How to increase Harmonic distortion for clean sounds

[Dave H]

The results of the VOX amp showing predominantly odd order harmonics is interesting. Would that be odd numbered harmonics or actually odd order harmonics? I ask because all of a sudden I'm not even sure what even order and odd order mean. I always assumed that odd order harmonics were dissonant rather than harmonious. Now I'm not at all sure. Because loud AC30's sound plenty musical to me. I don't think it was specified whether this particular AC30 was a TB being used on the TB channel or not. Obviously this makes a difference since the cathode follower is said to add even order harmonics!?!

Chuck, I only know enough music theory to play the three chords necessary to rock but as I understand it the fundamental is called the ‘first harmonic’ and the second, third, fourth, fifth etc. harmonics are 2, 3, 4, 5 times the frequency of the fundamental or first harmonic. The 2nd, 4th, 8th etc harmonics are octaves and the 3rd is a ‘fifth’ an octave up i.e. if the fundamental is an A the 3rd harmonic is an E. If your amp produces lots of 2nd, 3rd and 4th harmonics then it’s generating power chords (root + fifth) so it should be ready to rock but higher order odd harmonics are not as nicely musically related.

It’s not that simple of course as there is also lots of IMD. Have a look at my plot below. The fundamental is 500Hz and you can see the harmonics at 1000, 1500, 2000, 2500 etc. but look at all the distortion between the harmonics! A lot of that has to be due to IMD with the 100Hz ripple of the power supply.

The ‘AC30’ I measured only has the Normal channel i.e. it only has one gain stage before the PI and no cathode follower (and I removed two EL84s). The odd order harmonics must be generated in the PI/output stage. I don’t think there is any significant distortion from the first stage as the input signal was only 250mV pk but I can check that.

J20-15 scope.pdf
J20-15 Spectrum.pdf

[Alan0354]

The ‘AC30’ I measured only has the Normal channel i.e. it only has one gain stage before the PI and no cathode follower (and I removed two EL84s). The odd order harmonics must be generated in the PI/output stage. I don’t think there is any significant distortion from the first stage as the input signal was only 250mV pk but I can check that.

[ATTACH]27128[/ATTACH]
[ATTACH]27129[/ATTACH]

I have three different schematic version of AC30, only the one AC301960 has one preamp stage in the normal channel, the others have two. The stage uses 220K plate resistor which from my calculation, give much lower distortion. But it is offset by using very low +B ( HT6) of 135V, which make the rp higher and more prone to variation as current increases. So I don't know what is the final outcome until using a simulation.

[Dave H]

The Top Boost circuit was added to the Brilliant channel. As far as I know the Normal channel has always had only ½ 12AX7 before the PI.

[Alan0354]

The Top Boost circuit was added to the Brilliant channel. As far as I know the Normal channel has always had only ½ 12AX7 before the PI.

Check out the AC301978 and AC301990, they both have two stages.

But I was just come here to point out something to you. From the wave form, you are already in the soft clipping mode as both top and bottom of the sine wave is flattened. You are not in what we called the "clean" mode anymore. That's the reason you see strong 3rd and 5th harmonics. You really has to back down the input and redo the measurement again. At the signal level you simulated, you'll definitely can hear the distortion, that is in the touch sensitive arena where you get the break up by picking up harder.

The Paper you posted definitely not in the soft clip region, mainly talking about variation of rp with the input level.

[Dave H]

Check out the AC301978 and AC301990, they both have two stages.

No they don’t, check again The Normal channel has only one stage unless you are including the PI?

But I was just come here to point out something to you. From the wave form, you are already in the soft clipping mode as both top and bottom of the sine wave is flattened. You are not in what we called the "clean" mode anymore. That's the reason you see strong 3rd and 5th harmonics. You really has to back down the input and redo the measurement again. At the signal level you simulated, you'll definitely can hear the distortion, that is in the touch sensitive arena where you get the break up by picking up harder.

I did the measurement just below clipping because that’s where I play it with the band. I can hear it going in and out off clipping as I dig in. That’s just how the AC15/30 is. There’s a fair bit of compression from about half to full voltage output (you can see it rounding off the peaks of the sine wave). To turn it down to half voltage output the input has to be reduced to a quarter.

I’ll measure it at lower power when I have time. Where do you suggest, half power?

[Gingertube]

Then J20-15 Spectrum.pdf pfile above clearly shows the Harmonic Distortion Products AND the Intermodulation Distortion Products from IM between the 500Hz test tone and residual power supply ripple. It is a good example of why we call these IM products "sidebands". You can see them spaced equally, either side of the test tone and its harmonics. You also see that you get sidebands at the harmonics of the ripple frequency too. That is because there are harmonics of the ripple frequecy present on the power supply.

Thanks for posting that - a picture is worth a 1,000 words, particularly newbies wanting to know what the hell we are talking about
Cheers,
Ian

[Alan0354]

No they don’t, check again The Normal channel has only one stage unless you are including the PI?



I did the measurement just below clipping because that’s where I play it with the band. I can hear it going in and out off clipping as I dig in. That’s just how the AC15/30 is. There’s a fair bit of compression from about half to full voltage output (you can see it rounding off the peaks of the sine wave). To turn it down to half voltage output the input has to be reduced to a quarter.

I’ll measure it at lower power when I have time. Where do you suggest, half power?

Vox AC301990:http://www.webphix.com/schematic%20h...s/ac301990.pdf

V4 through 0.047uF and volume pot, through RC network to V7. That's two stages.

Vox AC1978:http://www.webphix.com/schematic%20h...s/ac301978.pdf

V4B through 0.047uF and volume pot, through 470K and RC network to V12B. Two stages.

You don't consider clean as when you are at the verge of breaking up, you are in the soft clipping mode. That's the reason the odd harmonics is so high. BTW, you can artificially raise the even harmonics by soft clipping if you intentionally do asymmetrical clipping by biasing the plate close to cut off or bottoming out. I would say at least lower the input voltage by half to one quarter.

[kg]

alan: i'm having a bit of trouble figuring out what it is exactly you're looking for. why would dave talk about a guitar amp's distortion operating at a signal level he doesn't (and probably most people don't) play at?

[Alan0354]

alan: i'm having a bit of trouble figuring out what it is exactly you're looking for. why would dave talk about a guitar amp's distortion operating at a signal level he doesn't (and probably most people don't) play at?

If you look at the original post #1, OP talked about adding harmonics in the clean channel. So I confined to "clean" sound, the article of using 300A as the driver stage definitely was talking about "clean" sound also. Dave H measure is in soft clip condition. The two is not related. That's what I tried to point out.

Clipping distortion is a totally different ball game from distortion due to change of rp. I don't think people should lump them together. Soft clip is mostly due to running into the limit of swing. small signal distortion in tube mostly due to change of internal plate resistance of the tube.

[kg]

If you look at the original post #1, OP talked about adding harmonics in the clean channel. So I confined to "clean" sound, the article of using 300A as the driver stage definitely was talking about "clean" sound also. Dave H measure is in soft clip condition. The two is not related. That's what I tried to point out.

Clipping distortion is a totally different ball game from distortion due to change of rp. I don't think people should lump them together. Soft clip is mostly due to running into the limit of swing. small signal distortion in tube mostly due to change of internal plate resistance of the tube.

the reason i posted the article was to define what happens to even harmonic distortion when it is cascaded. though the relative values (and orders) of said even harmonic distortion may be different, the same result will be achieved: distortions of distortions, pushed higher in order. that CAN be extrapolated even into extreme hard clipping scenarios.

the reason dave posted his test results was to demonstrate that the generally very desirable "chimey," "swirly" clean tone of a vox do NOT rely on extreme levels of even harmonic distortion. i'm sure we can agree that 1) the vox has a pretty damned good clean tone, and 2) that dave's test represented a typical use scenario.

you got into a bit of a tussle with enzo because of your insistence that different distortion is inherently different. i do not agree with you: distortion is distortion. the method and mechanism of that distortion generation may vary, as well as its severity, but any situation in which output does not change in precise proportion to input is distortion.

[Tahoebrian5]

Since when do people (especially guitar players) really know what they want?

We usually just know we want something. The what doesn't get answered until we get our mind blown by amazing tone someone else has achieved.. Then we end up lurking in forums like this one

[Tahoebrian5]

AC 30 uses 220K plate resistance, as you can see in my calculation in post #43, you get much lower distortion with load resistance much higher than the plate resistance of the tube.

As I described in post #50, if you have even number of preamp triodes in the signal chain, the even harmonics cancel out.

Very interesting. So you are saying if one side of the wave gets stretched you get harmonics, but then if the other side gets stretched the same way in the next stage they then cancel? It seems like it would have to be near perfect circumstances to achieve the canceling.. Same component setup, but also same input voltage swing. Also if there was a tone stack between them it seems like the canceling would not occur. This leads me to wonder if the greatest amount of harmonics could be achieved by leaving one side of the wave as linear as possible.

[eschertron]

Very interesting. So you are saying if one side of the wave gets stretched you get harmonics, but then if the other side gets stretched the same way in the next stage they then cancel? It seems like it would have to be near perfect circumstances to achieve the canceling.. Same component setup, but also same input voltage swing. Also if there was a tone stack between them it seems like the canceling would not occur. This leads me to wonder if the greatest amount of harmonics could be achieved by leaving one side of the wave as linear as possible.

You are wise to be skeptical. Validate with *your* ears and *your* instruments.

[Alan0354]

First of all, I don't mean to offend anyone. I am still quite new in this guitar game, I am just very interested in the discussion. I don't mean to be offensive. I just present my understanding and open for discussion and ask if I still don't agree. Please don't take it as an attack of any sort. If I offended anyone, I apologize as I don't mean to.

[Alan0354]

the reason i posted the article was to define what happens to even harmonic distortion when it is cascaded. though the relative values (and orders) of said even harmonic distortion may be different, the same result will be achieved: distortions of distortions, pushed higher in order. that CAN be extrapolated even into extreme hard clipping scenarios.

the reason dave posted his test results was to demonstrate that the generally very desirable "chimey," "swirly" clean tone of a vox do NOT rely on extreme levels of even harmonic distortion. i'm sure we can agree that 1) the vox has a pretty damned good clean tone, and 2) that dave's test represented a typical use scenario.

you got into a bit of a tussle with enzo because of your insistence that different distortion is inherently different. i do not agree with you: distortion is distortion. the method and mechanism of that distortion generation may vary, as well as its severity, but any situation in which output does not change in precise proportion to input is distortion.

If you define distortion as creation of side band and IMD, then yes, distortion is distortion. But I look at it in the sound and the source of distortion. OP asked about introducing harmonic in clean sound, that's what we are talking about. Your article on compensating of 2nd harmonics are all on the "clean" sound. When signal starts to clip, it will sound totally different, it start to sound fuzzy. To me, the result sound is totally different, and the mechanism of distortion is totally different. I choose to look at it as separate cases.

Clipping is signal run into the limit of each rail, "clean" distortion is about changing the internal plate resistance of the triode.

This is getting into definition of distortion and I really not interested in what is right or wrong.