Subsonic Filters & Film Caps

[SY]

I think you're plugging in the wrong numbers. Unless there's an output coupling cap (rare), the speaker load is irrelevant- what counts is the resistance load following the coupling cap, which is likely to be 10k-1M ohm, depending on the circuit and where in the circuit the cap is used.

Personal note from someone who's been doing this stuff for longer than most people here have been alive: You are unlikely to find a real improvement in the sound (as opposed to the expected placebo after shelling out hundreds of dollars) of a decently engineered circuit by just popping in an ultraexpensive component. If you have a good quality polypropylene cap (a couple of dollars), further expenditure is for bragging rights, not actual performance. Willy-nilly replacement of components is a popular audiophile hobby since it doesn't require any actual electronics knowledge. The quality and honesty of your writing suggests that you could easily learn the electronics basics, and thus armed, could make the sorts of circuit changes that might actually increase performance rather than just costing money.

[AmCan]

Originally Posted by SY

I think you're plugging in the wrong numbers. Unless there's an output coupling cap (rare), the speaker load is irrelevant- what counts is the resistance load following the coupling cap, which is likely to be 10k-1M ohm, depending on the circuit and where in the circuit the cap is used.

Personal note from someone who's been doing this stuff for longer than most people here have been alive: You are unlikely to find a real improvement in the sound (as opposed to the expected placebo after shelling out hundreds of dollars) of a decently engineered circuit by just popping in an ultraexpensive component. If you have a good quality polypropylene cap (a couple of dollars), further expenditure is for bragging rights, not actual performance. Willy-nilly replacement of components is a popular audiophile hobby since it doesn't require any actual electronics knowledge. The quality and honesty of your writing suggests that you could easily learn the electronics basics, and thus armed, could make the sorts of circuit changes that might actually increase performance rather than just costing money.

Thanks. Your point is well taken. When one first gets started in DIY, forums they are littered with suggestions of upgrades and my intuition is to ask not just how but also why. I will stick to basic value swap upgrades for now and revisit this when I can make my own determinations. Not to belabor the point but is this an output coupling cap in the schematic on the shaping board?

[SY]

If you're talking about C1, it's a Zobel cap, not coupling. Unless it's really, really bad, changing it will have close to zero effect on the performance. You don't want to change the value- what it does is helps keep the amp stable at very high frequencies under some odd (but possible) reactive loads. A cheap polypropylene is more than good enough.

[AndrewT]

a lossy MKT, or MKS would do in the Zobel location.
It should be completely inaudible if it is doing it's stability ensuring job correctly.

[AmCan]

Thank you Andrew and Sy. That kept me from spending about $60 bucks!

[AmCan]

Originally Posted by picowallspeaker

Great question , thou !
I also changed some caps in my preamplifier , leaving only the 1 µF electrolytic ,which is...in the subsonic filter !! Then lately I also re-discovered the beauty and perfection of vynil and turntable . And with some records at medium volume it really becomes unbearable to see those cones do the HOP ;
but putting the Subsonic filter it really does nasty things , no good ,you know ,it puts a veil...
So I think it's a real problem ,when listening at medium/ high volume ,but it has to be related with the whole : mechanical damping /isolation of the turntable =>electronic chain =>speakers and feedback make the circle again.
So the first things to do would be to eliminate or correct the possible causes ,in other words to put every piece of the chain outside audio band resonance.

There is no getting around using this filter so I would at least like to try to change the cutoff frequencies to include as much of the audible frequency band as possible. First, I have no idea where the hi & low frequency cut off is occurring and 2nd what values I could substitute to extend the cut off. I would appreciate it if someone could help me modify this circuit. Having listened to this circuit the low cut off is not too bad considering my speakers are only flat down to 38 hertz. But the high frequency cut off is way too low. I would like to extend it to 20khz. Maybe 35hz to 20khz. I have attached the schematic. Any help or advice is appreciated.

[DF96]

You have there two Sallen-Key filters, one high pass (low cut) and and low pass (high cut).

The high pass uses an opamp, wired as a follower. The frequency response is set by R21,R22, C17,C18.

The low pass uses a BJT emitter follower. Frequency response set by R15,R16, C14,C15.

Do a web search for Sallen-Key filter. The standard circuits will refer to R1, R2, C1, C2 so you need to match these to your circuit. There are always two components in series feeding the follower input (Rs or Cs). Then there is a component to ground at the follower input (C or R), and a component (C or R) from the follower output back to the junction of the two series components. The actual numbering may vary from one author to another so I won't confuse you by trying to second-guess here.

PS if you move the filter frequencies outside the audio band then all they will do is add phase shift, without actually reducing any unwanted signals - better then to simply omit them. With the Sallen-Key you need to calculate both frequency and Q.

[AmCan]

Originally Posted by DF96

You have there two Sallen-Key filters, one high pass (low cut) and and low pass (high cut).

The high pass uses an opamp, wired as a follower. The frequency response is set by R21,R22, C17,C18.

The low pass uses a BJT emitter follower. Frequency response set by R15,R16, C14,C15.

Do a web search for Sallen-Key filter. The standard circuits will refer to R1, R2, C1, C2 so you need to match these to your circuit. There are always two components in series feeding the follower input (Rs or Cs). Then there is a component to ground at the follower input (C or R), and a component (C or R) from the follower output back to the junction of the two series components. The actual numbering may vary from one author to another so I won't confuse you by trying to second-guess here.

PS if you move the filter frequencies outside the audio band then all they will do is add phase shift, without actually reducing any unwanted signals - better then to simply omit them. With the Sallen-Key you need to calculate both frequency and Q.

Thanks for pointing me in the right direction. There are several calculators available to help determine values. I need to spend some time absorbing this information. The concept seems simple but I am not an engineer and some of the language is beyond my knowledge base. Meaning I still don't know what to do but I would like to change the resistor values because I have already selected some nice capacitors for this circuit.

[tvrgeek]

Originally Posted by DF96

Apart from an all-pass filter.

Or a digital filter

[picowallspeaker]

Low-cut and hi-cut are (were ) also known as rumble filter and scratch.
Guess why ...