Resettable overcurrent protection for ADSL equipment

Telecommunications network equipment must comply with regional safety agency standards that mandate the use of overcurrent and overvoltage protection to help achieve safe and reliable performance. Resettable polymeric positive temperature coefficient (PPTC) devices have become an accepted circuit protection solution for analog and digital linecards, ADSL and xDSL modems, T1/E1 and voice over IP (VoIP) equipment that must comply with ITU-T recommendations for both customer premise and central office applications.

The PPTC device offers equipment manufacturers and network operators a number of performance advantages, including:
• Improved reliability via resettable protection,
• Tighter protection against overcurrent faults, and
• Lower power let-through to protect sensitive electronics.
PPTC devices are commonly employed to meet telecom equipment standards, and to improve system reliability. However, some ADSL designers have expressed concern that the additional resistance of PPTC devices may affect performance in bit rate and reach, and have resorted to non-resettable fuse protection, thereby reducing the reliability and robustness of the ADSL connection. Recent testing on reach-rate performance with PPTC devices was performed at the University of New Hampshire ADSL Interoperability Laboratory. The test data demonstrate that the resistance added by the PPTC devices to tipand-ring has minimal impact on the impedance of the line.

<%@ LANGUAGE="VBSCRIPT" %>
<% Randomize: ord=int(rnd*1000000000) %>


Current limiting devices used to protect equipment from AC power induction and power faults may include resistors, fuses, or
PPTC devices. In telecom applications, the advantages of resettable functionality, faster activation time, lower energy
let-through, and lower surface temperature under fault conditions have made the PPTC device a popular alternative to lightning robust telecom fuses.


FUSES
Because fuses are one-use devices and have a low thermal mass, they must be specified with a very high current rating for
telecom applications. For example, fuses commonly used to assist in meeting the Telcordia GR-1089 requirements for protection of network equipment in North America are typically rated at 1.25A to 1.50A, while PPTC devices designed to meet the requirements of the same specification are rated at 160 to 200mA. The greater than 10-to-1 fuse rating (1.5A) to operating current (150mA) ratio is necessary for fuse technology since the specifications require a defined level of survivability and operability of telecom equipment after being subjected to various overcurrent and lightning events.



These lightning-robust fuses often utilize time delay features to improve survivability. The disadvantage of time delay is that the longer the overcurrent protection device takes to react to a fault event, the more energy will be let through, which can damage electronic components or provide a source of ignition.

The thermal mass and trip temperature of the PPTC device permits a closer match to the damage current of the equipment, thus reducing activation time in lower current fault events. Figure 1 compares the activation time (time to-trip) of a commonly used telecom fuse to that of several PPTC devices designed for telecom applications. At a given fault current, PPTC devices activate faster than surface mount fuses. The device’s activation time affects energy let-through and surface temperature of the overcurrent protection device.

Energy let-through (I2t) is a parameter commonly used in fuse technology to measure the ability of a fuse’s resistance to “weakening� by transient events. The I2t value of a lightning-robust fuse is typically in the double digits. In the same application the PPTC device can have a much lower I2t, thus preventing excess energy let through.

A consequence of using an oversized fuse is the potential for elevated surface temperature in a fault condition, leading to damaged equipment or unsafe conditions. When exposed to a fault condition that exceeds the operating current of the system, but is not high enough to cause timely activation, the lightning-robust fuse can reach undesirable temperatures. In contrast, the PPTC device activates relatively quickly and stabilizes in temperature, so the fault current has little effect on its surface temperature.






REACH-RATE TESTING WITH PPTC DEVICES
Recent testing on reach-rate performance with PPTC devices, performed at the University of New Hampshire ADSL Interoperability Laboratory, demonstrates that the resistance added by the PPTC devices to tip-and-ring has minimal impact on the impedance of the line.

In these tests, the upstream and downstream bit rates were established between an ADSL DSLAM and an Alcatel Home Speed
Pro CPE modem. Test boards containing PPTC devices were placed between the DSLAM and the input to a Spirent DLS 400HN Wire
Line Simulator. Each board included a PPTC device on the tip-and-ring lines as shown in Figure 2.

Connection rate was determined for line lengths of 0 to 18kft in 1kft increments. Connection rates were measured with –140dBm of white noise imposed on the line. A Sony/ Tektronix AWG2005 Arbitrary Waveform Generator generated the white noise, and input impedance, phase, and attenuation were measured using a DCM 2XLD Automated Cable Tester. This piece of equipment is designed to measure transmission line characteristics of 100Ω twisted pair cables. The test boards were connected with 24
AWG Cat 3 cable.

The bit rates measured were not significantly different from the rates measured without the devices on tip-and-ring, except at 16,000 feet. All conditions tested, including the control, resulted in showtime (a successful connection) to a maximum line length of 17,500 feet. As shown in Figure 3, the input impedance of the line does change slightly with the addition of
the PPTC devices. The difference increases as the resistance of the part increases.



It is important to note, however, that the additional resistance added to the line by the PPTC devices does not cause input
impedance of the line to increase by the amount of the total resistance of the parts used. Because the input impedance is a transmission line parameter and the same resistance added to both tip-and-ring lines is basically balanced, the impedance remains relatively unchanged. The presence of the PPTC devices does affect the minimum and maximum impedance values observed at the resonant frequencies as shown in Figure 4.



As shown in Figure 5, the PPTC devices do add some attenuation to the line, but the attenuation is flat and frequency independent over the ADSL frequency range. In this figure, the attenuation values presented are the difference between the “cable only� and the “cable with devices� test boards.




TRANSMISSION RATE WITH PPTC DEVICES
The data in Figures 6 through 8 show that the presence of a PPTC device on tip-and-ring of the DSLAM has no significant effect on the connection rate between the DSLAM and CPE modem. There is essentially no effect on the downstream rate except at 16,000 feet. The greatest decrease in downstream rate, 96kb/s, was observed at 16,000 feet when examining both the minimum
and average connection rate differences.

This may be due to a reduction in power coupled onto tip-and-ring from reflections caused by the mismatch between the expected line impedance and actual line impedance. If the decrease were caused by increased power losses due to reflections, a decrease in the upstream rate would be expected. However, a decrease in the upstream rate at 16,000 feet was not observed. It is interesting to note that increasing the resistance of the PPTC device mitigated the decrease observed.

For this particular DSLAM, the 3.5Ω devices on tip-and-ring test condition appear to provide the best overall performance. At
16,000 feet, performance is best with 6Ω devices on tipand- ring. Upstream rates for 3.5 and 6Ω devices on tip-and-ring
tests are ±32kb/s of the control rates for each line length. No decrease in reach is observed when the PPTC tests are
compared to the control with no PPTC device included, and all conditions achieved a successful connection at a maximum length of 17,500 feet.


SUMMARY
Because a resettable PPTC device can be matched more closely to the operating currents of the ADSL system than a lightning
resistant fuse, the amount of power let-through to the equipment during an overcurrent fault is considerably lower. This may allow the use of smaller, lower power-rated, less expensive components downstream. ADSL designs using PPTC devices can provide ADSL rate and reach performance comparable to designs using fuses or other non-resettable protection. More importantly, their resettable functionality helps improve system reliability.