История изменений

V. CONCLUSION The 50 Hz injection test has shown that concerns about ground loop 50 Hz CM currents over cable screens (enabled by earthing both screen ends) are unfounded since the CM current will not disturb the signal. For currents higher than 7 A there is a risk that the wires will melt, but these levels are not to be expected in real life. During testing this was observed, however data traffic (goodput) did not seem to be Fig. 17: Test results of all the cables used during the CW test Fig. 18: Results of all the equipment used during the CW test affected by the high CM current that was running over the cable screen. The EFT-B test results have been compared for tests on UTP and STP cables in order to determine which type of cable can better withstand high frequency disturbances. For STP cables, both methods of earthing are also compared. The results show that STP cables with earthing at both-ends can withstand higher disturbance levels than all other tested configurations. The single-end earthed STP cables give similar results as an UTP cable. That means the screen of the cable, which was added for increased immunity is useless against EFT-B. This experiment proves the theory that earthing shielded cables on both cable ends, increases its immunity. With the continuous wave test all the cables pass at a test level of 10 V, but with a higher disturbance level like 30 V, 50 V or even higher, the STP cables with two side earthing perform better than UTP cables or STP cables with only one sided earthing. The overall conclusion of this paper is that the risk of interference on networking cables can be significantly reduced by using two sided earthing of cable screen. There still consist situations in practice however, where disconnecting the cable screen at one side leads to better results. In these cases however, we have the assumption that the cable screen is terminated at the PCB and not on the enclosure, which means that the crosstalk would occur within the equipment and not on the cable. The cable screens in the measurements performed in this paper were all terminated correctly on the enclosures to prevent crosstalk within the equipment. These seemingly contradicting situations should be correctly substantiated with further research

PS: https://ris.utwente.nl/ws/portalfiles/portal/263929833/2019_EMC_Europe_Barcelona_Two_Sided_Earthing.pdf

V. CONCLUSION The 50 Hz injection test has shown that concerns about ground loop 50 Hz CM currents over cable screens (enabled by earthing both screen ends) are unfounded since the CM current will not disturb the signal. For currents higher than 7 A there is a risk that the wires will melt, but these levels are not to be expected in real life. During testing this was observed, however data traffic (goodput) did not seem to be Fig. 17: Test results of all the cables used during the CW test Fig. 18: Results of all the equipment used during the CW test affected by the high CM current that was running over the cable screen. The EFT-B test results have been compared for tests on UTP and STP cables in order to determine which type of cable can better withstand high frequency disturbances. For STP cables, both methods of earthing are also compared. The results show that STP cables with earthing at both-ends can withstand higher disturbance levels than all other tested configurations. The single-end earthed STP cables give similar results as an UTP cable. That means the screen of the cable, which was added for increased immunity is useless against EFT-B. This experiment proves the theory that earthing shielded cables on both cable ends, increases its immunity. With the continuous wave test all the cables pass at a test level of 10 V, but with a higher disturbance level like 30 V, 50 V or even higher, the STP cables with two side earthing perform better than UTP cables or STP cables with only one sided earthing. The overall conclusion of this paper is that the risk of interference on networking cables can be significantly reduced by using two sided earthing of cable screen. There still consist situations in practice however, where disconnecting the cable screen at one side leads to better results. In these cases however, we have the assumption that the cable screen is terminated at the PCB and not on the enclosure, which means that the crosstalk would occur within the equipment and not on the cable. The cable screens in the measurements performed in this paper were all terminated correctly on the enclosures to prevent crosstalk within the equipment. These seemingly contradicting situations should be correctly substantiated with further resea