Distance-to-fault: it ain't 'pass/fail'

While the 'distance-to-fault' (DTF) transmission line test is irreplaceable as a site diagnostic tool, the growing trend to specify threshold DTF performance levels is worrying RF practitioners the world over.

Bruce Holsted is, by his own description, 'a tower dog'. His business, Holsted Corporation, troubleshoots and repairs antennas and feed lines for a wide range of wireless telephony, power utility and government clients across the state of Arkansas, US.

Recent trends in site test requirements have thrown a new challenge at Holsted - the quantitative distance-to-fault (DTF) test.

The trend among a handful of carriers and consultants to specify a minimum level of performance in the transmission line DTF test concerns Holsted. He believes it demonstrates a distinct misunderstanding of the nature of the DTF test - one that he believes is inherently qualitative rather than quantitative. His company's ruling on this is firm.

"We ask at the front end about the customer's testing specification and if they demand a particular DTF figure with the antenna connected, then we generally don't do the work. We just back out of it," he says.

While maturity and experience allow Holsted and his team to make such decisions, he points out that younger and more easily convinced RF test and maintenance technicians are demonstrating worrying trends.

"I have heard that one of the techniques [to meet the DTF 'pass/fail' requirement] used by some young men is to start loosening the connectors until such time as they 'pass' that specification. This is absolutely insane," Holsted says.

Perhaps more disturbing, it is suspected that some carriers' site construction representatives are actually encouraging site crews to use this erroneous 'loosen-to-achieve-DTF-pass' procedure. This is contrary to the purpose of the DTF test: to find possible problems with the cable/connector installation.

To fully understand just what is behind the DTF threshold debate, it is necessary to go back to basics and explore the very source and nature of the test - and its near cousin, the voltage standing wave ratio (VSWR) test.

The VSWR, or 'return loss' sweep test, measures the ability of the entire transmission line (including jumpers, connectors and any other inline elements) to deliver maximum power to and from the antenna with minimum signal distortion.

A ratio of transmitted versus reflected voltage, the VSWR is a vital quantitative measure of the line's 'impedance match' with the transmitter or receiver at one end and the antenna at the other. It is a clear measure of the line's ability to do what is intended - to transmit RF power or receive an RF signal.

The VSWR test response is frequency-dependent so the test is undertaken across a range of frequencies corresponding to the operational requirements of the installation. From this perspective, VSWR testing is purely frequency-domain based. The results are a plot of either return loss (measured in dB) or VSWR, against the sweep frequency (see Figure 1).

DTF testing, on the other hand, is an inherently diagnostic test - one that is measured in the time - (rather than frequency) domain.

"It works on the same principle as radar," explains Charlie Spellman, applications engineering manager with wireless technology group, Radio Frequency Systems (RFS).

"The earliest time domain reflectometer (TDR) testers output a step or pulse into the cable. When the pulse encountered a discontinuity, some of the signal was reflected back. The amount of time it takes for the signal to return can be converted to distance along the line and provides an approximate location of the reflection point."

Discontinuities that can cause significant reflections include damaged cable, improperly installed connectors, improperly mated connectors or water ingress. The shape and nature of the DTF pulse response plotted against line distance (see Figure 2) can provide valuable transmission line fault diagnostic.

"When a high VSWR is encountered, a DTF test can provide valuable information to help locate the source," Spellman says. "This avoids randomly changing components to find the fault."

The DTF test can also provide a valuable 'birth record' for the transmission line, allowing maintenance crews to compare the line's current performance with that achieved at its initial commissioning.

Spellman further explains that advances in the data processing power of site instrumentation have permitted the two test procedures - VSWR and DTF - to be combined into a single frequency-domain-based test instrument.

"The modern network analyser uses frequency-domain reflectometry to actually simulate the time-domain DTF test," he says. "The basis is a mathematical function called the inverse fast Fourier transform."

The principle of the inverse fast Fourier transform is to combine a set of sinusoids together to produce an approximation to a pulse train (see Figure 3). In short, this allows the modern network analyser to use the data obtained during a return loss or VSWR sweep test to 'back calculate' the transmission line's time-domain DTF response.

The 'on-the-ground' procedure for the VSWR or return loss sweep test is fairly straightforward and well-understood.

"What you see in the VSWR test is a functional check of the entire transmission system as it shall be operated in its later life," says Gerhard Wunder, Radio Frequency System's global product manager transmission lines. "RFS recommends first visually checking the installation for any abnormalities. Then we recommend sweeping across the frequency band that the installation is supposed to operate at - for example, 890 to 960 MHz for European global system for mobile communications (GSM) - and observe that reflected power is within specification. It's a highly repeatable test."

Wunder describes the method used to determine the line's return loss - with a 50 ohm standard load in place of the antenna. "Typically, with the 50 ohm load, we'd expect to see a return loss of better than 23 dB. This depends, of course, on the length of the line. If we don't achieve these minimums, then we VSWR-test the line in sections to isolate the problem."

Holsted concurs with this approach.

"We first generally check for tight on the cable connectors and for presence of moisture. Then we disconnect the antenna and any tower-top amplifiers, then prove [via VSWR] the feed line. We then put an RF series adapter in between the jumpers and prove the antenna. If everything checks out, then we've got a good antenna network."

While the all-in-one site network analyser is convenient, it is important to recognise the limitations of quantified DTF test results obtained - particularly the erroneous application of pass/fail threshold.

"There are parameters that impact on the quantitative results that come out of a modern DTF test," says Wunder.

He points out that the length of the cable itself will impact on the value of the DTF reading. "The frequency-domain instrument is limited in the bandwidth sweep it can perform over longer cables. The longer the cable, the narrower the sweep bandwidth. This reduction also reduces the resolution of the instrument and diminishes its ability to resolve closely spaced reflections," Wunder says.

"If you enter a centre frequency and length of cable into the analyser, the instrument auto-corrects the DTF test bandwidth to suit its capability. This means that transmission lines of identical diameter and installation quality, but of different lengths, can produce quite different DTF results."

But most important is the very nature of the test. The DTF is not being directly measured, but calculated, so what is shown is an estimate of the average return loss over the frequency range being swept.

"This really explains the 'non-repeatable' nature of DTF," Wunder says. "It is this inherent calculated and averaged nature of the DTF test that can result in dramatic differences in the measurements taken just by varying test conditions such as test bandwidth or cable length."

Spellman points out that one of the most tedious aspects of all this is the somewhat nominal selection of the DTF threshold. Some carriers, he says, are applying a DTF measure that can be difficult to achieve with the specified transmission line components.

"The cable connectors might be specified at 30 dB. Now our connectors generally perform better than spec, so an average connector DTF performance level might be say, 36 dB," Spellman says. "The problem is that some carriers are taking this performance level of 36 dB and applying it as a blanket threshold!"

He also points out that gaining another 1 or 2 dB of DTF performance at such a low level has little or no impact on the overall system performance.

He cites a recent experience with an installation in New Jersey, US, where he and his team were called down to view a number of connections that were apparently 'failing' DTF. "The site DTF threshold specification was '34 dB or better', so the installation crew was cutting these connectors and changing them out," he says. "When they opened the 'failed' connectors they found they were perfect."

Under Spellman's instructions, a DTF sweep of all transmission lines on the site demonstrated the true purpose and value of the test.

"The connectors were all running at the same sort of level - between 32 and 34 dB. Then we hit one running at 26 dB. Sure enough, when we opened it up, the flare inside the connector was all bent over and crushed. Now that is what the DTF test is supposed to be used for!"

In light of the nature of the DTF test, Holsted isn't backward in expressing his views on the current trend to specify minimum DTF performance thresholds. "This is the bane of the industry right now. Why they establish this DTF threshold is beyond me - I don't know whose idea that was. It's just ignorance."

Spellman emphasises that the DTF test has a vital diagnostic role to play, but that it is misused when a threshold is simplistically applied as a pass/fail criteria. "The threshold should not be a pass/fail requirement, but one that triggers an investigation into whether connectors are properly installed or if there is cable damage," he says.

Wunder concurs, emphasising the cost and waste associated with the misuse and misinterpretation of the DTF test results. "These DTF pass/fail thresholds really have site testing crews out there 'chasing ghosts', trying to achieve DTF values that are impossible or unachievable," he concludes.

"DTF is a powerful diagnostic test, pure and simple. The fact that it can now be instantly viewed on a network analyser hasn't changed a thing!"

Holsted cites the 'push-button' nature of the modern network analyser as the root cause of some of this ignorance - that it is simply too easy to achieve what appears to be an accurate plot of DTF versus line length, at the push of a button.

The tip from Holsted is to go back to basics and truly understand the test being carried out. "I grew up in the industry with a simple inline wattmeter - a Bird 43 Thruline," he recalls. "So I'm used to measuring forward and reverse power, then calculating VSWR - that's the way I approach it and understand it. It's a shame there isn't more of this level of understanding today."