One of the prerequisites to obtain certification and launch a product on the European market is meeting the requirements of electromagnetic compatibility. This text presents a case study of compatibility on the example of a device for medical applications. The device is equipped with a TFT display and a capacitive touch screen. Researchers conducted tests on electrostatic discharge and electromagnetic emission. There appeared two problems. The device was hanging up during ESD tests and it didn’t fulfill the emission requirements in the measurement range of 30 MHz – 1 GHz.
ESD resistance – research, results, solutions
The tested device should meet the requirements of electromagnetic compatibility PN-EN 61000-4-2 standard. The goal was to meet the grade ‘B’ criterion. First tests caused TFT display’s hanging up. Sometimes the graphic controller was self-resetting. As a result, the device was still running but without graphic interface. Restarting the entire device restored regular work mode. The described test result gained grade ‘C’ criterion. It was not sufficient according to the standard’s guidelines.
Further research resulted in pointing out a specific measuring point where the discharge caused display’s turning off. The edge near the FFC tape of the capacitive panel turned out to be a critical point. At first, engineers were searching for software solution as the main control board continued to work despite LCD display turning off. To receive a continuous workflow of the display, there was used a reinitialization of display’s controller. Unfortunately, this solution was again insufficient. According to PN-EN 61000-4-2 standard, the device should be resistant to tenfold discharge in one measuring point. The series of discharges completely stopped the device work. The problem was solved by using shielded FFC tape that connected the motherboard with the graphic controller. The carried-out tests were positive. However, the usage of FFC tape generated some additional problems. The R&D team focused on finding an alternative method.
Isolation of specific discharge area allowed to locate the place where the impulse penetrated the device’s interior. Further tests showed that the electric charge was transferred via touch screen signal lines. The housing had a socket for the display along with protective glass. However, engineers didn’t use an insulating material during assembly and that created a thin air gap (figure 2). At best, the air has a breakdown voltage of 2.8 kV/mm. The standard required resistance to discharges at the level of 8 kV.
| Material | Breakdown voltage [kV/mm] | Minimum thickness for 8kV [mm] |
|---|---|---|
| Air | 1,2-2,8 | 6,7 |
| Glass | 7,9 | 1 |
| ABS | 16 | 0,5 |
| Akrylic tape | 16,7 | 0,47 |
Researchers used an acrylic tape to fill the gap. It ensured proper insulation and eliminated the problem (figure 3). Further tests showed that this solution was right. The device received a degree ‘A’ criterion and the goal was achieved. There was no reset or device’s workflow suspension during the tests. Proper insulation hardened the control panel against airborne electric discharges and enabled to resign from using shielded FFC tape.
Interferences emission – research, results, solutions
Research on EMI emission was so called ‘engineers tests’. They were carried out with use of antenna set at 3 meters from the device. According to the standard, the distance should be set at 10 meters. The difference between distances was compensated by raising the emission limit up to 40 dBμV/m. This method is a standard practice.
The first step in solving problems with excessive level of electromagnetic emission was to search a circuit generating the biggest interferences. Researchers carried out tests with disconnected peripherals. It turned out that the main source of emission was the TFT block (figure 1). Figure 4 shows device’s emission graph with turned on (blue color) and turned off TFT block (green color).
After the analysis of the TFT block diagram, engineers pointed out possible sources of interferences: LED backlight voltage converter, RGB signal lines and SPI interface lines for communication between the TFT block and the main board. There were conducted a series of tests with various combinations of passive elements until obtaining a satisfactory result. Finally, researchers implemented the below-mentioned modifications.
They added a series choke coils on power supply of LED converter and on coupling between mass of converter and RGB controller. This action aimed to increase the suppression of interferences caused by internal switching of the converter. The emission decrease that results from RGB signal was obtained by adding in series the resistance on each RGB and clock line. Engineers made also improvements to the communication interface between motherboard and display block. They used ferrite suppressors on SPI bus. The improvements were successful. The result was in the worst case 3 dBμV/m lower than the targeted limit (figure 5). The applied changes allowed the device to meet the requirements of interference emission. The satisfactory result was confirmed by further tests in authorized laboratory.
Conclusions
If one thinks about launching a new electronic device on European market, research on electromagnetic compatibility is nowadays a must. The level of EMC requirements grows, if the final device is to be installed in demanding environment, such as railway systems, industrial automation and medical applications. Tests carried out in authorized laboratory are usually quite expensive. Tested device do not always meet the standard what means that there is a need to apply corrections and extends the overall certification time. When finding solutions, it is worth considering not only the achievement of the targeted result but also the solution’s economy and efficiency.
In the case study mentioned above, the use of shielded FFC tape gave the required result. However, this method generated additional costs when considering mass production. That’s why there was a need to invest in alternative solution. Knowledge obtained during the tests may improve solving problems occurring in similar devices. It also worth to remember that the research should be held during the prototyping phase.
