Symmetrical vs Asymmetrical Clipping

[jazbo8]

Guys,

Just came across this article by Rod Elliot >> Intermodulation - Something 'New' To Ponder

He claimed that he found little on the subject and only sited Crowhurst's paper. May be I am missing something, but using the diode clippers as examples, aren't the results exactly what one would expect from fourier transforms of the fullwave and halfwave signals?

Halfwave/asymmetrical clipping, has more even harmonic components than the fullwave/symmetrical clipping, just from looking at the respective fourier transform formulas and/or graphs.

So isn't the difference between the two types well known as opposed to "un-known"? Or perhaps I simply missed the whole point that Rod is trying to make which is quite possible ;-) If so, can someone enlighten me...

Jaz

[Enzo]

Um... link?

[jazbo8]

Sorry...fixed, see above.

[Steve Conner]

I think it should be read as Rod's personal epiphany. The implication is basically just another poke in the eye for the single-ended triode hi-fi crowd, as it suggests that their amps produce a higher level of IMD than more complicated push-pull circuits.

Radio engineers are mainly interested in third-order intermodulation products, because they appear close to the desired signal, with the potential to splatter over adjacent channels. These are of the form 2f1 +- f2. In the case of the 19+20kHz test, they would be 18 and 21kHz.

As far as I know, all electronic circuits produce the third-order and other odd-order products, symmetrical or not. It is only the even-order products, such as Rod's f1+f2 and f1-f2, that are cancelled by balanced circuitry.

If you think of a distortion harmonic as an IM product due to a single frequency intermodulating with itself, then you can see that a balanced circuit cancels even-order IMD for the same reason that it cancels even-order harmonic distortion.

My own efforts at measuring IMD have mostly been spoiled because the audio ADCs, DACs and op-amps I used produced levels similar to the power amp circuits I was working on. At the end of the day I got more useful information out of an old Tektronix distortion analyser.

[jazbo8]

@Steve - Perhaps that is the case, I always thought of Rod as a good engineer and writer (his snide remarks on tube amps and SE amp aside ), so I was somewhat taken aback when he said he found the IMD results to be counter-intuitive... Anyway, instead of running a sim or breadboarding it, just run this java app, you can make a halfwave/asymmetrical clipper by dragging the mouse across either half of the sinewave, and compare the resultant Fourier series components vs the fullwave clipped version - no surprise really...

Also I forgot to provide the link to the first part of Rod's writeup.

Jaz

[EDIT] someone pointed out that the java app is only for a single frequency, so not applicable to IMD, but it does show the difference between symmetrical vs asymmetrical clipping and the effect on their Fourier transforms.

[Gingertube]

When I was designing my Baby Huey HiFi Tube Amp I came across something very interesting.
There was suggestion that a common resistor in the cathodes of the output tubes in a push pull amp could supress 3rd and other odd harmonic distortion by generating some common mode feeback.
A value equal to about 10 to 15% of the normal single tube cathode bias resistor value was suggested (e.g. for EL84 which would normally uses 270 Ohms per tube, then add a 39R Common resistor). If fixed bias then just that resisrtor goes between the cathodes and 0V, if cathode biased then lift the ground side of the bypass caps and fit this resitor between 0V and the old 0V side of the bypass caps.
One of the guys posted the spectrum analyser results from this mod. What was noticeable is that while there WAS a slight reduction in 3rd and other odd oder harmonics what really stood out was an almost complete elimination of all the IM sidebands from power supply ripple.

For those of you who are also DIYAudio Members then the Spectrum Analyser plots are here:
http://www.diyaudio.com/forums/tubes...y-huey-46.html
See the posts by Sheldon

Cheers,
Ian

[bob p]

[Gingertube]

Bob,
Thanks for draging them across.
Bottom trace is the unmodified response, the top one is with the common resistor added, note all the 1KHz +/- 60Hz, 1kHz +/- 120Hz etc intermodulation sidebands have gone!!!
Cheers,
Ian

[jazbo8]

Thanks, Ian & Bob. Very interesting stuff, most of it still being digested I will try the circuit one of these days - all my tools, and test gear are packed up by the movers, so all I can do these days, is read and sim stuff

Anyway, found an explanation from Steve Bench on the Harmonic Equalizer:

"On the overall concept. As everyone knows, push pull operation cancels even order distortion components through symmetry. Thus, you can push the tubes harder and achieve more power before distortion becomes objectionable. Well, those clever Western Electric folks came up with a "Harmonic Equalizer" concept in the WE86 amp. This exploits the fact that even order distortion is cancelled in PP to come up with a way to cancel the odd order products too. What they did was to take some of the even order products (developed at the common cathodes of the output stage) and feed them into the bias. This is not feedback: the injected frequency is the harmonic energy, not a copy of the desired signal. This allows these components to be mixed through intermodulation action producing odd order distortion components that just happen to be phased properly to cancel the odd order distortion produced. This lowers overall distortion even more. Note that this is a wideband effect. Distortion is lowered over a wide range."

Jaz