Jim's Blog: Pure Evoke 1 Crackly Burbling Audio

Warning: while this radio is low voltage around 12V dc no mains voltage appears in the actual radio, so it is relatively safe to work on. The wall-wart power supply itself however is mains voltage ( in my case it's 230V AC ) so this needs to be treated with respect. Also while soldering be careful of solder fumes, wear eye protection and don't burn yourself. You have been warned, I am not responsible for your actions.

I have had this Pure Evoke 1 DAB radio for about 7 years but it started having problems. The audio became crackly and it also had an associated burbling noise. The signal quality displayed for the radio station I listen to normally which previously was 97 was now wavering between 40 up to about 70.
The first thing I tested was the most accessible thing the wall-wart power supply. This power supply seemed to be ok, measuring around 17.4 volts dc on my Fluke multimeter.

Evoke 1 Amplifier Board

I opened the case of the Evoke (remove all the screws on the back of the case) this is where I noticed that the antenna f connector was loose. I tightened the f connector using an adjustable spanner and pliers. Thinking it was a poor signal because of the loose connector I switched it on again but the audio problem was still present. I then proceeded to measure the regulator voltages on the main audio board. Most voltages appeared ok, however I noticed that the 3.3v supply which supplies the RF board was not solid but was wavering round quite a bit. I examined the electrolytic capacitors on the board but found no evidence of them being swollen due to localised heating from the various heat-sinks on the board.
So I started removing the capacitors on the amplifier board one-by-one to test them on my Peak capacitance and equivalent series resistance meter. All capacitors measured well within 10% of their marked values and low ESR apart from one capacitor, C7 a 100uF 16V electrolytic which is located very close to a regulator heatsink that runs very warm. Its value measured low and its ESR was higher than it should be so it appears that it is effectively being dried out over time.

Faulty Capacitor C7 on the tester

I replaced this capacitor with one of equivalent value but its voltage rating was higher at 25V this is not a problem in this application.

When I soldered the new capacitor into the board I bent the new capacitor body away from the offending 3.3 V regulator heat-sink to prevent the new capacitor being dried out with the heat.