Much as the binkum in the Wikipedia "valve sound" article grates due to its poor/selective/non-existent quoting of references, the regularly touted pleasing nature of second harmonic distortion is interesting.
Once one moves away from a global loop feedback circuit to a design with zero feedback then the possibility to dial in second harmonic distortion deliberately raises its head.
For instance, in the PP el84 amp I'm working on, simply by increasing the negative bias voltage on the input tube (6CG7) from 5v to around 10v I slide along the load line from a point where the lines are pretty evenly spaced to a point where there's a decent wad of 2nd harmonic being produced.
So, while in theory accuracy is what we strive for, do any of you deliberately engineer second harmonic into your designs?
(I'm not saying that I intend to, merely curious to see if anyone does)
Once one moves away from a global loop feedback circuit to a design with zero feedback then the possibility to dial in second harmonic distortion deliberately raises its head.
For instance, in the PP el84 amp I'm working on, simply by increasing the negative bias voltage on the input tube (6CG7) from 5v to around 10v I slide along the load line from a point where the lines are pretty evenly spaced to a point where there's a decent wad of 2nd harmonic being produced.
So, while in theory accuracy is what we strive for, do any of you deliberately engineer second harmonic into your designs?
(I'm not saying that I intend to, merely curious to see if anyone does)
I've been playing around with this for a while. I've found that dialing in a little extra 2nd harmonic can make 'dry' speakers, like many metal coned drivers, sound more listenable.
An interesting arrangement is to use starved filaments and at the same time using 0V bias. Here is Steve Benchs work... DHT with starved filaments.
Using the 01A as a driver with fil of 3.2v, anode 60v and bias 0v is a surprisingly enjoyable listen, albeit high distortion. The imaging is spectacular, likely because of the directly grounded cathode and no cap in the signal path on the grid or cathode of the tube.
An interesting arrangement is to use starved filaments and at the same time using 0V bias. Here is Steve Benchs work... DHT with starved filaments.
Using the 01A as a driver with fil of 3.2v, anode 60v and bias 0v is a surprisingly enjoyable listen, albeit high distortion. The imaging is spectacular, likely because of the directly grounded cathode and no cap in the signal path on the grid or cathode of the tube.
This is according to my understanding, a key point to why quite many people claim to prefer tubes, in particular SE. If you take a look at the Zen articles, you will find, that those amps are very popular. They sure are the SS pendant to a SE tube amp.
As for the last part, yes most guys would rather admit to have problems satisfying their woman , than that they actually like distortion.
However, in the tests I have performed, it was evident that many actually prefers some distortion, over say 0.0002% THD.
Magura
As for the last part, yes most guys would rather admit to have problems satisfying their woman , than that they actually like distortion.
However, in the tests I have performed, it was evident that many actually prefers some distortion, over say 0.0002% THD.
Magura
Hey now, what are you insinuating? Lets just say the mature man is confident in his ability... to appreciate distortion.
Hey now, what are you insinuating? Lets just say the mature man is confident in his ability... to appreciate distortion.
SS guys are the ones that need "feedback" to know that they are men
A little distortion never hurt a tube manIMO, it's a matter of distortion spectrum. That's why I strongly advocate the use of 12AT7s in PP designs. The NET HD spectrum is an ear pleasing "waterfall" of 2nd > 3rd > 4th ... The 12AT7's bias towards 2nd order products meshes very well with the internal even order product cancellation inherent to PP "finals".
IM distortion is an altogether different matter. As IM products are unrelated to the desired signal, they are EXTREMELY irritating. Therefore, getting the IM distortion of a circuit as close to zero as can be is a good thing. To a degree, music masks HD and underscores IMD.
IM distortion is an altogether different matter. As IM products are unrelated to the desired signal, they are EXTREMELY irritating. Therefore, getting the IM distortion of a circuit as close to zero as can be is a good thing. To a degree, music masks HD and underscores IMD.
Thank you Eli. People rarely discuss IMD and its affect on music reproduction.
Take any "tube output" Chinese made CD/SACD player. Put in a test disc and measure the audio distortion before (very low) and after the tube - VERY high, and mostly 2nd h!
Yet some people just love the sound of these sonic attrocities.
But the worst I have ever seen/heard is a German CD player based on a Playstation mechanism, with a EL34 as an output amp, with something like a 100k anode load R.
Over 25% 2nd harmonic! And when I corrected it - the owner demanded it be put back as it was!
Chocolate sauce is chocolate sauce...and some peole can't get enough of it!
Regards, Allen
Yet some people just love the sound of these sonic attrocities.
But the worst I have ever seen/heard is a German CD player based on a Playstation mechanism, with a EL34 as an output amp, with something like a 100k anode load R.
Over 25% 2nd harmonic! And when I corrected it - the owner demanded it be put back as it was!
Chocolate sauce is chocolate sauce...and some peole can't get enough of it!
Regards, Allen
LOL, and just a year ago my Hept'AU7 was denigrated as an abomination. Don't tell me 12AU7s are suddenly back?
Some day, I've been meaning to build a squaring circuit. Take something like a MC1496 balanced modulator, feed your signal into both inputs thereby generating the square, then mix with the original signal, thereby adding second harmonic. The modulator is linear, so I suppose there will be a minimum of IMD, but only relative to other possibilities (e.g., a diode mixer adds its own distortion, being exponential rather than linear). By definition, any frequencies in the source will be intermixed with each other, otherwise known as IMD. If one assumes that 2nd H is potentially appealing, while IMD is not, then one might compare the mixture to a wine, where the 2nd H is body and flavor, and the IMD is sour and bitter; you always get a mixture of both, but the goal is to optimize that mixture.
Higher harmonics can also be generated by chaining multiplier stages. So you get x * x = x^2, then multiply that by x again to get x^3 (which contains mostly third harmonics), and etc.
Oh, and as long as we're trashing audio signals for the perceptive appeal of it, a popcorn noise generator might be useful. If you can find a component that makes noise very much like dust on a record, it should do an excellent job of "bringing your CDs to life".
Tim
Some day, I've been meaning to build a squaring circuit. Take something like a MC1496 balanced modulator, feed your signal into both inputs thereby generating the square, then mix with the original signal, thereby adding second harmonic. The modulator is linear, so I suppose there will be a minimum of IMD, but only relative to other possibilities (e.g., a diode mixer adds its own distortion, being exponential rather than linear). By definition, any frequencies in the source will be intermixed with each other, otherwise known as IMD. If one assumes that 2nd H is potentially appealing, while IMD is not, then one might compare the mixture to a wine, where the 2nd H is body and flavor, and the IMD is sour and bitter; you always get a mixture of both, but the goal is to optimize that mixture.
Higher harmonics can also be generated by chaining multiplier stages. So you get x * x = x^2, then multiply that by x again to get x^3 (which contains mostly third harmonics), and etc.
Oh, and as long as we're trashing audio signals for the perceptive appeal of it, a popcorn noise generator might be useful. If you can find a component that makes noise very much like dust on a record, it should do an excellent job of "bringing your CDs to life".
Tim
2nd harmonic cards is widely used in studio desks to the low frequencies where bookshelf speakers have problems with lower bass notes, and the 2nd harmonic inclusion gives a fuller sound although somewhat artifical. For large systems with very low frequencies,one hears the lower notes as normal. The trick is easily accomplished; all transformer iron will go into 2nd harmonic when overloaded at LF.The rest is controlling it.
richy
richy
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Could be a LPF, could be something else but who knows.
Oh yes I forgot the source: here - http://www.angelfire.com/biz/bizzyb/TADAC.html
Last edited:
Adding 2nd (and higher) tube generated harmonics is very common in solid state microphone pre-amplifiers.. I own a Presonus Bluetube stereo mic pre and it actually allows you to add tube distortion to the signal in order to warm things up. Surprisingly or not this is commonly done to warm up particular mic sound for digital recording purposes. You might call this an "effect"
I've heavily modified this mic pre to improve its performance, but I have yet to record with it so I have no idea how well this works in practice.
I've heavily modified this mic pre to improve its performance, but I have yet to record with it so I have no idea how well this works in practice.
SS generation of 2nd harmonic
There's really no need to go that fancy if all you want is 2nd harmonic. Second harmonic can be generated by asymmetrical clipping. An op-amp in an inverting amplifier configuration with a diode across the feedback resistor will do that once the output voltage reaches the forward voltage of the diode (0.6~0.7 V for silicon diodes). You can adjust the amount of distortion by adjusting the input level to the amplifier. Depending on how hard the diode turns on, I would imagine that you would get some amount of odd order harmonic products as well. I would imagine that if you used a circuit akin to half of a sine shaper, you could probably tune it to deliver whatever harmonics you wanted.
~Tom
There's really no need to go that fancy if all you want is 2nd harmonic. Second harmonic can be generated by asymmetrical clipping. An op-amp in an inverting amplifier configuration with a diode across the feedback resistor will do that once the output voltage reaches the forward voltage of the diode (0.6~0.7 V for silicon diodes). You can adjust the amount of distortion by adjusting the input level to the amplifier. Depending on how hard the diode turns on, I would imagine that you would get some amount of odd order harmonic products as well. I would imagine that if you used a circuit akin to half of a sine shaper, you could probably tune it to deliver whatever harmonics you wanted.
~Tom
That's a pretty easy way, and it allows plausible deniability, i.e. you can say it sounds transistor-ey and strained at the higher current bias point and the sound just opens up and flows at the lower (more f2) point.
Kevin:
I have 2 bluetubes and would be interested in your mods. I use them mostly as active DI boxes.
Michael
Tube pre question
Kevin -
May I open another thread about these mods? I have one of these and would really appreciate your insight. thanks.
Kevin -
May I open another thread about these mods? I have one of these and would really appreciate your insight. thanks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.