Posts Tagged ‘repair’

Audio Distortion Analyzer Added

/ January 18th, 2014 / Comments Off on Audio Distortion Analyzer Added

On a lark, I bought this audio distortion analyzer from eBay for $100 including shipping.  Similar working units on eBay run anywhere from $150 to $900, so this was a fair price for what I got.  Convenience really drove this purchase as to make FM receiver measurements (Signal plus noise and distortion, or SINAD) the distortion of the demodulated audio tone is measured at the speaker output.  Questiny has digital instruments that can do this in easy fashion, but the set-up can take some time, so I thought I would buy a used distortion analyzer and see how it worked.  This unit came last week, and was DOA; plus the feet had been damaged in shipping.  This is not altogether unexpected as the seller stated that the unit powered up, but had not been tested in any other fashion .

$_57I confirmed that the unit did light up the power indicator light, but checking the service manual showed that the light is actually right across the input mains, so it is no indication of function.  In the end, three electrolytic capacitors had failed.  One on the positive rail of the power supply, and two in the meter circuit.  Upon replacing the capacitors, the unit sprang to life.  A few hours with the manual walking through the calibration procedure, and the unit seems to be functioning as expected.  The only issue remaining is that the frequency dial is off, but the set screws have been stripped.  That’s going to take some effort to figure how to get this off without further damage.  I may have to resort to buying a sacrificial unit on eBay.  The unit is also capable of AM detection up to 65 MHz, but that doesn’t seem to working either.  Likely just a diode failure.

In the end, I have to be impressed by HP’s design.  This unit is over 40 years old, and the construction used discrete transistors.  It has probably 30 electrolytic capacitors and only three of them failed after forty years.  Not bad.  Further, during the calibration procedure I was able to measure distortion down to .03%.  That’s about a 70 dB null from 40 year-old analog circuitry!  Not bad.  Even as a just voltmeter, it has a 10 MOhm input impedance and a frequency response of nearly 3 MHz.

The bottom-line is that this is a handy addition to the lab, and repairing it really gave me a chance to check out some of the other equipment in the lab such as the in-circuit ESR  (equivalent series resistance) meter, the LCR (inductance, capacitance, resistance) meter, the variable isolation transformer, the function generator, and the digital oscilloscope.  All of these units provided invaluable contributions to the troubleshooting and justified their procurement.  I have to make special note of the LCR meter.  Typically, electrolytic caps open as the dry out, but in this case, two of the capacitors actually shorted.  The ESR meter showed a small amount of resistance indicating a good cap, but a dc continuity test showed them as shorted.  One of the capacitors in the meter circuit was in between an open and a short.  Here is where the LCR meter really shined.  The capacitor was rated at 50 uF, but it measured at 800 uF!  Not shorted, and not open, but not operating as rated.  Without the meter, I would either have not replaced this one.  In the end, I will re-cap the whole unit.  If 10% of the caps have failed, it’s likely others will fail soon.  More to the Mouser order!