Tektronix AA-5001 (1983)
A programmable distortion analyzer that promised .0032% residual distortion - and quietly asked you to buy its oscillator too.
Overview
The AA-5001 was never a standalone instrument. Tektronix built it as one half of a pair: the AA-5001 Programmable Distortion Analyzer joined with the SG-5010 Programmable Oscillator to form an automated audio test system. Tektronix's own advertising in DB-Magazine (August 1984, p.15) billed the combination as "built to test the best," aimed at technicians chasing both faster throughput and more comprehensive measurement coverage. The same ad copy said that even buyers not ready for full automation had options from Tek. The intended buyers were broadcast, communications, and high-quality consumer equipment manufacturers, per the company's promotional material. This was bench gear for people measuring other people's gear.
The headline number Tek pushed was a residual distortion of .0032%, with all the standard test regimes covered - CCIF, SMPTE, CCIR. On paper that was a serious claim for 1983. Mix-Magazine (February 1984, p.66) noted the pair had first been shown at NAB '83 and were by then available, with US base prices of $4,000 for the SG-5010 oscillator and $3,300 for the AA-5001 analyzer. That puts a working two-box system north of seven thousand 1984 dollars before you've bought the TM-5000 mainframe to plug them into. Both are TM-5000 plug-in products. Neither box does anything on its own.
According to the same Mix report, the system ran harmonic distortion to IHF A202 and intermodulation distortion to SMPTE TH 22.51, DIN 45043, and IEC 268.3 - automatically. The standards coverage is the real story. The AA-5001 wasn't selling a magic number so much as a repeatable procedure: set up the test, let the pair execute it, and get results that conformed to the specs your customers or regulators cared about. For a broadcast shop running the same distortion checks day after day, that automation was the product.
The caveat comes from a comparative table published in Studio Sound (August 1985, p.52), which lined the Tek pair up against the Amber 5500, Audio Precision System One, Hewlett-Packard 8903A, Wandel & Goltermann NFA-1, and others. In that table the AA-5001 looks capable but not dominant. Its amplitude measurement error was listed at 0.2 dB, its flatness across 20 Hz-20 kHz at 0.1 dB, and its residual noise in an 80 kHz bandwidth at -108 dBu. Those are respectable figures, but the Audio Precision System One posted -108 dBu noise with 0.05 dB flatness, and the Wandel & Goltermann NFA-1 hit -115 dBu. The same table shows the AA-5001's total harmonic distortion measurement residual at -90 dB, against -95 dB for the Audio Precision and -100 dB for the 3200A. The AA-5001's full system bandwidth was listed at 300 kHz, narrower than the 500 kHz several rivals claimed. None of this makes it a bad instrument. It makes it a specific one.
One detail from that comparison is easy to misread. The AA-5001's frequency measurement error and resolution were both marked "NA" - not available - in the Studio Sound table, while nearly every competitor listed a frequency-measurement figure. The table did credit the AA-5001 with two detectors, two lowpass filters, and two highpass filters, a common-mode rejection ratio of 50 dB, and a crest factor of 3. Workmanlike rather than class-leading.
The Studio Sound feature that accompanied the comparison gave some design context that applies squarely to this kind of instrument. Packing microprocessors and noisy logic into a box where sub-microvolt noise levels set the performance floor is difficult, and providing computer-variable analog control elements with adequately low distortion is harder still. That's the tightrope the AA-5001 was walking. The .0032% residual figure and the -108 dBu noise number come from careful layout and grounding in a chassis that also contains digital control circuitry. Anyone repairing one today should treat that grounding with respect.
As a collectible, the AA-5001 sits in an awkward spot. It's a TM-5000 plug-in, so it's useless without a mainframe and largely pointless without its SG-5010 partner. That two- or three-piece commitment keeps casual buyers away and prices (where they surface at all) modest. The people who want one are the people who remember running one: broadcast engineers, production test techs, and the small cohort of audio metrology enthusiasts who appreciate a period instrument that automated standards-conformant distortion testing. It has a documented place in the lineage of automated audio test, a price history from the period press, and a spec sheet that a 1985 trade comparison set beside the instruments that eventually ate its lunch.
eBay Listings
As an eBay Partner, we earn from qualifying purchases. This helps support our independent vintage technology research.