1970s Design Indulgence

Couplingcapphobic?

The following is a quote from a designer who is what I'd call a reasonable and approachable fellow to get on with:

"Because these amplifiers are almost invariably considered "lo-fi" and will normally drive small speakers in horrible small plastic boxes, the coupling cap doesn't make much difference"

I doubt the Quad 303 could be classed as "lo-fi"? Rather than being used with small speakers in horrible small plastic boxes, it was/is often used with electrostatics, or conventional high quality speakers.

The 303 has output capacitors and so had several other well-respected 1970's amplifiers. In the case of the Quad 303 the output coupling capacitor value is 2,000 uF.

My use here of the humble output capacitor is in trying to get away from today's "click-bang" amplifiers. "Click-bang" was a phrase used by my colleague and old friend Paul from about 27 years ago. These days Paul is one of the guys responsible for making commercial radio transmitters work and knows one or two things about stability... at FM frequencies and above!

Almost invariably (to borrow the above phrase) DC coupled amplifiers are at risk of "click-bang", which as most amplifier designers will know (but never admit) is the sequence of switching the thing on, followed swiftly by it self-destructing.

Output protection circuits are often misused to prevent this condition, which is cheating by the way. And I have worked on the development of some amplifiers where they compound the problem.

To DC couple requires a level of complication which is just asking for trouble, and should the trouble occur "in the field" causes the expensive speakers such amplifiers are used with to have their bass driver voice coils melt! This is probably the most common dispute between customer and seller.

Can you imagine working in one room of a factory on a DC coupled amplifier whose output keeps failing to one rail or the other, whilst in another room a customer on the phone is being told his destroyed speaker is not the amplifier's fault?

Back to the output capacitor then: if its value results in cut at 4 or 5 Hz, its audible effects will have long disappeared by the time the output reaches 40 - 50 Hz, and might even lead to that "grunt" one magazine keeps extoling on behalf of its favoured manufacturer.

The problem we have is in finding a capacitor which treats all frequencies as equals. In the past that would have been easy because the capacitor manufacturer would have been aware that his product was being used as an output capacitor in addition to its other uses. But today such high value capacitors are made to make the manufacturer money in the switched-mode power supplies he hopes his product will be used in.

I will add to the above that I have so far not seen any 100V switched-mode power supplies: this voltage rating being the one used here.

People replacing Quad 303 output capacitors beware!

Very Nice! I do hope you are still considering doing this as a Kit..? 

DogBox wrote: Apparently Neville Theil did something similar on the output coming up with 6.8uH in parallel with 6.8ohm [1W] and a 150nF 250VAC cap tying it to ground was able to overcome the cable capacitance.