In my last column, I described how to test for pin-1 problems, using a test device dubbed "the Hummer," which injects a 60 Hz signal into the ground system of a device under test. An output is monitored and any increase in the noise level with the Hummer signal applied indicates a pin-1 problem.
If you are evaluating equipment for purchase, then you may want to reconsider, especially if another piece of gear has similar features and no pin-1 problem. It will make system interconnection easier. If not, then at least you will know what you will be up against.
Notify the equipment manufacturers of pin-1 problems. They may not be aware of them, but if they are, they may already have a fix, like modifications to the circuit board or other internal wiring.
Some people have made internal changes to reduce or eliminate the pin-1 problem without direction from a manufacturer, but there be warranty issues in doing this; most users will probably not want to go this route.
Barring fixing the equipment itself, take care in cable interconnection. Find a way to connect the shield of the signal cable to the equipment chassis instead of through the connector. This means breaking out the shield from the cable connector and finding a convenient place on the chassis, ideally near the mating connector, for a chassis connection. You may need to scrape paint and add a toothed washer beneath a screw to get a solid connection.
Remember that the goal is to keep any shield connection (and thus shield noise) from entering the chassis and connecting to the signal reference.
Microphones are special cases, as they often need phantom power from an audio mixer or other device. For phantom power to work, the cable shield has to be connected to signal reference because that is the wire used for phantom-power return.
But microphones with active circuitry can present pin-1 problems, as Chicago-based consultant Jim Brown described in two AES presentations.
In these cases, something beyond a standard audio connector offers a solution. Neutrik is working on a special XLR connector that contains an annular capacitor to connect the cable shield wire to the connector shell. This provides a low-impedance path for RF to the shell, and subsequently to the equipment chassis when solidly connected to a proper mating receptacle.
(Brown found that very often, XLR connectors don't mate solidly enough with their receptacles for the shells to make good contact with each other. This is something to watch out for.)
The cable shield is also connected to pin-1 of this new type of XLR, but with a ferrite bead surrounding this lead to lower the Q of the capacitance-series inductance resonator. This provides the DC path to pin-1 needed for phantom power, keeping pin-1 isolated from the shell.
Brown reported that using this connector solved RF pin-1 problems with mics that were susceptible to RF interference. It certainly has applications for line-level signals as well.
PIN-1 RF INTERFERENCE
This brings up the point that pin-1 problems may not always manifest themselves at audio frequencies. If the shield is connected to the chassis by a small length of wire or trace, it may act as an antenna at RF frequencies, but not cause problems in the audio range. Also, pin-1 problems could cause the internal clock or other types of noise to be radiated outside the chassis and be a potential source of interference to other pieces of gear. This can easily snowball.
I've used analog-balanced interconnections for the examples presented here, but pin-1 problems can lurk in unbalanced inputs and outputs, as well as in digital equipment.
Although induced noise may not be heard directly in digital equipment with pin-1 problems, that noise could be enough to prevent the digital signal from being received/decoded properly somewhere else in the signal chain. The cliff effect, where the signal suddenly disappears, can even be intermittent, depending on the nature of the induced noise--a troubleshooting headache.
What can an equipment manufacturer do to eliminate the pin-1 problem? Restructure the grounding scheme inside the gear, which is easier said than done. Help is now available in the newly ratified "AES48-2005: AES Standard on Interconnections--Grounding and EMC Practices--Shields of Connectors in Audio Equipment Containing Active Circuitry."
Developed by the AES SC-05-05 Working Group on Grounding and EMC Practices chaired by Bruce C. Olson and vice-chaired by Jim Brown, this standard shows how to connect designated shield contacts to the shielding enclosure (chassis) through the lowest possible impedance path.
It provides a diagram of a star-point connection to equipment ground (the ground prong of a three-prong AC power plug), the internal circuit reference and the chassis. The star-point approach provides a single point of reference. The standard also indicates a designated shield contact on the enclosure.
This scheme allows shield currents to flow through the chassis to shield connections in the cable and equipment ground, but not to the internal circuit reference. If this standard is followed correctly, there will be no pin-1 problem.
The standard also describes how to deal with unshielded connectors, connectors built into microphone cases and situations where the chassis doesn't provide a shielding enclosure (not encouraged).
Annex B to AES48-2005 includes examples of what not to do and why, and how to fix problems. These examples of the pin-1 problem, which do not meet the standard, are unfortunately very common today. Annex B could be used to help identify what mechanism is causing a pin-1 problem in a piece of equipment, should the user (and hopefully the manufacturer) wish to investigate further.
MORE AES STANDARDS
Having AES48-compliant equipment is the first step in creating a system that is free of hum, buzzes and RF interference. The AES SC-05-05 Working Group is continuing its efforts to standardize other pieces of the puzzle.
This work includes "AES-X027: Test Methods for Measuring Electromag-netic Interference Susceptibility in Balanced Line-Level Interconnections;" "AES-X125: Input Filtering for Electromagnetic Compatibility;" "AES-X147A: Connection of Cable Shields Within Connectors Attached to Balanced Audio Cables;" "AES-X147B: Shielding of Balanced Audio Wiring Within Passive Connector Panels and Passive Mic Splitters;" and "AES-X147C: Shields of Microphone Level Outputs of Active Equipment Other Than Microphones."
This all-volunteer group welcomes input from the industry.
The first person that I heard refer to the pin-1 problem was consultant Neil Muncy, but I later was informed by Bruce Olson that Neutrik CEO Werner Bachmann coined the term.
There are many good references for further study available on the Web, including www.aes.org for copies of AES standards and reports; www.audiosystemsgroup.com ; www.rane.com ; www.jensentransformers.com for excellent white papers and applications notes; www.synaudcon.com for information about the workshop "Hum, Buzzes, and RF: Understanding, Designing, and Troubleshooting Real-World Audio and AV Systems," to be held in August in Chicago.
Next time you look for new equipment, ask the manufacturer, "Are you AES48-compliant?"