Matchless Chieftan - WTH

[TimmyP1955]

Depends on where you look it up. Some sources say that comprises and comprised of are both correct. To me, comprises doesn't make sense:A bunch does no make a number of things, a number of things make a bunch.

Back to my post: The speaker does not spend time at one extreme or the other when reproducing a square wave. Forget what it looks like on an oscilloscope.

[bob p]

Back to my post: The speaker does not spend time at one extreme or the other when reproducing a square wave. Forget what it looks like on an oscilloscope.

Try hooking a 9V battery up to a woofer and let me know what it looks like when driven by a really low frequency (DC) square wave.

[Enzo]

Depends on where you look it up. Some sources say that comprises and comprised of are both correct. To me, comprises doesn't make sense:A bunch does no make a number of things, a number of things make a bunch.

Makes sense when you realize it means "includes".

Or "covers."
My team covers the infield, the outfield, and behind the plate. Versus: my team covers of the outfield, the infield, and behind the plate.

[bob p]

I don't want to edit my post after someone "liked" it, as that would be like putting words into their mouth, so I'll just make a new post.

Maybe some of the difference in opinion here stems from the difference between real world observations and how we like to mathematically model real world situations for study. Fourier theory tells us that any square waveform can be approximated and/or modeled by convolving a group of sine waves of the required amplitude and frequency, but that's not the same as asserting that that's how square waves are actually produced in the real world, outside of the realm of mathematical models, pencil and paper.

We can debate whether or not real world phenomea like square waves actually exist as convolutions of sine waves (the theory) or whether square waves actually exist in nature (observation). To the observer, DC is DC. It's square. A 9V battery and a speaker sure look like square wave behavior. At the level of battery chemistry, we're looking at a combination of high frequency and low frequency events in the battery's matrix, and the power spectrum of those reactions is definitely in the low frequency side approximating DC. But the observed electrical result in the speaker looks pretty square-wave in it's behavior, as long as the chemical reaction can hold out.

Energizing a speaker with a battery is like energizing a relay or a solenoid. You can feed a relay with AC and it will drive it's armature to move and stay put, acting as a Boolean Gate. It sure looks like square wave behavior. Of course, these the behavior of these things looks different if you look at them on the macro vs. micro levels.

[Leo_Gnardo]

I don't want to edit my post after someone "liked" it, as that would be like putting words into their mouth, so I'll just make a new post.

Maybe some of the difference in opinion here stems from the difference between real world observations and how we like to mathematically model real world situations for study. Fourier theory tells us that any square waveform can be approximated and/or modeled by convolving a group of sine waves of the required amplitude and frequency, but that's not the same as asserting that that's how square waves are actually produced in the real world, outside of the realm of mathematical models, pencil and paper.

We can debate whether or not real world phenomea like square waves actually exist as convolutions of sine waves (the theory) or whether square waves actually exist in nature (observation). To the observer, DC is DC. It's square. A 9V battery and a speaker sure look like square wave behavior. At the level of battery chemistry, we're looking at a combination of high frequency and low frequency events in the battery's matrix, and the power spectrum of those reactions is definitely in the low frequency side approximating DC. But the observed electrical result in the speaker looks pretty square-wave in it's behavior, as long as the chemical reaction can hold out.

Energizing a speaker with a battery is like energizing a relay or a solenoid. You can feed a relay with AC and it will drive it's armature to move and stay put, acting as a Boolean Gate. It sure looks like square wave behavior. Of course, these the behavior of these things looks different if you look at them on the macro vs. micro levels.

Good observations. By Fourier, even the battery example can be described by a series of sine waves. The fancy term for it is Heaviside Step Function. The derivitave of that, a spike of presumably infinite height, Dirac Delta Function. Both theoretically contain information including all frequencies. I had to deal with all that stuff in "Introduction to Partial Differential Equations" nearly half a century ago in math classes then again in Quantum. Funny what sticks in the head, even some minor remnant. Starting in the 1970's computerized acoustic room analysis started to employ approximations of the Dirac spike as a method for presenting "all frequencies at once" to a space under test, with a "click machine." Prior to that they tried using starter pistols. And we've all encountered the "hand clap" method for getting a feel for the reverberant nature of a room.

Meanwhile in the biophysics department we had a brilliant professor who had a study going testing the reactions of frog and clam muscles to being stretched. He used a modified JBL 12 inch speaker - cone removed - as the muscle pulling and sensing device. A square wave applied to a DC amp pulled the cone, and its return to rest position, pulled by the muscles under test was monitored on a scope. Kool frankenstein stuff! One of my apartment mates worked on this project. For a while we got all the leftover clam bits to make chowder once a week. That guy went on to be a pioneering biophysics research prof at U Virginia in Charlottesville. Lesson: clams make you smart!

[The Dude]

That story reminds of the the old ServoDrive speakers. The cone movement was done with a motor and belts. No voice coil.

[bob p]

as a side note to Leo's comment about impulse response transfer functions, we were once talking on this site about how modelling amps worked, and a few of us diverted off into a discussion about DSP and i commented about how I thought that modelling amps could never be made to sound like tube amps for some technical reasons...

the next day I got a phone call from a forum member (can't remember who it was) who wanted to talk to me about the impulse-response transfer function theory, specifically relating to my comment that an infinitely high and narrow pulse contained all frequencies. He was an RC model car enthusiast who was trying to solve a battery charging problem, and ended up using high frequency pulses to quick-charge NiCD batteries for the cars. The idea was that the HF charging could more effectively populate the battery matrix, and he got a patent for it.

I can't remember who it was. Anyone want to claim that one?

[potatofarmer]

Yes, sticking DC on a speaker's terminals will cause the coil to jump in one direction and stay there, but the suspension is going to cause it to bounce a bit before it settles. Particularly when we're talking about woofers, the inherent low-pass keeps them from recreating any sort of pure square wave at all. It's basically the same idea as the "underpowering" myth, which is silly enough in the HiFi and PA worlds, but with guitar amps/speakers it's just ridiculous.

[TimmyP1955]

Try hooking a 9V battery up to a woofer and let me know what it looks like when driven by a really low frequency (DC) square wave.

Apples and oranges. An oscilloscope does not show you the frequency spectrum of the square wave - you need an FFT for that. Putting a bunch of sine waves together does not create a DC component.

[The Dude]

Respectfully!

......"An oscilloscope does not show you the frequency spectrum of the square wave."......

Exactly. And neither does a speaker. By comparison to a scope, the speaker is just crude mechanical junk unable to respond to Mhz.

......"Putting a bunch of sine waves together does not create a DC component."

As far as the speaker is concerned, that is exactly what happens.

I'm not trying to argue (and I'm not going to), and I do understand your theory, but it doesn't hold up in practice.

[J M Fahey]

Apples and oranges. An oscilloscope does not show you the frequency spectrum of the square wave - you need an FFT for that.

Sorry but wrong on both.
A squarewave waveform shows you the frequency response, just does not crunch numbers which is something else.

At the most basic, square wave edge shows you :

* flat response extending to very high frequencies, way higher than test squarewave: perfect sharp edge squarewave.

* rising highs: spike at the front edge.

Spike width and height gives you an idea of how much treble boost you have , and whether it includes low order or high order harmonics.

* ringing front edge shows you high frequency resonances.

You can even know what frequency is ringing by counting small peaks and seeing how many fit in the main squarewave.

* rounded front edge speaks of high frequency attenuation.
How rounded gives you a good idea of how much.
and also:
* rising or drooping top/roof/horizontal part of waveform shows whether there is a bass boost or cut.

There are whole books dedicated to squarewave (and other waveforms) interpretation and use in analysis, design or servicing.

Putting a bunch of sine waves together does not create a DC component.

I beg to differ.
You wonīt create pure DC , but DC components, any day of the week.

Look at it the other way:

*any* waveform, that means any shape, can be decomposed into (or made up of) a mix of various sinewaves.
You may need a single one (a sinewave is made out of *one* sinewave) or a very complex mix of wildly varying ones but the basic principle is the same.

An assymmetrical waveform will have a DC component, whether it is a poorly balanced tube power amp (poorly balanced PI or unmatched power tubes) which would be the mildest case to a narrow high voltage single polarity spike which would be the extreme and can be decomposed in a bunch of sinewaves.

So since you can reconstruct them using the sinewaves we found just above, by the same token we can create them, using the same recipe.

As a side comment, saying "you need an FFT for that" while dissing other methods sounds akin to saying "you canīt tune a guitar without a digital tuner" , or even "you canīt fly NY to Washington without VOR/ILS/GPS/Inertials/Radar" or any other Electronic Navigation aids.

As they often repeat here, "thereīs more than one way to skin a cat"

[bob p]

Apples and oranges. An oscilloscope does not show you the frequency spectrum of the square wave - you need an FFT for that.

I must be really stupid because I'm completely missing something. I never said anything about looking at an oscilloscope to view a power spectrum, so I don't know where your "apples and oranges" comment comes from, as it seems to imply that I suggested something that I never said.

Who would have said something like that?!? Looking at Post #55, it seems that you did:

A square wave is comprised of a bunch of sine waves. Look at one on a scope displaying FFT.

When I talked about a power spectrum I took it for granted that anyone reading the post would know that you need a spectrum analyzer to do that, not an oscilloscope. I'm at a loss to explain why you're accusing me of saying something that stupid.

Putting a bunch of sine waves together does not create a DC component.

I think you need to share your opinion with Joseph Fourier and tell him that his theory that any waveform can be approximated by a convolution of sine waves is wrong.

[Chuck H]

How are you deploying the Aiken thingy? I'm thinking that to get the most bang out of it you'd need to have the ability to variably divide output between it and a real speaker.

Sorry for the late response Bob.

I mentioned that my design is partly resistive. I'm using a power rheostat wired like a pot to blend/shunt between the speaker and the Aiken load design. I did have trouble with the .5mH air core inductor in that there is no core to mitigate EMF. This caused my single coil guitars to squeal whenever I was within a few feet of the attenuator. I corrected for this by using two .25mH inductors wired in series/out of phase with a copper plate sandwiched in between. The copper plate reduces mutual inductance and circuit function cancellation but since the EMF from each inductor is out of phase it cancels out in the radiant field. IIRC I just epoxied them together and wrapped the thing in electrical tape. You can probably skip this fussiness if you can find a suitably rated .5mH cored inductor with a low enough DCR.