Which parameter of the filtering caps do influence the sound of a gainclone???

Status
Not open for further replies.
Jump to Latest
#22
All this discussion of Caps and changing the Frequency response of the amp - Bah - I agree with jackinnj - do what the LM3886 datasheet says - why wouldn't they know what's best? Everyone's "Opinion" is purely subjective. Do you really think those offering their opinions have their amps in Anechoic chambers? Not me.

Want to really modify the amps sound and frequency response? Move the couch or other furniture in the room. Much easier, no chance of electrocuting yourself, and has much more of an effect over changing input caps from .5uF to 4.7 uF.

Purely my subjective opinion.
 
#23
sharpi31 said:
I_F,

The psu caps are both smoothing the rectified AC and delivering current as the amp demands it. While the voltage swings required by the amp may not require the full slew rate of the capacitor, a higher slew rate would still be advantageous for rapidly responding to the noise and ripples from the rectifier (which extend way beyond 20KHz).

In other words, slew rate will effect how well the cap can smooth voltage irregularities on the supply rail.
Click to expand...

Audio requires about 2V/us. The previous poster said data showed 30-40 V/us from even the cheapest cap. That's a margin of 15-20X what is required. One could argue that there is no need to smooth input side ripple/noise beyond the audio band.

I_F
 
#26
BWRX said:
You could attempt to determine the "slew rate" of a capacitor by charging it up to its rated voltage then measuring how much current it is capable of dumping when its leads are shorted out -> (dv/dt)=i/C, (Volts/second=Amperes/Farads). The more inductance a cap has the slower (and lower in magnitude) the current pulse will be...
Click to expand...

Maybe use a very fast high current MOSFET or SCR which could be switched on in under 100 nS and record the voltage waveform on the cap terminals using a fast digital storage oscilloscope. Its a pretty severe test to completely discharge a fully charged cap in this manner, and it would never occur in normal operation of an amplifier. Maybe a 1V step function would be better, but you would need a 'stiffer' supply to discharge the 1V into.

I am skeptical of the usefulness of knowing the "slew rate" of a power supply electrolytic capacitor and how it would contribute to overall sound and performance of an amplifier, but I would like to continue this line of discussion to see what comes out.
 
#27
Well, forcing a constant current into a capacitor ought to give a constant dV/dt across it (for a LITTLE while, anyway). And since we would want to know what is USABLE with a particular cap (i.e. within specs), it seems fair to use only <= the max rated ripple current, or whatever the manufacturer's max rated current of any kind might be, for the particular cap-under-test (or, if not given or not clear, use <= whatever current keeps it within temperature specs).

So maybe one could bang square waves or pulse trains from a really-stiff current source into a cap, and read the voltage slew-rate off their scope.

And how does ESR affect a capacitor's theoretical maximum voltage slew-rate, anyway? If we use the simplest model, with C in series with ESR, and look at the exponential equations for voltage and current vs time, does that tell us anything about this "slew rate" stuff? And probably the ESL (Equivalent Series Inductance) should also be included in the model. I can't find my slide rule. Maybe somebody should Spice that.

- Tom Gootee

http://www.fullnet.com/~tomg/index.html
 
#28
gootee said:
Well, forcing a constant current into a capacitor ought to give a constant dV/dt across it (for a LITTLE while, anyway). And since we would want to know what is USABLE with a particular cap (i.e. within specs), it seems fair to use only <= the max rated ripple current, or whatever the manufacturer's max rated current of any kind might be, for the particular cap-under-test (or, if not given or not clear, use <= whatever current keeps it within temperature specs).

So maybe one could bang square waves or pulse trains from a really-stiff current source into a cap, and read the voltage slew-rate off their scope.
Click to expand...

If you bang 10mF 63v Sikorel with its rated ripple current it slews 0.003V/us. Not exactly same as quoted 100v/us fiqure. Still doesnt make any sense.
Even if you bang it with 3500A(short-circuit current) it "slews" only 0.35v/us
 
#29
Hi,
if we approach the numbers from the opposite direction, we arrive at 1MA for the peak current at the 100V/uS limit.
It is obvious that ESR & ESL will not allow Mega amps to be sourced or sunk however short the charge/discharge period.

It then follows that the manufacturer's stated slew rate must include allowance for the ESR/ESL effect.
If we reduce the peak current to 1kA and use ESR ~ 30mohm then the volt drop across the non capacitive portion is already up to 30V, that's a significant proportion of the maximum slew rate.

Any further thoughts? My logic can't see how to turn this idea into a useful conclusion.
 
#30
AndrewT said:



It then follows that the manufacturer's stated slew rate must include allowance for the ESR/ESL effect.

Any further thoughts? My logic can't see how to turn this idea into a useful conclusion.
Click to expand...
There is no manufacturer stated slew rate, just some audio-nutcase throwing fiqures to TNT-audio article.

And I think thats the main problem in trying to make any usefull conclusion.
 
#31
EPCOS states a dv/dt rating for some of their big snubber/damper foil caps, to be calculated as follows:
max dv/dt (repetitive) = Ip/C
max dv/dt (non-repetitive) = Is/C
Peak current Ip and surge current Is are given in the tables.

Kind of a "I-will-explode-if-you-stress-me-above-that" rating.

For the B25855 (a true beast of a power cap), say the 40uF/1200V, we have
Ip=8kA, Is=20kA, thus a dv/dt=200V/us and 500V/us resp.
Continuous RMS ripple current is "only" 120A, ESR/ESL/DF=1.5mR/50nH/0.2m

It is a safety rating, not a performance rating...

- Klaus
 
#32
KSTR said:
EPCOS states a dv/dt rating for some of their big snubber/damper foil caps, to be calculated as follows:
Click to expand...
Many other film caps are also dv/dt rated in same way, most polypropylene caps like KP, FKP, FKP1 -series and so on. But it still doesnt explain those funny 100V/us fiqures for electrolytics.
 
#35
mzzj said:

There is no manufacturer stated slew rate, just some audio-nutcase throwing fiqures to TNT-audio article.

And I think thats the main problem in trying to make any usefull conclusion.
Click to expand...

"Audiophile Special"

(From about halfway down the page, here: http://archive.chipcenter.com/circuitcellar/november00/c1100rp58.htm . Link to below-mentioned site included.)

-----------------------------------------------------------------------------------------------------------
"General Electric Industrial Systems has capacitors and Surge Protection Devices the sizes of railroad box cars."
-----------------------------------------------------------------------------------------------------------
 
Status
Not open for further replies.
Top Bottom