GB2522689A - Power regulation of LED lighting used to replace fluorescent lighting powered by electronic ballasts - Google Patents

Abstract

A means to enable the direct replacement of an electronically ballasted fluorescent lighting element with a replacement device 23 containing light emitting diodes 21, 22 that are driven via a controller or processor unit 25 to simulate the load characteristic of the fluorescent lighting element being replaced to allow the LED array power to be maintained at a level lower than that of the original fluorescent element. The original heater elements may be substituted by inductors 12 and steering diode arrays 16, 17 to allow best routing of power from any pin combination and provide the necessary low impedance sense paths to the source electronic ballast 1. The invention makes possible the direct substitution of a fluorescent lighting element without the need to make changes to the wiring so lowering the power consumption, maintaining the original light fitting warranty and minimising the cost of installation.

Description

Power regulation of LED lighting used to replace fluorescent lighting powered by electronic ballasts.

Background

Fluorescent lighting has been the predominant source of area illumination for many decades. It is highly efficient compared to incandescent lighting converting about 5-6 times more of the energy consumed into visible light. There has been no other real rival to its efficiency barring the fairly recent increase in Metal Halide lighting used for quality industrial illumination. Metal Halide has an efficiency which is about 6-7 times that of incandescent lighting. Metal Halide is the best current member of the group of light sources known as High Intensity Discharge or HID. The low pressure sodium used on street lighting is the most common but it is slightly less efficient and suffers badly because it produces a yellowish light. The efficiencies of both fluorescent and metal halide gradually reduce as the source ages until at the end of their useful life they generally consume around twice the energy and produce about 50% of their original light output.

The invention relates to the latest lighting source known as Light Emitting Diodes or LED's. LED's have been available since the mid 1960's and have been used as function indicators in a wide variety of products. They are generally very reliable and usually survive the life of the products they are used in. In general LED's are able to operate for over look hours. They were, until recently, able to produce only very low levels of light and therefore could not be used in general illumination.

However over the last few years LED's have improved and now produce much more light due to advances in both the manufacturing and chemical processes used to create them. It is now possible for an LED to easily outperform even a Metal Halide light. The improvement in LED lighting efficiency continues and is by no means at its limit. It is technically possible that the LED will be able to reach efficiency levels of between 12-16 times that of a traditional incandescent lighting source within the very near future.

Also of importance is that the LED does not age in the same way as both fluorescent and metal halide. An LED still losses some of its total illumination output over its life, which is about 3 times that of either fluorescent or metal halide, however its power consumption remains the same.

Another important advantage to an LED is that it reaches its full operating illumination instantly and only improves at low temperatures. All other efficient lighting technologies require considerable time to heat to reach full operating temperature to produce their best illumination and this falls off quickly as the temperature is lowered. In fact operating a modern fluorescent in a freezing environment will lower its light output by as much as 70% and shorten its life expectancy by about half. A final point relevant to fluorescent lighting is that its light is generated using harmful chemicals such as mercury which are released into the atmosphere when the lamp is broken. An LED has no such issue.

The greatest reason for replacing fluorescent lighting with LED lighting is efficiency. LED lamps can deliver an equivalent light source which requires less than 50% of the energy compared to a fluorescent light source. LED lamps have a life expectancy of many times that of a fluorescent.

Fluorescent lighting is generally, but not always, formed in linear tubular lengths. Such tubes are mounted within, or on, purpose built chassis's which have within them the components required to allow the tube to operate. Fluorescent tubes have to be driven by means which regulate their power to a fairly constant level. This regulation is achieved by an electrical circuit which is generally known as a ballast. Until recent times the primary ballast component was a large inductor which was cheap and simple and is usually described as an electro-magnetic or switch-start ballast. In recent times this ballast has been superseded by a much more complex electronic circuit known as an electronic ballast. The electronic ballast, whilst more costly and less reliable, drives the fluorescent tube in a more efficient way achieving a reduced power requirement of some 10% over its electro-magnetic predecessor. Such is the improvement in fluorescent performance that EU legislation has implemented a gradual requirement for all fluorescent lamps to be powered by electronic ballasts.

The invention makes possible a means to allow the direct replace of electronically ballasted fluorescent lighting with compatible lamps containing low energy LED (light emitting diodes) elements without the need to remove the electronic ballast itself or change any of its wiring. This replacement lamp can work directly with the electronic ballast without modification and lower the power that this electronic ballast consumes by a significant and controllable amount determined by the replacement LED unit itself. This replacement lighting unit is created by the functions of the invention working with an array of LED's elements. The invention makes possible this simple replacement of the legacy fluorescent based lighting unit with a low energy LED based lighting source and requires no technical or professional assistance. Once inserted it delivers the power saving and other benefits associated with LED lighting immediately.

Follows is a brief overview of the operation of both electro-magnetic and electronic fluorescent tube lighting units. All fluorescent lighting cannot be directly connected to the power source due to the physical properties of the fluorescent lighting state known as plasma. To allow this to be safety regulated a reactive element must be put in series with the fluorescent lighting element itself. This element effectively limits the current within the series circuit to a safe value. It is also the job of this reactive element to provide a sufficiently high initial voltage, usually well above the supply voltage, to instigate this plasma gas state within the fluorescent lighting element.

Looking firstly at the basic electro-magnetic unit, this has changed little since its inception in the early 1930's. There are really only two physical elements to the operation which is the starter switch and the large inductor. The starter switch is mounted in a socket as it requires replacement about as often as the tube itself. The inductor is a permanent part and will last for the life of the unit which can be several decades. There is an optional third element which is a corrective capacitor connected across the input supply to the unit. This capacitor does not affect the operation of the unit itself but is normally fitted in an effort to apply power factor correction to the supply as required by current international regulations.

The inductor performs two roles. Firstly as power is applied to the circuit the starter reacts to the voltage and becomes a short circuit. This causes the series energy, through the inductor and fluorescent heaters, to rise and as the voltage is lost across the starter it returns to its open circuit state which releases the energy from the inductor in the form of a high voltage pulse which is presented across the fluorescent element. This high voltage will cause the mercury vapour within the tube to becomes a plasma and emit light. The inductor limits the energy within the tube during normal operation compensating for the negative resistance nature of the tubes plasma.

Next the electronic powered fluorescent unit. This is a modern replacement for the original large power inductor of the electro-magnetic ballast. Within the electronic ballast there are two distinct sections. Firstly there is an optional power factor correction stage that provides the function required to correctly profile the current drawn from the supply such that it complies to the requirements of international standards. The output of this first stage is then converted by a second inverter stage which performs the three fluorescent tube drive needs. Namely: to provide power to the optional heaters at each end of the fluorescent lighting unit, to create a ramped strike voltage of sufficient size to trigger the initial plasma ionisation within the tube and thirdly to regulate the power within the ignited fluorescent unit to overcome its negative resistive characteristic.

The power factor correction stage, while very important, is not actually required for the correct operation of the fluorescent lighting unit and can be ignored for the purposes of this explanation.

The inverter stage has connections to the heaters at each end of the fluorescent lighting element so there are normally therefore four connections. It should be noted that not all fluorescent lighting units use heaters. Those without are called rapid or quick start fluorescent lighting units. A very important second feature of the heaters is that most electronic ballasts will measure the low initial resistance of the heater elements before the inverter is started to confirm that the lighting element is correctly in circuit. This is not only to protect the inverter from an unloaded voltage ramp but is also a useful safety feature.

The inverter therefore has several rolls to play within the electronic ballast. Firstly it must optionally detect the presence of the physical fluorescent lighting element by measurement of the low heater resistances. Next, it must deliver power to the heaters while maintaining a relatively low voltage across the unit to prevent any chance of premature ionisation from occurring. Then once enough time has elapsed for the heaters to heat sufficiently to assist the ionisation by lowering the voltage required to ignite the plasma it must ramp the voltage across the fluorescent element until ionisation takes place. Finally it must regulate the current through the fluorescent element by varying the current to maintain the required operating power of the fluorescent element.

Directly replacing an electronically ballasted fluorescent element with an LED lamp has to overcome several problems. Firstly it has to present a low resistance across its heater pins (if used) so as to allow the detection function of the electronic ballast to register the existence of the unit. Secondly as the phase and polarisation of the three drive voltages, the two heater voltages and the overall element voltage, cannot be known to the replacement LED lamp it must be able to draw current successfully from all possible sources. Thirdly it must withstand the electronic ballast's initial ionisation strike voltage ramp and present such a dynamic load to the electronic ballast that will be seen as if the plasma strike phase has correctly occurred. Finally it must achieve a load characteristic that drives the LED array at its target power level) which will be naturally much lower than the original fluorescent element, in such as way as to allow the electronic ballast to regulate the load as if it was the original high power fluorescent element.

Statement of Invention

The invention makes possible a direct means by which LED(s) can directly replace a fluorescent lighting element in an electronically ballasted light fitting without requiring the removal of the electronic ballast itself or any of its associated wiring. The current state of the art is that there is no known method to correctly deliver this characteristic for electronically ballasted fluorescent lighting.

It is known that there exist LED drive interfaces that can survive being connected into electro-magnetic fluorescent units but in the more complex situation of electronically ballasted fittings the power is not matched to the actual operating profile required by the electronic ballast itself.

The present invention provides all the necessary requirements to deliver a means by which an electronically ballasted fluorescent lighting element can successfully be directly replaced by a unit containing LED(s) and the invention.

Whilst driving energy into an LED to enable it to produce light is well understood, it is very different from the way energy is driven into fluorescent lighting elements. In fluorescent technology the operating voltages used to drive the lighting unit are between SOy and 300v (the initial strike voltage is much higher), depending on power, length and diameter. Fluorescent elements have what is known as a negative resistance characteristic, this means that as the voltage increases the current decreases and visa versa. Conversely an LED needs between 3-4v per device and its current characteristic is positive while the voltage is relatively static.

Advantages of the Invention The invention allows the replacement of a traditional electronically ballasted fluorescent lighting element with a device containing light emitting diode(s). It achieves this by replacing the heaters with inductors and coupling all input terminals via appropriate steering diodes to a capacitive storage element which feeds by a regulatory means the LED array in such as way as to simulate the load characteristic of the fluorescent lighting element being replaced. There are several important features of the invention. Firstly the three distinct modes of operation of the electronic ballast power source are accommodated by the invention such that they are seen as being within the characteristic expected by the ballast. Secondly the invention wastes very little energy converting the electronic ballast power into power usable by the LED array. Thirdly the components required to form the invention only occupy a small physical space and therefore are able to be fitted within the tight confines available within the replacement lamp. But most important of all is that the invention enables the opportunity for simple direct replacement of electronically ballasted fluorescent lighting by low energy LED lighting without any need to modify the wiring of the light fitting itself which minimises the cost of installation and maintains the original warranty of the lighting unit.

The invention is not limited to working solely from an electronic ballast source it is perfectly useable when connecting the replacement lamp to direct or inducted power sources. This is accomplished by simply detecting the presence of the power source type with the processor unit (25) and switching the control method of the stabilisation sequence such that the switching unit (27) reverts to positive resistance regulation as opposed to negative resistance regulation.

Introduction to Drawings

The drawings show the wiring of a traditional electronically ballasted fluorescent tube and the functional blocks of the invention when replacing the same. All functional blocks of the invention are able to be contained within the physical dimensions of the original fluorescent lighting unit so are exactly the same diameter and length as the replaced unit and also are well within the maximum weight limits required by international & EU standards. As a general rule the replacement is very little different to the weight of the unit it is replacing due to the compact nature of the invention.

Detailed Description

The description given here is for a tubular lamp which is normal use of fluorescent lighting elements.

With reference to Figure 1 it can be seen how an electronically ballasted fluorescent lighting tube is wired. For simplicity a single driven tube is shown although the invention is able to be used in multi tube configurations with one to one, series or parallel configuration wiring. The electronic ballast (1) is connected to the mains supply and provides the three drive voltages for the fluorescent tube (2) at its output terminals. These are the two heater (7 & 8) voltages (9 & 10) applied between the heater terminals of the fluorescent tube (3-4 & 5-6) and the main drive voltage (11).

The actual individual connections between the ballast (1) and the tube (2) will always be indeterminate as the wiring between the sockets that mount the tubes are not polarised and moreover the ballast terminals never indentify which in each of the heaters pairs (3-4 & 5-6) are the source of the main drive voltage (11). Therefore the invention has to be able to take its primary power from either terminal at either end of the tube. Also as the heaters (7 & 8) are generally used by the ballast (1) internally to detect the presence of the tube (2) the invention must allow a means by which the detection circuitry can be serviced correctly. The voltage from the ballast (1) to the heaters (7 & 8) may be in phase with the primary drive voltage (11) or not, so the invention works with whichever phase is presented.

In the invention (Figure 2) this function is achieved by replacing the heaters with inductors (12 & 13) that are of such a value as to present a low resistance path to DC or low frequency signals as used by the ballast (1) to detect the presence of the tube (23) but which will become high resistance when the ballast has enabled its internal high frequency inverter during initial ionisation strike phase and normal running mode. The voltage across the tube heaters (9 & 10) delivered from the ballast (1) is much lower in value than the primary voltage (11) as the heaters of a normal fluorescent tube (2) are of a low resistance. Therefore the actual active current flow within the inductors (12 & 13) is extremely low so providing a very efficient means of substitution.

There is an optional pair of DC blocking capacitors (14 & 15). These are not essential for the correct operation of the invention but provide a means for protecting a replacement tube (23), powered by the invention, from damage if it was wrongly fitted into an electro-magnetically ballasted light fitting. To achieve the requirement for the primary power (11) to be un-polarised the invention uses two diode arrays (1 & 17). This configuration of diodes creates a current path for the primary voltage (11) from any source to any destination of the four end pins (3-6). By this combination of steering diodes the invention overcomes the problem of unknown voltage phase and source. Also by using this method any in-phase voltage that is available at the heaters (9 & 10) will be included by the invention as part of its LED power source and therefore maximise the efficiency of the drive.

After the steering diode blocks (16 & 17) there is a small bulk storage capacitor (18). Its value can be very small as the electronic ballast (1) operates at a high frequency, usually between 40kHz and 100kHz, so the storage time required is very short.

The electronic ballast (Figure 1)(1) varies the power to the tube (2) by adjusting the voltage across it (11) this control loop is fairly simple in construct being a closed stabilisation loop which tries to maintain a constant current through the tube (2). If the current drops the electronic ballast (1) increases the voltage across it (11) which increases the current flow bringing the voltage back down again until the current returns to the target value. Equally if the current rises, the electronic ballast (1) lowers the voltage across it (11) which reduces the current which increases the voltage (11) again until the current returns to the target value. So it can be seen that the power is the inverse function of voltage (11) across the tube (2) the effective dynamic resistance of the tube (2) lowers with increasing current flow through it. If there were no ballast controlling the flow through the fluorescent tube then the current would keep increasing leading to rapid and catastrophic failure of the fluorescent tube.

In a normal power control for LED(s) the input voltage source would be relatively constant and the current driven to the LED(s) would be regulated such as to maintain a constant power. For efficiency this regulation is generally done via a high frequency switching circuit which includes an inductor and recovery diode that regulates the energy driven to the LED(s) by varying the average time that the switch is on within each cycle of the switching frequency. This therefore regulates by adjusting for changes of the power within the connected LED array or device.

The power delivered in each cycle of the switching period is determined by two significant parameters, the voltage at the switch input and the time that it is on compared to the time that it is off. Therefore should the LED(s) current increase then the switch must reduce its switching on time to correct for the increase. This is effectively an increase in the impedance of the circuit loading.

Conversely should the LED(s) current decrease then the switch must increase its switching on time to correct for the decrease. This is effectively a decrease in the impedance of the circuit loading.

So the effective impedance of the traditional LED(s) power regulator increases to adjust for increased load power and decreases to adjust for decreased load power so it behaves like a normal positive value resistance. Whereas the fluorescent tubes effective impedance is negative as it decreases to adjust for increased load power and increases to adjust for decreased load power so it behaves like a negative value resistance which is the opposite to that of a normal LED(s) power regulator circuit.

The invention is a means by which this inverse regulation can be used to allow the LED(s) to be power stabilised when the source of the power is from the output of electronic ballast designed to power fluorescent lighting elements. With reference to Figure 2 the power stored in the capacitor (18) is not directly coupled to the actual LED(s) load (21 & 22) instead it is routed via a control unit (27) that comprises an inductor, high speed recovery diode and power switch. Its output is directly connected to the storage capacitor (28) which is across the LED(s) load formed by the matrix (21 & 22) which comprises a variable series/parallel arrangement (19 & 20) that best suits the maximum power of the LED unit verses the optimum power per LED device. The switching control unit (27) is driven by a signal from the processing unit (25). The processing unit (25) has monitoring signals supplied to it from the voltage across the LED array (21 & 21), the current through the LED array via a resistive element (29) and the input voltage monitor and conditioning unit (24). The power required to operate the processor unit (25) is derived from a low power regulator unit (26). The low power regulator unit (26) can obtain its input supply from either the input stored energy capacitor (18) or from the voltage across LED storage capacitor (28).

The sequence of operation is as follows. When power is first applied to the electronic ballast (1) the ballast can determine the presence of the replacement LED lamp by sensing the low impedance circuits formed via the inductors (12 & 13) across the heater circuits (3 & 4, S & 6). After the electronic ballast (1) senses the presence of the lamp it will drive energy into the original heater circuits (3 & 4, 5 & 6). This is steered via diode arrays (16 & 17) to the storage capacitor (18). This then drives the initial input source of the low power regulator unit (26) to provide power to the processor unit (25) and also to supply the driver control voltage required by the switching unit (27).

The processor will now monitor the voltage across the storage capacitor (18) to determine if the value is within a safe range. This monitoring process continues constantly while the power is present as a safety function of the LED replacement lamp (23). If the processor unit (25) has not sensed the correct operating conditions to begin normal drive of the LED array (21 & 22) but the sensed storage capacitor voltage (18) rises to a predetermined maximum upper limit then the processor unit (25) will instigate an immediate and rapid discharge of the capacitor (18) via the switching control unit (27) into the LED array (21 & 22) for a sufficient period to reduce the voltage across the storage capacitor to a safe level. This safety feature is also performed manually by a secondary circuit within the switching control unit (27). This is required to protect against any period when the processor unit (25) is not ready or able to provide the protection signal required to the switching unit (27).

After the electronic ballast (1) has passed its heater period, about 1 second, it enters the ramp ignition period. During this time an increasing voltage is driven into the replacement LED lamp (23).

This voltage rises rapidly and can easily reach 1-2kv if no load is presented. To protect the LED circuitry from this very high potential voltage the processor unit (25) enters a regulation phase were the switching unit (27) is driven with a variable high frequency switching signal such that the upper voltage is limited by forcing any excessive power into the LED storage capacitor (28) and therefore uses the LED array as a means of excess power dumping. As the LED voltage storage capacitor (28) is normally at a low state during this period the dumping only creates an initial pre-charge boost for the capacitor (28). This limiting function is also important for the electronic ballast as many suppliers of these products detect a high upper voltage to determine if the connected lamp has failed. If a value considered too high is detected by the electronic ballast (1) then the ballast will shut down. So by the processor controlling this upper voltage this fault condition is never reached which allows the electronic ballast (1) to continue working.

The electronic ballast (1) will detect the limiting upper voltage caused by the processor unit (25) as the point at which a successful ignition strike of the original fluorescent tube has been detected and enter into the running power stabilisation phase of its operation. The LED lamp processor unit (25) now detects this state being entered and begins to regulate the LED load to the predetermined value set for the replacement lamp. By way of example if the original fluorescent tube was a 36w 1200mm size then the LED replacement would be set to run at the much lower value of 19w. To achieve this the processor determines the power within the LED array by means of the current sensed across resistor (29) and the voltage sensed across the LED storage capacitor (28). If the processor determines this power is too high it reduces the on/off switching ratio signal supplied to the switching unit (27) which is counter to normal regulation of power in normal LED power supplies.

This increase in ratio causes the effective impedance of the LED lamp (23) to reduce. This causes the electronic ballast (1) to lower its power to maintain the target current it is designed to deliver. This means the lowering of impedance by the LED lamp (23) causes the power to be reduced if the power sensed by the processor unit (25) was too high. The converse is also true that if the power sensed by the processor unit (25) is too low it increases the on/off switching ratio signal supplied to the switching unit (27). This increase in ratio causes the effective impedance of the LED lamp (23) to increase. This causes the electronic ballast (1) to raise its power to maintain the target current it is a designed to deliver. This means the raising of impedance by the LED lamp (23) causes the power to be increased if the power sensed by the processor unit (25) was too low.

By careful application of this principal the processor unit (25) can smoothly regulate the LED power to the lower target value than the electronic ballast normally supplies. What this principal is therefore achieving is reverse power regulation, normally the fluorescent element is regulated by the electronic ballast in the invention it is the LED lamp that regulates the electronic ballast. A further complication is introduced when the electronic ballast is designed to be able to dim the fluorescent lamp. In this scenario the value of the current regulation of the electronic ballast (1) is internally varied. The LED regulation loop must change to anticipate this additional requirement by entering into a second control method that changes to include the detection of the power being sourced from the electronic ballast compared to the actual power being used from a set on/off period of the switching unit (27). The input voltage monitor and conditioning unit (24) provides some of these additional parameters by observing the voltage state from the electronic ballast (1) in two temporal modes both medium and long. These two states provide a differential comparison to aid determination of the average electronic ballast (1) power delivery value.

By including these additional parameters the control method can apply a constant to the loop that is determined by the average power being delivered by the electronic ballast (1). Therefore if the electronic ballast (1) begins to regulate a lower current value the processor unit (25) can sense this and apply a lower constant to the on/off switching unit (27) to lower the LED power being controlled. This method is not as stable as is possible without the additional parameters but it does provide a means to allow even the most advanced electronic ballast to have its fluorescent element replaced by a lower power LED lamp constructed using the invention.

Claims (9)

1. Claims 1. A device that contains LED (light emitting diode) elements and control circuitry which implements a reverse impedance regulation technique to match the negative resistive properties of a fluorescent lighting element to the positive resistive properties of LED devices.
2. 2. A device as in claim 1, where the LED power regulation is varied in accordance to the needs of an electronic ballast source that has dimming capability.
3. 3. A device as in claim 1 or 2, where the power is sourced from the input pins by a combination of steering diodes arranged in such a way as to route the highest power source from any two input connections to an LED array thus allowing the replacement of a fluorescent element or load that is driven by an electronic control ballast or regulator without the need to make changes to the connections or wiring.
4. 4. A device as in claim 1, 2 or 3, where the power drawn by the device is less than the stated output power that the electronic ballast source was designed to deliver to a fluorescent element or load.
5. S. A device as in any of the preceding claims, where the arrangement of the array of light emitting diodes is passed through a transforming element that allows the effective power profile to be modified in a such way as to improve the effective power presented by the device to the source electronic ballast.
6. 6. A device as in any of the preceding claims, which contains a control means by which the power profile of the device can be varied during its operation so as to vary the power drawn by the device from the source electronic ballast.
7. 7. A device as in claim 6, were the control means is dependent on the measurement of heat or light level or presence detection.
8. 8. A device as in any of the preceding claims, were there are included direct current (DC) blocking components within the construction inserted so as to protect the device from incorrect operation or use.
9. 9. A device as in any of the preceding claims) were the invention also supports by additional means the use of the replacement device from direct power or inducted power sources.