Pulse width modulation is achieved by controlling the illumination time of the LED to achieve gray scale. The photoelectric response speed of the LED is relatively fast, and can reach several tens of megabytes. Therefore, we can use the control circuit to illuminate and extinguish the LED at high speed to get a fine gray level. Thus, the current driving the LED becomes a series of pulses, and the width of the pulse is modulated by the image data, so it is called pulse width modulation. This method has been widely used because of its low cost control, high control precision and good linearity.
1. Weighting method
The method is explained by taking the corrected 12-bit image data D11⋯D0 as an example. First, the parallel data is converted into serial, and then the output LEDs are turned on in turn. When the highest bit D11 is output, it is necessary to light (or extinguish) the LED lamp for 2 048 clock cycles (the LED lamp is lit when D11 is high level, and the LED lamp is turned off when D11 is low level); when D10 is output. It takes 1 024 clock cycles; and so on, it takes only one clock cycle to output D0. Thus, it takes 4 096 clock cycles to complete the gray scale formation of one pixel. Obviously, this method is too expensive in terms of time, and this method is generally not used in practice.
2, the weight extinction method
The method is still illustrated by taking the corrected 12-bit image data D1 1⋯D0 as an example. First, the parallel data is still converted into serial, and then the output LEDs are turned on in turn. However, this method can complete the gray scale formation of one pixel in only 12 clock cycles. This method defines the output of D11 to light (or extinguish) the LED light for 1 clock cycle; then, DIO is 1/2 clock cycle; D9 is 1, 4 clock cycles; and so on, D0 is l/2048 Clock cycle. It should be noted that the clock mentioned here refers to the locked clock (LAT) of the LED driver outputted to the LED display , which is generally around 1 000-3 000 Hz (the internal clock of the main control chip is generally above 100 MHz). Each time one bit of image data is output and locked on the screen, the master chip outputs a signal (OE) called "blanking pulse", which is generally defined when OE is high. Lights up when OE is low. The width of the OE pulse should be precisely controlled. When D11 is output, OE outputs 1 clock cycle: when DO is output, OE outputs 1/2 048 clock cycles. Although this method has a small overhead in time, it greatly sacrifices the brightness of the display. If the image data is fully lit (ie, D11⋯ D0 is all 1), the brightness of the display is only 17% (myopia) of the weighting method, and the calculation is as follows:
Percentage of brightness = l/12 (1+1/2+1/4+1/8+1/16+1/32+1/64+1/128+1/256+1/512+1/1 024+ 1/2 048) = 17%
Obviously, such brightness is unacceptable.
3, mixing method
In order to increase the brightness of the screen without wasting too much time, the two methods can be mixed. For example, we can make D11 repeat 8 times, D10 repeat 4 times, D9 repeats 2 times, D8 outputs 1 time, starting from D7, OE pulse width is halved, and to D0 output, OE pulse width is 1/256. This method requires 23 clock cycles to complete the gray scale formation of one pixel. The weighting method is much less than 4,096 clock cycles, and the brightness of the screen can reach 70% (approximate) of the weighting method. The formula is as follows:
Brightness percentage = 1/23 (8+4+2+1+1/2+1/4+l/8+1/16+1/32+1/64+1/128+l/256)=70%
In general, such brightness should be acceptable. If you want to increase the brightness, you can repeat D11 16 times or more; otherwise. If you want to reduce the time overhead, you can reduce the repetition of D11 a little. The two cannot be combined and need to be considered together.
Mini Substation,Solar Power Substation,Step-Up Solar Power Substation,Step-Up Transformer
Hangzhou Qiantang River Electric Group Co., Ltd.(QRE) , https://www.qretransformer.com