| Author |
Topic: Tube Amp: Crossover Distortion |
Tom Gorr
From:
Three Hills, Alberta
|
Posted 21 Oct 2007 5:24 pm |
|
| Could someone describe the "sound" of crossover distortion so I can evaluate whether my tube power amp is having this problem. |
|
|
 |
b0b
From:
Cloverdale, CA, USA
|
|
|
|
John Groover McDuffie
From:
LA California, USA
|
Posted 22 Oct 2007 2:32 am crossover distortion
|
|
I must, with all due respect, correct what Bob says in the above post.
Crossover distortion happens in a push-pull amplifier when the bias is set incorrectly to the extent that the transition from when the "pushing" tube is carrying signal to when the "pulling" tube is carrying signal does not happen smoothly. This will appear on an oscilloscope trace as a roughness, glitch, or in more extreme as a kind of sawtooth disturbance in the sinewave as the voltage crosses zero when going from positive to negative or vice versa, hence the name.
Compare this with the trace of clipping distortion, where the distortion occurs at the peaks and negative peaks of the wave.
This was demonstrated for me by the late great Red Rhodes when I worked at his amp shop in the late 70s.
It will be audible as a (usually subtle but not pleasing) distortion that does not go away even when the amp is turned down way below clipping levels. Other than that it's been so long since I have specifically listened to the sound that I can't accurately describe the specific character of the distortion.
The phenomenon that Bob describes is actually a sum/difference frequency and does not originate in the amplifier, but is inherent to the physics of combining waves of two different frequencies. Because of constructive and destructive interference between the two waves, two more waves are generated, one equal to the sum of the two and one equal to the difference. When discussed in the context of electronic circuits this is referred to as intermodulation distortion.
Musicians experience and actually take advantage of this phenomenon when they tune by listening to the "beats" caused by an out-of-tune unison interval. The beat frequency is the difference frequency caused by the two waves. The closer the two frequencies are to one another, the lower the diff. freq. is, hence as the two notes get closer to the same pitch (hence the diff. is decr.) the beats get slower until the notes are perfectly in tune, the diff=0, and the beats stop.
Difference frequency is heard in another context when using a lot of gain for a distorted sound and playing more than one note. If one plays an interval, for example a maj or min 3rd one will also hear another note lower in pitch added to the sound. The high gain and compression of an overdriven signal make the diff. freq. (which is always present as a part of the complex wave) more apparent in the signal than it is when using a cleaner sound.
As Bob describes above, if you bend one note while holding another note at a steady pitch the diff freq. will move in the following manner: if the two notes are getting closer together the new note will descend and vice versa. Once again it is more apparent at higher gain settings.
If you are wondering "What about the sum frequency?" Well since it is higher in pitch than the fundamental frequencies it just blends with the overtone series, so we don't hear it as a distinct part of the sound. |
|
|
|
Jay Fagerlie
From:
Lotus, California, USA
|
Posted 22 Oct 2007 5:52 am
|
|
Crossover distortion-The "Marshall" sound.
Jay |
|
|
|
b0b
From:
Cloverdale, CA, USA
|
|
|
|
Tom Gorr
From:
Three Hills, Alberta
|
Posted 22 Oct 2007 10:00 am
|
|
The distortion I'm getting is not pleasing. Its happening in a Mesa Boogie amp on the lead channel.
Basically, there's a low frequency sound (lower than the fundamental) that sounds "like someone clearing their throat". It is there, and changes character as the note fades away, but it disappears when the note is mostly faded. Its a very "upfront sound" and present sound - like you'd expect if the speaker was fried. I don't think the speaker is fried.
What makes this difficult to diagnose by ear is that the clean sound doesn't appear to have this issue. Only the "Boogie Lead" channel.
Any ideas? |
|
|
|
b0b
From:
Cloverdale, CA, USA
|
Posted 22 Oct 2007 10:12 am
|
|
| Tom Gorr wrote: |
The distortion I'm getting is not pleasing. Its happening in a Mesa Boogie amp on the lead channel.
...
What makes this difficult to diagnose by ear is that the clean sound doesn't appear to have this issue. Only the "Boogie Lead" channel.
Any ideas? |
Start swapping out the 12AX7 preamp tubes until it goes away.
_________________
-𝕓𝕆𝕓- (admin) - Robert P. Lee - Recordings - Breathe - D6th - Video |
|
|
|
Tom Gorr
From:
Three Hills, Alberta
|
Posted 22 Oct 2007 10:49 am
|
|
| b0b wrote: |
| Tom Gorr wrote: |
The distortion I'm getting is not pleasing. Its happening in a Mesa Boogie amp on the lead channel.
...
What makes this difficult to diagnose by ear is that the clean sound doesn't appear to have this issue. Only the "Boogie Lead" channel.
Any ideas? |
Start swapping out the 12AX7 preamp tubes until it goes away. |
Do you think it may be a "microphonic tube" ? |
|
|
|
b0b
From:
Cloverdale, CA, USA
|
Posted 22 Oct 2007 10:56 am
|
|
| I think it could be a tube gone bad. That's the simplest thing to try, anyway. When something goes wrong with one of my tube amps, tube-swapping usually solves the problem. Since it's only in the lead channel, it can't be the power tubes, the 12AT7 driver or the reverb tube. Start with the 12AX7 closest to the input jack. |
|
|
|
Tom Gorr
From:
Three Hills, Alberta
|
Posted 22 Oct 2007 2:41 pm
|
|
I swapped V1 from a 12AX7 to a 12AT7, and things seem much better, and just 'hint' of that distortion remains.
The signature boogie lead sound seems more balanced and controlled than before.
May even try a 12AY7 in V1 because the directionality is definately the right one.
In terms of diagnosis, is it possible that by having an unusually high gain tube in V1 - may the cathode follower or phase inverter stages been driven 'to the rail', causing this harsh and ungodly distortion to come through ?
I also heard that old fenders tone stack caused clipping under certain conditions...what may have caused that? The boogies have similar style tone stacks. |
|
|
|
John Groover McDuffie
From:
LA California, USA
|
Posted 22 Oct 2007 5:49 pm
|
|
Fender amp tone stacks are passive circuits, (they only work by subtracting frequencies) therefore I don't think they alone can cause clipping. But I could be wrong.
If you had all the tone controls on a Fender (must be one with bass, treble and MIDRANGE controls) set to very low setting they will work as a master volume. If you have all 3 all the way down no signal will be present at the output. So I suppose it is possible that with them set very low, you could them be motivated to turn up the output of something feeding the input of the amp to the point that the first input stage clips. The volume control in the Fender circuit is after the tone stack, so you wouldn't be able to do what I describe by turning up the volume control of the amp.
Of course you could make the amp clip by turning up the volume control, but probably couldn't say that the clipping was CAUSED by the tone stack. |
|
|
|
Tom Gorr
From:
Three Hills, Alberta
|
Posted 22 Oct 2007 7:46 pm
|
|
| IT does seem counter intuitive that clipping would directly occur in the tone stack....I did read it on an internet thread somewhere...but who says the internet is the depository of fact ? |
|
|
|
