freestompboxes.org

How easy or hard is it?

My main problem is that they are mechanical devices and so they can't be simulated on a computer with a program so easily. The manafactures post a frequency response curve on their websites most of the time and so this would need to be replicated and then tweaked.

There are two other properties that I am interested in - distortion and harmonic content. In an electrical world a Fourier analysis would allow one to look at the harmonic content via a simulation but obviously this doesn't apply to the situation. Also the distortion, I'm guessing that speakers will have some sort of harmonic distortion due to their physical properties but what would this look like? What is the harmonic content of a guitar speaker like?

Any help would be greatly appreciated! Maybe we could have a good discussion and share some knowledge It may be that the only way to simulate this is with a sig gen, scope, closed system, a clean power amp and the speaker but that's slightly tricky for an average DIYer.

Hi jimbob_400,

Short answer, don't make something needlessly complex when it is not. Just use a circuit that dramatically cuts frequencies outside of the bandwidth of a guitar and it will be close enough.

IMO, all that tripe about speaker breakup,, never failing to add every kind of teckninal sounding nerdy name for speakers moving outside of normal,, get up my nose.

Sure some of it may well be true BUT it Might pay to ask yourself how much of all this teckno babble you can actually hear?
If you believe all the write ups then you will find yourself hearing the 17th harmonic of the 49th fret. Winky

Trust me,, TRUE speaker distortion sounds like crap.

The trick is all about finding the speaker output curve/ tube compression/tone shaping that suits your playing style.
Speaker simulators are just a complex name for tone shaping devices which are far more powerfull that the normal tone controls.

The distortion is the hard part to nail down because it's very dependant on the tone shaping that happens when outputvalves are drivin into compression.
So one is kinda dependant on the other ,,dog chase tail stuff.

There is a neat little circuit hidden inside the Marshall JCM30 or 60 Amps,,, start by building one of those.
I added that circuit after my old Quadraverb and it made one hell of a difference to my recordings. Just rips off all the extended freq that just gets in the way of making good recordings.
The last thing you want for great hard driven guitar is extended bandwidth.

google Ed Rembolds Cab Sim,, should still be floating around.
Or look at the Lxh2 circuits,,forgotten the links.
Phil.

The loudspeaker distortion is also a confusingly technical subject because the phenomenon is extremely unlinear (no pun intented). What I'm trying to say is that the type of distortion depends on extensive list of different types of variables and is by no means as predictable as distortion created by, eg. diode clipping.

First of all you have the ungodly amount of variations in the speaker design itself; such as is the voice coil overhung or underhung, damping, excursion, materials, cone nodes, back EMF, etc. Then you have the effects from another ungodly amount of enclosure types that vary all these parameters, and finally you have the effects of the overall amp design and the signal driving the speaker (eg. amp's damping factor, frequency, amplitude of drive current and voltage, and so on) that again interact with all these parameters.

So, it is damn near impossible to present any universal data on the matter or characterize the phenomenon with some simple function. The discussion is of entire, complex systems, each usually having its individual form of distortion, which naturally varies at different frequencies and so on...

Several hundreds of pages of whitepapers and books have been written about the subject and you can easily find weeks worth of reading simply by doing Google search. For example, with a keyword "loudspeaker distortion". But it's widely technical and complex issue and extremely hard to digest.

Oh, there is plenty of discussion, but most of it easily goes over the head of an average person, which kinda puts it off the scope of a handful of DIY enthusiasts. I've studies these things several months and I still don't usually understand even half of what's written. Guys struggling to cope how their "Tube Screamer" pedal works shouldn't even bother. You need to read, read, read, read... Basically, the guys discussing these things are often EE professors working for universities or big companies like JVC, Sony, ...

It's an extremely complex subject.

So, in the end of practically applying this stuff to applications such as guitar amps or effects for guitars, it really just falls down to how complex simulation you wish to make. For example, you can simulate the speaker response with simple band-pass filter and the distortions from cone excursion limits with some crude diode clipping scheme but all this barely scratches the surface. Once you begin to add detail to these approximations the circuit's will begin to bloat like a bread dough.

phatt wrote:Hi jimbob_400,

Short answer, don't make something needlessly complex when it is not. Just use a circuit that dramatically cuts frequencies outside of the bandwidth of a guitar and it will be close enough.

IMO, all that tripe about speaker breakup,, never failing to add every kind of teckninal sounding nerdy name for speakers moving outside of normal,, get up my nose.

Sure some of it may well be true BUT it Might pay to ask yourself how much of all this teckno babble you can actually hear?
If you believe all the write ups then you will find yourself hearing the 17th harmonic of the 49th fret. Winky

Trust me,, TRUE speaker distortion sounds like crap.

The trick is all about finding the speaker output curve/ tube compression/tone shaping that suits your playing style.
Speaker simulators are just a complex name for tone shaping devices which are far more powerfull that the normal tone controls.

The distortion is the hard part to nail down because it's very dependant on the tone shaping that happens when outputvalves are drivin into compression.
So one is kinda dependant on the other ,,dog chase tail stuff.

There is a neat little circuit hidden inside the Marshall JCM30 or 60 Amps,,, start by building one of those.
I added that circuit after my old Quadraverb and it made one hell of a difference to my recordings. Just rips off all the extended freq that just gets in the way of making good recordings.
The last thing you want for great hard driven guitar is extended bandwidth.

google Ed Rembolds Cab Sim,, should still be floating around.
Or look at the Lxh2 circuits,,forgotten the links.
Phil.

Thank you. I've tried a few speaker simulations but nothing ever seemed convincing enough for me! I will have a look at the marshall one.

teemuk wrote:The loudspeaker distortion is also a confusingly technical subject because the phenomenon is extremely unlinear (no pun intented). What I'm trying to say is that the type of distortion depends on extensive list of different types of variables and is by no means as predictable as distortion created by, eg. diode clipping.

First of all you have the ungodly amount of variations in the speaker design itself; such as is the voice coil overhung or underhung, damping, excursion, materials, cone nodes, back EMF, etc. Then you have the effects from another ungodly amount of enclosure types that vary all these parameters, and finally you have the effects of the overall amp design and the signal driving the speaker (eg. amp's damping factor, frequency, amplitude of drive current and voltage, and so on) that again interact with all these parameters.

So, it is damn near impossible to present any universal data on the matter or characterize the phenomenon with some simple function. The discussion is of entire, complex systems, each usually having its individual form of distortion, which naturally varies at different frequencies and so on...

Several hundreds of pages of whitepapers and books have been written about the subject and you can easily find weeks worth of reading simply by doing Google search. For example, with a keyword "loudspeaker distortion". But it's widely technical and complex issue and extremely hard to digest.

Oh, there is plenty of discussion, but most of it easily goes over the head of an average person, which kinda puts it off the scope of a handful of DIY enthusiasts. I've studies these things several months and I still don't usually understand even half of what's written. Guys struggling to cope how their "Tube Screamer" pedal works shouldn't even bother. You need to read, read, read, read... Basically, the guys discussing these things are often EE professors working for universities or big companies like JVC, Sony, ...

It's an extremely complex subject.

So, in the end of practically applying this stuff to applications such as guitar amps or effects for guitars, it really just falls down to how complex simulation you wish to make. For example, you can simulate the speaker response with simple band-pass filter and the distortions from cone excursion limits with some crude diode clipping scheme but all this barely scratches the surface. Once you begin to add detail to these approximations the circuit's will begin to bloat like a bread dough.

I see, it does sound very complicated!

I must state that I know a fair amount of electronics, I will be studying EE full time by the end of this year. I am very interested in designing something to suit my needs as a guitarist.

Surely quite a few of the features of speakers which affect the overall sound of the device will just be affecting the frequency response of the device and so can just be replicated by a network of filters? Of course the nature of the input signal will have an effect on the response but this issue can also be addressed in the design process. I guess the distortion is often negligable anyway in guitar amps? I mean maybe not with a greenback speaker and its 'flabby-ness' but with other speakers such as an alnico gold in a low wattage amplifier?

Don't guitar speakers generally not add much harmonic content to a guitar signal?

At the moment I'm thinking of maybe running an optical compressor into a network of filters to replicate the response curve on the manafactures website and then thinking up some sort of cabinet simulation Thank you for your input on this subject!

James