The bottom line is the higher the filter capacitance the better the amplifier maintains voltage (and power). But there is a limit and any capacitance above that limit is a waste of money.
Power is determined by the voltage in the rails and the filter cap keeps that voltage from sagging during low impedance loading that increases current draw.
The principle is the same as a water pressure tank. If you have well water with a pump but no pressure tank, the water pressure will fluctuate during the time the tap is turned on full and the pump kicks in and then turns off at the pressure set point and then on again as the pressure falls. That fluctuation is the same as power supply ripple.
With a pressure tank, the water piping volume increases and the water velocity slows down, creating pressure. Now, as the water is turned on, the pump feeds the tank and the "ripple" is absorbed by the tank and the water comes out of the tap at a constant pressure, regardless of how much the tap is open.
A pressure tank is just a big pipe, so as the tank size increases the pipe size increases, the pressure becomes more constant. However, there is a limit to the tank size for a given volume of water system to suppress the surges and any size bigger tank does nothing. Just like a filter cap does to the power supply. |
I apologize if I’m veering this thread off course, but wanted to add that the power supply in the Cocci tube amps is unregulated in order to provide speedy power delivery to the audio circuit, while the input section of the audio circuit includes a 3 Henry choke and the output section of the audio circuit includes a 8 Henry choke. Massive power and (200 watt rated) output transformers and Dueland VSFs in the audio circuit. I believe the high speed and great frequency extension of these amps are products of this strategy of quick power delivery to the audio circuit and ’smoothing’ through inductance rather than capacitance.
More than one way to skin a cat... |
I'm no engineer but, transformer size, filter capacitance, output devices, rail voltage, etc. all play a huge roll in amplifier power.
You can have an amp with a larger transformer, medium capacitance, and more output devices that will "perform" the "same" as an amp with a smaller transformer, large capacitance, and medium output devices.
Bill |
First, I STRONGLY recommend you read this: https://sound-au.com/power-supplies.htmHere is the summary: Filter capacitance has roughly 2 functions: - It smooths the waveform from a jagged form to a smoother one. This is called the ripple voltage. The higher the capacitance, the lower the ripple which IS what you want, up to a certain point. Above a certain value, you do not get much improvement. - It supplies instantaneous burst current (energy) when needed. Bottom line is: - Get the highest capacitance you can afford which fits into your physical space. - Its voltage rating should be at least about 20 % higher than your amplifier rail voltage measured when the amplifier is powered on but NOT producing any output. - Get the longest lasting one which are typically around 20,000 hrs at 85 degrees C. |
Lets say you are a tailor and s costumer enters your shop and wants a tailored suit . He decides on some specific cloth then he leaves to return in two weeks to do final adjustments on the client.
The power supply in a amplifier is the responsible to give you the size and type of clothing to be used as a blueprint for the signal that is coming threw the input cercuit . Then the signal modulates ( cuts and forms the suit )
what it receives from the power supply ( the cloth elected by the client )
You can easily understand the great important of a properly design power supply .
Any distortion ( aberration in the cloth rinkles or whatever) or ripples in the converted dc to ac will create problems when the input signal uses that cloth to cut the suit.
Now that you have an idea how an amplfier works lets see what does capacitance fit in all of this . Now remember the signal that modulates the product of the power supply to create a exact copy of the signal that is coming in but at higher levels this signal is not static , it constanly changed in level . It could reach a level higher then what the power supply can acomodate without distorting . The needed energy is warehoused in the capacitors and when needed they comei in , do their job and are recharged again for when needed
This is a very important aspect of an amplifier the capacity to give extra current when needed and is expressed in amps or current .
One more thing is the type of capacitor as they give amplifiers a particular sound and how they are used in the circuit . Do they discharge fast and do they charge fast enought for the next burst . Hope this helps , take care and be safe . |
Filter capacitance is instrumental for speed and the quality of the performance capacitors do count Nichicon super through, are excellent as well as Nippon ,Rubicon .I have owned highly modified electronics and speakers for years, coupling capsif circuit uses them also as well as 1% resistors and very noise regulators like top ofthe line Bellison ,even Solder like WBT , or Cardas
effect how natural it sounds as well as quality wiring ,
pin oututs get rid of the brass connectors on electronics and speakers,over 85% are gold Copper 3 x better conductor much more natural sounding from a purist perspective. |
Thanks George! I really think a lot of people have really missed the boat with Matrix 801s, myself included for a long time....the oft repeated 'need for big solid state amplification' just isn't so in my experience...and in fact, is contraindicated for best sound in my experience.
Still puzzled by the total of 94mF capacitance in the power supply of these killer amps...
|
By the way, the tube amps absolutely kill the Threshold in bass response...much better control of the big 12" woofers in the NorthCreek crossovered Matrix 801 S2s Understandable when you look at the very "benign impedance load" of the 801’s https://www.stereophile.com/images/archivesart/bwll801fig01.jpgBut put something like a Focal No3 Sopra with a magnitude of 2.75 ohms at 96Hz and a combination of 4 ohms and a –56° negative phase angle, then you find the Threshold will be the better. https://www.stereophile.com/images/417FSop3fig1.jpgIt's all horses for courses. Cheers George |
Hoping Ralph will comment on the situation with my custom built 100 watt 4 x KT77 monoblocks (built by Charlie Cocci): when I took off the bottom plate to admire Charlie's work, I was shocked to see only 2 x 47mF 630 volt capacitors in the power supply section. By comparison, my Jon Soderberg updated (200 w/ch.) Threshold Stasis 2 has 4 x 63,000mF capacitors in the power supply section!
When I asked Charlie about this, he told me (paraphrasing) 'When you've got enough voltage, you don't need as much capacitance.' I'm thinking that this may be a statement of the Work equation that Ralph posted.
By the way, the tube amps absolutely kill the Threshold in bass response...much better control of the big 12" woofers in the NorthCreek crossovered Matrix 801 S2s and what sounds like an extra half octave of response. The tube amps kill the Threshold in every other way as well, but the bass response was a real surprise.
Thanks for any insight from anyone into how this can be! |
|
OP, you ask logical enough questions. Except it doesn’t exactly work quite the way you think.
There’s two main ideas behind power supplies. The first one is the one people tend to get fixated on, power. To a certain extent this does work. Amps with big power supplies probably do tend to have better bass and slam and things you would expect to hear from a more powerful amp. I say "probably do tend" because there’s a lot more to it and no shortage of 30 watt tube amps that have more authority and control than 200 watt amps with monster power supplies. Hardly ever helps to focus too much on any one thing.
The other main reason is ripple. Not just with amps but with all components its vitally important to have a nice smooth steady supply of DC power. DC comes from running AC through rectifier diodes. These diodes don’t ever output nice smooth DC. They always have some spikes or ripple in their output. The idea is we use this ripply DC to fill the filter caps which then let out nice smooth DC. The water in a dam metaphor.
Which kinda sorta works. Problem being there are no caps that do this perfectly. This is why you see all these guys asking which caps to use to get what kind of sound in their crossovers, etc. No cap is perfect. They all have some kind of sonic signature.
Put it all together and you can see the way you’re looking at it is pretty much meaningless. First off since no caps are perfect the diodes have as much influence on the sound as the caps. Since ripple matters and ripple is minuscule compared to power then total capacitance hardly matters either. This is why you sometimes will find a tiny little high quality cap in there. The idea is the big caps provide the oomph, the tiny one smooths out the ripple.
Main thing to learn from all this: Its the sum total quality of the whole package that matters. Which you can only evaluate by listening.
The difference by the way with better caps and diodes is a deeper and much more focused image, with less grain and glare, greater resolution of subtle dynamic shadings, and a lowering of the noise floor that leads to a sense of greater power. Even though power as measured by standard measurements is the same.
|
My understanding of capacitors is that they store energy. Does this mean
that my Due amplifier is just as powerful as the on paper more
powerful Atoll in300? No. It just means that the energy storage for the output section will have less noise so less intermodulation at full power. But one thing left out is the voltage of the power supply with all these different amps. Its important because in the formula for electron storage in a capacitor, the voltage dominates the equation W (work) =1/2 (C xVsquared) where C is capacitance and V is voltage. So by raising the voltage a little, you raise the storage a lot! BTW, its the storage in the supply that is usually what is being quoted when the manufacturer states the 'amps' it has- for example 80 amps. If we use that figure we see that its obviously not output power, since using the power formula P=Isquared x R (where P is power in watts, I is current in amps and R is resistance in ohms) If we give the amp the benefit of the doubt and state a 1 ohm load, then the power is simply the current squared. No-one makes a 8400 watt amp. Into 2 ohms this would be 16,800 watts.... So a value like that is actually the current that flows for 10mS when the power supply is shorted out. Its an easier way of understanding the energy there instead of the 1st formula I showed above. Our MA-2 amplifier has 72,000uf at 150 Volts... it can easily do that 80 amps and its a 220 watt tube amp. Capacitive values like this are there solely because the amp sounds better when they are. |
My understanding of capacitors is that they store energy. Does this mean that my Due amplifier is just as powerful as the on paper more powerful Atoll in300?
For comparison sake with my Due amplifier I demoed a Plinius Hautonga integrated amp with 200wpc in 8 ohm and 280wpc in 4 ohm. I did not hear any appreciable difference in bass or other frequencies. The ability for an amp to drive hard loads, also comes down to very much what the output stage is and how much current is availible (transistors amount if they are Bi-polar or mosfet, and yes capacitor storage amount ect ect) If you look at old Mark Levinson ML2 monoblocks they are only 25w into 8ohm, yet can double into 4ohm (50w), double again into 2ohm (100w) and again into (1ohm 200w) This means they can drive ANYTHING to a certain volume level. Too many of todays manufactures "under quote" their 8ohm wattage, so it looks like the 4ohm is doubling (this should be eradicated from the industry and advertising as it’s fraud!!) That why it’s always good to read Stereophiles bench test as they expose this "under quoting" of the 8 ohm figure so the 4ohm looks much better. Cheers George |
If I was trying to sell you an amp, and that is what you wanted, wouldn't that be GOOD selling point?
If my product is VERY shiny, wouldn't that be what I would want to turn your eye towards?
The build quality, the noise floor, how fast, how slow, how big, how small are thing I look for. BUT I'm not a designer... The application, then quality control are KEY issues for me, from REBUILD or repair perspective.
More is not always better, to much can be a problem too.. I've screwed around and doubled the capacitance in certain things...Power supplies in certain SS amps....
Tough to fire some of my contraptions up.... Fuse blowers...Room warmers. But once they are going.. LOL # 10 SC copper wire (20 amp breaker and feed wire) fixed that.
A LOT of BASS a lot of CAP, kinda go hand in hand...
Look at my Mac valve amps... Caps aren't to big at all.. VTL, HUGE!!
Regard
|
