He first tunnels were constructed some 4,000 years ago to allow pedestrians and chariots to pass under the Euphrates River in what is now Iraq. 2,000 years ago the Romans became very successful as tunnel builders and built a tunnel around 5.6km long. When the Roman Empire died the tunnelling industry died for another 1,000 years. It was revived in Moscow around 1,500 AD.
The construction of these tunnels was very labour intensive. Then the industrial revolution in England needed a good transport system and so tunnels were made through hills and mountains by labourers using machines. While not the first tunnel built under a river, the Thames Tunnel in London was the first such tunnel in England. The tunnel was built by Sir Isambard Brunel and his perhaps more famous son, Isambard Kingdom Brunel. The Thames Tunnel was being built through soft clay and thus resulted in many accidents and deaths. Marc Isambard Brunel invented the tunnelling shield to construct the tunnel; the principle is still used today. It was not until the advent of the motor car and the lorry that the number of tunnels increased even more dramatically. With the increase came the use of explosives and TBMs to make tunnels through mountains.
There are now two more types of tunnel, the floating tunnel and the immersed tunnel.
The original tunnels built were done so for the task of overcoming geological problems namely rivers and mountains.
Now they are also used to overcome manmade obstacles, cities. The vehicular traffic has increased to such a degree that to travel through a city by car takes as long in the 21st century as it did in the 19th century. Using London as a Western example of vehicle speed. In the 19th century a horse and carriage would travel at 13.6km per hour. This was the same speed of travel until the congestion charge was introduced in 2003 (which charges certain types of vehicle to use the road in central London) the speed then went to 16km per hour. In Jakarta the speed is 12km per hour.
When we drive into road tunnels we can see and smell the pollution. We know it is dangerous. We know that to live close to a tunnel portal is not good but we must do it. It is possible to forecast the volume of traffic in a tunnel but many times that estimate is too low. We now have the ability to filter the air in tunnels so as to make driving in those tunnels safer and also the city itself healthier.
Basics
There are three basic types of installations used in the filtration of air in a road tunnel. These are ‘bypass’ (built into a side tunnel), ‘stack system’ (chimney) and ‘ceiling mounted’ (built on to a platform in the roof).
The first tunnel filter systems were installed in Japan and were mechanical filters normally ‘bag filters’. With all mechanical filtration systems (a household vacuum cleaner is one such) as the bag collects more and more dust, the fan pulling the air through the system must do more work. The energy requirement increases so when the fan is sized for the system the fan must be able to still collect dust when the bag is full and requires changing. The day-to- day running cost therefore of this system is very high and due to inflation would only get higher.
A system that had been growing in popularity since the 1940s is the electrostatic precipitator (EP).
The EP is a filtration system which uses a high electrical direct current to give an electrical charge to particles that are then collected on oppositely charged flat plates.
This EP system was designed and installed to overcome the problems with the mechanical filter and be as, or more efficient. Among the advantages of the EP was that as it collected the particles the pressure drop, which was low at the start of a collecting cycle, did not get significantly higher throughout that collecting cycle.
This meant that the fan could be smaller than the one specified for a bag filter and could be size for a low pressure drop. The cost advantage of an EP system is very significant.
The Japanese system still had one major flaw, which was that it had wires for the ionizer (the section giving the high electrical charge to the particles). These wires would break due to the vibration caused by the high voltage and to electro-discharge machining. The problem is that it is impossible to calculate when a wire would break. Any wire that broke then required replacing, which therefore entailed a high maintenance cost.
The first tunnel filter designed by the author used what is now called a ‘Saw Tooth Ionizer’ and had negative ionization (both were later installed in Japanese tunnel systems). This saw an increase in the life expectancy of the ionizer and, due to the negative ionization a much higher efficiency when compared to the positive charging system. It was generally accepted that there were still some disadvantages to what has been called the European system’ so it was decided to investigate the disadvantages and design a filter system which would overcome the disadvantages.
With the standard system the filter could crash meaning that no particulate would be collected. Also if one filter crashed then many others would also crash. As tunnels get longer there is a requirement for removing the exhaust gases by using activated carbon. Activated carbon has minute holes that increase the surface area used in collecting the gas. As an example a 10mm cube can have a surface area of circa 1500m². These minute holes get blocked with the particulate. If an EP crashes then the holes get blocked and the carbon would need to be changed prematurely.
OBJECTIVE
The object of the study was to develop an electrostatic precipitator filter system that has a higher efficiency against velocity, is electrically more stable than any other systems and would not collapse if the collector section of the filter cell was short circuited. All the filter systems presently available were susceptible to the latter. It was determined that the method of testing to be used should be Eurovent 4/9. The method should determine the weight efficiency for a single electrostatic precipitator cell. The method should also be able give a particle count to determine the efficiency for varying particle sizes.
METHOD
Electrostatic precipitators have changed over the years. The original precipitators had wires in the ionising section to generate the corona discharge which is necessary for the precipitator to work. The problem with using wires was that they could break due to the high voltage vibrating the wires and electro-discharge machining. Other disadvantages are that if the collector cell is short circuited, for whatever reason, then that cell and any other connected to it also fails. We wanted to address all these problems and design a filter which could be adjusted for efficiency without the need to increase the cells dimensions.
To do this we had to analyse the way that electrostatic precipitators worked and how to improve them. A system was devised to test the filter system and to determine the disadvantages of the present filters. The test system comprised of outdoor air mixed with particles from a particle generator (diesel generator), a duct system, a filter housing which could take different sizes of filter and a fan with adjustable air flow. The duct was equipped with sensors before and after the filter to measure the airflow and particle weight and particle counting instruments (Figure 1).
Filter arrestance (A) is calculated by the following equation: Wd — weight of dust downstream of the filter [µg/m³] Wu — weight of dust upstream of the filter [µg/m³] A fluke particle counter was used in conjunction with the pDR 1200 to ascertain the particle size and the efficiency against particle size.
The Eurovent 4/9 fractional efficiency method uses a laser particle counter to count particles within specified ranges upstream and downstream the test device.
A given particle size range means all particles between two specified diameter values. The number of ranges is equipment specific, for instance, the Fluke counters have six ranges, (0.3µm-0.5µm, 0.5µm-1.0µm, 1.0µm-2.0µm, 2.0µm-5.0µm, 5.0µm-10µm and >10µm).
The basic expression of the fractional efficiency for a given particle size range, is the ratio of the number of particles retained by the filter to the number of particles fed upstream of the filter. The efficiency measurement is done by a series of 12 counts of one minute, conducted successively upstream and downstream of the test device. Between each count transfer lines are purged for one minute.
The fractional efficiency (E1) for one repetition is calculated with the following equation:
N1 — downstream count at time 1, N2 — upstream count at time 2, N3 — downstream count at time 3. Results showed that for a standard filter which is charged on the ioniser and collector the efficiency varies with the air velocity this is known and used to determine the efficiency of a system. A filter system is designed so that it is very close to arcing. This gives the highest efficiency. The problem with this is that with a system where more than one cell is powered with the same power supply when a cell discharges the power is lost in all interconnected cells for a fraction of a second. There is a time lag between the discharge and the cell/s regaining the previous charge. During this time there is a lowering of the efficiency in the system. This is a big disadvantage especially if one cell has a short circuit. If a short circuit happens, the total system shuts down. This is perhaps the one main area where the efficiency and filter cost could be improved.
As the collector is a big capacitor and that the EP was a capacitor which had a controlled discharge charging the capacitor was considered. The ioniser charges the particle, however it was the intention to quantify the effect of the ionizer with a varying velocity and a constant collector voltage.
Normally due to the way the power generator is designed as the ioniser voltage is increased so the collector voltage also increases. A generator was used for the ioniser and a separate generator for the collector and the results are shown in Table 1.
As can be seen the velocity affects the efficiency. The higher the velocity the lower the efficiency.
The ionising voltage was varied with a constant airflow in Table 2.
As can be seen to increase the ionising voltage we increase the efficiency and we can have a higher velocity with a high efficiency. The collector voltage was varied with the ionising and the air flow constant (see Table 3).
As can be seen with at lower collector voltage the efficiency was lower but due to the high ionising voltage the collection efficiency was not dramatically decreased. Below 3kV however the efficiency dropped dramatically.
The results showed that with a high collector voltage the collector could arc, which caused a drop in efficiency. Looking at the design of the ioniser it was possible to cause the collector to accept a charge without having a power connection. The principle used was that of inducing a voltage in the collector. The faraday cage uses this principle. The final stage was to design an ioniser and test the filter with varying voltages and air speed. The induction electrostatic filter was borne (US Patent 7,942,952). The principle is shown in Figure 2.
The new filter was tested for efficiency and general performance. The filter had to be better than other filters available. The filter efficiency was compared with the standard filter and was seen to be of a magnitude higher (see Table 4). When the study short-circuited the filter cell it was found that, as expected, the ioniser continued to operate and the filter cell continued to collect particulate. The collection of the shorted filter cell did decrease but the other filter cells continued to collect at their rated efficiency. This was the biggest improvement since the introduction of negative ionisation in 1989.
CONCLUSIONS
The tests highlighted that the efficiency of the system was much higher than the existing filters and used less energy to attain the same efficiency thus offering a saving in running costs. The installation of the IEP system showed other cost savings due to only requiring one power generator.
Operational Advantages
In the old system if one cell was shorted then all the cells connected to that filter will shut down. With the IEP should one cell short circuit, only that filter cell will be affected. The IEP system allows the shorted filter cell to collect particulate but at a lower efficiency, when the cause of the short is removed the cell immediately attains full efficiency. The IEP system was less prone to arcing even when operating at a higher voltage. It could also be run at a lower ionizing voltage and maintain the same efficiency.
This is the problem. The vast majority of tunnels in the World were built before either the health problems were fully understood or built with an underestimation of potential traffic flow through the tunnel.
The Lane Cove Tunnel in Sydney, Australia, is a classic example of this. The tunnel was finished in 2007 and almost immediately started to have problems with the pollution due to the quantity of traffic. The stated requirement for a single pass filter was 80 per cent.
This is easily achievable with the IEP filter system as shown earlier. After every installation of a filtration system, the system is tested by at least one accredited test authority. In each case the required efficiency has been exceeded.
There is however a major problem which is that if a filtration system is required to be installed it is impossible without disruptive excavation to the tunnel. So it was impossible to install a filtration system.
The author and his colleagues recognised that this is a problem with many roads that didn’t seem to be being addressed by the filtration industry. While looking at the problem two disastrous European tunnel fires were discussed These being the Mont Blanc Tunnel, and the St Gotthard Tunnel fires. In these fires there were 38 and 11 fatalities respectively all due to smoke inhalation.
It is known that smoke kills but it also disorientated the victims in the tunnels.
This was understood but in was not that evident until, by chance, looking at the Great Smog of London in 1952. During a four-day period 20,000 people died. What was surprising was that some died through drowning. The reason was that visibility was 1m. they could not see their feet so could not see where they were walking.
In the Mont Blanc Tunnel this was 0.5m. So it was thought that there are two objectives. Firstly to remove exhaust pollution on a daily basis, and secondly to remove smoke in the event of a tunnel fire.
Having specified the problem, to devise a system to remove exhaust pollution and smoke from a tunnel fire. It was decided to investigate the problem of fitting an EP into an existing road tunnel.
The requirements were to have a filter to clean particulate and a fan to move the air through the filter. It was realised that there were two basic problems with the filter system, the maximum economical speed was circa 10m/sec and that of the fan which required space.
It was soon understood that there was a fan available, which was the jet fan. The problem was that the jet fan moves the air at up to 35m/sec. To use the jet fan it would require a filter measuring in at at least three times the size of the face area of the Jet Fan.
This is without taking into consideration the pressure drop. The concept of spinning the air was discussed. The thought of spinning the air and ionising the particulate at the same time was addressed. In theory it meant that the faster the air, the bigger the force to throw the particulate to the side of what had to be a tube.
A ‘Saw Tooth Ioniser’ was made and twisted through 360 degrees and fitted into a 100mm tube. This one tube prototype was tested in an existing tunnel and it was shown that the concept was viable. A number of tubes were fitted together and it was shown that the system could be scaled. This meant that the filter system could be sized to the jet fan.
The principle was therefore that the filter system would be attached to the jet fans in the tunnel, using the same installation method, and the cleaning of the air would be progressive. This meant that the air cleaned by one system would entrain the pollution in the tunnel and this would be drawn into the next filter which cleaned the air and then the cycle would be repeated. Therefore for the majority of the time the filter system would be used to clean the air of vehicle exhaust. The filtration system can be fitted to all existing road tunnels as well as submersed and floating road tunnels. There is one occasion when the filter has an emergency response.
There have been many deaths due to the smoke generated by a tunnel fire. As has been seen the filter system is fitted to the Jet Fans. In the event of a fire, the filtration system and the jet fans would be activated and the smoke drawn through the filters.
The smoke would be removed and this would allow the vehicle occupants time to escape and the fire fighters to tackle the fire.
The filters are 100 per cent reversible without any modification to the filter. This means that the tunnel control centre can direct the air movement and clean the smoke at the same time.
