[renormalised]

Looks like it's performing fairly well...what sort of resolution are you getting on the spectra??

[theodog]

Here are the specs the unit should have, from simspecV3.2.
However, there was no 'telescope' lens used, I have no idea of the slit width, the collimating lens was only approx focus and the camera lens was just sitting in position (so the angle and distance to the grating are approx.):sadeyes:
So.... these numbers represent what I am aiming to achieve with this system.:)

[sheeny]

Cool! 1.75Å/pixel. Looking good.

Al.

[Merlin66]

Jeff,
Certainly looking good!! You're well on your way to having a creditable slit spectroscope..
A more accurate method of getting the "actual" resolution ( when you consider your unknowns...) is to use the FWHM measurement of a prominent line.
On your spectrum BEFORE normalising, take the 4358 Hg line and measure from Vspec the maximum intensity, take half this value and move the cursor across the line at this intensity value and note the blue and red edge intersection value, the difference will give you the FWHM value in A. Check it against other lines (5769/5789?) to confirm.
Better still use a Neon lamp, and do the same.

Don't mix up A/pixel which is the dispersion or plate scale with resolution. The resolution, as you see from SimSpec is at least twice this value and depends on the pixel size, slit size, collimator focal length etc etc etc.

The raw image shows the camera is not aligned with the grating, seems to also be sitting a bit off centre ( too low?) and the illumination is not central to the slit. What exposure did you use? There seems to be some light getting in on the LHS which will affect contrast and s/n. All these are easy fixes and will improve the resolution even more!!
Any images of the "final" construction??

[Heian]

now that is impressive, :eyepop::thumbsup:
I had collected some old camera bodies and lenses to do something similar with the bodies being cut down to use them as lens mounts, but it ended up beyond my skills....
congrat's

[theodog]

Thanks Ken.
Nothing was set-up, just thrown together.
No alignment or even a cover for the optics -just set up on the kitchen table (proped up on books), with all lights off except the target in another room.
In all, just to see if it may work. Seems so.
:)

[Merlin66]

Under those conditions you've managed extremely well!!!
Shows great promise of things to come!!
Keep it going, your on the final straight.........

[theodog]

OK, I have some images of the set-up so far.
1. Camera lens fixed onto back plate.
2. Spectrum more evenly illuminated.
Progress has been slow due to an accidental palm spike to the elbow, a piece is still there beside the funny-bone nerve. Hurt like hell but modern pain-killers and antibiotics are teriffic.
:)

[Merlin66]

A bit tighter on the focus and you're winning prizes!!
I see what you mean about "work in progress"
Don't get tetnus!

[theodog]

Another investigation, this time of the sky through a window this morning.
It is referenced against an image I found on the net, I don't have the site (my apologies).:sadeyes:
Seems to be a good match. Still some work to do.
:)

[Merlin66]

Jeff,
the good news is that on the sun it will get even better!!
A 6mm diameter aperture mask on your telescope will give MORE than enough light to close down the slit (+/-20 micron) and get close to the design resolution.
There are lots of good solar spectrum references:

http://bass2000.obspm.fr/solar_spect.php?step=1
http://fermi.jhuapl.edu/liege/s00_0000.html

[theodog]

Ok, I've fixed the camera lens and narrowed the slit. Placed the unit in a film change bag and used a 6 sec exposure, still no telescope 'though.
1x6sec exp Cannon 300d unmodded. Processed in Maxim, Iris & VSpec.
Colour is synthetic spec from VSpec.
Camera is variable in sensitivity.
:)

[Merlin66]

The dip between the green and red responce around 560nm is the same as I get with the Canon 300.
Still a bit noisy and the resolution will improve as you tune it further.

I've got some 300D (modded) solar spectra images with the 600lpm grating and 135mm telelens. First couple shows the responce in UV and the resolution, the others are at Ha and beyond ( the 2nd order starts to kick in). about 1 sec exposure with a 4mm aperture stop on the ST80.

[theodog]

It's late and the arm is sore again but this is where I got up to.
I have attached all components solidly to the backing plate and aligned them. Slit can be rotated to align with the field.
Fitted a neon callibration globe under the beamsplitter soas to reflect onto the slit.
Seems to work OK.
FHMW seems to be 10A:question:.
Average results from several Ne lines.
:)

[Merlin66]

Jeff,
Not quite right yet...
You should be getting far clearer and tighter focus on the Neon lines..
The resolution FWHM I'd expect to be down around 3A. Running the numbers through SimSpec V3.2 it gives 2.7A
The slit alignment to the grating and the slit width may be an issue as well as the focusing of the camera lens
Don't take my comments as being critical of your work; this is very typical of the testing and fine tuning cycle that goes on with EVERY new spectroscope - believe me!!!!
You and I both know that in the end you'll have a very good capable spectroscope....you just got to get there.
( looking closely at your photo..why is the collimator lens not showing a focus on infinite? trying to compensate for the back focus to the slit??? May have an effect....)

[theodog]

I haven't read anything as critical, just advice.

I found the old Pentax, apparently I didn't throw it out, Measured the focal point of the 200mm M42 at 45mm. I have 51mm lens-slit. I will set about adjusting this.:thumbsup:
:)

[theodog]

Next test,
Attaching the spectroscope to the telescope.
Image of Jupiter positioned over the slit, superimposed by an image of the spectra taken.
I see some lines. :D
Slit needs to be aligned with the frame.
Not measured but the slit ran approx. N-S on Jupiter.
Next step is to align guide image and slit.

[Merlin66]

For such a wide slit, you did well to record the detail you did!
Looks like you're moving well up the learning curve!!
BTW now look at the moons of Jupiter - see the difference between the sulphur of Io and the ice of Europa???

[sheeny]

Looking good Jeff! :thumbsup:

Al.

[theodog]

Thanks Al.
I have finally finished the cover and attached a ST7.
It took 8 frames to cover the sky spect with a refocus for each frame.
Some work needed to re-enforce the structure for mounting on the scope due to its mass.
:D

[Merlin66]

Looks more and more like a spectroscope every day!
What's the final weight, excluding the camera??
The alignment of the slit/ grating/ camera horizontal pixel needs just a little improvement.
What slit width do you think you're using?
The response down in the UV (H&K lines) looks promising!!
Gaffa tape around the joints will help.
Take a few flat images to verify no light leaks and minimum vignetting.
This can be done by setting it up on the scope, cover objective with white T shirt and illuminate with a Quartz Hologen lamp; you'll need one for each setting of the grating.
Looking forward to some good results!
For the record, what do you think it's cost so far?

[theodog]

Unit without camera's is 3.0kg

I have no idea. I just close it down and then open it until I get a 0 image. I am experimenting with finding a way by using the image. I need this to complete a scale for the slit adjustment screw.

Probably $600. Both lenses could be sourced from e-bay for ~$100 each, slit US$15 surplus shed, Flip mirror $150?, Bintel and 600l/mm grating US$150 Optimetrics, Aluminium and wood from scrap ~$50, Screws and threads ~$20-$30. And plenty of paper, patience to make the bits.
Great project for those inclined 'though.
:D

[Merlin66]

So, for the price of a "Hot Shot" eyepiece and a bit of work, you get a scientific instrument which will give more useful information than 100 astro-photographs!!!!
BTW there's a link in my project article on how to measure the width of the slit using a laser torch.
http://www.iceinspace.com.au/index.php?id=63,339,0,0,1,0

[theodog]

:sadeyes: Oh my apologies, the most important part was the help and support from IIS (in particular Ken & Al) and the Astronomical Spectra Forum.
:hi::thumbsup:
:D

[sheeny]

Looks good Jeff!

I'll have to do a project like that myself one day:P, given the price for buying an SGS or LHires.

Al.