ElectroHolography Laboratory

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Custom Polygon Mirror ======

===== Background =====

One of the objectives of the HoloMonitor is to be cheap. To this end, consumer hardware is repurposed as much as possible to lower costs. The horizontally-scanning optical element of the Mark V (as of Tue 23 May 2017) is a polygon mirror from laser printers:

{{monitor:laser_printer_back.jpg?400}}

{{monitor:polygon_mirror_with_vibration_absorber.jpg?400}}

As shown, this uses a six-sided polygon mirror, resulting in a 120° sweep of incoming lasers. We don't utilize all of this sweep range. Testing shows that we only utilize the center third section of this range with current apertures (as of Tue 23 May 2017).

The laser scanner's facets measure 20 x 2.85 mm, and the vertex-to-vertex diameter measures 77.2 mm.

----
===== Custom-Machined Polygon Mirror =====

Having the capability to machine custom polygon mirrors would open up a useful degree of design flexibility to our research group.

For the Monitor in particular, having a higher-count polygon mirror would provide two benefits:

  - a greater number of vertical lines that can be scanned with minimal changes to hardware/software
  - a reduction in unused HoloMonitor output area

The angular range that light is swept across is equal to ''720°/N_facets'', where ''N_facets'' is the count of facets in the polygon mirror. As an example, polygon mirrors of 6 and 12 facets are shown below, with the resulting sweep range:

{{monitor:sweep_range_example.png}}

We want to maintain the facet size so that we can maintain the angular output range of the AOMs. Although a 12-sided polygon mirror wouldn't completely eliminate wasted output, it would double the number of scan lines compared to a 6-sided polygon mirror, and would be smaller than a 24-sided mirror with equal facet size. 

Polygon mirrors are difficult to manufacture for three requirements:

  - a near-mirror finish
  - regular facet geometries
  - radially-symmetric balance

1) is required to prevent scattering of the AOM output, which would add to background noise.\\
2) is required to have lines be regularly spaced (vertical component of facet normal) and start at the same position (horizontal component of facet normal)\\
3) is required to prevent vibrations at high rotational speed (~15,600 RPM).

Files and folders related to this development are in the J:\ drive under ''J:\groups\holography\~Monitor Development\~Custom Polygon Mirrors''.

----
===== Personnel / Shops =====

==== Kevin Farr ====

{{monitor:kevin_farr_diamond_tooler_contact_info.jpg?400}}

Kevin Farr is an application engineer for [[https://www.kennametal.com/en/home.html|Kennametal]], a company specializing in industrial machining solutions. He was referred to us by Nick Hawkins (see below). Kevin has worked with polishing metal before.

==== Therin Garrett ====

{{https://pml.et.byu.edu/sites/default/files/therin_-_new_med.jpg?150}}

''therin@byu.edu''

[[https://pml.et.byu.edu/content/therin-garrett|Therin Garrett]] is manager of the [[https://pml.et.byu.edu/|Precision Machining Laboratory (PML)]], a machine shop which is on campus but is a separate entity than BYU. It is staffed by professionals, and students are not allowed to work on the equipment.

==== Nick Hawkins ====

{{https://me.byu.edu/sites/default/files/nhawks.jpg}}

''nick.hawkins@byu.edu''\\
''801-422-6297''
''CB 150''

Nick Hawkins is the Projects Lab Supervisor for the Mechanical Engineering Department's Clyde Building [[https://me.byu.edu/content/cb-150-projects-lab|Project Lab]] (Machine Shop), in CB 150.

----
===== Development History =====

We first started looking into upgrading the polygon mirror in February 2016, and began looking into making them ourselves the month after.

Methods we've attempted to some degree:

  * 3D printing
  * casting
  * grinding
  * machining
  * laser-cutting

**As of Tue 23 May 2017, machining is the most promising solution so far**.

A [[http://www.lincolnlaser.com/files/Lincoln-Laser-PolygonsTech-Paper-2-17-2015.pdf|Handbook of Optical and Laser Scanning]], published by Lincoln Laser, talks about polygon mirror fabrication techniques in Section 4.4, pg. 253-4 (pg. 9-11 of the PDF). It discusses two techniques:

  * Conventional polishing - uses a polishing compound and a grinding wheel
  * Single Point Diamond Turning - uses a single-crystal diamond-cutting tool with air-bearing mills

Videos of polishing examples:

**[[https://www.youtube.com/watch?v=XwKCALyJ7r8&list=PLpwEnRRY3DEcRmjLL8cQK1Gntx07cBmjr&index=60|Single Crystal Diamond Tool Cutting on CNC Ultra precision Lathe]]**\\
[[https://youtu.be/niop1HkuzdY|Kugler Flycutter F3000-2X]]\\
[[https://www.youtube.com/watch?v=jxWbz0Wf2rM&list=PLpwEnRRY3DEcRmjLL8cQK1Gntx07cBmjr&index=36|Diamond turning an acrylic dome]]\\
[[https://www.youtube.com/watch?v=xkiKyc0Bf9k&index=21&list=PLpwEnRRY3DEcRmjLL8cQK1Gntx07cBmjr|Diamond flycutting of plane mirror surface]]\\
[[https://www.youtube.com/watch?v=7na8kQ78vkQ|Optical finish for acrylic -- vapor polishing and other techniques]]

==== 3D Printing ====

We printed a couple of different polygon mirrors using the FormLabs 1 printer. Unfortunately this left little bumps of extra resin on the prints.

The FormLabs 2 has higher resin exposure precision than the FormLabs 1 printer, but individual ridges are still visible. The FormLabs 2 has a minimum layer thickness of 25 micrometers.

==== Casting ====

We printed polygon mirrors in castable FormLabs 1 resin for casting with aluminum, but bumps in the resin and in the casting process didn't merit further confidence.

[[monitor:drew_log#wed_2_mar_2016_-_new_optics_table|Drew's Log - Wed 2 Mar 2016]]

==== Grinding / Lapping ====

We're able to regularly polish Lithium Niobate crystals and glass to optical finishes with the Cleanroom's polisher / lapper.

Concerns with the Cleanroom's polisher:

  * polishing planes are different between facets, pads :?:
  * depth isn't easily controlled, so it could become unbalanced
  * the amount of time needed to completely grind grows the finer the grit becomes

[[monitor:drew_log#fri_2_sep_2016|Drew's Log - Fri 2 Sep 2016]]\\
[[monitor:drew_log#thu_26_may_2016|Drew's Log - Thu 26 May 2016]]\\
[[monitor:drew_log#thu_19_may_2016|Drew's Log - Thu 19 May 2016]]\\
[[monitor:drew_log#wed_18_may_2016|Drew's Log - Wed 18 May 2016]]\\
[[monitor:drew_log#mon_16_may_2016|Drew's Log - Mon 16 May 2016]]\\
[[monitor:drew_log#thu_12_may_2016|Drew's Log - Thu 12 May 2016]]\\
[[monitor:drew_log#tue_29_mar_2016|Drew's Log - Tue 29 Mar 2016]]\\
[[monitor:drew_log#fri_25_mar_2016|Drew's Log - Fri 25 Mar 2016]]\\
[[monitor:drew_log#thu_24_mar_2016|Drew's Log - Thu 24 Mar 2016]]\\
[[monitor:drew_log#wed_2_mar_2016_-_new_optics_table|Drew's Log - Wed 2 Mar 2016]]

A good polishing compound is **Mothers® Mag & Aluminum Polish** and can provide near-mirror-like quality. Unfortunately finding the right material to polish against is difficult - too soft and the uneven pressure can cause the aluminum to be polished unevenly into a curve, too hard and it can scratch the aluminum.

==== Machining ====

This is the way most factories produce polygon mirrors.

Drew had Therin Garrett mill a 12-sided polygon mirror in Aug 2016. While the quality was very high, we weren't able to obtain a mirror finish with metal-tipped bits (although some reflectivity could be seen). Therin recommended that we purchase a diamond-tipped tool. Their 12-sided polygon mirror attempt is shown below:

{{monitor:img_9038.jpg?400}}

=== Tue 23 May 2017 ===

Kevin Farr visited BYU and was consulted for recommendations given this projects requirements and the tools available. We discussed various tool heads he had available and the pros/cons of each.

He recommended us using a [[https://www.google.com/search?q=face+mill&safe=active&source=lnms&tbm=isch&sa=X&ved=0ahUKEwjnxt6wt4fUAhWCilQKHfV6BsAQ_AUICygC&biw=1680&bih=944#imgrc=_|face mill]] with at least two tips (aka inserts). The tips would have a height difference of 0.0002" (two-ten-thousandths of an inch). According to him, Aluminum and Titanium have memory, meaning they'll spring back up a little bit after being milled. The 1st mill tip does the normal cutting, and the 2nd mill tip cuts off the part that springs back up again.

Each diamond-tipped insert costs $70-90, and a 4-insert mill head costs $350?. Kevin thinks he could loan us a head for us to use to see if this process would work without air bearings.

He also recommended cooling the metal being cut with **rubbing alcohol** during the polishing pass, something that the PML staff didn't know about.

He also talked about rake angles along the lines of [[https://youtu.be/0b34pkLjIuM|this discussion]] (he showed me those same source videos as well). Aluminum needs to have a positive rake angle of more than 12°.
==== Laser-Cutting ====

Pros:

  * Very easy to cut
  * Very symmetric
  * Can deposit aluminum in the E-Beam evaporator to create a mirrored surface as long as that surface is already mirror smooth

Cons:

  * Jitter from the stepper motors causes the facet surface to be slightly uneven - still requires post-cut finishing before depositing aluminum

=== Wed 24 May 2017 ===
Jesse tried cutting a 12-sided polygon out of acrylic while intentionally overheating the material (cutting at 100% power but only 7% speed instead of the normal 12-15%). Problem is, that causes the center hole to be over cut (so it doesn't mount tightly on the motor and so vibrates when spinning) and actually makes the uneven reforming of the edges even more apparent. If we're going to pursue this route, we should cut as normal, then sand, then flame polish (the shop should have a torch we can use for that). 

----
===== Suppliers =====

  * [[https://www.alibaba.com/product-detail/Original-Polygon-motor-polygon-mirror-motor_60295449081.html?spm=a2700.7724838.30.87.rRPNU8|12-sided Alibaba]]
  * [[https://www.alibaba.com/product-detail/Polygon-mirror-motor-RMOTP0063QSPZ-for-sharp_1938115990.html?spm=a2700.7724857.29.100.Ue2pHa|7-sided Alibaba]]
  * [[https://spanish.alibaba.com/product-detail/fm4-7176-000-ncu-board-for-canon-mf4450-network-board-1938142826.html|7-sided Alibaba]]
  * [[https://www.nidec-copal-electronics.com/e/product/list/017/|Nidec Copal Electronics]]
ElectroHolography Laboratory

User Tools

Site Tools

Page Tools
