I try to replace the DC blocking capacitor of my Revel B15a internal amp. and found that it is with 330uf , of course , no film possible , I have good 220uf in hand , is it OK to replace 330uf with 220uf in this occasion ? Will it impair its performance ? I heard this big value is important for low end response .
I have replaced my BAT amp. before and it is only with 3.3uf film capacitor only , why that big difference ? Is BAT not intended as bass amp. It is almost flat to Zero Hz .
Yours kind reply to this is appreciated .
Larry
I have replaced my BAT amp. before and it is only with 3.3uf film capacitor only , why that big difference ? Is BAT not intended as bass amp. It is almost flat to Zero Hz .
Yours kind reply to this is appreciated .
Larry
Very odd to have that large a value for a low lvl high impedance input since most line level sources wouldn't drive an impedance as low as what that size would imply. Are you sure these aren't maybe rail filters that happen to be close to the input traces? Are there smaller ones very close to them that may be them instead?
Thanks all for reply .
jerluwoo - Not power rail capacitors , there are 2 of them for balance input , connected directly to Hot & Cold pins , 3rd one connect to tap of hot pin - for RCA input .
RJM1 - thanks for suggestion , but I am not that good on this , just DIY with general knowledge .
Osvaldo - I have some Panasonic Premium Cap , 220uf - 25v , can I pair 2 to 440 uf , will it be as good as one single one .
Thanks all again .
Larry
jerluwoo - Not power rail capacitors , there are 2 of them for balance input , connected directly to Hot & Cold pins , 3rd one connect to tap of hot pin - for RCA input .
RJM1 - thanks for suggestion , but I am not that good on this , just DIY with general knowledge .
Osvaldo - I have some Panasonic Premium Cap , 220uf - 25v , can I pair 2 to 440 uf , will it be as good as one single one .
Thanks all again .
Larry
The capacitance does not determine the frequency response, by itself. It is part of a passive RC high-pass filter, which is formed whenever a series capacitor is followed by a "shunt" resistance to ground.
A high-pass filter attenuates low frequencies, not letting them pass as well as it does higher frequencies.
It's just like when two resistors are used to make a voltage divider, except here the series one is replaced by a capacitor, which then makes it a "frequency-dependent voltage divider".
By making the R be the series component and putting the C after it, to ground, you could have a low-pass filter (not that you want that, here).
The roll-off in amplitude as the frequency gets lower is somewhat gradual, but people often specify RC filters by what they call their "cutoff frequency" or corner frequency, which is the freequency where the output voltage is 3 dB lower than the input voltage (-3dB is a factor of 0.7071 if we're talking voltage, or a factor of 0.5 if we're talking power in Watts. The cutoff frequency is given by
f = 1 / ( 2 π R C )
Looking at that equation, where the 2, Pi, R, and C are all multiplied together in the denominator, you can see that any C could make any cutoff frequency, because the R could be changed so that R times C was whatever you wanted.
So a very large C with a very small R could still not give good bass, for example.
Usually there is a resistor to ground, just before the first actual amplifying component (like a transistor or opamp). You would have to determine the resistance between the downstream lead of the capacitor and the signal ground, and then use that, AND the C value, to calculate the filter's cutoff frequency.
Typically, the cutoff frequency is set to be very low, often to less than 1 Hertz (one cycle per second). Since the attenuation by the passive RC high-pass filter gradually increases as the frequency gets lower, you typically should pick a cutoff frequency that is at MOST 1/10th of the lowest frequency that you don't want the filter to affect, although many people go 10x further down than that.
Example: If the input-to-ground resistor in your amp is 10,000 Ohms, i.e. 10kOhms, and you want a 0.1 Hz cutoff frquency, then solve the equation for C and plug in the values:
C = 1 / ( 2 π R f ) = 1 / ( 2 x 3.14 x 10000 x 0.1) = 0.001590 Farads
That's 1590 microFarads.
But if the input R was 100,000 Ohms (100k), i.e. ten times more than before, then C = 159 microFarads would give exactly the same cutoff frequency of 0.1 Hz.
There are usually other reasons for picking the value of either the R or the C, which then dictates the value of the other one, for any given desired cutoff frequency.
For example, you "could" use ten million Ohms (10 Meg) and then you'd only need 1.59 uF to get the same 0.1 Hz cutoff frequency. That capacitor would either be a lot cheaper or you could get a really good-quality cap for the same price as a crappy much-larger one.
But we don't use 10 Meg because it would pick up a lot of noise, and would also generate a lot more noise of its own. Lower resistor values are less noisy. But we can't go TOO low or else the source device might have trouble supplying enough current to keep the voltage up at the right level. So R usually gets picked and then C is calculated based on what f we want.
Cheers,
Tom
Last edited:
this is what Rane suggest. a 220uF panasonic non-polar on each signal line with a film in parallel.
Find the Rane note.
Although the more recent Bateman cap report strongly suggests that the reduced distortion of using back to back non-polar would be worth investigating.
Tom , thanks very much , I now understand why we can find some input blocking capacitors with as low as 1 uf and some , as in my Revel B15a sub amp. as high as 330uf . I learn a lot from this forum , reading other questions and related replies allow me to upgrade myself considerably .
Thanks again with appreciation .
Larry
Thanks again with appreciation .
Larry
It's gootee, not Goottee. Thank you.
The font is New Times Roman, which is a standard MS Windows font, and which is offered by this forum's drop-down font selection menu. So it is a standard font for this forum. Selecting different fonts is standard, in this forum. It is not the default font. But it is a standard font.
If that font doesn't display correctly for you, please let me know. Otherwise, please understand that the default font does not correctly show Greek characters, which bothers me when I am posting equations that need them. I do try to only use a non-default font when I post equations.
Oh alright. I will go back to the more-laborious method that I used in the past: I will try to remember to insert the codes that will turn the font on only at the start of each equation and then turn it back off and revert back to the default font at the end of each equation.
Damn. I am getting far too easy to get along with.
you may be misremembering - Bateman specfically "debunks" back-to-back polar electros - showing non-polar with full oxide thickness on both foils measures lower in distortion
he does recommend when blocking more than just a couple of Vdc that you should use 2x series non-polar Al electros of very much higher V rating
I really don't believe paralleled low value films affect audio when good low ESL, ESR electros will have 10-100 mOhms Z over all audio frequencies
if that amp was meant to be driven off the power amplifier outputs, then higher values can be used...what happened to the old caps? were they blown? try 470 at higher voltages say 100v dc...what was the original rated voltage?
sorry Andrew, I misread you - "back-to-back" immediately suggests to me that the devices have orientation, for electrolytic Cap means polar
so I assume when I see "back-to-back" in these discussions that someone is bringing up the older, discredited suggestion of opposed orientation polar electros - sometimes with a polarizing V applied thru a MegOhm to the middle
so I assume when I see "back-to-back" in these discussions that someone is bringing up the older, discredited suggestion of opposed orientation polar electros - sometimes with a polarizing V applied thru a MegOhm to the middle
Yes, Bateman clearly states that the series pair of Bipolar Nitai, perform better than a single bipolar and better than a series pair of back to back polar electrolytics.
But Bateman also states that the back to back polar electrolytics perform better, by a big margin, than a single polar, but generally not quite as good as a single bipolar.
Some of the DC biased testing of the silmic and blackgate came out a lot worse than the dual polar electrolytics.
Those conclusions are not what I would refer to as
But Bateman also states that the back to back polar electrolytics perform better, by a big margin, than a single polar, but generally not quite as good as a single bipolar.
Some of the DC biased testing of the silmic and blackgate came out a lot worse than the dual polar electrolytics.
Those conclusions are not what I would refer to as
Hi guys !
In circuit diagrams of both voltage controllers and amplifiers, you often see a film capacitor of a few nanoFarads in parallel with a much larger electrolytic.
What is the function of this film capacitor, how does it work in this situation, and why is it that the electrolytic's capacitance seems "unable" to do what the tiny film capacitor can ? - This has baffled me for yonks!
Thanks
In circuit diagrams of both voltage controllers and amplifiers, you often see a film capacitor of a few nanoFarads in parallel with a much larger electrolytic.
What is the function of this film capacitor, how does it work in this situation, and why is it that the electrolytic's capacitance seems "unable" to do what the tiny film capacitor can ? - This has baffled me for yonks!
Thanks
There are nonlinearities in electrolytics that allegedly (probably) affect the sound, especially if the electrolytic isn't biased. Film caps have a much wider bandwidth and better linearity. It is debatable whether the extra bandwidth of the film cap is necessary for audio frequency coupling, but using the film cap in parallel does reduce some of the nonlinearity of the electrolytic, and it's cheaper than just using a big film capacitor.
So the film cap can pass some of the higher frequencies with more linearity than the electrolytic, which results in arguably cleaner sound. Bateman (which I am reading for the first time thanks to the folks here) addresses this and many other capacitor topics in his writings. Check it out.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.