UK Vintage Radio Repair and Restoration Discussion Forum - View Single Post

Craig Sawyers · 3rd Dec 2021, 5:22 pm

You can't just scale capacitors here. the 330nF are film capacitors. Replacing these with electrolytics you need an extra factor of at least 5, and preferably 10 to prevent distortion. Douglas Self (a complete objectivist) presents a whole bunch of measurement evidence about that. Small Signal Audio Design, 2nd Ed, p71-74.

He measures the distortion of a 47uF electrolytic looking into 1k (nominally a 3.5Hz break point - so looks OK). But he distortion starts to rise from a base level of 0.0002% at about 80Hz, and by 10Hz is 0.04%.

The solution is simple, he says - increase the value. At 10Hz into 1k: 100uF, 0.008%. 220u 0.0015%. By 470uF it is close to the 0.0002% baseline.

So 33uF electrolytic into 1k (instead of 330n into 100k) will guarantee increased distortion. You need at least 220uF.

Craig

3rd Dec 2021, 5:22 pm	#53
Craig Sawyers Dekatron Join Date: Nov 2010 Location: Oxford, UK. Posts: 4,991	Re: Quad 34 preamp volume knob clicks You can't just scale capacitors here. the 330nF are film capacitors. Replacing these with electrolytics you need an extra factor of at least 5, and preferably 10 to prevent distortion. Douglas Self (a complete objectivist) presents a whole bunch of measurement evidence about that. Small Signal Audio Design, 2nd Ed, p71-74. He measures the distortion of a 47uF electrolytic looking into 1k (nominally a 3.5Hz break point - so looks OK). But he distortion starts to rise from a base level of 0.0002% at about 80Hz, and by 10Hz is 0.04%. The solution is simple, he says - increase the value. At 10Hz into 1k: 100uF, 0.008%. 220u 0.0015%. By 470uF it is close to the 0.0002% baseline. So 33uF electrolytic into 1k (instead of 330n into 100k) will guarantee increased distortion. You need at least 220uF. Craig __________________ Doomed for a certain term to walk the night