The usual choice are film capacitors. Some people claim that the differences due to manufacturing tolerances are bigger than the differences from one capacitor type to another. Others put capacitors there that take up more space than the entire rest of the amp and also cost more and everybody has his preferences. Older engineers swear by polystyrol, younger by teflon and you will find devotees to any available boutique component.
It rather depends on the mechanical resilience than on the frequency response. Ideally the slope of the filter should correspond to the slope the speaker's maximum excursion produces. The filter is then set to the frequency below which the amplifier would be able to drive the speaker above its excursion limit. By that you make sure that the speaker is never damaged and that the frequency response remains the same at all listening levels.
You should also adapt it to your listening habits. For loud listening the filter should mimic the maximum mechanical excursion at maximum power. For civilised listening it should mimic the maximum linear excursion at the highest power you usually listen at. You can use http://www.linkwitzlab.com/closed-box1.xls to determine the corresponding slope.
Your taste can of course be different and you may prefer a lower filter frequency with the corresponding compression it produces, because that is what most of us are used to. Most amplifiers go far beyond 20 Hz to be able to drive even the biggest and most robust speakers to their limit. And the additional tactile information we get from subsonic content often appeals more to us than undistorted sound.
thank you PB just what i wanted to know and more. I can tell you know what your talking about based on your answer to jbernst5's question above. people who know just enough about a subject usually recommend top of the price range products and say everything else is junk. people who know alot will usually give a recommendation on a middle/low range product and explain why you dont need to pay 5 times more for a 5% better product.
Wow, this is great stuff. Thanks.
I just was noticing that without the HP filter, the highs sound better at low levels, and when I crank it up more, they tend to get lost. Is this what you are talking about with respect to choosing the cutoff frequency depending on listening levels?
Am I correct in saying that with a resistance of 23K, and a 3uF cap, the cutoff frequency is 2.3 hz, while if I use a 1uF cap, the cutoff frequency would be 6.9 hz? And if so, would it be right to say that for smaller speakers, the 1uF would tend to sound better at low AND high volumes?
Thanks for your explanations!
I just was noticing that without the HP filter, the highs sound better at low levels, and when I crank it up more, they tend to get lost. Is this what you are talking about with respect to choosing the cutoff frequency depending on listening levels?
Am I correct in saying that with a resistance of 23K, and a 3uF cap, the cutoff frequency is 2.3 hz, while if I use a 1uF cap, the cutoff frequency would be 6.9 hz? And if so, would it be right to say that for smaller speakers, the 1uF would tend to sound better at low AND high volumes?
Thanks for your explanations!
Thats exactly why I asked that.
the formula to calculate the HP filter is 1/(R * C)=hz
In your case with 23k input impedance and 1uf dc blocking capacitor 1/(23000ohms * 0.000001farads) = 43.48hz
If you know what frequency you are targeting you can calculate for the capacitor too. say I wanted to filter at 20hz and below first convert frequency to time 1/hz = t or 1/20hz = 0.05seconds then divide time by input impedance t/R=C or 0.05s/23000ohms = 0.00000217farads or 2.2uf
Last edited:
I'm pretty ignorant about this stuff, and am just going from good old Wikipedia's explanation of a simple RC high pass filter. According to them,
f = 1 / (2 x pi x R x C)
So I was using another factor of 6.28 in there for 2 x pi (how DO I make that symbol here?).
If I'm right, for a 20 hz filter, you would use a 7.8 uF capacitor...
Am I confused? Do you know the answer to the question about which cutoff frequency would tend to work better with small speakers, on the high side, or on the low side?
I did read Pacific Blue's explanation about choosing the cutoff based on the resilience of the speaker rather than the frequency per se, but I must say I'm still lost on that.
f = 1 / (2 x pi x R x C)
So I was using another factor of 6.28 in there for 2 x pi (how DO I make that symbol here?).
If I'm right, for a 20 hz filter, you would use a 7.8 uF capacitor...
Am I confused? Do you know the answer to the question about which cutoff frequency would tend to work better with small speakers, on the high side, or on the low side?
I did read Pacific Blue's explanation about choosing the cutoff based on the resilience of the speaker rather than the frequency per se, but I must say I'm still lost on that.
I got my wires crossed with my calculations. Now that I do it again, to get a cutoff frequency of 20 Hz, I was thinking I would have to use a 0.35 uF cap, since 1 / (2 x 3.14 x 230000 x .00000035) = 19.8 hz.
I am coming back to the above comment. A smaller cap will have a higher cutoff frequency and thus may be better for smaller speakers, especially at high volumes? Am I right to say that a larger cap will have a lower cutoff frequency, but will still be more filtering than what I have now, which is none?
I am coming back to the above comment. A smaller cap will have a higher cutoff frequency and thus may be better for smaller speakers, especially at high volumes? Am I right to say that a larger cap will have a lower cutoff frequency, but will still be more filtering than what I have now, which is none?
The formula quoted F= 1 / [ 2* Pi * R * C] gives the turn over or rolloff frequency where the signal is down by 3dB from the reference level, i.e. F(-3dB).
22k and 1uF have F-3dB = 7.2Hz
The frequency at which the level is 1dB down from ref level is ~ two times that, ~14Hz
Below 7Hz the H.P. passive filter continues at -6dB/octave, (the same as -20dB/decade).
22k and 1uF have F-3dB = 7.2Hz
The frequency at which the level is 1dB down from ref level is ~ two times that, ~14Hz
Below 7Hz the H.P. passive filter continues at -6dB/octave, (the same as -20dB/decade).
I have not read all posts on this thread thouigh this question caught my attention.
Yes input cap does matter a lot , in my opinion. You need to try out the types you can get. Film generally being superior to electrolytics but some low cost films are not so great either.
Note that what sounds great on this amp might not sound so great in another application. So trial and error is the way to go about it !
By a mechanism that I don't see conclusively explained anywhere , it can affect how the bass and / or HF sounds. Try an electrolytic and a decent film type and see if you can hear a difference.
If you can't hear a difference it's great. You don't need to spend much money !
There is "nothing" that's not worth a try !
I have some caps that cost 25 times less than some botique caps that I had bought long ago. Both sound the same in my application !
I also have some low cost film 1uF/250V caps like what you probably have. They sound good by themselves. There are better caps than that but the difference in most cases is not 'night and day ' !
Try to stick to polypropylene types. They generally do seem to sound better. They are also more expensive though I find lots of low cost polypropylene types in the catalogues of many manufacturers. Hard to find them in the retail market !
Use the X2 rated caps ( used for mains application and rated 275 V etc.). They are generally all polypropylene and freely available at reasonable cost.
Cheers.
I have some caps that cost 25 times less than some botique caps that I had bought long ago. Both sound the same in my application !
I also have some low cost film 1uF/250V caps like what you probably have. They sound good by themselves. There are better caps than that but the difference in most cases is not 'night and day ' !
Try to stick to polypropylene types. They generally do seem to sound better. They are also more expensive though I find lots of low cost polypropylene types in the catalogues of many manufacturers. Hard to find them in the retail market !
Use the X2 rated caps ( used for mains application and rated 275 V etc.). They are generally all polypropylene and freely available at reasonable cost.
Cheers.
No. You may be experiencing the effect a variable resistance has on a low-pass filter. The better sound without the HP filter is why people choose DC coupled amplifiers or why they experiment with different capacitor types.
Or you notice the effect of human equal loudness perception. Our hearing is less sensitive at extremely high and low frequencies and that worsens at low volume. That led to the invention of loudness switches, but they are nearly extinct now, because manufacturers never got them right.
Yes.
No. If you hear any difference at all the 3 µF cap should sound better at low to medium volumes.
The standard setup uses a -3 dB point well below the audible range. When you turn the volume up far enough you may find that the sound becomes thin or blaring as the woofer reaches its excursion limit and low bass cannot keep up with the rest of the music. To counteract that you would need a -3 dB point well in the audio range. That will perceptibly worsen the sound at low volumes, because a woofer will not reach its excursion limit there even without the filter and the additional low frequency content will improve the sound. E.g. for a 160 mm woofer you would need a filter at 50-100 Hz. That means more than one octave at the bottom will be missing. The advantage of such a filter is that the sound does not change with the volume which will make it appear more relaxed at high volumes. The woofer will also distort less at high volumes with such a filter, because it remains in its linear working range.
It may be easier to understand when you think of a tweeter. Nobody would expect a 20 mm dome tweeter to reproduce 500 Hz. Usually there is a high-pass filter and another driver will take over below that. A woofer should equally be relieved of tasks it cannot perform. The reason why that is not done is that there is usually no further driver below the woofer. Having more low bass at normal listening level is habitually given preference over having less distortion and a consistent frequency response at higher listening levels. That must not necessarily be what suits you best.
Thanks for your explanations and patience. I'm most grateful.
Now I added a 1uF film cap between the pot and the amp board. I then tried removing R1, which was the 1K resistor in series with the new input cap. I did that based on a post I found here:
Mick Feuerbacher Audio Projects
Am I right to think I don't need that 1K resistor? I think it sounded a little better after I removed it.
I am still missing the highs. I listed to jazz and lose the detail in the ride cymbals, for example. This is driving me nuts!
I just came across this snippet from Danielwritesbac 4 years ago:
"You can also add a very small value ceramic cap or polyester cap as a "bypass cap" to your input filter cap, CIN. See the discussions on bypass cap practice and extra "air" effects. That can make the upper mids less apparent by boosting frequencies that are higher than the upper mids. About 4.7nF (code 472) is a good spot to start, and then try progressively larger values until the "highs" appear."
So, now I am wondering, is THIS the answer to my issue?! Does this simply mean putting the tiny cap in parallel with the coupling cap?
The same post by Danielwritesbac goes on to say:
"Choose between a pair of 220uF, or 330uF, or 470uF for the power caps at the amplifier board.
This will "make a difference" in the audio; however, since capacitors vary greatly, then it is up to you to figure it out. "
The BrianGT kit uses 100uF parallel with 100nF caps for Cs, I think for the snubber power supply. The National Datasheet mentions using 2200uF caps for Cs. Could this be part of my problem. Are these the same caps that the above suggestions pertain to?
Thanks again for continued input. While I am dense, I'm starting to learn some things...
Now I added a 1uF film cap between the pot and the amp board. I then tried removing R1, which was the 1K resistor in series with the new input cap. I did that based on a post I found here:
Mick Feuerbacher Audio Projects
Am I right to think I don't need that 1K resistor? I think it sounded a little better after I removed it.
I am still missing the highs. I listed to jazz and lose the detail in the ride cymbals, for example. This is driving me nuts!
I just came across this snippet from Danielwritesbac 4 years ago:
"You can also add a very small value ceramic cap or polyester cap as a "bypass cap" to your input filter cap, CIN. See the discussions on bypass cap practice and extra "air" effects. That can make the upper mids less apparent by boosting frequencies that are higher than the upper mids. About 4.7nF (code 472) is a good spot to start, and then try progressively larger values until the "highs" appear."
So, now I am wondering, is THIS the answer to my issue?! Does this simply mean putting the tiny cap in parallel with the coupling cap?
The same post by Danielwritesbac goes on to say:
"Choose between a pair of 220uF, or 330uF, or 470uF for the power caps at the amplifier board.
This will "make a difference" in the audio; however, since capacitors vary greatly, then it is up to you to figure it out. "
The BrianGT kit uses 100uF parallel with 100nF caps for Cs, I think for the snubber power supply. The National Datasheet mentions using 2200uF caps for Cs. Could this be part of my problem. Are these the same caps that the above suggestions pertain to?
Thanks again for continued input. While I am dense, I'm starting to learn some things...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.