TFT touch screen power supply problem

chenbingjy

TFT touch screen power supply problem [Copy link]

AT070TN94 Pre-spec(V01)_0707 M.pdf (810.52 KB, downloads: 4)

2018-12-5 17:21 上传

点击文件名下载附件

The circuit I used to power the TFT LCD screen is shown above. If the screen is not connected, all four voltages are normal. When the screen is connected, only -7V is normal, and the other three power supplies have dropped a lot. 10.4V becomes 8.6V, 9.3 becomes 8.5V, and 16V becomes 10V. I am very surprised. The load capacity of LM317 is good. Why is it like this? Experts, please give me some advice. Thank you!

chenbingjy

The first attachment is the LCD screen data

maychang

Did you only measure the output voltage of each 317 and the output voltage of MC34063? Did you carefully measure the 24V input of U3 to see if it dropped or was not enough?

chunyang

Test the voltage of the front-stage power supply under load. If the front-stage power supply has insufficient load capacity, then the 317 will be useless even if it is 5A or 10A.

chunyang

Supplement: Also pay attention to the heating situation. If the input and output voltage difference of 317 is large, under a certain current, if the temperature of 317 is too high, its internal protection circuit will work and limit the output, which can also be manifested as a drop in output voltage. Whether this is the reason, a simple calculation can tell.

chenbingjy

Maximum 260mA

topwon

The voltage drop on U3 LM317 = 24-16 = 8V, the current is 0.26A, then the power consumption of LM317 is 8x0.26 = 2.08W. I don't know what package you use, and whether you have a heat sink. The θj-A parameter of the TO-220 package chip is 50℃/W, and the temperature difference is 104℃; the θj-A parameter of the D2PAK package chip is 62.5℃/W, and the temperature difference is 130℃. So even at room temperature 25℃, the junction temperature will exceed 125℃.

chenbingjy

Package is SOT-223

topwon

The thermal resistance θj-A of the SOT223 package of TI's LM317 data is 66.8℃/W, and considering that these data are obtained on a standard test board (seems to be 3x3 inches), generally speaking, the heat dissipation conditions are better than those on the actual PCB. The power consumption of 2W should be too large for this package. I think it would be better to control it below 1W. From the rough calculation results, it is recommended to connect a 22 ohm 3W metal film resistor in series with the input of U3 to divide the voltage. Or simply use DCDC to do it. In addition, U1 should also have similar problems, but the heat generation is slightly lower, and it may not occur at room temperature (and the current input 16V is pulled down by U3, and the power consumption is not large). If U3 resumes outputting 16V and the ambient temperature rises to, for example, more than 40℃, it may also overheat (it is also better to add a 22 ohm 3W resistor in series to divide the voltage).