John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
<snip>

I never realised that there was so much confusion about such a simple point. Ohm's law is true for ohmic materials. An ohmic material is one for which Ohm's law is true, to a considerable extent i.e. voltage and current are approximately proportional. A non-ohmic material is one for which Ohm's law is not true. If, as some seem to be saying, Ohm's law is simply an equation with a ratio which is not necessarily constant then how do you define ohmic? Ohm's law becomes a name for an equation instead of a name for a common phenomenon. <snip>

Although it is most likely not of relevance for our normal use (for example in audio) but generally ohmic resistors are ohmic if the material is used in the region where it follows/obeys Ohm's law.
In general it does not only depend on the specific material used but on the conditions under which it is used.
 
Last edited:
As to impedance compensation for individual drivers, this may be useful in order to make a passive xover work properly. For example, the rising impedance of a woofer renders the coil in the xover less effective and makes it impossible to get the desired filter slope. A zobel is your answer. Another situation would be a tweeter. When filtered relatively close to the resonance frequency, an lcr in parallel may be required to present the correct impedance to the xover. In these situations, impedance compensation may be a useful tool, but only at the individual driver level.

Impedance compensation before the xover is putting the horse behind the carriage.

Yah.....I ended up removing it from before the xo (made the mid/vocals too harsh) was better without, then for the heck of it I tried it up on the tweeter after the xo and I’m getting that focus and better stage without the harshness......still gonna take some more critical listening to see if it’s a permanent addition.
 
I’ve got a Dayton imm-6 calibrated mic and audio tool app.....I’ve been correlating for about a year. And to tell you the truth there weren’t many surprises, i’d been playing around with audio for years.....more than 4 decades actually.

But yah I know my approach to this makes you objectivists a bit nuts but it’s just how I’m wired!

I really do appreciate everyone’s help (and patience for that matter) :p
 
Last edited:
But yah I know my approach to this makes you objectivists a bit nuts but it’s just how I’m wired!

Why would you say that? I've watched folks randomly mixing and matching components for years, they sometimes end up with something that makes them happy. I like some and hate others and have yet to have faeries carry each note lovingly to my ears.
 
www.hifisonix.com
Joined 2003
Paid Member
Yes, the return energy is transduced in the RC resistor and converted to heat. (a perfect short cannot convert electrical energy to heat energy) The usual situation is that the amplifier has to deal with this return energy, my experience is that amplifiers sound cleaner and run cooler with resistive load/no return energy.

Surely the return energy splits between the damper (Joe’s parallel Network to the speaker) and the amplifier. So the amplifier will not be dealing with the full return energy but that left over from the total less that absorbed by the parallel network. For this not to be the case, the amplifier output impedance would have to be infinite. Clearly if it’s delivering (high) currents into the load, that’s not possible.

What happens is you drive the reactive speaker load with a current output amplifier so that no return energy can be absorbed by the amplifier?
 
Last edited:
Dan has said a number of times NO energy gets back to the amp, however it would be interesting to know how this can be so without using an "all knowing magic" shunt network
"NO" energy would certainly be the wrong claim. What is true is that the AC energy the amp has to deliver appears as an ohmic load but the total load is increased as well, so a two-sided sword. Basically we load the amp down with an effective low ohmic resistor and hope that it does better, with a more stable and perceptually benign distortion profile though even with higher total distortion. Might have some merits with rather high distortion amps, certainly still open to investigation.
 
Max Headroom said:
So if the voice coil inductively stored energy does not transfer to the amplifier, where the hell does it go Einstein ?.
You don't need to be Einstein to realise that almost all of the voice coil inductively stored energy goes into heating the voice coil, since this is likely to be the highest resistance in that current loop.

Bonsai said:
Surely the return energy splits between the damper (Joe’s parallel Network to the speaker) and the amplifier.
It would be better to say that the 'return current' splits in this way. If the ampifier is a voltage source, or approximately so, then almost none of the energy ends up there or in the damper. Almost all of the energy ends up in the voice coil resistance.

What happens is you drive the reactive speaker load with a current output amplifier so that no return energy can be absorbed by the amplifier?
Misleading question, as the same is true for a voltage output amplifier.

There may be some confusion between current and energy? To get energy produced or dissipated you need current and voltage together.
 
When an amp is discharging a reactive element, energy is returned to the amp's pass elements. This only occurs during second and fourth quadrant operation, and the pass element voltage will be rail plus output voltage. During charging, pass element voltage is rail minus output, easier on dissipation.

A speaker stores energy in it's locked coil inductance, in it's position (air pressure times displaced volume / air pressure), in it's velocity (momentum of air and cone), and by gaining or shedding magfield energy as Le changes with position.

Where the storage acts as a pure reactance, a parallel reactive network may reduce the amplifier operation in quadrants two and four. But no non active network is capable of compensations the non linear elements of storage within the speaker.

In this day and age, it is trivial to get an amp to behave as it's pass elements transit even to odd quadrant operation.

And as ETM said, the parallel network does not alter the speaker current, just the amp.
Jn
 
Last edited:
Re Ohm's 'law', it's actually not a law but rather a definition, as resistance has not yet been defined before the 'law' is stated. Real laws are relations between already existing quantities under given conditions and assumptions. (for example, the energy stored in a capacitor: W=½CU^2). I can remember also my high school teacher admonishing not to call the defining equation of resistance 'Ohm's law', and I understand that, for if we do, it's kinda chicken and egg logic IMO. Maybe 'Ohm's relation' would be a more appropriate term.
 
Last edited:
You don't need to be Einstein to realise that almost all of the voice coil inductively stored energy goes into heating the voice coil, since this is likely to be the highest resistance in that current loop.
There will be a time constant accociated with the loop. If the signal require the system move faster than that time constant, the amp has do work absorbing energy.
Above resonance, the amp will be needed in Q2 and 4 to discharge the energy. Below resonance, the amp is needed to support the energy as the vc is dissipating it, Q1 and 3.

Jn
 
www.hifisonix.com
Joined 2003
Paid Member
You don't need to be Einstein to realise that almost all of the voice coil inductively stored energy goes into heating the voice coil, since this is likely to be the highest resistance in that current loop.

It would be better to say that the 'return current' splits in this way. If the ampifier is a voltage source, or approximately so, then almost none of the energy ends up there or in the damper. Almost all of the energy ends up in the voice coil resistance.

Misleading question, as the same is true for a voltage output amplifier.

There may be some confusion between current and energy? To get energy produced or dissipated you need current and voltage together.
No confusion. The point I was making is that the amplifier in voltage mode must absorb some of the return energy (volts x current x time) even with Joes damper network. And in the big scheme of things any decent solid state amplifier will deal with this quite adequately (feedback - awesome stuff!) so you really don’t need the damper. For example, one of my amps will comfortable handle > 60 amps for short periods what will a damper bring to the party? Not much IMV.

The current drive comment was simply saying if you have an very high output impedance in parallel with the speaker and damping network how does the system deal with the return energy? A prompt to think a bit differently about the damper to Dan et al.

Why would you say that? I've watched folks randomly mixing and matching components for years, they sometimes end up with something that makes them happy. I like some and hate others and have yet to have faeries carry each note lovingly to my ears.

You’re not smoking the right stuff!

Watched a video of one of the Stereophile reviewers (a popular one in almost every issue) discuss his (perhaps occasional) weed habit despite his advancing years and I thought to myself ‘hmmm, there’s a music loving hippy . . . ‘
 
Wow!

This forum is finally talking about speakers.

The comments I see are often a bit off the mark, but that is OK. For example, with current drive (which I am not promoting) there is no reactive current, no matter how reactive you might consider the load. Think about it. What happens to the inductance in the driver? It becomes something measurable with a voltmeter. This is interesting, I would like to know what others might think, but please tone down the aggro. OK?
 
Status
Not open for further replies.