WIYN One Degree Imager

1
WIYN One Degree Imager
2
Filter Changer
3
Demonstration Filters

• Sloan g r and i delivered from Infinite Optics
• Johnson U on order from Barr
• Filter below is a r - 146 mm demo of the 400 mm
  filter process

4
IO gri filters
5
Infinite Optics Experience Summary

• IO is new to the filter business, but has a lot
  of experience with complex and exotic thin film
  designs (e.g. broad band AR coatings)
• Small company in southern California, surviving
  largely on diverse commercial contracts (e.g.
  colored lighting displays for Disneyland)
• Very interested in the astronomy filter and AR
  coating business
• Did the AR coatings for PanSTARRS corrector
• The WIYN filters probably cost IO 2-3 times what
  we paid (typically 5000 per filter) for their
  investment in engineering and testing
• Capable of producing good filters, but still
  developing the art
• Should be kept in the loop for RFPs

6
Strengths of IO

• Filters are extremely uniform across the field
  only in the very corner can a transmission
  deviation be seen
• Filters do not suffer loss of performance at high
  angles of incidence though the usual blue
  shift is unavoidable (i.e., good for fast beam
  like Blanco and non-telecentric optics)
• Excellent red leak suppression critical for
  DECs red detectors
• Adequate in-band performance
• Worked hard to achieve transmission and reduce
  wiggles
• Meets the out-of-band spec
• Their AR coating work (as opposed to filters) was
  very fast (3 weeks) and high quality, as they
  have been doing that for years

7
Weaknesses of IO

• Slow to deliver as they learn about filter
  manufacture
• Some of the filters were domed, up to about 10
  waves across the 146 mm clear aperture
• These were later replaced by filters that were
  good to 1-2 waves.
• Partly, this is WIYNs problem by requiring thin
  filters (10 mm)
• Small shop can be distracted by other contracts
  unless large dollars are involved
• Larger filters are still an unknown item they
  purchased a large chamber that can do the job (as
  well as AR coating of large optics) but it has
  not been turned on yet.

8
Plots of the IO r filter
Filter sampling points p8 is in the far lower
right beyond p7
9
Specifications Bandpasses

• We may wish to modify the y filter since our
  detectors are not so red sensitive

10
Low Resistivity Detector QE
11
High Resistivity Detector QE
12
Total ODI Throughput

• Includes 3 mirrors, optics and coatings, low
  resistivity CCDs, no filter

Blue edge is defined by ADC glass and coatings,
red by CCDs
13
More Specs Out of Band Blocking

• Outside the main bandpasses, the following
  maximum transmissions cannot be exceeded.
• Between 300 and 1200 nm filter transmission must
  be less than 10-2 (1 maximum transmission)
  outside the filter passband.
• Between 300 and 1200 nm, the total out of band
  light cannot exceed 0.25 of the total in-band
  light.
• That is The integral of the transmission
  blueward and redward of the passband must be
  smaller than 0.0025 times the integral of the
  passband transmission. We define the limits of
  the integral at the passband as the point where
  the filter transmission is 0.1 (0.001)
• This definition is more meaningful from an
  astronomy standpoint than a simple peak
  violation, and is easier for the vendors to meet.
• See full WIYN specs at http//www.wiyn.org/ODI/fi
  lters.htm

14
Testing

• Filters need to be tested!
• For acceptance
• About every 1-2 years as they age (drift to the
  blue)
• PanSTARRS has not tested their Barr filters
  only witness samples.
• Kitt Peaks Coude Feed telescope can serve as
  test facility it is very rarely used, and KPNO
  will allow us to use it
• Provides a perfect platform for filter testing
  it is essentially a giant spectrograph with an
  f/13 sampling beam.
• Need the following
• XY stage to obtain spectra through filter at any
  spatial position
• Rotational stage to obtain spectra at various
  incidence angles allows simulation of f/3 beam
• Red-sensitive CCD to check for red leaks (can be
  cosmetically poor)
• Software to automate process

15
PanSTARRS Filters g
16
PanSTARRS Filters r
17
PanSTARRS Good Enough?

• Spatial variations are a concern
• Red leaks become serious at 1100 nm is that
  okay?
• Both DEC and WIYN can put a second filter in
  parallel to block leaks
• But, changes optical design optimization and
  reduces throughput
• Cost seems acceptable 60-80K
• Delivery times seem acceptable 1 year

18
Next Steps

• WIYN wishes to initiate RFP in 3-4 months
• Does DEC wish to coordinate?
• Multiple filters of same type are always cheaper
• Must agree on specs
• WIYN complement 8 filters
• griz minimum first batch
• Narrow band 500.7 and 656.3 nm (5 nm FWHM)
  second batch
• Then, Y and U bands third batch

19
Internal Ghosting

• Multiple ghosts extreme example
• Seen with g and i, but not r
• Example at f/3 at WIYN f/6, not as bad
• Evident only around bright stars
• Diameters indicate separations of 1 mm
• Filter are 10 mm in laminated layers
• g has 5 1 mm layers
• r and i have 4 layers
• All have external covers, 2-3 mm
• LESSON! need specification for internal ghosting
• Has been seen occasionally over the years with
  interference filters at Kitt Peak