Do the filter caps in class A amps wear out faster than they would in class AB? In other words, in generalities knowing there are a variety of ways you can implement a PSU, does class A put a heavier demand on filter caps than a class AB amp would?
The ripple current is larger, and the internal temperature is higher.
But you would use parts that are more suitable for class A, to preserve reliability.
Like capacitors with low ESR, high ripple current, 105C.
But you would use parts that are more suitable for class A, to preserve reliability.
Like capacitors with low ESR, high ripple current, 105C.
Last edited:
For amplifiers with the same power output, the total thermal power on the capacitors will be higher for class A (taking into account the crest factor). Other parts of the class A amplifier (transformer, transistors) also get hotter.
Therefore, it is difficult to design a class A amplifier for a power greater than 10-20 watts. Difficult, expensive, hot 🙁
Therefore, it is difficult to design a class A amplifier for a power greater than 10-20 watts. Difficult, expensive, hot 🙁
Last edited:
FWIW i've recapped big class A amps around their 20th anniversary. Not a single old cap was off speck, main filters as well. None of the caps had a 105deg rating.
Ironically, doing the same service to Jeff Rowland class AB monoblocks of roughly the same vintage was more illuminating. About 90% of the PS caps were either dry or leaking. No idea about the brand as they were proudly JR re-badged.
Krell, in their standard $2k service for old amps used to replace all polarised caps with the exception of the main filters.
As far as i am concerned increased ripple does nothing much to high quality caps. Temperature is a lot more important, yet again, good caps last for decades and not so good fail for no obvious reason in just a few years.
Ironically, doing the same service to Jeff Rowland class AB monoblocks of roughly the same vintage was more illuminating. About 90% of the PS caps were either dry or leaking. No idea about the brand as they were proudly JR re-badged.
Krell, in their standard $2k service for old amps used to replace all polarised caps with the exception of the main filters.
As far as i am concerned increased ripple does nothing much to high quality caps. Temperature is a lot more important, yet again, good caps last for decades and not so good fail for no obvious reason in just a few years.
Heat clobbers 'lytics. Class A draws constant current out of the power supply, so ripple current is the only concern. Best thing to do is to site the caps in the coolest area of the amp possible.
High ripple current causes high internal temperature.
A 105C part will be more reliable in the long run than a similar part rated at 85C.
Every 10C decrease in internal temperature from rated roughly doubles the useful life.
Ripple Current and its Effects on the Performance of Capacitors – Passive Components Blog
Last edited:
The ripple current of keeping the caps maintained is very far below the ripple current rating of the caps. There is never surge current drawn by the amplifiers, just a constant DC current.
The rectifier of the power supply unit operates to charge the capacitor bank in a pulsed mode (short pulses) with the mains frequency. Ripple current of capacitors with double frequency.
Last edited:
The voltage curve is like a saw: charging occurs in a short time at a switching angle close to 90 degrees. Then, during a half-period, there is a discharge with a load current. To reduce ripple, high-capacity capacitors are used, which imposes additional requirements on the diodes and the transformer (power reserve).
That doesn't change my point that the average ripple current is far less than the rating on the caps. And far under the current rating of the rectifiers. I know how power supplies work.
Most of what kills caps is being operated too close to their rated voltage. I've done some switching DC-DC converters where the primary ripple current was too close to the rating, and I had to change the cap, but that is not typical for caps on the secondary side.
Most of what kills caps is being operated too close to their rated voltage. I've done some switching DC-DC converters where the primary ripple current was too close to the rating, and I had to change the cap, but that is not typical for caps on the secondary side.
Old DIY's point was that the rectifiers charge the capacitors in pulses. The volt drop in the rectifier diodes is larger as a result, as well. The point is that the RMS power is larger with pulses than if the current were smoother.