Corner frequency question

[Strikkflypilot]

If I am not completely wrong, the coupling caps e.g. in the Crack is part of a highpass filter on the output. The other part is the 2.49kohm resistor on the jack?

So if I want to keep the cutoff frequency and swap out the cap for a slightly larger or smaller one, I could if I also change the jack resistor, thus keeping the corner frequency but altering output impedance and time constant?

[Strikkflypilot]

Thanks for the reply Grainger.
Slowly I am grasping elements of understanding:)

[Paul Birkeland]

If I am not completely wrong, the coupling caps e.g. in the Crack is part of a highpass filter on the output.

Yes.

The other part is the 2.49kohm resistor on the jack?

No - see Grainger's post.

So if I want to keep the cutoff frequency and swap out the cap for a slightly larger or smaller one, I could if I also change the jack resistor,

The resistor on the jack will be sized to keep the voltage spike at the output of the amplifier decent when turning it on.  If you put in a higher value capacitor in, that resistor value should go down.  Doing so will load the amplifier a little bit more, regardless of which headphones you use.

thus keeping the corner frequency but altering output impedance and time constant?

The corner frequency is the headphone impedance in parallel with the resistor value in conjunction with the C value.  Output impedance is is not altered with these modifications.  (See the 6080 data sheet for the relationship between operating points and impedance)

[Strikkflypilot]

This is a great forum!
I really appreciate You guys taking Your time with this.
I will compare what You are writing with some schematics so I can visualize.

I hope it is ok to post this image here, it doesnt give away much I think. I thought this was the high pass filter.

[Doc B.]

The point here is that the impedance of the headphones is so much lower than the 2.49K resistor in parallel with it that it's value will dominate in the filter calculation. For example, a 2.49K parallel with 300 ohm headphone yields an impedance of about 268 ohms. so your HP filter is composed of the output coupling cap and a 268 ohm resistor.

[Strikkflypilot]

Thanks Doc.

That was the final piece!
Now i got it.

[theGopher]

Doc - doesn't the output impedance of the Crack also come into play in determining the -3dB corner frequency or is it only the bleeder resistance and headphone impedance after the output coupling capacitor that matter?

[Paul Birkeland]

Doc - doesn't the output impedance of the Crack also come into play in determining the -3dB corner frequency

The output impedance can influence where the -3dB corner is, but you have to take into account the actual impedance curve of each headphone you are trying to analyze (and the impedance usually moves pretty wildly at low frequencies, so you may dip down -5dB, then come back up to your reference level, then dip back down again).

or is it only the bleeder resistance and headphone impedance after the output coupling capacitor that matter?

This is a sufficient approximation, and is accurate if you simplify the load that your headphones present as a resistive one. 

[theGopher]

Thanks Paul.

I'd seen mention elsewhere in the forums about the effect of Crack's ~120 Ohms output impedance on corner frequency. I could see how it might affect the overall gain at DC since it would form a voltage divider with the output cap's bleeder resistors in parallel with the headphones, but not how it could affect the AC frequency response of the C-R high pass filter former by the output cap and the output cap's bleeder resistors in parallel with the headphones.