The No.1 Website for Pro Audio
Is guitar distortion more than clipping?
Old 30th June 2015
  #1
Gear Head
 
🎧 5 years
Is guitar distortion more than clipping?

As far as I understand things, guitarists achieve distortion by overdriving their signal in the amp, causing it to clip. Is there more going on to give it that particular sound?

The reason I ask, is because I've accidentally clipped clean guitar sounds when I was first learning how to record. They sounded distorted, but not like the sounds I associate with heavy metal guitar distortion. It basically sounded like the clean guitar sound, but with some crackling layered over it. I imagine that if I ran a clean guitar signal into a mixer through a DI, and then boosted the gain until it clipped, it would probably sound similar to what I was getting when I was making those early recordings. As far as I can tell, the process is the same (a signal is being clipped) but the results are different. The only thing I can think is that a mixer or audio interface doesn't have vacuum tubes like a guitar amp.

Sorry if this is kind of confusing, it's hard to explain exactly with words.
Old 30th June 2015
  #2
Lives for gear
 
mcgruff's Avatar
 
🎧 10 years
Yes it's just clipping. Click the "triode gain stage" link here at valve wizard for a detailed explanation of what's going on inside a tube (pentode tubes are pretty much the same cathode-grid-anode design with an extra turbo boost from the screen). The section on Harmonic Distortion is towards the end.

Note that clipping adds extra harmonics into the signal before you can hear audible crunch. Playing a clean tone just under the break-up point can add a lot of tube "warmth" to the sound.

Tubes create a special kind of soft clipping which is very difficult to match with solid state circuits in pedals etc. Many old tube amp designs from the 50's and 60's are still some of the best guitar amps ever made - and always will be.
Old 30th June 2015
  #3
Gear Addict
 
🎧 10 years
Making the assumption that all distortion is the same is like making the assumption that all people are the same.

The reality is that "distortion" is any change to a signal. That change can be dynamic in nature or it could be phase or frequency. To create "good" distortion in a guitar amplification circuit, the designer will "shape" the frequency response into various gain stages to cause different parts of the frequency spectrum to have different amounts of distortion. Each of these gain stages could feed any number of clipping devices, or simply overload another gain stage.

The dynamic signal response characteristics of different devices vary widely. LED's clip differently than silicon diodes which clip differently than germanium diodes, which clip differently than opamps, which clip different than triodes, which clip different than pentodes. Amp designers will use the characteristics of these "clipping" devices in various combinations with the above mentioned frequency response shaping to achieve their sonic goals.
Old 30th June 2015
  #4
Lives for gear
 
mcgruff's Avatar
 
🎧 10 years
It is specifically clipping which creates what is called crunch/overdrive/distortion when we're talking about electric guitars.
Old 30th June 2015
  #5
Lives for gear
 
ewegogetemtiger's Avatar
 
🎧 15 years
OP, if you want a simpler explanation a lot of it can be boiled down to this: tube amps generate even-order harmonic distortion whereas most solid state amplifiers produce odd-order harmonic distortion when pushed too hard. A valve amp design will also produce considerably more Total Harmonic Distortion (THD) than efficient ss designs (like, 10 to 20 times more).
Old 30th June 2015
  #6
Gear Addict
 
🎧 10 years
Something else to consider...

The distortion in a guitar amp is filtered through the sound of the speaker in the cabinet. The speaker acts as a complex resonant EQ with distortion characteristics of it's own when pushed beyond its capabilities. The EQ consists of a high-pass filter, a low pass filter, and ton of boost/cuts in between.

Clipping an input on a mixer or preamp won't have the benefit of that filter which will sound "buzzy" "fizzy" "nasty" "crackly" because when a signal is clipped, it generates harmonics. The more severely it's clipped, the more harmonics it generates. Solid-state devices (transistors/opamps) generate higher order harmonics (more buzzy/fizzy stuff) in greater abundance than tubes do because their clipping characteristics are harder/faster.
Old 30th June 2015
  #7
Gear Head
 
🎧 5 years
Thanks for the all the replies, everyone. I love learning more about these things.
Old 30th June 2015
  #8
Lives for gear
 
cavern's Avatar
 
🎧 10 years
What about saturation, the build up to clipping. Does that not play a part in the type of distortion though?
Old 30th June 2015
  #9
Lives for gear
 
mdme_sadie's Avatar
Calling it clipping is somewhat of a simplification. Clipping to most people these days equates to digital clipping, this is a ruler flat sawn off waveform at a set volume, the result is a rigid squaring off of the waveform and a full range of unpleasant harmonics. You get a similar effect with transistors, it's the "fuzz" pedal sound there, it's not perfectly flat and cut off due to the materials used and their reaction to overload.

When you clip with a valve however something a little more complex happens, you try to drain the valves supply of electrons faster than it can generate them, this creates a more complex waveform with a softer leading edge and drop back to parity, due to peculiarities of the valve design it enhances the more pleasant sounding even order harmonics more than the odd.

Don't forget that distortion is just one small part of an amp, it's important, but everything matters and is a part of the sound, from the power source through to the speaker enclosure. Your pure clipped signal is going to sound more like a real guitar (although it'll be more like fuzz than a guitar amp), once you apply a decent speaker simulation.
Old 30th June 2015 | Show parent
  #10
Lives for gear
 
mcgruff's Avatar
 
🎧 10 years
Quote:
Originally Posted by cavern ➑️
What about saturation, the build up to clipping. Does that not play a part in the type of distortion though?
Tubes aren't actually operated anywhere near to saturation in the technical sense (the maximum theoretical anode-cathode current). When people talk about saturation they are just talking about clipping to an extent which causes audible crunch.
Old 30th June 2015 | Show parent
  #11
Lives for gear
 
mcgruff's Avatar
 
🎧 10 years
Quote:
Originally Posted by mdme_sadie ➑️
Calling it clipping is somewhat of a simplification.
Guitarists use lots of words to describe the sound: overdrive, saturation, distortion, crunch, break-up, etc but what is actually happening in the tube is clipping.

Quote:
Originally Posted by mdme_sadie ➑️
When you clip with a valve however something a little more complex happens, you try to drain the valves supply of electrons faster than it can generate them
Soft clipping is actually caused by a phenomenon known as island effect. When the control grid starts to move into cut-off, some of the current can still sneak around the sides. Eventually this is cut off too.

Maybe you're thinking about sag? This is caused by limitations in the power supply part of the circuit eg a tube rectifier with a big voltage drop or power supply capacitors literally running out of charge when the output tubes demand a large current.

Last edited by mcgruff; 30th June 2015 at 07:20 AM..
Old 30th June 2015 | Show parent
  #12
Lives for gear
 
cavern's Avatar
 
🎧 10 years
Quote:
Originally Posted by mcgruff ➑️
Tubes aren't actually operated anywhere near to saturation in the technical sense (the maximum theoretical anode-cathode current). When people talk about saturation they are just talking about clipping to an extent which causes audible crunch.
Ah I see.
So its not like when I get saturated with alcohol and my vision starts to distort.
Old 30th June 2015
  #13
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by RSager ➑️
As far as I understand things, guitarists achieve distortion by overdriving their signal in the amp, causing it to clip. Is there more going on to give it that particular sound?

The reason I ask, is because I've accidentally clipped clean guitar sounds when I was first learning how to record. They sounded distorted, but not like the sounds I associate with heavy metal guitar distortion. It basically sounded like the clean guitar sound, but with some crackling layered over it. I imagine that if I ran a clean guitar signal into a mixer through a DI, and then boosted the gain until it clipped, it would probably sound similar to what I was getting when I was making those early recordings. As far as I can tell, the process is the same (a signal is being clipped) but the results are different. The only thing I can think is that a mixer or audio interface doesn't have vacuum tubes like a guitar amp.

Sorry if this is kind of confusing, it's hard to explain exactly with words.
distortion is too much of generic term perhaps?? Others have addressed the tube attributes but in many cases speakers with "early breakup" as well as transformer saturation are part of the distorted guitar sound. As well hotter pickups driving/overloading the input..

lots of amps, for example the JCM800 2205... Get (part) of their distortion not only from the tube gain stage(s) but they have additional clipping diodes. Germanium diodes? not sure

A lot of amps have a form of attenuators for lack of a better word. If you notice tube amps with a gain knob a volume knob and a master volume usually have a type of attenuation going on where you can not only crank the preamp with the "gain" but you can also crank the power amp tubes with the "volume" .. then the master controls or attenuates the collective output. So this on it's own can create a lots of distortion from different parts of the amp's circuitry.

Many modern high gain amps have numerous gain stages. Generally getting most of the distortion from preamp tubes as well as clipping diodes and other types of circuits. Not to complicate things.. but 'distortion', 'fuzz', 'overdrive', 'saturation'... are not necessarily the same meaning. There are lots of different ways to get "that sound" whatever you call it. Different people prefer the different circuits and implementations. I don't think its fair to compare the distortion from an Engl fireball to the distortion/overdrive from an old 100watt plexi on 10 overloading 30watt greenbacks... Totally different sound and totally different reasons why they are distorting, respectively.

You could fill up a 100 pages on this thread of how different amps or different guitarists get their "distortion". While all simple concepts on their own, they are sort of involved since it's not always one thing that's getting the tone. Preamp tubes, power tubes, transformers, speakers, attenuators, gain-stages, input clipping, pedals, pickups..... blah blah.........
Old 1st July 2015 | Show parent
  #14
Gear Guru
 
kafka's Avatar
 
🎧 10 years
Quote:
Originally Posted by cavern ➑️
Ah I see.
So its not like when I get saturated with alcohol and my vision starts to distort.
I find that in that case, it's the time and space around me that distorts. It's different, because with tubes it's the ceiling that hard clips, and in this case it's usually the floor.

Last edited by kafka; 1st July 2015 at 02:19 AM..
Old 1st July 2015 | Show parent
  #15
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
Quote:
Originally Posted by mcgruff ➑️
It is specifically clipping which creates what is called crunch/overdrive/distortion when we're talking about electric guitars.
That's an overly simplistic answer and actually not really entirely true.

There's transformer saturation (not the same as clipping the power tubes) and power supply sag, just to name a couple of things...

Last edited by John Eppstein; 1st July 2015 at 06:12 AM..
Old 1st July 2015
  #16
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
To answer the question, yes, it's clipping, BUT - that's not all there is to it.

Plain clipping* happens when an amplification stage runs out of voltage headroom, so that when the signal tries to exceed the available voltage the top gets cut off ("clipped") flat. This is what happens if you clip a normal solid state device like a mic preamp. In a tube guitar amp there's more to it.

First, since most tube amps don't hit the maximum available voltage at clipping, but rather exceed the amount of gain the tube can provide in linear operation, tubes don't immediately start clipping hard - they round off the leading and trailing edges of the "flat top" a bit which gives a softer, warmer sound with fewer dissonant harmonics. Second, when you start hitting a tube REALLY hard it doesn't actually maintain a flat top on the wave, there's something of a dip between the leading and trailing ends of the flat-top, with the leading edge generally a bit higher than the trailing edge. That gives another sort of distinctive harmonic character. There's more too it than that, concerning how tubes interact with the power supply, the difference in clipping between single ended and double ended (push-pull) circuits, and magnetic saturation effects in the output transformer which are distinct and separate from actually clipping the tubes, but involved in the whole overall effect. There's also a difference between clipping caused by overdriving the grid (input) of an tube (excessive input gain) and clipping caused by exceeding the tube's rated output.

There are people like Randall Smith (Mesa Boogie) and Alexander "Howard" Dumble who have spent lifetimes studying the fine points of the different distortion modes in guitar amps.

To really understand what's going on beyond this (admittedly simplified) explanation, the easiest way is to look at the waveform on an oscilloscope at different points in a tube amp circuit as you progressively overdrive it. The way the waveform changes is quite interesting.

Another interesting thing you'll find when comparing waveforms on a scope is that many classic guitar amps (e.g. lots of vintage Fenders) never really output a true sine wave even when they're "clean", there's always a bit of rounding of the sine wave going on.

Solid state amps aren't like this - they'll go right up to rated power and then start cutting the waveform off clean.

Clipping of digital devices is another matter - when you exceed 0dBfs on a digital device it starts generating aliasing tones which are really nasty and unmusical.

So there's clipping and there's clipping - it's not always a clear cut thing. And in a tube guitar amp there are other sources of distortion wrapped up in the process - which is why clipping you mic pre doesn't sound like clipping your guitar amp and clipping your converter just sounds horrible.


* - as it's usually described in entry level texts on the subject.

Last edited by John Eppstein; 1st July 2015 at 06:09 AM..
Old 1st July 2015 | Show parent
  #17
Lives for gear
 
mcgruff's Avatar
 
🎧 10 years
Quote:
Originally Posted by John Eppstein ➑️
To answer the question, yes, it's clipping, BUT - that's not all there is to it.
There's lots of stuff going on in the circuit (and speaker) - OK - but the OP asked about distortion and we should be clear that clipping is what creates the crunchy, sizzling sound of an overdriven electric guitar.

Imagine for argument's sake a circuit design which does everything else as normal but doesn't ever clip. Where will the crunch come from..?

Quote:
Originally Posted by John Eppstein ➑️
First, since most tube amps don't hit the maximum available voltage at clipping, but rather exceed the amount of gain the tube can provide in linear operation, tubes don't immediately start clipping hard - they round off the leading and trailing edges of the "flat top" a bit which gives a softer, warmer sound with fewer dissonant harmonics.
In a gain stage the control grid voltage gets imprinted on the (much larger) anode-cathode current. The bias voltage determines where - or if - a given signal will clip.

So the way a tube is set up determines where clipping will begin but "soft" clipping occurs specifically because of the physical properties of the control grid - the island effect. Different bits of the grid cut off at different voltages and that's what makes it a gradual rather than a sudden process not the anode-cathode or bias voltages.

Quote:
Originally Posted by John Eppstein ➑️
Second, when you start hitting a tube REALLY hard it doesn't actually maintain a flat top on the wave, there's something of a dip between the leading and trailing ends of the flat-top, with the leading edge generally a bit higher than the trailing edge. That gives another sort of distinctive harmonic character.
= clipping.

Quote:
Originally Posted by John Eppstein ➑️
There's more too it than that, concerning how tubes interact with the power supply
Are we back to sag again?

Quote:
Originally Posted by John Eppstein ➑️
the difference in clipping between single ended and double ended (push-pull) circuits
= clipping.

Quote:
Originally Posted by John Eppstein ➑️
and magnetic saturation effects in the output transformer
OK but we're kinda fudging the importance of tube clipping. A Matchless amp with enough iron to build a ship will still crunch with the best of them because transformer saturation is really a secondary effect on top of the fundamental crunchy sound of clipping.

Quote:
Originally Posted by John Eppstein ➑️
There's also a difference between clipping caused by overdriving the grid (input) of an tube (excessive input gain) and clipping caused by exceeding the tube's rated output.
Ratings can be conservative but I guess you mean tubes pushed into actual, thermally-limited saturation? Who does that?
Old 1st July 2015 | Show parent
  #18
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
Quote:
Originally Posted by mcgruff ➑️
There's lots of stuff going on in the circuit (and speaker) - OK - but the OP asked about distortion and we should be clear that clipping is what creates the crunchy, sizzling sound of an overdriven electric guitar.

Imagine for argument's sake a circuit design which does everything else as normal but doesn't ever clip. Where will the crunch come from..?



In a gain stage the control grid voltage gets imprinted on the (much larger) anode-cathode current. The bias voltage determines where - or if - a given signal will clip.

So the way a tube is set up determines where clipping will begin but "soft" clipping occurs specifically because of the physical properties of the control grid - the island effect. Different bits of the grid cut off at different voltages and that's what makes it a gradual rather than a sudden process not the anode-cathode or bias voltages.



= clipping.



Are we back to sag again?



= clipping.



OK but we're kinda fudging the importance of tube clipping. A Matchless amp with enough iron to build a ship will still crunch with the best of them because transformer saturation is really a secondary effect on top of the fundamental crunchy sound of clipping.



Ratings can be conservative but I guess you mean tubes pushed into actual, thermally-limited saturation? Who does that?
It's late and I'm not going to bother going through a point by point response at this time, I have to get up in the morning and do a bunch of stuff (bank, lawyer, rent, band practice, etc.)

However, the question was " Is guitar distortion more than clipping? ".

The honest answer is, "Yes, it is". If it wasn't there would be no difference in the sound of a guitar amp distorting and a solid state hi-fi amp driven into clipping - and there obviously is. There are a number of factors involved, which I've tried to touch on at least a bit. I don't think aq deeply technical discussion is warranted and frankly I'm not going to waste hours looking up references to indulge you.

Your comments like "Are we back to sag again?" are nothing but pure trolling, and you didn't even understand the difference between "sag" in a clipped waveform (which is something more than pure clipping because it results in something other that a flat top) and power supply sag (the one you hear a lot about these days) that is a totally different effect that results in compression of the signal.

I wonder how many hours you've logged in front of a test bench with a proper oscilloscope, signal generator, and various types of amplifiers? If you had much time you'd understand it's not that simple.

The waveform of e.g. a Crown DC300 driven into hard clipping (which looks more or less like a square wave) is totally different than the waveform of a tube guitar amp hit equivalently hard (which has nothing "square" about it.)

Unfortunately I no longer possess a functioning oscilloscope camera, so I can't take pictures to post. (You can't take pictures of a scope trace properly without one, as you need to sync the shutter to the sweep rate of the scope.) I could hand draw some waveforms but that wouldn't really mean much, would it?

To get back to answering the question, yes, clipping is part of it - but there's a whole lot more to it than that.
Old 1st July 2015
  #19
Lives for gear
 
mcgruff's Avatar
 
🎧 10 years
I'm sure you do have a lot more practical experience with tube amps than me and the forum is a much more interesting and informative place because of that but the OP seemed to be asking a specific question: where does crunch come from?
Old 2nd July 2015 | Show parent
  #20
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
Quote:
Originally Posted by mcgruff ➑️
I'm sure you do have a lot more practical experience with tube amps than me and the forum is a much more interesting and informative place because of that but the OP seemed to be asking a specific question: where does crunch come from?
What do you mean by "crunch". Different people use the term somewhat differently. But I'd answer that by saying it's a combination of distortion products and playing technique.

Playing technique is really outside the boundaries of the question.

(If I wanted to be really facetious I'd say that "Crunch" comes out of a breakfast cereal box, but I'll attempt to resist the temptation...)

I'd also have to point out that solid state amplifiers that clip cleanly* don't generally "crunch" very well by anybody's definition unless maybe you run over them with a large truck.

The thing is that most distorted tones from tube guitar amps will involve a certain amount of clipping - but they also involve other factors that are not themselves clipping.

Read the question again:

"Is guitar distortion more than clipping?"

The answer is "Yes, it is."

That's not to say that clipping isn't involved. But saying that's all there is to it is like saying that all there is to a popsicle is frozen water.

And clipping in most tube guitar amps is not itself "only" clipping because other things go on that are associated with it.

* - meaning that they produce a nice straight flat-top waveform when driven into clipping, without any attending secondary distortion effects.
Old 2nd July 2015
  #21
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
And this whole discussion so far has totally ignored the influence of guitar speakers on the sound - speakers are a huge part of the sound of any distorted guitar amp and speakers, being passive devices, don't clip. They do, however, distort.
Old 2nd July 2015
  #22
Gear Head
 
🎧 5 years
Quote:
However, the question was " Is guitar distortion more than clipping? ".

The honest answer is, "Yes, it is". If it wasn't there would be no difference in the sound of a guitar amp distorting and a solid state hi-fi amp driven into clipping - and there obviously is. There are a number of factors involved, which I've tried to touch on at least a bit. I don't think aq deeply technical discussion is warranted and frankly I'm not going to waste hours looking up references to indulge you.
Thank you very much for your informative response. It's exactly the information I was looking for.

And guys, let's not argue about this. I'm always looking to learn more about audio engineering, so any additional information that goes beyond my original question is also appreciated.

Quote:
And this whole discussion so far has totally ignored the influence of guitar speakers on the sound - speakers are a huge part of the sound of any distorted guitar amp and speakers, being passive devices, don't clip. They do, however, distort.
Yes, that's true. Though, it wasn't really what I was initially asking about. I was mainly just curious about what goes on inside the amp before it hits the speakers.

However, now that you mention it, that reminds me of another question I have. I've heard you shouldn't send a clipped signal to passive speakers in a PA system because it can damage them. So if an overdriven guitar signal is essentially clipped, then why doesn't it damage the speakers in a guitar cabinet? Are they designed differently than PA speakers?
Old 2nd July 2015 | Show parent
  #23
Gear Guru
 
kafka's Avatar
 
🎧 10 years
Quote:
Originally Posted by RSager ➑️
However, now that you mention it, that reminds me of another question I have. I've heard you shouldn't send a clipped signal to passive speakers in a PA system because it can damage them. So if an overdriven guitar signal is essentially clipped, then why doesn't it damage the speakers in a guitar cabinet? Are they designed differently than PA speakers?
Yes. Clipping generates harmonics. A PA tweeter doesn't nearly have the nearly the power capacity as a driver. So if you start generating more high frequency components than in the normal music spectrum, the tweeter likely won't have the capacity to handle them and you'll blow it out. There's no tweeter in a guitar cab to blow out. It's all driver, so clip away!
Old 2nd July 2015 | Show parent
  #24
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
Quote:
Originally Posted by RSager ➑️
However, now that you mention it, that reminds me of another question I have. I've heard you shouldn't send a clipped signal to passive speakers in a PA system because it can damage them. So if an overdriven guitar signal is essentially clipped, then why doesn't it damage the speakers in a guitar cabinet? Are they designed differently than PA speakers?
The short (but not really correct) answer would be "Yes, they are. And Spec'd differently as well in many cases."

But of course there's a bit more to it than that, starting with the fact that most "clipped" guitar signals are not actually pure flat-top clipping.

The thing about clipping that kills PA speakers is that when a hi-fidelity type audio system clips the waveform goes flat at the top of the waveform (Well on bothe top and bottom on a scope trace but we'll assume a symmetrical wave and just call it "top" for simplicity's sake.) When the waveform goes flat at the top, that's DC, at the maximum voltage the amp can put out.

A speaker is an oscillating linear electric motor. The energy input into the voice coil is converted to two things - mechanical energy, which is what we want, and heat, which we don't want because it damages voice coils as well as being a waste of energy. The movement of the voice coil in the magnetic gap also creates a pumping effect in the air surrounding the coil, which is essential in keeping it cool and getting rid of that nasty heat. That means you want to keep the voice coil in constant motion.

When you apply DC to a voice coil it moves to one direction and stops, because the current isn't alternating. Since the voice coil isn't moving during that period two things happen - first, electrical energy is not being converted to mechanical energy so it has no choice but to turn into heat. Second, since the coil isn't moving there is no pumping effect and the heat has nowhere to go. Double whammy.

Now of course these periods of non-motion are very brief, only when the waveform is flat, but it's still enough to cause a significant heat build-up in the voice coil which will eventually soften the glue (causing out-of-spec performance and even rattle) and in time may even burn out the wire of the voice coil.

That's how clipping kills speakers and even an amp rated at less power than the speaker can do it if clipping is prolonged - like when a loud band is trying to push an inadequately powered PA over their din.

Why doesn't this happen in guitar amps?

Well, that's a good question. A lot of it has to do with what we discussed in the earlier part of this thread - in most guitar amps "clipping" isn't really "pure" clipping, it's a combination of distortion effects and what it boils down to is that the waveform out of a guitar amp usually isn't flat-topped. There's some sag and some spikiness to it, which means that the voice coil is still moving a bit. But it depends on the nature of the distortion - if you're running a fuzz box that puts out a near square wave through a solid state amp with no output transformer you have a much greater danger of a blown speaker than if you're just cranking a Marshall tube head. It all comes down to waveforms.

One thing to understand about tube amps, BTW, is that a tube amp with an output transformer is incapable of producing a clean clipped waveform because of the simple fact that transformers are AC only devices - they will not pass DC. So even if your output tubes are emitting a "pure" clipped waveform it can't get past the transformer.
Old 2nd July 2015
  #25
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
There's more to why PA speakers are more easily killed by clipping as well. One thing to understand is that most modern high frequency drivers are very delicate as they employ thin metal diaphragms that shatter easily when subjected to certain types of distortion. More rugged diaphragms made of phenolic impregnated cloth are more sturdy but don't have the extended HF response.
Old 3rd July 2015 | Show parent
  #26
Lives for gear
 
cavern's Avatar
 
🎧 10 years
Quote:
Originally Posted by John Eppstein ➑️
The short (but not really correct) answer would be "Yes, they are. And Spec'd differently as well in many cases."

But of course there's a bit more to it than that, starting with the fact that most "clipped" guitar signals are not actually pure flat-top clipping.

The thing about clipping that kills PA speakers is that when a hi-fidelity type audio system clips the waveform goes flat at the top of the waveform (Well on bothe top and bottom on a scope trace but we'll assume a symmetrical wave and just call it "top" for simplicity's sake.) When the waveform goes flat at the top, that's DC, at the maximum voltage the amp can put out.

A speaker is an oscillating linear electric motor. The energy input into the voice coil is converted to two things - mechanical energy, which is what we want, and heat, which we don't want because it damages voice coils as well as being a waste of energy. The movement of the voice coil in the magnetic gap also creates a pumping effect in the air surrounding the coil, which is essential in keeping it cool and getting rid of that nasty heat. That means you want to keep the voice coil in constant motion.

When you apply DC to a voice coil it moves to one direction and stops, because the current isn't alternating. Since the voice coil isn't moving during that period two things happen - first, electrical energy is not being converted to mechanical energy so it has no choice but to turn into heat. Second, since the coil isn't moving there is no pumping effect and the heat has nowhere to go. Double whammy.

Now of course these periods of non-motion are very brief, only when the waveform is flat, but it's still enough to cause a significant heat build-up in the voice coil which will eventually soften the glue (causing out-of-spec performance and even rattle) and in time may even burn out the wire of the voice coil.

That's how clipping kills speakers and even an amp rated at less power than the speaker can do it if clipping is prolonged - like when a loud band is trying to push an inadequately powered PA over their din.

Why doesn't this happen in guitar amps?

Well, that's a good question. A lot of it has to do with what we discussed in the earlier part of this thread - in most guitar amps "clipping" isn't really "pure" clipping, it's a combination of distortion effects and what it boils down to is that the waveform out of a guitar amp usually isn't flat-topped. There's some sag and some spikiness to it, which means that the voice coil is still moving a bit. But it depends on the nature of the distortion - if you're running a fuzz box that puts out a near square wave through a solid state amp with no output transformer you have a much greater danger of a blown speaker than if you're just cranking a Marshall tube head. It all comes down to waveforms.

One thing to understand about tube amps, BTW, is that a tube amp with an output transformer is incapable of producing a clean clipped waveform because of the simple fact that transformers are AC only devices - they will not pass DC. So even if your output tubes are emitting a "pure" clipped waveform it can't get past the transformer.
I had a Belles amp in the 80's with a pair of Celestions DL-10's and when I took it in for repair, the tech told me that side went DC.I still don't know what he meant.

What happened is we were listening to Level 42..haha and the woofer cone on the left side lit up bright red and BAM, only on channel left.
When I touched it with my finger, it crumbled and the coil was fused.
Cost me $400.

Last edited by cavern; 3rd July 2015 at 04:44 AM..
Old 3rd July 2015 | Show parent
  #27
Gear Guru
 
John Eppstein's Avatar
 
58 Reviews written
🎧 10 years
Quote:
Originally Posted by cavern ➑️
I had a Belles amp in the 80's with a pair of Celestions DL-10's and when I took it in for repair, the tech told me that side went DC.I still don't know what he meant.

What happened is we were listening to Level 42..haha and the woofer cone on the left side lit up bright red and BAM, only on channel left.
When I touched it with my finger, it crumbled and the coil was fused.
Cost me $400.
What he mean was that one or more power transistors blew and went dead short, dumping the full voltage of one side of the power supply into your speakers, which is generally a recipe for the magic smoke. The amp must have been rather poorly designed without adequate speaker protection. By the '80s most amps had a crowbar circuit to protect the speakers if that happened.

Some nosebleed audiophile types claimed that the protection circuitry had an adverse effect on the sound. While that case might be arguable for VI current limiter circuits found in some amps like the Crown DC300A, I have never heard any audible effects caused by a crowbar circuit, which simply senses if DC appears on the speaker terminals for more than an instant and, if it does, throws a short circuit ("crowbar") across the output to protect the speaker. It generally consists of a triac (functioning as an electronic switch), a zener diode (providing a voltage reference for turning the switch on) and about 2 resistors. If DC is present above threshold the triac turns on an shorts the output until the DC is removed.

Amps designed earlier, in the '70s, didn't always have that. The early Phase Linear amps, which lacked all protection circuitry earned the nickname "Flame Linear" by actually causing speakers to catch fire in the amp blew. Fortunately for people using them in professional application there was an easy mod consisting of adding 4 or 5 parts to the speaker terminals that took care of the problem.
πŸ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 92 views: 9190
Avatar for vtone
vtone 1st February 2008
replies: 63 views: 12326
Avatar for Ari88
Ari88 20th October 2017
replies: 0 views: 559
Avatar for MUSICMUSIC
MUSICMUSIC 12th November 2017
replies: 113 views: 13996
Avatar for oldgearguy
oldgearguy 2nd September 2020
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump