At the request of several forum members, i've tried in the post below to bring together all the most important and relevant things you need to know about night vision.
WELCOME TO THE WEIRD AND WONDERFUL WORLD OF NIGHT VISION
This night vision primer is directed at people who already shoot with rifles and want to extend that capability into the hours of darkness without using visible light.
NOTE: This post is very long and quite detailed, so I don't really expect people to read it all at one sitting. Feel free to dip in and out as you please, hopefully picking up little titbits of knowledge as you go.
Night vision (NV) systems break down into two main technologies: tubed and digital.
This is based on amplifying any existing light using analog devices called image intensifiers.
These devices are glass electronic valves (tubes to our American buddies).
Light enters one end of the tube and each photon (packet) of light causes an electron to be emitted from the inner face of the tube.
The electron is accelerated by a high voltage within the tube so that it hits another electrode in the tube causing more electrons to be emitted.
This cascading process continues along the tube such that a single photon of light hitting one end of the tube could produce up to 50,000 electrons at the other end of the tube.
The electrons eventually hit a phosphor coating at the other end of the tube.
The pattern of photons entering the tube is replicated on the phosphor surface at the other end of the tube producing an image of what the tube is pointing at.
In practice, a lens or lenses is placed in front of the tube to focus the scene onto the entry end of the tube and another set of lenses is placed at the phosphor end of the tube to focus the image so it can be seen clearly.
Tubed NV is the original type of night vision and was initially developed towards the end of World War 2 and this development continues to the present day.
Tubed NV is generally categorised by Generation (GEN) ranging from GEN1 thru GEN 2 to GEN 3 with several subcategories in between.
GEN 1 is the cheapest and least capable type of the tubed NV generations.
It has the lowest light amplification, the most distortion of the viewed image and always requires additional illumination.
GEN 3 is the most expensive and most capable tubed NV.
The best GEN 3 equipment requires only the smallest amount of ambient light to produce a clear, sharp image of the viewed scene.
Since tubed NV was, and continues to be, developed primarily for military purposes there are restrictions on the availability of the best GEN 3 equipment and the prices of the better equipment are very high.
Most tubed NV (except the better GEN 3 units) can be easily damaged by exposing the tube to excessive amounts of light, such as daylight or a flashlight, so it's essential that tubes be protected from such light sources, and to this end it's normal to see tubed NV equipment with tightly fitting lens covers which have a pinhole at their centre to allow a small amount of light into the tube so that users can test them during daytime without damaging the tube.
Tubed NV equipment can be purchased as a complete dedicated unit which attaches to the scope rail of the rifle or as a front or rear add-on which attaches to an existing riflescope.
Obvioulsy, a dedicated unit is permanently mounted on the rifle while the add-ons can be removed from the rifle so that it can be used as normal during daylight hours.
Some add-ons can also be used as spotters when removed from the rifle.
Because of the difficulty in obtaining individual parts and the requirements for a very high voltage supply for the tube, DIY tubed NV is almost non existent
Digital NV is a recent technology and is based on the fact that digital cameras have some sensitivity to infra red (IR) light as well as visible light.
In a digital camera sensor, a photon hitting the sensor surface can cause an electric charge in the sensor material to be produced.
This is known as the "Photo Electric Effect"
If enough electric charge is accumulated, an electric current can flow.
How often an electric charge is produced when the sensor is hit by a photon of light is known as the "quantum efficiency" of the sensor and this varies with both the sensor material and the wavelength of the incoming light.
Light is electromagnetic radiation - the same as radio waves, microwaves, nuclear radiation, X rays and cosmic rays.
The only thing the differentiates all of these is the wavelength of the radiation.
Visible light has wavelengths extending from around 400 nanometres (nm) at the violet end of the spectrum to around 750nm at the deep red end of the spectrum.
Wavelengths longer than the red end of the spectrum are what we call infra red (IR)
For digital NV the two most common wavelengths used are 850nm and 940nm.
IR light sources for tubed NV usually have shorter wavelengths, typically 795 and 808nm.
Most digital cameras cannot "see" IR with wavelengths longer than 1100nm
Digital cameras are everywhere, and the technology has advanced rapidly meaning that cameras with very high sensitivity to IR are available at very reasonable prices.
The cameras used in digital NV are essentially cctv cameras designed for use in various security applications.
Digital cameras are described by following main features:
This is usually the diagonal distance from one corner of the sensor to the other corner.
The most common size is 1/3".
There are some 1/2" cameras and also lots of 1/4" cameras.
Bigger is better, but manufactures are producing fewer and fewer 1/2" cameras.
Number of pixels
The camera sensor is made up of many individual light sensing elements called pixels.
A sensor with many pixels will give a sharper, higher resolution image than a camera with fewer pixels.
The sensitivity of the sensor to IR depends on the size of the individual pixels.
Big pixels are more sensitive than small pixels.
Since the sensor is a fixed size (1/3", 1/2" etc), having a lot of pixels to get high resolution means the pixels have to be small, resulting in lower sensitivity.
Conversely, a sensor with big pixels will have good light sensitivity, but will not have particularly high resolution.
At present, the trend in the cameras we use for digital NV is for 1/3" sensors and increasing numbers of pixels combined with digital signal processing of the signals coming from the sensor to try to make up for the loss of sensitivity caused by having small pixels.
Colour or black and white (mono)
All camera sensors respond primarily to the intensity of the light hitting them.
Although light at certain wavelengths will produce more electric charges than at other wavelengths, these variations are not enough to make the sensor distinguish individual colours and shades.
To produce a colour image, a red/green/blue filter mask is placed in front of the sensor which allows only certain colours to hit certain specific pixels.
The sensors' signal processing electronics uses this information to generate a colour video signal.
However, placing anything in front of the sensor will reduce the intensity of light hitting the sensor meaning that mono cameras are generally more light sensitive than colour cameras.
When used with IR, all cameras will produce an mono image since there is no colour information in 850 or 940nm light.
CCD or CMOS
These are the 2 technologies of camera sensor which dominate the cctv camera market today and all the cameras we use in digital NV use either CCD or CMOS sensors.
CCD (charge coupled device) was the first digital light sensing technology developed and is still very common in many of the cameras we use today.
CMOS (complimentary metal oxide silicon) sensors were developed because they use less power and are theoretically cheaper to manufacture.
Early CMOS sensors had poor sensitivty compared to CCD, but are now approaching the performance levels of CCD sensors and in some cases surpassing them.
Both CCD and CMOS sensors convert light to electrical signals in exactly the same way, the differences are in how those signal are processed by the sensors' electronics.
TVL (television lines) is another way of describing the resolution of a cctv camera. It originates from analog television and descibes how many individual lines can be seen in a specified screen area.
Higher TVL means the camera produces a sharper higher resolution image.
In practice, more camera pixels gives a higher TVL value.
Typical TVL values for the cameras we commonly use for NV range from 380-700 TVL with anything from 550-700TVL being regarded as more than good enough resolution.
Since high TVL values mean lots of pixels, it can be the case that a camera with a high TVL value may not have good light sensitivity
Lux is the name of the unit of measuremnt of illuminance.
In simple terms, Lux is a measure of how much light a sensor needs to produce an image.
Obviously a camera with good sensitivity in low light conditions should have as low a Lux figure as possible, and camera sellers frequently quote figures like 0.1 lux, 0.01 lux, 0.0001, lux, 0.000001 lux and even 0 lux.
Clearly these values need to taken with a pinch of salt and only figures quoted by well regarded manufacturers such as Watec and KT&C and qualified by stating the aperture at which the measurement is made are worth placing much trust in.
A common "trick" to claim extremely low lux values is to use a feature of some modern cameras known as "sense up" , "sumlight", "Digital slow shutter (DSS)" or some such marketing name.
The trick is to slow down how often the camera produces a single image frame.
Normally the camera will be producing 25 or 30 frames per second.
At this frame rate, the image produced will be smooth video, just like your TV at home, the cinema or a DVD.
Since each frame takes 1/25 or 1/30 second to create, it's the amount of light hitting the sensor in that time period which will control how bright the image is.
If the intensity of light hitting the sensor is very low, 1/25 or 1/30 second will not be enough time for a decent image to be formed.
Reducing the frame rate will allow more time for each image to be formed, so a lower intensity of light hitting the sensor can now produce a decent image.
The major downside of this trick is that slowing down the frame rate makes the video jerky because now the camera is producing fewer than 25 or 30 frames per second.
Claims of very low lux levels such as 0.000001 lux are usually obtained when the camera frame rate is very low, sometimes as low as 1 frame every 10 seconds.
This is clearly useless when there is any movement within the area at which the camera is pointed
Effio is a name commonly used by cctv camera sellers.
It refers to a set of electonic "chips" produced by Sony, and used in conjunction with certain specific Sony ccd sensors.
An Effio chipset is not a sensor.
An Effio chipset is a set of electronic circuits known as a digital signal processor.
The effio chipset takes the signals from the sensor and manipulates them to provide the user with a wide range of adjustments such as "sense up" and digital zoom.
There are different models of Effio chipset varying by price and the range of functions they provide.
Most, if not all cameras, which use Effio chipsets will have an "On Screen Display" (OSD) capability which allows the user to access the various functions created by the Effio chipset.
Sony are the largest manufacturer of sensors (both CCD and CMOS) but Panasonic, Sharp and a few other also manufacture sensors and in certain cases also have their own proprietary chipsets somewhat equivalent to the Effio chipsets
All the cameras we use are video cameras and they produce a video signal which can be connected to some sort of viewing device such as a screen and/or to a video recorder to produce a permanent record of what the camera sees.
Video signals come in many flavours but the most commonly used in our area of interest is known as composite video (CVBS).
This is a standard definition analog video signal with 480 (NTSC) or 576(PAL) resolution.
It is NOT high definition (HD) and although it can be electronically upscaled to display on HD devices, the quality can never be as good as true HD.
This limitation on video signal resolution means that cctv camers with CVBS outputs are limited to sensors containing no more than about 0.6 megapixels (600,000 pixels)
CCTV cameras with sensors containing many more than 0.6 megapixels are readily available, primarily for use in network based security systems where each camera has a discrete IP address.
However the video signal produced by these cameras is transmitted in an entirely different format to CVBS and normal screens and other viewing devices cannot display the images from the camera in real time.
The time delay between the signal being sent from the camera and being visible on the screen (known as "latency") is usually too long when observing moving targets, and for a gun mounted set up, there is no simple direct connection between the camera and the screen.
A more recent and possibly useful video system is HD-SDI.
Signals from HD-SDI cameras are passed along cables similar to those currently used and can be seen in real time on screens which have an HD-SDI input.
HD-SDI cameras with 2.1 megapixels are available at similar prices to the cameras in common use at present, although there are only a few screens with HD-SDI inputs available and they are significantly more expensive than the screens in common use.
Undoubtedly more screens at lower prices will become available.
Of more concern is the sensitivity of HD-SDI cameras.
2.1 megapixels on a 1/3" sensor means small pixels and it remains to be seen how the sensitivty of these cameras compare to the "best in class" at this time.
The cctv cameras used in digital NV equipment come in 3 basic forms.
a. Board cameras (AKA module cameras)
As the name suggests, this form of camera consists of one or more exposed circuit boards containing the sensor, all the related electronic circuitry and proprietary connectors for DC power input, video output and possibly another connector for the OSD functions.
The sensor is normally covered by a plastic lens holder into which may be fitted an M12 board lens of a suitable focal length.
These cameras have no protection from the surrounding environment and are always built into some sort of housing.
b. Module cameras (AKA square cameras or mini cameras)
These come in two types, distinguished by size and connector type.
The smaller types (eg E700) are essentially board cameras installed in a small closely fitting metal housing to give them some protection from the surrounding environment.
The housing contains a threaded M12 lens holder and cut outs for the connectors on the board.
The larger types (EG Watec 902H) are threaded to take larger C or CS threaded lenses and have standard power and video connectors (phono or BNC).
They often also have a connector used to control the aperture (auto iris) of the attached lens.
c. Bullet cameras
The name comes from their shape - long and with a small diameter relative to their length. The best known of these is the EJ230.
This form of camera normally has a trailing cable fitted with standard power and video connectors
Many cameras have IR blocking filters fitted during manufacture.
This helps to produce a sharper daylight image and improve colour rendition.
These filters usually take the form of a pinkish coloured glass square bonded to the face of the sensor.
The filter blocks the passage of IR onto the sensor so it must be removed before the camera becomes usable for NV work.
Note that the difficulty in removing the filter varies greatly.
The filters on KT&C cameras (E700, EJ230 etc) are easily removed, whereas some cheaper board cameras have been destroyed whilst attempting to remove the filter.
Digital cameras used for NV applications always require a lens to focus the incoming IR onto the camera sensor.
The focal length of the lens is chosen to suit the camera and to produce an acceptable size of image on the viewing device.
Too short a focal length lens will produce a small image of the scope reticle surrounded by a large black area.
Too long a focal length lens will only allow the centre of the scopes' field of view to be visible, and since the scopes' field of view is already small, target aquisition can become very difficult.
Board, small module and bullet cameras typically use lenses with an M12 thread. Larger module cameras typically use lenses with C or CS threads.
Common focal lengths for gun mounted NV equipment range from 12-25mm
Digital NV spotters use longer focal length lenses, typically 75-100mm.
Most lenses are coated to reduce reflections at their surface and improve light transmission through the lens.
Some lenses have coatings which are optimised for IR and some which are optimised for visible light.
In practice it's very difficult to notice much difference between them.
It's important to note that the focal position of an image produced by visible light will be different from the same image produced using IR.
For example, during daylight, a target at 100 yards distance when viewed through an adjustable focus scope will require the adjustment (side focus wheel or adjustable objective) to be set at or near the 100 yard mark.
When the same target, at the same range, is viewed in darkness using IR, the range mark will have to be set at around 20 yards before a focussed image is obtained.
ILLUMINATION FOR DIGITAL NV
It must be understood that digital cameras do not amplify light and therefore they will only work during darkness if the target at which they are pointed is illuminated with IR.
Generally speaking, the more IR hitting the target, the better the quality of the image the camera can produce.
Although human eyes are not sensitive to 850 and 940nm IR, most IR sources do not produce all of their light at exactly those wavelengths and there is usually a small amount of light produced at wavelengths visible to the human eye.
This is more noticeable with 850nm than with 940nm.
There is much discussion about how well our potential quarry can "see" IR at these wavelengths and there is little doubt that they can detect it, probably in a similar way to ourselves.
The issue is actually not how well they can detect it , but rather how they respond when they do detect it.
In most cases is seems that they are not spooked by it.
There are 3 commonly available types of IR illuminator
a. White (halogen) light and filter.
Power hungry and inefficient because IR is only a small part of the total light produced by a halogen lamp and the filter has to block (and waste) most of the light which is being produced.
Typically uses 12volt 60 watt lamps plus a dark coloured filter - the dinosaur of IR sources, but still used by some of those who have never had the good fortune to become members of this forum.
Note - White LEDs and a coloured filter don't work because white LEDs don't produce any IR.
b. Infra red lasers.
Small and very efficient.
Still popular with tubed NV users.
Now looked upon with disdain by many digital NV users because most of them produce a "dirty" beam with lots of artifacts and variations in brightness.
There are also serious safety issues with IR lasers in that eyesight damge can easily occur if the laser beam is looked at directly or enters the eye via a reflection from a smooth surface.
c. IR LEDs (light emitting diodes)
Currently the preferred source of IR for digital NV work, and one IR LED in particular, the Osram Oslon (part no SFH4715S) provides the best "bang for your buck"
With the correct optical arrangement this LED will easily illuminate potential targets at more than 300 metres, yet only consume 3 watts of electrical power
3. Complete IR illuminators
The oslon produces most of it's light in a 90 degree cone and to be useful as a long range illuminator as much of that light as possible must be converted into a tight parallel beam.
The most common way of doing that is using an aspheric lens mounted in front of the LED.
Adjusting the distance from the lens to the LED controls beam width.
At present, several fourum members manufacture and sell IR illuminators based around standard flashlight types.
For close range work the 501b flashlight fitted with an 850nm oslon and an aspheric lens is an excellent choice.
Forum member "some bloke" sells these.
For longer range work, the T20 flashlight, again using an 850nm oslon but with a somewhat larger aspheric lens is very popular.
Forum members "Sika Stag" and "Marky 610" sell these.
In general, increasing the size of the aspheric lens will increase the usable range of the illuminator and aspheric lenses up to 78mm diameter have been used with success, although 66mm is the most common of the larger sizes.
The illuminator "lights up" the target and the lens focusses the IR reflected from the target onto the camera sensor so that a sharp clear image of the target can be created in the camera electronics and converted into a video signal which can be sent from the camera along a cable to a viewing screen and/or video recording device.
The following types of viewing device are in use:
a. LCD screen
These are the most common viewing devices and for gun mounted use typically have 3.5" or 4.3" or 5" diagonal size screens.
These screens are cheap and readily available.
In the same way that the camera sensor is made up of individual light sensing pixels, so the LCD screen is made up of individual light emitting pixels.
The sharpest and highest resolution screen image will be obtained with screens which have the highest number of pixels per inch (PPI).
Note that this is not the same as simply having the highest number of pixels in the screen.
Although small LCD screens are relatively cheap, many sellers do not state the screen resolution correctly and claims of screens having higher resolutions than is actually the case are very common.
Each pixel in an LCD screen is made up of 3 sub-pixels, one red, one green and one blue and sellers often to refer to each sub pixels as a full pixel.
This has the effect of inflating the number of pixels claimed for the screen by a factor of 3.
For example a screen with 480x272 full pixels will be described as having 1440x272 pixels - the number of horizontal pixels having been multiplied by 3.
At present, for gun mounted use, the 5" 800x480 pixel screen is generally regarded as best in terms of price vs performance.
Note that the number of pixels on the LCD screen is frequently less than the number of pixels in the camera sensor and this inevitably means that some of the information obtained by the camera is lost when displayed on a lower resolution screen
For vehicle mounted spotters, larger screens can easily be used and with larger size comes a wider choice of screen resolution.
7"or 8" screens with HD resolutions of 1024x768 or even 1280x800 are available at reasonable prices
b. Near eye displays
These are displays with diagonal sizes of less than 1", and with resolutions between 320x240 and 800x600 pixels.
The display requires optical magnification before the image it produces can be viewed clearly by the human eye
Currently, there are two types of near eye display: LCOS and OLED
LCOS (liquid crystal on silicon) is essentially a micro miniaturised LCD display.
The main manufacturer of these displays is KOPIN and they are often referred to simply as KOPIN near eye displays.
They typically have resolutions of 320x240 and 640x480.
The latter produces an excellent image more than good enough for most NV work.
These displays are 3-5 times more expensive than standard LCD screens.
The display generally comes in 2 parts - the display with it's associated LED backlight all mounted in a plastic housing with adjustable lens and a separate circuit board containing all the electronics required to convert the incoming CVBS signal to the form required by the display.
These displays can be rather fragile and hence tricky to incorporate into DIY NV builds
OLED (Organic Light Emitting Diode) displays produce the highest resolution images (800x600) and also the brightest and highest quality images.
They are much less fragile than Kopin displays and easy to use in a DIY build and since the display produces it's own light, there is no need for a separate backlight.
Unfortunately, at present, they cost around 4 times as much as Kopin displays.
An analog near eye display used in some DIY NV builds is the crt viewfinder from an old camcorder.
These produce good images with resolutions supposedly around 400-450 TVL.
These viewfiders can be obtained cheaply from the many old camcorders available on E Bay or similar sites.
The trick is finding a camcorder with simple viewfinder wiring which makes connection to power and camera video signal easy.
Unfortunately, many camcorder viewfinders use ribbon cable with many wires and it can be very difficult, if not impossible to figure out which wire does what.
PUTTING IT ALL TOGETHER
To summarise, a basic digital NV system consists of a camera, lens, viewing device and IR illuminator.
A power source for these items is also needed and, because of it's small size and weight, a 12volt lithium ion battery is frequently used.
Digital NV systems can be broken down into 2 main types, add-on or scopeless.
As the name suggests, an add-on is a device which attaches to a normal day scope allowing it to be used in darkness.
Add-ons may fitted to the front (objective)end of the scope, but are far more commonly fitted to the rear (ocular) end of the scope.
In a rear add-on, the camera lens is in the position occupied by the eye when the scope is used during daylight.
Rear add-ons are probably the most common type of DIY NV project because the cost of parts is relatively low, construction is fairly simple with lots of opportunities for personalisation and the results can be better than some commercially available NV equipment.
Some forum members with DIY NV rear add-ons are regularly shooting foxes at ranges exceeding 200 yards.
There are several commercially available NV rear add-ons, including models manufactured by forum member Marky610 and also by others including Ward Optical Sytems (War-d-vision), NiteSite and Starlight.
The main advantages of a rear add-on are:
Can be moved easily from gun to gun allowing use on quarry requiring different calibre rifles
Does not interfere with the use of the rifle during normal daylight hours.
Relatively cheap and easy to build.
Can have great performance - significantly better than some comparable commercially available equipment.
Does not affect the zero of rifle to which is attached.
The main disadvantages of of a rear add-on are:
In most cases the rifle cannot be mounted normally because of the length of the add-on behind the scope. Many rear add-ons work best when used with sticks or when shooting from a vehicle.
Because IR reflected from the target has to pass through all the scope lenses and the camera lens before reaching the camera sensor, the sensitivity can be less than other types of Digital NV equipment which have fewer glass/air interfaces.
The scope used with a rear add-on can have a significant effect on overall performance.
The most expensive scopes don't always perform better than cheaper scopes when used with an NV rear add-on
Choice of scope for NV rear add-on
Because visible light and IR come to a focus at different positions, some means of adjusting the focus of the scope is needed.
Side wheel focus and adjustable objective scopes are readily available and both can provide the focus adjustment required.
Side wheel focus controls are easier to reach and use in darkness.
Scopes should be able to focus on targets as close as 10 metres in daylight to make sure they can focus on targets less than 50 metres away when using IR
MTC and Bushnell scopes are being used successfully by forum members with the MTC Mamba and the Bushnell 6500 Elite being the pick of the bunch.
I'd describe any NV kit which is not an add-on as a scopless device.
By that, I mean that the light is passing through many fewer glass/air interfaces than in an add-on before reaching the camera sensor.
This means that potentially, scopless NV units can be more sensitive than add-ons.
There are several commercially available dedicated scopeless units such as the Armasight DronePro, the Yukon N550 and N750 and the Yukon Photon,
All these devices have a single adjustable focus objective lens placed in front of a camera sensor and linked to some form of near eye display.
With the exception of the Photon which uses a scope tube, mechanical reticle and ocular lens to produce the final image, the others have some form of electronically generated reticle which acts as the aiming point.
In terms of performance, the DronePro is generally regarded most highly and is of course, also the most expensive.
DIY scopless builds do exist and are becoming more common as more affordable cameras and near eye displays become available. Forum members including "Some Bloke", "genesis 22" and this writer have all built scopeless NV units and given the improvements in sensitivity which can be achieved, I'm sure more will follow.
HOW DOES DIGITAL NV PERFORMANCE COMPARE WITH TUBED NV
Without getting into to much detail, the best digital NV kit is equivalent to good quality GEN 2 tubed kit.
For ultimate NV performance GEN 3 is still way ahead of digital and although digital continues to improve I don't see it reaching GEN 3 performance levels anytime soon
If you've read all of this post at one sitting then your brain will now be well and truly fried and you should go away and give it a rest and let all the information slowly sink in.
Come back and re-read the bits that particularly interest you or your're not sure you understand completely, and if after that you stil don't get it, then post a question on the forum - we're usually a friendly bunch so you should get a sensible answer fairly quickly
A discussion group for all nightvision users. Find 'Best scope for nv?' here also.
1 post • Page 1 of 1