Simple and cheap near IR conversion of the wide-angle HD 808#16D camera

edited February 2013 in General
I thought that this was worth pulling out of the HD keychain cam thread:,
especially as it is easy to confuse the different keychain models.


A HD 808#16D video still with the camera's rear hot mirror replaced with a £7, 6mm diameter 830nm IR pass filter (LPF-830-0601) as identified by David Sykes:


You screw out the lens, prize out the hot mirror and then push in the 6mm IR filter and then screw back the lens. Focusing obviously varies as you screw the lens in and out. Keep everything clean!!!

I will add a small focusing scale to the lens, to sharpen up the focus.

I did this with the camera taken apart but it should be possible to do it without dismantling the body of the camera (I will have a go soon).

I have worked with a range of near IR cameras but, despite its limited resolution, I find this camera most pleasing and, weighing only 18g, of amazing potential!!!

Added 16 Feb:
Without the sun and better focused (look at the windows in the distant houses).

Near IR KAP is now a technique that can be used with kids without worrying about cost.

The converted camera is ideal as a stick-on auxiliary camera.
You will need a class 4 (or above), micro SDHC card.

Manual and software for interval mode, exposure etc:


  • Impressive John
  • Remarkable IR sensitivity. T get this sort of shot best I could do with my old Olympus was 1/2 sec. And this is a frame grab.
  • I can get my D100 nikon up to 1/20th on a sunny day with r72 but its close to a kilo in weight ;-(
  • Hey, thanks for this info! And thanks for the links to the particular bits that went into it. I've been hemming and hawing on a #16 keychain camera with the 120 degree lens for another project. This is about to tip the scales. And the 6mm IR filter looks wonderfully affordable!

  • John, any chance of a how to video or photos showing how you did the convertion. Indicating where and what the hot mirror is and the adding of the filter. I may give this a go in the near future but I've never played with the insides of a any camera before!
  • edited February 2013
    I will photograph the next conversion (unless someone beats me to it), hopefully without taking the camera apart.

    When the circular part of the lens assembly, which is lightly glued in place, is screwed out of the sensor block, a pink square filter can be seen at the back of the removed lens. This is removed as cleanly as possible and the 6mm IR pass filter pushed in to replace it.

    When the lens is removed it should be pointing downwards so nothing falls onto the sensor.

    If the lens or sensor is damaged, a replacement lens/sensor unit is cheap ($12/£8 inc. postage):

    After doing this, you will have no idea how far to screw in the lens to get images into focus.
    I drew some lines 2mm apart along the edge of a sticky label.
    I then cut off the edge of the label and stuck it to the lens:
    The lens was then turned 2mm (and then later 0.5 mm) at a time while recording a video (you can use stills).
    The video was then viewed and the position corresponding to optimum focus at infinity obtained.
  • edited February 2013
    Not much on the TV tonight, so here goes:

    Above, the lens has been screwed out of the camera and placed rear (hot mirror) side upwards.
    The 830nm IR filter is on the left in the bag.

    Lens with hot mirror in place

    Lens after the hot mirror (which broke) has been prized out with a small screwdriver.

    Lens with 830nm near IR pass filter ready to be pushed into place...looking a bit grubby but easily cleaned....make sure that the lower surface is spotless before pushing into place!
    The exposed sensor can just be seen in the camera.

    Lens screwed back into place with the back of the camera removed (after taking out two screws at the rear end).

    I tried to screw the lens back into place without opening the camera but, as luck would have it, I did not have it perfectly aligned and the thread stuck. So, off came the back enabling me to visually align the lens!

    You then need to calibrate the lens for focusing as described in my posting before this one.

    In general use, I do not know if the filter will remain in place without using glue.

    17 February
    I tested the above second conversion today.
    Focusing was not uniform across the frame.
    I do not know if this is because I displaced the lens when removing the hot mirror or because of off-alignment threading.
    I took the 830nm filter out of the lens and put it in a spare lens assembly. All worked fine.

    Follow David's advice and use a scalpel (or scalpel blade as I did this 3rd, last time - the hot mirror came out in one piece) to gently remove the hot mirror.
    Screw back the lens into the sensor mount very gently and with the back off the camera.

  • Not yet taken mine apart, but it might be easier to adjust the focus by connecting the camera's live video-out to the video-in on a TV.
  • edited February 2013
    Very true if you can get something suitable to focus on, or have an extension lead ;o)
  • edited February 2013
    I Initially only received one of the two filters from
    The salesman explained someone else had also ordered that filter and he got the paperwork mixed up.
    I think I know who ordered it.
    My 720nm filter has still not arrived nine days after allegedly being dispatched, I have informed them.

    So, in order of difficulty of removing the IR-blocking filter are (easiest first) the B, D and A.
    For the D lens you need care and a brand new scalpel blade.
    For the A, you need extreme care, a stereo microscope and a brand new scalpel blade.

    Received another camera today.
    I have two connected to a single 3.5" car reversing monitor.
    They are initially synchronised for movie capture to 1/5000 sec.
    That drifts to 1/1000 sec after three minutes.

  • edited February 2013
    I ordered two filters and only received one too.
    When questioned they said they had found the other and wondered who it belonged to!!

    Where did you order the 720nm from?
  • It is Ebay item # 190731082763

    They have added £60 to usual price presumably because they are out of stock, a typical Ebay practice.

    I will try to cut it with a diamond wheel on a Dremel-type tool or use a diamond scriber.

    With the A and B lens, the (intact) IR-blocking filter can be mounted in a lens cap to restore normal function.
  • edited February 2013
    John many thanks for the instructable, seems simple enough.
  • edited February 2013
    If it was difficult I would not do it ;o)
    Macular degeneration and arthritis limit my dexterity.

    Just watch out for that easily damaged thread!
  • This is tedious, have still not received the filter so they are refunding my money.

    Local post has just been reorganised so it may well have been lost.
  • edited February 2013
    .....and flown velcroed to a helium-filled rubbish bag by West Lothian Archaeology Group member, Heidi Walker:


  • That is powerfully cool.

  • Hello, I have a question about 808 and IR filters.
    I see your aerial images are very detailed.
    But I want to ask if, with your IR-pass filter, you noticed some tipical "magic tricks" of IR photograpy. For example the famous "x-ray effect" described here:

    Most of the amazing facts are the capability to see througt the fog, or to see an object covered by a paper fold... did you noticed something like this with your modified 808?
  • I haven't tried that with a modified 808, but I don't see any reason why it shouldn't work. It's just a function of wavelength. Fog scatters heavily in the blue, and less in the red. Ansel Adams used Kodak HIE high speed infrared film to take advantage of this and cut through haze at least a decade before I was born. Same thing works great now using IR converted detectors.

    I liked the one with the tinted car window. I hadn't thought about it, but yeah, I bet most window tints stop just past the visible, around 700nm. I'll have to play with that.

    But it's not going to be universally true, of course. Different tints will cut out at different wavelengths. I got my butt kicked earlier this week for something like this. We're putting two cameras together at work, and just installed the final detectors earlier this week. It was my job to measure the cameras, adjust the detector-to-reference distance, and tilt them in flat. Easy! Stick the thing on our laser metrology station, and let 'er rip! But... The anti-reflection coating on the CCDs was so efficient, even out at the IR wavelength of our laser, I got no return spot AT ALL. It's like the thing was a black hole. I'm still trying to find a way out of this quandry.

    But hey, it's Friday. I could seriously use a weekend of playing with IR cameras.

  • Infragram: the Infrared Photography Project

    What is Infragram?

    «Infragram is a simple, affordable near-infrared camera produced by the Public Laboratory community in a series of collaborative experiments over the last few years. We originally developed this technology to monitor wetlands damages in the wake of the BP oil spill, but its simplicity of use and easy-to-modify open-source hardware & software makes it a useful tool for home gardeners, hikers, makers, farmers, amateur scientists, teachers, artists, and anyone curious about the secret lives of plants.»

    They also modify compact cameras to near infrared, replacing the IR sensor filter, with a "special" cheap polyester filter ("Superblue"), apparently with good IR blocking capabilities :
  • edited June 2013
    I am looking forward to a visit here in Scotland by Mathew Lippincott, the Public Lab's Director of Production, in about 3 months time.

    I have not tried any special effects with the converted 808.

    I remember reading in a childrens encyclopedia, around 1958, that the fire service had units for use in fog! They leaned out of the window to use it....... if I remember correctly.

    The long awaited 6mm, ultra violet filters (UG11) have now arrived in the UK and will be available after they have been given a protective coating. I will report if the filter works with the 808 as soon as I have used it.

    We hope to fly three 808s together (~55g total weight), near IR, near UV and one visible (red, green and blue) for a total cost of about £100.

    Following on from the previous posting, I will have a look to see if there is a 6mm diameter BG3 out there...or have you seen one David (Sykes)?
  • Cool! Looking forward to your report, John.

  • UV still a bit problematic. The UG 11 filter is letting through more near IR than I expected.
    Mainly using the 808#16D IR as an auxiliary, low resolution camera for picking up non-visible features.
  • Might be another filter you can combine with the UG11 to take out everything above 650nm. I don't know of one off the top of my head, but a look through a Kodak gel filter catalog might turn something up. A single 3" square gel filter would give you enough material to convert dozens of cameras.

    (I just wish there was a gel equivalent of UG11...)

  • benedict
    Some BG filters will cut out the near IR (
    I would have to mount it on the front of the lens, so something a bit bigger than 6mm dia would be ideal, or as you suggest cut gel.

    I seem to spend most of my time these days preparing KAP kits for distribution, but will eventually get back to more camera work and flying.
  • Hi guys,

    Has anyone done the same with the Mobius Actioncam ?
  • Carl Lipo modified a Mobius ActionCam with a Rosco blue filter and took some photos of plants. It looked like there was a lot of NIR getting into the other color channels, so computing NDVI was going to be hard. This is a problem with a lot of CMOS cameras. There might be a way to get some good plant health information from such a camera, but NDVI might not be the way to present it.
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion