Here is a description of the equipment utilized for our AMS in its first version. We tried to do our best with the means that we had at our disposal. Part of the equipment was bought from several French or foreign companies. Another part was used equipment that was found in a storage area. It took a long time to come up with something operational, not counting our hours of tinkering to set-up this first prototype.
here is still much to do before the station is finished. The next big improvement involves adding other detectors. But before this, we have to improve the existing system, particularly the need for the system to be autonomous electrically.
If you wish to have additional information on a particular detail or element, please do not hesitate to contact us.
On the left, the case during modifications, on the right the final product.
It was originally a parking meter ticket unit. We chose this unit for practical reasons: the volume corresponded to the space where we wanted to put it. And for security reason: the thickness of the steel casing of 5 mm, 3-point armoured lock, etc.
The stand supports the entire apparatus. It was "homemade" by our friend Mirko ("homemade" meaning using an angle grinder and soldering iron). The stand raised the AMS off the ground and is sufficiently solid to satisfy most of the mechanical constraints (weight of the AMS, wind, shock, etc.). It has a ratcheted head that rotates 360░ which allows the AMS to be positioned in any direction.
Concerning the base of the stand, an iron reinforcement (four bolts welded with pieces of concrete-reinforcing steel) was fixed in a small concrete foundation block. The bolts protruding from the stand are used to attach the stand using large nuts.
We use 12v lead-acid gel electrolyte batteries. They power the system (especially the PC and camera) during the night. They are recharged during the day by the solar panels.
The computer accepts 12v directly, therefore it is not necessary to adapt the battery output voltage (only a few fuses are necessary; a minor adjustment is made through the solar panel charger unit which balances the input/output currents).
There are two main batteries of 27 Ah each. Three other smaller batteries were added to run the fan for the housing unit (helped by a small independent solar panel). The batteries were placed at the bottom of the housing unit and are covered by a metallic grill (which also serves as a support for the element above it).
The AMS was originally designed to be autonomous energy-wise. It was necessary to add solar panels to recharge the batteries during the day. For the AMS V1, this solar power was not sufficient to ensure 100% of the power needed: the batteries went dead after a few days (maximum about one week) because the energy needs are constant at night (the AMS functions 8-10 hours non-stop at night) and the sunlight during the day is not always constant (clouds, daylight saving time, etc.). In spite of the components requiring little energy, the solar panels must be more efficient in order to mitigate these disadvantages. This is the case for the AMS V2 (operational soon) where the problem has been solved by using larger surface area solar panels (approximately 1.5 m▓. These new panels also have a better efficiency.
A ventilation system is necessary to evacuate the excess heat. The heat comes from the AMS itself (the computer is used a lot and has a tendency to heat up). During the day, the interior of the AMS housing unit heats up considerably with the sunlight and even if the equipment is deactivated during the day, this high temperature (especially in summer) risks to damage the equipment.
We installed three fansá: One extractor fan at the top of the housing unit under the top of the main solar panel, one aspiration fan at the bottom of the housing unit between the batteries, and one in the interior of the unit to circulate air around the computer. This last fan operates during the day and is powered by a small solar panel at the back of the housing unit. The sunnier it is, the more power is created and the fan turns faster and therefore the ventilation is better.
This electronic unit ensures the recharging of the batteries via the solar panels. It is controlled, is used as a switch when the batteries are recharged and also repowers the components (computer, camera) by taking the output current of the batteries (12v).
Several solar panels can be connected to it but currently only one is connected at the top of the housing unit. It is fixed to the housing unit with small magnets allowing it to be moved elsewhere as needed.
A small battery powered LED lamp has also been fixed to the housing unit with magnets (touch activated). It lights the interior of the housing unit when there is not enough natural light.
The camera is one of the most important elements of the AMS. We have chosen an analog 12v camera, highly sensitive, used generally in the field of astronomy. This camera uses a black and white CCD (Charge-Coupled Device) sensor which takes visible light spectrum and near infrared range. It is a WAT-902H2 Ultimate and the specifications and characteristics can be consulted here. At the time of purchase, this camera was among the best of its kind. Minimum illumination 0.0002 lx ! It is very compact (35.5 x 40 x 63 mm) and lightweight <100g).
The camera itself is nothing without a good lens. In fact the camera is only a CCD sensor that transforms light into an analog signal.
The lens added to the sensor is essential because it will determine the quality of the image especially in terms of luminosity.
The lens chosen is the Computar, 6 mm in F0.08. The angle is relatively wide and the luminosity is excellent.
All the characteristics are found here (you can also find them in the "Tools" section of this website).
The camera and lens are installed on an adjustable stand, a modified camera tripod. In this way the direction and angle of the camera can be easily adjusted. The stand is fixed to a small horizontal steel plate that can also be moved vertically. The plate is fixed to the door of the housing unit magnetically, facing the openings. It can freely move from top to bottom (the steel rails ensure the rigidity of the unit).
There are three openings on the front steel door. The door is covered with a thick plexiglass sheet which protects the openings.
The plexiglass is sufficiently translucent for the camera but we are going to change the plexiglass sheet to extra-transparent safety glass.
We inserted a short text in English and French, explaining the system. This document is a single sheet of paper that has been laminated and stuck between the steel door and the sheet of plexiglass that covers it.
The computer is the element which collects the data in numeric form from the camera sensors. It is via this computer that a software program will be used to record the data when the camera "detects" something. Therefore it is an evaluation in real-time of the environment and data recordings if the detection system is activated.
For the choice of the computer we had several requirements: a single machine that uses little energy, a computer with Windows platform to operate the software that we wanted, a computer powered by 12v (and not 220v) to avoid transforming the voltage from the batteries and finally a computer that can turn off and on automatically from a remote signal.
The ideal computer was foundá: It is a model used by PC- car enthusiasts. This computer is designed to be integrated in cars (DVD player, mp3, etc.). It is compact (size of a car amplifier), operates on 12v with a good margin of tolerance, turns on and off outside of the main power switch (this corresponds to a Neiman switch in cars) and uses little energy (approximately 50W). This type of computer doesn't exist in France and is very difficult to find. We bought it from a specialized German company: www.CarTFT.com.
Among the existing models, the one that we chose is a computer equipped with a PCMCIA connector so that at a later date a video card could be integrated.
The other parts are more standard and correspond to laptop partsá: hard drive 120 Go 2.5'', Intel Centrino processor 1.8 MHz 2Mo memory, 512 Mo RAM, etc. There is no DVD player, mouse, keyboard nor screen; all these are not necessary.
The computer is turned off during the day and operates only at night (via a relay and light-sensitive switch).
Data is relayed to the camera by an analog signal (by a composite cable). The analog signal must be transformed into a numeric signal so that it can be evaluated by the detection program. For this, a single PCMCIA data collection card was integrated. We wanted a raw uncompressed signal; a data collection card like the television analog/TNT receiver was enough.
Data is directly analyzed in real-time by our detection software. Data collection is only possible when the computer is on, i.e., at night.
Video images are not recorded continually but only when there is "something" to record. This is the role of the detection software: the software continually analyzes the video images that appear from the camera and when there is some type of movement or a light appears. It controls the recording of the video in the form of video files on the computer hard drive.
There exists a variety of software programs of this type, but one of the best at this time is UFOCapture which was developed by "Sonotaco". There is a freeware version and a version that you must buy. Other applications can be used with this program, for example, to analyze satellite orbits. We would like to thank those in charge of Sontaco for their help in choosing the appropriate camera and lens for the AMS.
Video files are recorded on the hard drive of the computer; therefore these files must be regularly collected to note any possible recorded phenomena. Also the size of the files is quite large (raw video files) and the capacity of the hard drive is limited. It is necessary to erase the date to free-up space.
To transfer the data, one could, of course, connect directly to the computer using a USB key or network cable but this involves a manipulation that is not very practical, therefore we use a Wi-Fi connection. Inside the housing unit we added a small USB Wi-Fi antenna just under the top of the solar panel which is connected to the computer.
We use a software program, Real VNC, which enables us to manipulate the AMS computer remotely ( by using another laptop for example). Access is secure (login, password) and can be operated at a distance of around 100-200 meters (the distance could be increased if a more powerful Wi-Fi system is installed). In this way, we don't need to directly access the AMS and the data can be collected from a distance.
Because of AMS (Version 1) is powered by solar panels, it is necessary that the elements that use energy are turned off during the day (computer, camera, fans). To turn off the power we constructed a light-sensitive relay switch. It is an electronic circuit that activates a relay thanks to a light sensor.
In function to the amount of light received by the sensor (adjustable and faces one of the plexiglass windows), the relay is, or is not, turned on. This switch is always powered but does not use very much energy. When the relay (a sort of switch) is activated, the devices are powered by 12v and a signal is sent to the computer to turn itself on 30 seconds later. Conversely, at daybreak the relay cuts the power to the devices. In addition for protection, we have included several fuses and manual switches which are housed in a case in the form of a large garden switch. The plans for the light-sensitive switch can be found in the "Tools" section.
The AMS (V1) is powered by batteries that are recharged during the day by the solar panels. However despite this energy supply, the batteries die after a certain amount of time. Therefore it is necessary to recharge the batteries using a charger connected to a power main.
It can be a time-consuming task to take out the batteries, replace them with new ones, recharge the dead batteries, etc.
Finally we modified the AMS to integrate a small charger. This charger is located directly in the housing unit between the 2 main batteries and can be connected by a cable to a power main (220v) in a building near the AMS.
The plug under the housing unit enables an extension cord to be easily plugged in as needed and also to unplug the cord when the batteries are fully charged (several hours).
The entire unit takes up a certain amount of space, but it is difficult to have both multiple functions and to be compact. The exterior is painted with a weather resistant paint over a coat of anti-rust paint (which gives it an orange color in some photos). One of the last problems to overcome was the problem of temperature within the housing unit. The interior of the housing unit can reach very high temperatures despite the use of fans for ventilation. Plus openings on the unit would have posed other problems in terms of waterproofing (water could infiltrate the unit) and/or security (more easily forced open). Another solution would have been to paint the unit with reflective paint, but this creates the problem of being too visible.
Several solutions are possible. For example, we plan to increase the Wi-Fi periphery in order to have nearby internet access. In this way with the owner's permission, we could set-up a way to control the AMS by internet. In the future surveillance in real-time for all internet users connected to the site is foreseeable.