Damping factor of 100 at 10KHz?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Agreed that the DF is immaterial if the cable has enough resistance but i have listened couple of amps having such high figures it even means DF at 100Hz is 10000 hence super grip even with cable resistance this will endup having a DF of in the order of 500 which is more than sufficient.

What are the ways to reduce output impedance adding more output pairs? How many required to get the DF at 10K to be 100. Is there any connection between the DF and Freq response of the amplifier? like extended upper freq like more than 100KHz has better grip at lower frequencies?
 
Moderator
Joined 2011
Agreed that the DF is immaterial if the cable has enough resistance but i have listened couple of amps having such high figures it even means DF at 100Hz is 10000 hence super grip even with cable resistance this will endup having a DF of in the order of 500 which is more than sufficient.

What are the ways to reduce output impedance adding more output pairs? How many required to get the DF at 10K to be 100. Is there any connection between the DF and Freq response of the amplifier? like extended upper freq like more than 100KHz has better grip at lower frequencies?

There's no way to do this without placing the amp right at the driver.
With #6 wire, using three feet long conductors, the loop resistance is 0.025 Ohms. (There is also inductance to consider at HF.)
With an 8 Ohm driver, this gives DF = 320. Even using remote sensing is unlikely to improve things much.
 
Last edited:
There's no way to do this without placing the amp right at the driver.
With #6 wire, using three feet long conductors, the loop resistance is 0.025 Ohms. (There is also inductance to consider at HF.)
With an 8 Ohm driver, this gives DF = 320. Even using remote sensing is unlikely to improve things much.

Agreed im talking only on the specification point of view not on resultant DF when wire used..

Just on the amp specwise...
 
Moderator
Joined 2011
But why? This is one of those specs where the engineering rule of ten applies, Once you get lower than 0.5 Ohm, nothing significant can be achieved and you are creating all manner of stability problems trying

Some of the improvements heard with a lower source resistance to the speaker are due to the crossover's impedance acting as a voltage divider.
See one of the Stereophile amplifier reviews for how this can add significant frequency response aberrations.
 
not a problem for a high feedback amp at its output terminals

Kelvin sensing to high audio frequency at home audio speaker cable lengths is also not really hard, cable construction isn't trivial though, "quadaxial" 4 signal conductor layer coax cables?
feedback may need some modification for the cable Z (Cload)

a 3rd way is synthesizing negative audio frequency impedance amplifier output with mixed feedback - such that it cancels most of the audio frequency cable R
again fun feedback problem - minor issue of amp output sticking to tails when not loaded...


but as others mention it seems a unlikely "solution" with real Loudspeaker loads, their XO components R
 
Last edited:
A couple of things:

Firstly, there is no difficulty in getting the output impedance down to something like 0.02 Ohm without any special measures or any problems with stability.

Secondly, the biggest resistance in the circuit is not the cabling but the resistance of the voice-coil. That will be around 7 Ohm for a unit with a nominal 8 Ohm impedance. This swamps all other resistances.
 
Last edited:
A couple of things:

Firstly, there is no difficulty in getting the output impedance down to something like 0.02 Ohm without any special measures or any problems with stability.

Secondly, the biggest resistance in the circuit is not the cabling but the resistance of the voice-coil. That will be around 7 Ohm for a unit with a nominal 8 Ohm impedance. This swamps all other resistances.

So what makes amp to get iron fist grip on bass?
 
Moderator
Joined 2011
How can u say that? Any mathematical equation?

The bulk capacitance C is where all the current for the load comes from. The larger the capacitance, the more current
that can flow to the load before the supply voltage droops too much. I = C*dV/dt
If a full wave rectified supply (1/120 charging period) with 10,000uF capacitance has 3V ripple, then I = 3.6A.

Also the larger the supply capacitance, the lower the supply impedance feeding the output devices. Z=1/(2*Pi*f*C)
At 20 Hz, a 10,000uF power supply capacitor has an impedance of about 0.8 Ohms.

Bear in mind that the power supply capacitance is in series with the speaker (unless there is an output transformer).
 
Last edited:
A couple of things:

Firstly, there is no difficulty in getting the output impedance down to something like 0.02 Ohm without any special measures or any problems with stability.

Secondly, the biggest resistance in the circuit is not the cabling but the resistance of the voice-coil. That will be around 7 Ohm for a unit with a nominal 8 Ohm impedance. This swamps all other resistances.
0.02 Ohms at lo frequencies is easy enough. The OP asked about at 10KHz, where most amplifiers have output inductors coming into play. The frequency response differences between 0.02 and 0.2 Ohm into even a 4R voice coil are negligible. Adding 4 wire remote sensing or negative output resistance to compensate cable resistance is asking for trouble unnecessarily
 
Bear in mind that the power supply capacitance is in series with the speaker (unless there is an output transformer).

When the amplifier is clipping, yes. Below clip, the effect of feedback dominates the output impedance - especially at low frequency. The feedback drives Zout to very low levels, even if the source impedance of the power supply is several ohms. When you clip the feedback is doing nothing and the speaker is indeed directly in series with the reservoir cap (with an output transistor fully turned on in between them).

The largest differences between amps with oversized vs. undersized caps can be heard when you're driving into clipping. With big caps you may not even notice it until it's very obvious on a scope. You can therefore turn it up louder before the sound degrades, and this gives you that 'iron fist' bass.
 
0.02 Ohms at lo frequencies is easy enough. The OP asked about at 10KHz, where most amplifiers have output inductors coming into play. The frequency response differences between 0.02 and 0.2 Ohm into even a 4R voice coil are negligible. Adding 4 wire remote sensing or negative output resistance to compensate cable resistance is asking for trouble unnecessarily

Permit me to clarify things. One of my Blameless amplifiers, with conventional Miller compensation, gives about 0.02 Ohm at 10 kHz. That does not include an output inductor.

At low frequencies (10 Hz - 2 kHz) the output impedance is flat at 0.009 Ohms. This is low even compared with the internal wiring of a typical power amplifier.

This subject is dealt with in detail on p356-358 of the 6th edition of my book Audio Power Amplifier Design.
 
Moderator
Joined 2011
When the amplifier is clipping, yes. Below clip, the effect of feedback dominates the output impedance - especially at low frequency. The feedback drives Zout to very low levels, even if the source impedance of the power supply is several ohms. When you clip the feedback is doing nothing and the speaker is indeed directly in series with the reservoir cap (with an output transistor fully turned on in between them).

Just a low output impedance does not guarantee an amplifier's ability to deliver the current that such an impedance would imply.
That depends on the power supply.
 
Just a low output impedance does not guarantee an amplifier's ability to deliver the current that such an impedance would imply.
That depends on the power supply.

Zout will remain low until the power supply is overloaded. The amp will then clip. At a much lower voltage than it "should".

It really isn't THAT hard to run up against the limits of your basic 100wpc home audio amp. If you have a nice stiff power supply at 50V or so you'll clip a little playing music "loud". If your supply is wimpy and bouncing around between 30 and 50V it will be clipping a whole lot more with the same volume setting. Even with the same amplifier circuit with .00008 ohm output impedance. Playing either one quietly you'd never hear the difference, but beat on it a little and things change in a hurry.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.