Replace resistor by low-inductance resistor - Is it audible?

[SIY]

Translation of Carver’s marketing speak: low output impedance.

 

[DualTriode]

@katch22i, and All,

“Same thing?”

I am not going to write on the white board that “the speakers listen to the room”. That is marketing copy.

Yes the load does interact with the (thinking voltage amplifier) with the voltage source.



@KSTR, and All,

Driver parameters include electrical Q and mechanical Q which together make up total Q for the driver. Install the driver in a box and the box will also contribute to the system Q. Q is all about the damping curve, so far without considering the contribution of the amplifier. Not that the amplifier does not play a part.



@All,

The typical amplifier we are speaking about is a voltage amplifier that works hard to maintain a constant output voltage for a given input voltage. Without the amplifier feedback loop the amplifier will have higher output impedance. The amplifier output voltage will increase when the load impedance increases and the other way about.

With the feedback loop in place and the load impedance increases, the output voltage will tend to rise. The feedback loop subtracts a portion of the voltage rise from the amplifier input voltage and maintains a “constant” output voltage for a given input voltage. The output voltage will remain “constant” even as the driver impedance varies with frequency.

So if damping is a ratio of load impedance to amplifier output impedance it is feedback that keeps the amplifier output impedance low.



@solderdude, and All,

I like your experiment.

Input one channel of a stereo amplifier, connect the output to a resistor.

Short the input of the second channel and connect the output to the same load as the first channel. This makes the output of channel number one the input of channel number two in fact.

Turn the whole thing on and watch the voltage dance in the feedback loop of channel number two.


Thanks DT

 

[SIY]

Hear that noise? It was the point whizzing over your head.

All of this is totally irrelevant to the question at hand, not to mention being grossly misunderstood.

 

[HammerSandwich]

Translation of Carver’s marketing speak: low output impedance.

Low?

 

[amirm]

Let's tone down personal remarks folks.

 

[SIY]

Low?

Seems like it, but it's so buried under Carver's carnival barker persona that it's tough to tell.

With a tube type amplifier, when back EMF appears at the output terminals, to the loudspeaker the output transformer’s secondary becomes a primary! The back EMF enters the amplifier through the transformer and affects frequency response downstream. With ordinary tube type amplifiers, speaker matching becomes critical. Our amplifiers are load-invariant.

Through decades of experimentation and some pretty heavy mathematics, Bob Carver has been able to use the Back-EMF to increase efficiency in the mighty SunFire subwoofers, and to improve the soundstage and openness of his full range amplifiers.

 

[HammerSandwich]

Thought you'd typoed, but I see where you're coming from. It seems that Bob has another trick, though. The copy still talks about how the amp "listens to the room," and the specs list a 1.7Ω output impedance. So, if FR is load-invariant, he must have another compensation in place, right? HTH that works with typical speakers' impedance variations is above my pay grade.

 

[SIY]

So, if FR is load-invariant, he must have another compensation in place, right?

Except he didn't quite say that. He said "Our amplifiers are load-invariant," which means... basically nothing. Carnival barker. Which is a pity because in some areas, he has done some remarkably innovative and creative design work. The tube stuff of his I've seen, though, is utterly conventional- the Silver Seven, for example, was just a scaled-up Mullard 5-20 from 1955.

 

[snake3276120]

as much as 0.02 dB! lol

 

[DualTriode]

Hello All,

While under house arrest just for fun I took a look at a Vifa XT25BG60-04 made in Denmark tweeter, Mills 4R 1% Non-Inductive wirewound resistor. Using a Bode100 VNA I looked at impedance, phase and inductance of both the resistor and driver and then in series.

As stated other places in this thread the driver inductance was much higher for the driver than for the resistor.

Most interesting for me was to see the phase angle for the resistor and driver in series was approximately half of the driver alone at 2k and 20kHz.

I have a handful of PP capacitors to do similar tests on while I am under house arrest tomorrow.


Just for fun DT


Driver without series resistor.

Driver without series resistor.

Mills Non-Inductive Resistor.

Mills Non-Inductive Resistor.

Driver with series resistor.

Driver with series resistor. Inductance

 

[RayDunzl]

Here's a little demonstration that might illustrate "damping" in some way.

Just consider the coil (voice coil) moving past the magnet, instead of the magnet moving past the coil (it's all relative, anyway).

The two choices of impedance here are "near infinite" and "near zero" ...

Or not, but the discussion reminded me of it.

 

[SIY]

As stated other places in this thread the driver inductance was much higher for the driver than for the resistor.

Most interesting for me was to see the phase angle for the resistor and driver in series was approximately half of the driver alone at 2k and 20kHz.

It's exactly what's expected. Z = R + iωL, where R includes VC DCR. So with a little algebra, you can put this in magnitude/phase notation for Z.

 

[DualTriode]

This house arrest continues.

Luckily I brought a couple of instruments to the house before the lockdown.

I have swept the impedance of a handful of Non-Inductive wirewound resistors and some without the NI label. Within the 20 hz to 20 k HZ band the impedance and phase curves look 99% the same. Up in the M Hz range bigger differences creep in but not large differences.

In terms of inductive impedance in the audio frequencies, non-inductive are good, parts without the NI label are much the same. Close tolerance parts are important. 10% parts need not apply. (today’s opine.)

I do have a pair of 8R 500Watt VFD buss brake resistors. The resistors measure ~ 8Ohms and 10 uH between 20 and 22k.

There are also a couple of power amplifiers on the shelf. I will attempt to trace out the feedback loop back from the output of one of them and see what the oscilloscope shows. (This is @solderdude experiment.) This will happen over the next few days.


Thanks DT

 

[pozz]

Just found a good talk on the topic of audio resistors:

 

[wje]

I was following another YouTube speaker upgrade video and those "dreaded" sand caste resistors were brought up again. Thankfully, for this thread, I was able to deprogram myself that such resistors were so horrific.

However, when I finish building those Klipsch RP600M crossovers that I purchased when I had my Klipsch speakers, you know I'll be making clarifications that the resistors utilized in the crossover were not sand caste resistors. Needless to say, I won't offer them for sale here on this forum.

 

[lamode]

Thanks to the OP for posting this. Great to see the real-world difference.

 

[kach22i]

Still reading this paper from a couple of people at Vishay

Title:
Selecting resistors for preamp, amplifier and other high-end audio applications

By Dr. Michael Belman (Vishay Intertechnology) and Yuval Hernik (Vishay Precision Group) 08.25.2010
https://www.eetimes.com/selecting-r...lifier-and-other-high-end-audio-applications/

Quote:
Overall resistor noise has several components. The most relevant for audio applications are thermal noise and current noise

End Quote.

 

[JohnYang1997]

Still reading this paper from a couple of people at Vishay

Title:
Selecting resistors for preamp, amplifier and other high-end audio applications

By Dr. Michael Belman (Vishay Intertechnology) and Yuval Hernik (Vishay Precision Group) 08.25.2010
https://www.eetimes.com/selecting-r...lifier-and-other-high-end-audio-applications/

Quote:
Overall resistor noise has several components. The most relevant for audio applications are thermal noise and current noise

End Quote.

Hence:
The resistor value(thermal noise) is the most important thing of a resistor in terms of noise performance. While many designers still use high value resistors in places that don't need them....

 

[ctrl]

Still reading this paper from a couple of people at Vishay ....
Quote:
Overall resistor noise has several components. The most relevant for audio applications are thermal noise and current noise

Above 1000Hz only thermal noise plays a role according to the link and this is what it says:

Thermal noise is notable for being independent of the resistive material. In fact, the thermal noise level is the same for any type of resistor provided the resistances and temperatures are the same.

If you then look at the order of magnitude (22 kHz frequency range @300K -> thermal noise −130,58 dBm), it is finally clear that this did not play any role at all in connection with the upgrade of resistors in crossovers.

Nice try!

 

[JohnYang1997]

Above 1000Hz only thermal noise plays a role according to the link and this is what it says:


If you then look at the order of magnitude (22 kHz frequency range @300K -> thermal noise −130,58 dBm), it is finally clear that this did not play any role at all in connection with the upgrade of resistors in crossovers.

Nice try!

wdym?