...
| focal length/f# | horizontal FOV | vertical FOV | arcsec/pixel | 900-1000 span (pixels) | filter threads | reasonable combo |
|---|---|---|---|---|---|---|
| 50/1.2 | 39.6 | 27 | 33 | 443*(lpmm/500) | 72 mm | 500 lpmm gives 443 pixels, gets zeroth and first order |
| 85/1.8 | 23.9 | 16 | 19.4 | 754*(lpmm/500) | 58 mm | 300 lpmm gives 452 pixels, gets zeroth and first order |
| 135//2.0 | 15.2 | 10 | 12.2 | 1198*(lpmm/500) | 72 | 300 lpmm gives 720 pixels, misses zeroth and first order 100 lpmm gives 240 pixels, gets zeroth and first order. Works well! |
...
Canon pixel size for 5D is 5616 pixels across 36mm so 6.4 microns per pixel, 5616 x 3744 format.
Powershot G6 is 2.3 microns per pixel, format is 3072x2304. 35-140 equivalent focal length. Threads are 58mm. Should be able to adjust zoom to get what we want. Lens is f/2-ish. But one quarter the number of pixels compared to 5D mark II.
Did a 5D test with 135mm lens and 100 l/mm grating, and it seems to work really well! Using 820 nm IR long pass gives a NIR spectrum that spans a few hundred pixels. Dispersion is about 1nm/pixel so that ought to work quite well for water vapor. The wider-field alternative, that might have comparable flux due to blaze on the grating, would be 300 l/mm in tandem with 85 mm lens.
Grating Behind the lens
We of course lose the collimation of the light but might be interesting to place grating behind the lens, so there are minimal chromatic vignetting effects. Guessing that back focal length is about 3 cm = 30 mm which isn't that different from lens focal length! At 300 lines/mm the full spectrum would span about 12 degrees which is 6 mm = 1000 pixels. Going to 150 lines/mm would produce a spectrum of around 500 pixels so R of a few hundred.
Grating fabrication:
...
On May 20, 2013, at 5:26 PM, Hickman, Steven wrote:Your masks are finished, and will ship tomorrow.
-steve
On 5/16/13 8:41 AM, "Christopher Stubbs" <stubbs@physics.harvard.edu> wrote:Hi and thanks for sending along the resolution information. Here's what I would like to do:
All on quartz substrates.
All with 50-50 stripe pattern, half transmissive and half blocked by metallic layer.
All with 4 inch x 4 inch coverage
qty 2 at 30 lines/mm, so a 16.667 micron wide open strip then a 16.667 micron wide metal strip then 16.667 micron gap....repeats
qty 1 at 100 lines/mm, so a 5 micron opening then a 5 micron metallic strip, repeats.
We do want crisp line edges so I'll leave it to your discretion as to what write head to use.
So
1) I need to do anything more formal than this?
2) turnaround time estimate?
Many thanks!
Chris
Christopher Stubbs
Samuel C. Moncher Professor of Physics and of Astronomy
17 Oxford Street
Harvard University
Cambridge MA 02138
stubbs@physics.harvard.edu
If you think this email is rather terse, see
http://emailcharter.org/
On May 15, 2013, at 11:42 AM, Hickman, Steven wrote:Re: Larger format gratings?
The tool has different lenses for different resolutions – at the upper end, we can do 2 micron wide lines with about 100nm of edge variation. But high resolution comes at the cost of slower writing speeds, and greatly increasing cost. I’ve attached a document that explains how to calculate writing speed based on resolution – all you really need is the data from the table on the first page. The tool + staff cost is $205/hr, and the staff time component does not fall under the monthly fee cap.
But as an example, going from a 3 x 3 exposed area to a 4 x 4 area for 30 micron lines would increase the writing time from ~0.5 hours to 1 hour – but going to 10 micron lines at 4 x 4 area would take 2 – 2.5 hours.
-steve
On 5/15/13 11:27 AM, "Christopher Stubbs" <stubbs@physics.harvard.edu <x-msg://2946/stubbs@physics.harvard.edu> > wrote:4 x 4 inches on quartz would be great.
Yes we want 50% metallic coverage, 50% open.
What's the thinnest line you can reliably lay down, with crisp edge?
Have asked for account re-activation, as you saw.
Many thanks,
Chris
Christopher Stubbs
Samuel C. Moncher Professor of Physics and of Astronomy
17 Oxford Street
Harvard University
Cambridge MA 02138
stubbs@physics.harvard.edu <x-msg://2946/stubbs@physics.harvard.edu>
If you think this email is rather terse, see
http://emailcharter.org/
On May 15, 2013, at 11:10 AM, Hickman, Steven wrote:Re: Larger format gratings?
Hello Chris – You are no longer active in our system, so you will need to contact Jim Reynolds (reynolds@cns.fas.harvard.edu <x-msg://2946/reynolds@cns.fas.harvard.edu> <x-msg://2479/reynolds@cns.fas.harvard.edu <x-msg://2479/reynolds@cns.fas.harvard.edu> > ) to reactivate your account – or let me know the name of any of your students who are current CNS users.
Do you want these gratings to be an equal size line and space pattern (I’m thinking of 33 micron lines on a 66 micron pitch)?
The standard mask material is Soda-lime glass; we have quartz available, but there is a material charge of $120/plate for quartz (vs $20/plate for soda-lime)
Also, the plate size will be 4 x 4 inches – although I can make the patterned area 3 x 3 inches, with opaque chrome covering the area outside of the 3 x 3 inch area.
Lastly, how many is a “couple”? - With the new billing structure for the maskwriter, I would estimate that each plate will cost $122 (0.5 hours tool + staff time, and $20 for a soda-lime plate).
-steve
...
Microstepping controller manual: manual, PDF file
Rotation stage manual: Newport_stage_manual.pdf
So to drive the rotator at sidereal rate would require about 106 Hz drive frequency, as a TTL square wave.
...
Wired up DC adapter and function generator. Worked best with microstepper controller switch 1 set to increase motor current one notch above the minimum. Seems to do a fine job of driving the turntable.
...
stuff to take:
- extension cord and power strip
- small canon camera
- turntable, wedge, power supply, function generator
- Tripod
- batteries, ac power supply, charger
- flat screen
- custom CTIO wedge: latitude is -30.2333 degrees. So pole is at a zenith angle of (90 minus 30.2333 ~ 60 degrees).