LCD CFAF320240F-035T screen corruption - EMI/EMC test

[dhavaljy]

We have used LCD CFAF320240F-035T in our new product. The product has a user interface with the LCD module. The LCD interface used is SPI running at 14 MHz and the data and clock signals are sent from main baord to user interface board through a 1.2m shileded cable. Functionally the machine and the LCD works fine.

During one of the EMI/EMC test - Radiated Susceptibility (RS) EN61000-4-3, we had observed lot of LCD screen corruption. In this test the machine is subjected to a electromagnetic field of 10V/m in frequency range from 80 MHz to 1000 MHz in Anechoic Chamber. The screen corruption observed were - inverted image, wrong colors, white lines or completely white display. This corruption was also observed when there was no data being transferred from main board to LCD

Is the Crystalfontz LCD CFAF320240F-035T EMI/EMC compliant? Is there any test report available ? Is transferring of high speed SPI clock and data through 1.2 m shileded cable correct ?

Request your help or suggestions for resolving the issue.

Crystalfontz LCD Blog

Looking for additional LCD resources? Check out our LCD blog for the latest developments in LCD technology.

 

[CF Tech]

. . . Is transferring of high speed SPI clock and data through 1.2 m shielded cable correct ? . . .

To my knowledge, SPI was only ever intended as a short distance interface.

To transmit 14MHz over 1.2 m will be a challenge. There are certainly more ways to do it wrong than right.

You would need to be aware of:

* ground path (shield should not be used as "return" ground path)
* slew rates (14MHz square wave clock will have harmonic frequency content much higher than 14MHz)
* probably want to use twisted pairs
* might want or need to use termination at the receivers and/or drivers. To do this you will need to know the characteristic impedance of the cable

I have attached an image of a possible cabling system that might work.

Note the Rt on the transmit side might not be the the same as RT on the receive side. As a guess, you might try 25 ~ 50 ohms in all six places.

Note that the shield is not connected on the display side. If you have a conductive case around the LCD, hook the shield to that conductive case, but leave the LCD signal ground unconnected from the shield + case.