CN102118910A - Method for reducing light attenuation of LED (light-emitting diode) illuminating light sources in matrix combination - Google Patents

Abstract

The invention discloses a method for reducing light attenuation of LED (light-emitting diode) illuminating light sources in matrix combination, which is used for determining in-series number N, in-parallel number M and total number Q in an in-series and in-parallel combination structure of the LED illuminating light source matrix. The method comprises the steps of: step 1, calculating the total number Q of LED lamps by the total power W of the needed light source and the power P of single used LED, wherein Q is equal to W/P; step 2, determining the lowest in-series number N in the light source matrix, wherein N is more than or equal to deltaVF/0.01, and if N is more than Q, the whole matrix is formed by one string, and N is equal to Q; step 3, calculating the in-parallel number M of the matrixes, wherein M is equal to Q/N; and step 4, implementing the matrix combination of LED light sources according to the determined values of in-series number N, in-parallel number M and the total number Q. In the method disclosed by the invention, the problem of equalization of all strings of working current in an LED light source matrix is solved obviously and effectively, the maximization of whole light effect is realized, the whole light attenuation is greatly reduced, and the effective service life of more than 50,000 hours of the LED is really realized.

Description

A kind of method that reduces the light decay of matrix combination LED lighting source

Technical field

The invention belongs to LED semiconductor solid state lighting technical field, relate to a kind of method that reduces the light decay of matrix combination LED lighting source.

Background technology

The high-power LED lamp light source can not be provided separately by a LED lamp, need carry out the matrix combination of some LED according to the demand of luminous power, realizes the light source power that do not wait, satisfies the lighting demand of various occasion.Elementary cell LED lamp wherein can adopt the 1W great power LED, also can adopt the 20mA low-power LED; Matrix form can be discrete LED assembly unit, also can be the integrated combination of chip-scale LED.After basic solution LED heat dissipation problem, it is found that, LED light fixture light decay does not thoroughly solve, influence the LED light decay and also have what factor? for LED, even if same batch, same chip (wafer), its forward voltage drop VF under rated current IF also has bigger dispersing, even if producer dispatches from the factory behind the stepping dispersion also more than 5%, VF can occur bigger discrete between each string of the LED matrix of being made up of them, cause existing LED lamp light source in actual use, it is unbalanced that each goes here and there the LED operating current, can not guarantee whole light fixture light efficiency, and light decay is remarkable, the problem of functional reliability and life-span difference.

Summary of the invention

The purpose of this invention is to provide a kind of method that reduces the light decay of matrix combination LED lighting source, solved the unbalanced problem of operating current of respectively going here and there in the existing led light source matrix.

The technical solution used in the present invention is, a kind of method that reduces the light decay of matrix combination LED lighting source according to following steps, is determined series connection number N, number M in parallel, total number Q in the LED lighting source matrix connection in series-parallel combining structure:

The first step is by the number Q of required light source gross power W and single the total LED lamp of power P calculating light source matrix of used LED, Q=W ÷ P;

Second step, determine the minimum series connection number N of every string in the light source matrix, N 〉=Δ VF ÷ 0.01, if N greater than Q, then whole matrix just is made of N=Q a string;

The 3rd step, compute matrix number M in parallel, M=Q ÷ N;

In the 4th step,, implement the matrix combination of led light source according to top definite series connection number N, number M in parallel, total number Q value.

The invention has the beneficial effects as follows, solved to significant effective and respectively gone here and there the operating current equalization problem in the led light source matrix, realize whole light efficiency maximization, reduce whole light decay greatly, realize that really the LED lamp reaches the useful life more than 50,000 hours.

Description of drawings

Fig. 1 is existing typical LED lamp 10 strings * 10 matrix structure schematic diagrames also;

Fig. 2 is the VF-IF curve of existing 350mA high-powered LED lamp;

Fig. 3 is the relative luminous intensity and the IF curve of existing 350mA high-powered LED lamp;

Fig. 4 has the VF of LED lamp now with the temperature drift curve;

Fig. 5 is that the relative luminous efficiency of existing LED lamp varies with temperature curve;

Fig. 6 is 100W street lamp 42 string * 2 and the matrix structure schematic diagrames that the embodiment of the invention 1 adopts;

Fig. 7 is T pipe 36 string * 8 and the matrix structure schematic diagrames of the 16W-18W of the embodiment of the invention 2 employings;

Fig. 8 is 50W street lamp 48 string * 19 and the matrix structure schematic diagrames that the embodiment of the invention 3 adopts.

Embodiment

The present invention is described in detail below in conjunction with accompanying drawing and concrete grammar.

Fig. 1 is the schematic diagram of existing a kind of typical LED matrix, and after basic solution LED heat dissipation problem, LED light fixture light decay does not thoroughly solve.In fact, for LED, even if same batch same chip, its forward voltage drop VF under rated current IF also has bigger dispersing.

Table 1 is existing a kind of 20mA low-power LED (Surface Mount 3528/3410,5mm straight cutting LED etc.) VF (under 25 ℃) stepping chart, and VF=2.8V changes to 3.8V under the IF=20mA rated current, and central value is 3.3V, have ± 15% discrete.By stepping, calculate with center UL shelves 3.2V-3.4V, consider ± 0.1V stepping precision that central value 3.3V just has ± 5% discrete (being the discrete of 0.2V+0.1V=0.3V).

Table 1, the VF parameter testing situation that a kind of 20mA low-power LED provides

Stepping	Minimum value (V of unit)	Maximum (V of unit)
			UJ	2.8	3.0
UK	3.0	3.2
			UL	3.2	3.4
UM	3.4	3.6
			UN	3.6	3.8

Annotate: test condition is IF=20mA, and the certainty of measurement of VF is ± 0.1V.

Table 2 is existing a kind of 350mA great power LED of producer (LED of 1W or 1-3W) VF (under 25 ℃) stepping chart, and VF=2.75V changes to 3.75V under the IF=350mA rated current, and central value is 3.25V, also has ± more than 15% disperse.By stepping, calculate with 4, the 5 grades of 3.0V-3.5V in center, consider ± 0.1V stepping precision that central value 3.25V also has the discrete of 0.25V+0.1V=0.35V, also have ± more than 5% disperse.

Table 2, the VF parameter testing situation that a kind of 350mA great power LED provides

Fig. 2 is the VF-IF curve of the LED lamp of existing a kind of 350mA, and from Fig. 2 as seen, the every variation of single VF 0.1V (3%), IF can change 50mA (15%), and relative luminous intensity changes 10%.

Fig. 3 is the relative luminous intensity and the IF curve of the LED lamp of existing a kind of 350mA, from Fig. 3 as seen, the IF electric current is big more, though relative luminous intensity is also increasing, but itself and the non-complete direct ratio of IF (is 15%: 10% as scheming to go up the ratio that shows), LED luminous efficiency (light efficiency) has reduced, IF (setting rated value 350mA be the basis) increase by 15% just, relative luminous intensity only increases by 10%, and light efficiency approximately reduces by 5%.

Know by above-mentioned LED stepping situation, typical case's stepping voltage is differential to be about 0.35V, LED within VF shelves is combined into matrix, consider that LED VF is distributed as normal distribution, know by accumulation of error probability, its bunchiness of connecting, VF (under the rated current IF) difference mean value is exactly Δ VF=0.35V between each string.LED work is the current drives pattern, and the associated working of LED matrix is adopted the constant-current source design with the matrix driving power supply, can guarantee that LED matrix total current is constant on total specified current value ± 0.5% precision.

LED matrix example structure with Fig. 1 is analyzed, and for independent every string, flowing through this, to go here and there the electric current of each LED identical, but between each string, current conditions then is not quite similar.Wherein intrinsic disperse (such as negative discrete) of certain string than other string average delta VF=0.35V (5%), to going here and there the VF of single LEDs the increase of 0.35V ÷ 10=0.035V is arranged on average, go here and there on the same operational voltage value of other string in parallel to compensate this, know by Fig. 2, can unify to increase more than 5% to going here and there IF, and thermal effect can further increase VF and reduce trend, IF is further increased, but IF and luminous light intensity are not proportional relationships, along with IF increases, the LED luminous power is increasing, but light efficiency is reducing (this string surpass 2%), and this is a string to have consumed more electric current and electrical power, and sends insufficient luminous power, owing to the supply current that the LED matrix is total is constant, global matrix just shows light decay so.

Each string of circuit structure shown in Figure 1 all has probability analogue to occur between any two, on average gets off, because this electric current is unbalanced, the light decay of 2% * 5 ÷ 2=5% will on average appear surpassing in whole matrix one start power supply.Simultaneously, Fig. 4 be single LEDs of providing of certain main flow LED producer under rated current VF with the temperature drift curve, VF reduces 0.2V (6%) in the time of 85 ℃, though and this variation is for all LED trend unanimities in the matrix, but have drift difference, this drift difference has further been aggravated the inconsistent trend of electric current between each string of matrix.Fig. 5 has provided the temperature variant trend curve of relative luminous efficiency of single LED lamp, and the light efficiency decay can reach more than 15% in the time of 85 ℃.

This imbalance between string and string in Fig. 1 structure is a unstable state balance, the possibility that also has aggravation with the LED long-term work, and run into the unexpected temperature rise of light fixture (when street lamp arrived in summer), perhaps certain burst inefficacy (PN junction puncture short), whole string VF will significantly reduce than other string under the rated current, if greater than 3V, corresponding single LEDs VF difference is greater than 0.3V (9% changes), cause this string to receive and surpass original electric current (increasing more than the 150mA) more than 45%, and other 9 string is worked under undercurrent, in conjunction with temperature effect, its global matrix light efficiency can very fast reduction by 30%, and the whole lamp life-span is also with very fast end.

As seen, the unbalanced of operating current is another main cause of LED matrix light decay between each string of matrix, has only to reduce each crosstalk stream difference of matrix as far as possible, just can make light fixture light efficiency maximum, and be difficult for light decay.

By aforementioned deduction as can be known, for single LEDs, only guaranteed VF changes within 0.3% (0.01V), and IF changes could guarantee that less than 1.5% relative light intensity changes could be less than 1%, and light efficiency changes just can ignore not to be remembered.

Therefore, respectively go here and there relatively evenly operating current of LED for guaranteeing the LED matrix, the series connection number N of matrix should be not less than the value of the VF stepping precision Δ VF of LED lamp divided by 0.01V, be N 〉=Δ VF ÷ 0.01, and N is big more, and each crosstalk stream difference can be more little, and each crosstalk stream is consistent more, the anti-individual parameter difference of whole string, temperature iso-stress changing capability are strong more, and whole light decay is also more little; The number M in parallel of matrix=Q ÷ N, Q wherein is the number of total LED lamp, and Q=W ÷ P, W are the lamp matrix gross power, and P is the rated power of single LED lamp.

The present invention proposes a kind of simple and feasible LED matrix current-sharing combination and determine method, in LED lighting source matrix connection in series-parallel combination, series connection number N, number M in parallel, total number Q determine that method is as follows:

The first step is by the number Q of required light source gross power W and single the total LED lamp of power P calculating light source matrix of used LED, Q=W ÷ P;

Second step, determine the minimum series connection number N of every string in the light source matrix, N 〉=Δ VF ÷ 0.01, if N greater than Q, whole matrix just is made up of a string, N=Q;

The 3rd step, compute matrix number M in parallel, M=Q ÷ N;

The 4th goes on foot, and preferentially guarantees the minimum value of N, does suitable adjustment according to practical application and equipped technology, finally determines matrix Q, N, M numerical value.

Embodiment 1

Road-light light-source gross power W=100W, adopt the general LED of P=1W-3W to form light source matrix, its VF stepping precision Δ VF gets 0.35V (0.25V+0.1V), and the number of connecting so N 〉=35 is in conjunction with installation manufacture crafts such as luminous intensity distributions, only get N=42, Q can get Q=84 (be equivalent to every and be operated in 1.2W), M=2 smaller or equal to 100, then the 100W matrix is 42 string * 2 and structures, sees Fig. 6.

Embodiment 2

The T pipe of road-light light-source gross power W=16W-18W, LED by the 20mA (P=0.06W) of 3528 (or 3410) encapsulation forms light source matrix, its VF stepping precision consideration surplus then Δ VF is got 0.25V, the number of connecting so N 〉=25, about Q=18 ÷ 0.06=300, in conjunction with manufacture craft is installed, desirable Q=288 only, get N=32/36/48, then M=9 also/8 also/6 also, so, this T manage available 288 3528 (or 3410) Surface Mount LED according to 32 the string 9 also or 36 the string * 8 also or 48 the string * 6 and three kind of mode carry out combinatorial matrix.Fig. 7 is 36 string * 8 and matrix-style wherein.

Embodiment 3

Road-light light-source gross power W=50W adopts the highlighted LED combination of 20mA (P=0.06W) 5mm straight cutting, and its VF stepping precision consideration surplus then Δ VF is got 0.3V, the number of then connecting N 〉=30 in conjunction with manufacture craft is installed, only gets N=48, about Q=50 ÷ 0.06=830, desirable M=18, Q=864, if light efficiency is not enough, desirable M=19, Q=912 only, this 50W light source can adopt 864/912 highlighted LED of 5mm straight cutting, goes here and there * 18/19 and form matrix by 48.Fig. 8 is 48 string * 19 and matrix-style wherein.

Claims (2)

1. a method that reduces the light decay of matrix combination LED lighting source is characterized in that, according to following steps, determines series connection number N, number M in parallel, total number Q in the LED lighting source matrix connection in series-parallel combining structure:

The first step is by the number Q of required light source gross power W and single the total LED lamp of power P calculating light source matrix of used LED, Q=W ÷ P;

Second step, determine the minimum series connection number N of every string in the light source matrix, N 〉=Δ VF ÷ 0.01, if N greater than Q, then whole matrix just is made of N=Q a string;

The 3rd step, compute matrix number M in parallel, M=Q ÷ N;

In the 4th step,, implement the matrix combination of led light source according to top definite series connection number N, number M in parallel, total number Q value.

2. according to claim 1, it is characterized in that: after described the 3rd step, also comprise, preferentially guarantee the minimum value of N, do suitable adjustment, finally determine matrix Q, N, M numerical value according to practical application and equipped technology.

Images