The NEID spectrograph – Optical Fiber Train

Optical fibers are thin waveguides made of glass and are used for transportation of light. Widely used in the telecommunication industry, in the 1980s they started to be used in astronomy to couple light from the telescope focus to highly specialized and sophisticated instruments located elsewhere, eg. in the observatory basement (Heacox 1986).

Fibers typically have three layers, the core, cladding and buffer (Figure 1).Fibers and operate on the principle of total internal reflection (TIR) : the core and cladding are made from glass, and the refractive index of the core is higher than that of the cladding to enable TIR. The role of the buffer, on the other hand, is to provide protection to these fragile glass strands.

Figure 1 : Optical fiber layout

For an extreme RV precision instrument like NEID, optical fibers offer various advantages, some of these are : –

  1. By decoupling the instrument from the telescope focus, they allow for the instrument to be quite heavy and sophisticated (NEID weighs about a couple metric tons!) . Further, it allows the instrument to be housed in the basement of the WIYN telescope in a room with temperature control to the level of 1 degree Celsius or better.

    Figure 2 : WIYN basement showing the location of NEID, the laser frequency comb and the electronics rack/s.

  2. Having the instrument in the basement also means a near constant gravitational vector for the instrument. An instrument at the Cassegrain focus of the telescope would move constantly as the telescope tracks the target. This is important to prevent a variable sag or bending of the optics or mounts in the instrument.
  3. The fiber provides a degree of scrambling for the light. This means that illumination at the output end of the fiber is less sensitive to the the location of the star on the input face. This is important to average over any illumination changes or guiding errors of the telescope (See Halverson & Roy et al. 2015).

NEID Fiber Train

Shown below is the entire fiber train for NEID. Here, we shall go over some of the components.

Figure 3 : Overview of the NEID fiber train

Fiber puck and polishing

NEID has two different operating modes, High Resolution (HR) for the brighter targets for maximum RV precision, and High Efficiency (HE) for the fainter targets, with higher throughput. To enable this, the HR mode has smaller fibers (64 micron core) while the HE mode fibers are bigger with ~100 micron cores to allow for higher efficiency.

For the HR mode we have three fibers – Sky, Science and Calibration ; whereas the HE mode has only Science and Sky (bigger fibers take up more detector real estate). The Science fiber, of course, is where the stellar light enters the instrument. The Calibration fiber in the HR mode is used for simultaneous wavelength calibration of the stellar spectrum. The Sky fiber is used for a variety of telluric corrections.

This combination of HE and HR mode fibers are inserted into a fused silica puck manufactured by Femtoprint using a combination of a femtosecond laser and hydrogen fluoride etching. This is shown in Figure 4.

Figure 4 : F. Silica puck from Femtoprint. It shows the 5 bore holes for the 3 HR and 2 HE fibers, along with their conical inserts to ease the process of fiber insertion.

After inserting the fibers in the puck they are epoxi-ed into place using Epotek 301 – 2;  once this epoxy dries, we polish the puck.  For the polishing we use a custom made polishing jig which attaches the puck shown above to the polishing arm on the table. It then moves in a ‘figure of 8’ pattern to polish the fiber. To do this we start with a coarse 60 um grit, and progressively move to smoother grits as 12, 5, 3, 1, 0.3 um grits of silicon carbide, aluminium oxide and diamond.

Figure 5 : Video clip showing the polishing mechanism

Once we polish the puck, and grind away all the epoxy, the resultant puck looks like shown below in Figure 6. We need to get a good surface smoothness to prevent FRD deterioration from roughness on the end surface of the fiber.

Further for NEID we plan to stick a plano – convex lens onto this polished puck surface as part of the input optics focal ratio changer, and to protect the fiber surface from damage.

Figure 6 : Polished NEID puck

Fiber Splicing

The fiber that goes inside the instrument in this puck is circular. However, the rest of the fiber train is octagonal. Therefore we splice the circular fiber on to the octagonal fiber face, which basically means fusing them together.

To do this, we first strip away the polyimide buffer shown in Figure 1. We do this using a 3SAE plasma arc fiber stripper. This creates a plasma arc between three electrodes located 120 degrees apart. The fiber is placed between them and slowly gets stripped off of its buffer. Shown in Figure 7 is the fiber being stripper for the HPF fiber.

Figure 7 : Clip showing an HPF fiber being stripped

After this we use a diamond blade cleaver to cleave (cut) the stripped fiber.  The stripped fiber has about a 4 -5 mm patch of exposed fiber (only core and cladding). This is then spliced on to the patch of fiber using a device similar to the stripper – it uses three electrodes to create a plasma arc which fuses the two ends together. Below in Figure 8 is shown a test fiber after being spliced. Figure 9 shows the spliced HPF fibers in a splice sleeve and heat shrink tubing to protect the splice.

Figure 8 : Splice test for NEID fibers

Figure 9 : Splice protection for the HPF fibers. NEID fibers will be similar

Modal Noise Agitator

The optical fibers being used for HPF and NEID are multi-mode, and hence support a finite number of modes. Hence the light emerging from the fiber contains an interference pattern between the many modes traversing across the fiber. This is called modal noise, or “speckle noise”. For big fibers, there are enough modes that the statistical variation in the change in energy distribution of modes is insignificant. However, in the presence of smaller fibers as in NEID, or the use of slits or fiber slicing, modal noise needs to be mitigated for the measurement of precise radial velocities.

The HPF science and sky fibers are agitated using the device shown below; whereas for the calibration fiber we use a combination of a mechanical agitator and an integrating sphere. This is because the problem gets exacerbated for the calibration fiber which uses the laser frequency comb – a highly coherent light source.

For an introduction to understand Modal Noise, refer to – Baudrand and Walker 2001,  McCoy et al. 2012, Mahadevan et al. 2014.

Figure 10 : Clip showing the fiber agitator for HPF. NEID will use a similar one.

Double Scrambler – Ball Lens

Fibers provide some degree of scrambling the spatial location of the input spot (Heacox 1986), however for precision RV application, this needs to be improved by many orders of magnitude.  For HPF and NEID we use the ball lens described in  Halverson & Roy et al. 2015. Basically, our scrambler is a simple ball lens with refractive index ~ 2, which exchanges the far and near field of the fiber.  The two ends of the fibers are placed inside a v – groove block (as shown in Figure 11), and butt coupled to the ball lens which is in the middle.

This ball lens is based on the double scrambler concept proposed early in the days of fiber-fed spectroscopy by Hunter and Ramsey (1992) at Penn State!

Figure 11 : Showing the ball lens setup during prototyping. Figure Credit : Halverson 2015

Figure 12 shows how the scrambling performance improves for an octagonal fiber over a circular, further the gains for the double scrambler, and for the spliced system.

Figure 12 : Showing the scrambling gain.
The top panel shows the input face of the test fiber. Whereas the bottom one shows the near field. Figure Credit : Halverson 2015.


This entire system forms the instrument optical fiber package. There are some additional systems involved for the acquisition camera as well as the exposure meter, which are not discussed here. Essential to ‘carry’ the photons from the telescope or calibration sources into the instrument, the fibers are like the circulatory system of the instrument! (In this analogy, the electronic system is perhaps the nervous system, communicating information from the sensors and heater to the system computer.)

Posted in NEID Hardware | Leave a comment

Under Construction Already!

It has been a busy summer for NEID and its team.  Now that the project has officially started, there are many tasks that must be done in a short amount of time.  Plans must be made, parts must be ordered, and blogs must be written!  Out of all this excitement and bustle, though, there is one important takeaway: NEID construction is underway!  Today we will briefly highlight the effort our summer students have put into fabricating NEID’s environment control system.


In order to achieve the sub-meter-per-second Doppler velocity measurement precision required of NEID, there are dozens of sources of instrumental error that must be controlled as tightly as possible.  One of those variables is the thermo-mechanical stability of the spectrograph and its optics.  As the temperature changes within the instrument, the glass optics or their mounts will change shape or position slightly.  Temperature and pressure changes will also affect the index of refraction of any air in the instrument, altering the motion of photons.  The impact of these environmental changes on our measurements of the stellar spectra can be much greater than the changes we expect to see from orbiting exoplanets, so they must be mitigated!

Like its predecessor spectrograph the Habitable-zone Planet Finder (HPF), NEID will be outfitted with an environmental control system (ECS for short) to keep the temperature and pressure within the spectrograph as stable as possible.  The entire spectrograph will be surrounded by a temperature-controlled aluminum box called a “radiation shield,” which itself is encased in a high-quality vacuum chamber.  The high vacuum eliminates heat transport by convection, making it much easier for us to control the temperature of the radiation shield—and thus the instrument itself—to a very precise degree.  On the HPF blog, we showed that by using the HPF ECS to simulate conditions for NEID, we could control the instrument’s temperature to better than a thousandth of a degree!  This level of precision is required to make the instrument sensitive to Earth-like planets.

The HPF and NEID instruments, with critical elements of the environmental control system highlighted.

The HPF and NEID instruments, with critical elements of the environmental control system highlighted.

Another requirement for NEID is that it must fit within approximately the same physical dimensions as HPF in order to be housed in the WIYN Telescope’s instrument room.  As a result, the NEID vacuum chamber and radiation shield will look very similar to HPF’s.  The image above shows the two instruments, highlighting some of the similarities and differences.

The vacuum chamber is the first major component that must be fabricated for the NEID project.  The instrument has to have a home, after all!  Thus, while the vacuum chamber itself is being built by our collaborators at Pulseray, the team at Penn State has been busy working on everything that goes inside it.  Let’s take a quick look.

Let’s build it—again!

Because NEID utilizes nearly all of the technology we developed for HPF, it is much quicker and easier to build a second system.  Many of the growing pains of achieving sub-milliKelvin temperature control have already been experienced, and we can simply get things right the first time.  With that said, this is still no easy job!  As shown in the figure above, NEID has twice as many active heaters as HPF, and the sheer amount of cabling required to operate so many heaters and thermometers is daunting.  The optical bench is also larger for NEID, leaving less space to work with during assembly, so the design and application of the multi-layer insulation must also be carefully considered.

Students Demetrius Tuggle, David Conran, and Joe Smolsky assemble a wiring harness for the NEID ECS.

Students Demetrius Tuggle, David Conran, and Joe Smolsky assemble a wiring harness for the NEID ECS.

One advantage of building an instrument at a world-class university like Penn State is that there is never a shortage of talented young scientists to help with major projects.  We are fortunate to have four undergraduates—David Conran and Adam Dykhouse from Penn State, Joe Smolsky from the University of Nebraska at Omaha, and Demetrius Tuggle from the Ohio State University—helping us to fabricate the NEID ECS over the summer.

With the help of these hardworking students, fabrication of the environment control system’s heaters, thermometers, control electronics, and the half-mile or so of associated cabling is nearly complete!  Assembly of the many MLI blankets is now underway as well.

Gudmundur Stefansson supervises the assembly of an MLI blanket.

Gudmundur Stefansson supervises the assembly of an MLI blanket.

Moving so fast on the first major component of the NEID instrument is not just critical for completing a subsystem before classes start again in the Fall.  It is essential for keeping to NEID’s aggressive project schedule and having the instrument ready to follow up all of the exciting exoplanets that NASA’s Transiting Exoplanet Survey Satellite (TESS), to be launched in 2017, will discover.

Posted in NEID Hardware, Project Development | Leave a comment

NEID Featured on the Many Worlds Blog


As part of NASA’s Nexus for Exoplanet Systems Science (NExSS), science journalist Marc Kaufman documents new and interesting research related to NASA’s various exoplanet science programs on the Many Worlds blog.  His latest post features NEID, and its role in the worldwide effort to discover and characterize Earthlike planets.  Have a look!

In addition to keeping you updated on the latest news related to the NEID spectrograph’s development, we will occasionally highlight some of the scientific research efforts from our team that are especially pertinent to the instrument.  For a broader look at state-of-the-art exoplanet research, add a bookmark for Many Worlds!

Posted in Media Coverage, NEID Background | Leave a comment

NEID: The Introduction

Welcome to NEID!  Here, we will document the development and deployment of a new high resolution planet-finding spectrograph that will be installed in 2019 on the 3.5-meter WIYN Telescope at Kitt Peak National Observatory near Tucson, Arizona. This instrument will be open to all astronomers, who can propose to use it through a peer-reviewed time allocation committee.

The instrument is named NEID – derived from the word meaning “to see” in the native language of the Tohono O’odham, on whose land Kitt Peak National Observatory is located. NEID also is short for “NN-EXPLORE Exoplanet Investigations with Doppler Spectroscopy.” NEID will detect planets by the tiny gravitational tug they exert on their stars.

NEID is being developed as part of a joint initiative between NASA and the National Science Foundation called NN-Explore, the purpose of which is to discover and characterize planets around nearby stars (so-called “exoplanets”) using the Doppler (or Radial Velocity) technique.  The Doppler method of exoplanet discovery relies on making ultra-precise measurements of a star’s line-of-sight motion to detect the tiny gravitational perturbations of the planets that orbit that star.

NEID is being designed and built by a team of astronomers and engineers led by Principal Investigator Suvrath Mahadevan at Penn State University.  The NEID instrument team is comprised of members from Penn State, in addition to collaborators at the University of Pennsylvania, Macquarie University, the Physical Research Laboratory in India, the National Institute of Standards and Technology/University of Colorado, and NASA’s Goddard Space Flight Center.

Team Photo

The NEID Instrument Team

Be sure to come back to this space soon, as we will be bringing a lot of exciting information about NEID’s purpose, design, and construction, as well as highlighting some of the cutting-edge exoplanet research that this instrument is designed to support.

Posted in NEID Background, Project Development | Leave a comment