compareeHub

YOU CAN BUILD A LENS FASTER THAN ANY CAMERA LENS EVER SOLD

A microscope objective oil-coupled to a bare camera sensor gathers light faster than any camera lens ever sold — at the cost of a few destroyed sensors along the way.

by Ben Krasnow

FULL CAD BOM FIRMWARE DOCS

ScienceOpen-hardware

difficulty
●●●●○
time
a weekend
license
license not specified
repo
repo 0 stars

WHAT YOU’LL NEED

Videos
1
Jump to section

COMPAREE VERDICT

This is not a project with a repository. It is an optics experiment documented on video, and you have to work backwards from what is shown on screen. Ben Krasnow removes the protective cover glass from a CMOS image sensor and uses immersion oil to couple the bare sensor directly to a 40X, 1.3 numerical aperture oil-immersion microscope objective, which works out to about f/0.38 — faster than the f/0.7 Zeiss lens Kubrick used for Barry Lyndon, the fastest lens used commercially. The objective sits in a kinematic mount on a fine-pitch slide, in a rigid 3D-printed carbon-PLA frame around an IDS industrial board camera. Side by side with an f/1.4 lens it is clearly brighter, but the surprise is that depth of field is huge, because the objective's focal length is only about 4 mm, so there is no dreamy cinematic look. The one thing most likely to go wrong is the glass removal: he went through about eight cameras and only around two survived, because the bond wires sit inside the epoxy bead.

GOOD TO KNOW

  • —This is a filmed experiment, not a project with files to download.
  • —The video explains the optics and shows the adapter assembly on camera.
  • —No CAD, no parts list, no bill of materials in a repository.
  • —The objective itself is a standard microscopy part; the adapter and camera mount are shown but not documented step-by-step.
  • —No licence applies because there is no repository.
  • —You are reverse-engineering an experiment from a 16-minute video.

Parts to buy

3 items

From our check of the build. Exact quantities and part numbers are in the creator’s BOM.

  • High numerical aperture oil-immersion microscope objectiveFind
  • Immersion oilFind
  • Camera whose sensor cover glass you can removeFind

BUILDS OF THE WEEK

Five open-source builds worth your weekend, every week.

Checked like this one: what’s really in the repo, what it costs, how hard it is. One email, unsubscribe anytime.

Can I build this?

PrintA rigid frame (he printed it in carbon PLA) that holds the board camera, a fine-pitch slide and a kinematic objective mount — you design it yourself from what is shown in the video.
BuyA high numerical aperture oil-immersion microscope objective (a 40X, NA 1.3 is used in the video), immersion oil, and a camera whose sensor cover glass you can remove (he used an IDS UI-5584LE industrial camera).
ToolsA 3D printer (stiff filament such as carbon-filled PLA) or machining tools for the frame, a fine-pitch slide and kinematic mount, a heat source for softening the cover-glass epoxy, and SMD rework tools if you replace sensors.
SkillsYou need to understand optical principles (numerical aperture, f-number equivalence, immersion oil), have machining or precision 3D printing experience, and be comfortable reverse-engineering mechanical assembly from video.
TimeA weekend — an afternoon to source and prototype the adapter, and the rest of a day to assemble, align, and test it.
CostMid-range budget — the oil-immersion objective is the main cost, plus several spare board cameras or sensors because glass removal often fails; the video gives no prices.
SafetyNone beyond ordinary workshop care. Immersion oil is safe to handle. No mains voltage, no lasers, no sharp blades.

Build at your own risk. Projects involve tools, electronics and sometimes mains voltage — follow the creator’s safety notes.

Videos

Creator’s build video

The entire experiment is explained and demonstrated on camera, including the side-by-side brightness comparison and the optical theory behind numerical aperture and f-number conversion.

More builds like this

All projects
You can build the microscope that films an embryo growing for daysScienceLaserTweezer: an open-source optical trap built from a DVD pickupScienceYou can build the machine that 3D prints heart tissue — FRESH bioprinter conversion for under 900 dollarsScience

Gallery

youtube.com
youtube.com
youtube.com
youtube.com

Start here

Navigation into the creator’s own docs — we don’t rewrite the guide, we route you to the source.

  1. 1.Watch the video and note the objective spec (40X, NA 1.3, oil immersion), the camera used, and how the frame, slide and kinematic mount are arranged. (The objective's specification and the theory are explained on camera, and the camera and sensor are named in the description. The adapter is shown in close-up but not documented with files.)
  2. 2.Source a 40X, 1.3 NA oil-immersion microscope objective and immersion oil.(The objective is a standard microscopy part. Used ones appear on eBay and lab surplus sites.)
  3. 3.Build a rigid frame with a fine-pitch slide and a kinematic mount to hold and align the objective over the bare sensor.(The adapter shown in the video is custom. You are making your own from what is visible on screen.)
  4. 4.Assemble, apply immersion oil between the objective and the bare sensor, and check focus and alignment.(The oil goes directly between the objective and the bare sensor. Use the kinematic mount to align the objective's axis with the sensor, and add an infrared filter in front as he did, because bare sensors are very IR-sensitive.)

KNOWN ISSUES

  • There is no repository, no parts list, and no adapter design to download — you are reverse-engineering an experiment from a video.
  • The mount has to be rigid and adjustable: he used a carbon-PLA printed frame, a fine-pitch slide for spacing and a kinematic mount to align the objective with the sensor; any misalignment shows up as blur or vignetting.
  • Removing the sensor's cover glass is a low-yield process: Ben went through about eight cameras and roughly two survived. Gentle heat to soften the epoxy and pushing the glass little by little worked best; cutting into the epoxy bead destroys the bond wires. Budget for spare sensors or cameras.
  • Do not expect a shallow, cinematic depth of field: with a focal length of about 4 mm the depth of field is huge, from about a foot to the far end of the room, even at f/0.38.
  • A high-NA objective is expensive, and a cheap one will not reach the same speed.
  • The immersion oil must stay between the objective and the bare sensor, and any air bubbles or dust in that layer scatter light and degrade the image.

Is there a repository with files?

No. This is a filmed experiment, not a project with downloadable files. The adapter is shown on camera but not documented.

What is the actual f-number?

About f/0.38, calculated from the 1.3 numerical aperture of the 40X oil-immersion microscope objective. The video explains the conversion.

Can I use this for normal photography?

No. The immersion oil sits between the objective and the bare camera sensor, so you have to strip the sensor cover glass and build a rigid mount. It is an optics demonstration, not a practical lens for everyday photography.

How much does the objective cost?

The video does not say. High-NA oil-immersion objectives appear used on eBay and lab-surplus sites; prices vary widely with brand and condition.

Community builds

No community builds yet — be the first, we feature the best ones.

Discussion1

FROM THE COMPAREE TEAM

Getting to f/0.38 means removing the sensor's cover glass and putting oil directly on the bare chip. Would you sacrifice a camera sensor to try it, and what would you point it at?

CompareeTEAM1mo agoedited

Practical notes from our verification: there is no repository, no parts list and no design files to download — this is a filmed experiment by Ben Krasnow (Applied Science), and the video description is the main written reference. What it does give you: the setup removes the protective glass from a CMOS image sensor and uses optical immersion oil to couple the bare sensor directly to a 40X, NA 1.3 microscope objective, which is what works out to about f/0.38. The camera and sensor are named in the description (an IDS UI-5584LE industrial camera with an onsemi MT9P006 sensor), along with background reading on converting between f-number and numerical aperture. The biggest challenge is not buying the objective, it is the step you cannot undo: stripping the cover glass off a sensor and putting oil on the bare chip. Treat the camera as a sacrifice. Correction (4 October 2026): we re-checked this page line by line against the project's own repository, documentation and videos, and fixed errors in earlier versions.

Ben Krasnow

Ben Krasnow is an engineer and the creator of Applied Science, a YouTube channel known for experimental builds that push the boundaries of optics, materials, and instrumentation. This video is one of many where he borrows a technique from one field — in this case, microscopy — and demonstrates it in another.

Star the project on GitHub

DISCLAIMER

  • Comparee is not the author of the projects featured here. All rights to each project belong to its creator — every page links to the original source, and we never host creators’ files.
  • Information is provided without warranty and may become outdated as projects evolve. Prices are indicative bands only — always check the creator’s parts list for current costs.
  • Building and operating any project is at your own responsibility. Protective equipment, safe workshop practice and compliance with local regulations are the builder’s responsibility.
GitHub repo