My son was unhappy with his NAD C 325BEE stereo amplifier purchased in 2009. He said that, while it played music ok, it produced an annoying, humming noise out of the loudspeakers which had become gradually stronger during the last few months. This noise was all the more disturbing when listening at low volume. He asked, should he trash the amplifier and buy a new one, but I answered it would be a pity to throw it out. NAD products usually have an extraordinarily good price/performance ratio, and I said I might, with some luck, be able to repair it.

As I already mentioned in at least one of my earlier Instructables, I'm into repairing and reusing (not only) electronic devices. The precondition for successfully troubleshooting any device is to find out where the trouble is located – the same way as a disease can best be cured when its cause is known.

Step 1: Diagnostics

To get a feeling what might be wrong with the amp, I played around with it for a while. With loudspeakers and a CD player connected, I checked whether the hum that came from the speakers changed when

turning the volume control
turning the balance control
turning the bass tone control
defeating the bass/treble tone controls
selecting different inputs
removing all input cables
removing the pre-/power amp links on the rear-panel

... but the hum always remained the same in both speakers. It disappeared in two cases only: When powering the amplifier off, and when plugging in headphones. In the second case the speakers got automatically switched off, of course, and the hum was audible in the headphones instead.

So I made a (more or less educated) guess that there might be a problem within the amplifier's power supply unit – I suspected that the cause of the hum was ripple voltage on one of the power supply rails.

After having found both PDF user and service manuals online and having studied the circuits for a while, I located the parts of the PSU unit that I suspected the most: Four electrolytic capacitors:

C517, C518: 15'000 uF / 50 V / 85 °C
C523: 3'300 uF / 50 V / 85 °C
C524: 2'200 uF / 50 V / 85 °C

In order to be on the safe side, I ordered spares with the higher (105 °C) temperature specifications for all of them - they are more reliable than the standard 85 °C versions while not being that much more expensive.

Step 2: Dismantling

While my son's family enjoyed their skiing holiday, I had his amp in my workshop and attempted to repair it. Removing the top cover was no problem (7 Phillips screws - 2 left, 2 right, 3 at the rear).

It is normal that a 15-year old amplifier is somewhat dusty inside. I carefully removed the dust with a vacuum cleaner.

The picture above shows a view of the amp with the cover removed. The location of the PCB in question is highlighted in yellow.

Step 3: The PSU PCB

The PSU PCB was a little difficult to remove. Fortunately, all the connections on this PCB are plugged rather than soldered. Before unplugging them I took some detail photographs of the connectors for later reference. There are eight (US version: nine) connectors on the PSU PCB in total, plus one flat cable that is plugged to the Main PCB. Getting this flat cable out from around the transformer is a bit tricky, but it can be done with some wiggling and some patience. There is another flat cable plugged into the small Headphones PCB that must be unplugged, too.

Once all connectors are unplugged, the PSU PCB can be removed by unscrewing 4 Phillips screws.

The component layout drawing shown above is copied from the amp's service manual. The four suspected capacitors are highlighted in yellow.

Step 4: The Culprits

I focused on the electrolytic capacitors first, before checking anything else. There are four of them – two large ones (15'000 uF/50 V), and two smaller ones (one 3'300 uF/50 V, and one 2'200 uF/50 V). I saw that the upper end of the two large ones looked somewhat bloated. They are not really easy to be removed since they are not only soldered to the PCB but also – being rather heavy components – glued to the PCB with silicone adhesive for improved mechanical stability (this is the white residue visible in the 1st picture). Therefor you have to heat the solder pads with your soldering iron while trying to pull them away from the PCB at the same time; I had some success by cutting the adhesive between the caps and the PCB with a knife having a long, narrow blade.

Once the two large ones were removed, I easily could see that one of them had leaked (and produced a residue looking like rust visible in the photographs above) and lost its capacitance – which was the reason for the hum. Diagnosis confirmed! The second one tested ok, but when shaking it I could hear something rattling inside; it wouldn't have taken long to break down, too. After removing the two smaller capacitors, they measured ok, but their capacitances were both at the lower limit of tolerance so I decided to trash them as well.

I replaced all four capacitors in question by 105 °C versions in order to improve the unit's reliability (the old, defective ones all were labeled '85 °C' only). The two large replacements fitted perfectly on the PCB. The two small ones, however, had a footprint slightly larger than the old ones – I had to install them with slightly bent pins and at some distance from the PCB, and fixed them with some hot glue for added stability.

Let me highlight that electrolytic capacitors are polarized components, so you need to strictly observe the correct polarity when installing them!

Then, after re-installing the PSU PCB within the amp and reconnecting it, it was time for...

Step 5: ... the Test

Powering up a device for the first time after a repair always requires some courage, so before venturing it, I double- and triple-checked the capacitor's polarities and all the connections. When everything seemed ok, I connected it to the mains and switched it on. All went well – no fireworks, no strange sound or light effects, no smoke signals!

I connected my headphones and carefully turned the volume control up in order to listen to any noise produced – but there was none at all. Then I connected an audio source (my mobile phone in this case), and the music played nicely.

So my repair was a success, and I only had to reinstall the top cover before doing a test for several hours.

Step 6: And at the End of the Day

The four capacitors were about 15 CHF/Euro/US$ in total, totally negligible compared with the cost of a new amplifier. My repair wasn't only economical, but also good for the environment – nothing had to be trashed, neither raw material nor energy was used for manufacturing a new amplifier. And the repaired unit will, hopefully, work ok for fifteen more years!

Last but not least: A successful repair is a great boost for your ego :-)