• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Tube amp design parameters (distortion)

What is a reasonable target for distortion in a Push-Pull tube amp of 4-8W? At the moment, I am excluding any OPT-induced distortion from the discussion as that will be very device dependent.

In some simulations I was able to achieve something like 0.2% at 1W rising to 1% at 5W. Is this par for the course, or are better figures achievable? I haven't seen distortion figures quoted much for tube amps, possibly because they are so often high.

FYI spice simulations when I was designing my tube line stages matched reality extremely closely, so for the moment I'll also pretend the simulations will match the real-world devices.
 

ashok

Member
2002-06-06 4:43 am
3RS
Distortion

Hello Tiroth,
I think you need not be too worried about the tube amps distortion figures when they are in the region you mentioned. Amps with much higher distortion figures at full output still sound great. I didn't quite understand why you are designing to meet a distortion figure target. Most people cannot hear the difference between 1% and 0.1%. You should try to achieve good sound. This you can confirm only by building the unit. IMO go ahead and build it and listen to it. You can tweak it after that.
Best of luck .
 

ashok

Member
2002-06-06 4:43 am
3RS
Distortion

Hi Tiroth,
I did a quick check with my ( tattered) RCA Tube manual. They have some circuit diagrams for 'Hi-Fidelity' amplifiers. There is a 50 watt amp with 0.1% harmonic distortion and 1% intermod distortion. A 30 watter has 0.7% distortion ( harmonic) and 1.5% intermod distortion. You could check the web for data on the Dynaco ST70 or the Willimason or Marantz tube amplifiers. All generally considered good.
I once built a tubed amp with no feedback with something like 2% distortion at full output ( 40 watts) and also with negative feedback with less than 0.2% distortion. I liked the no feedback version for midrange and highs but the feedback version had better bass ( tighter and deeper). This would most probably have been due to reduced output impedance and wider bandwidth due to feedback.
You should be OK. Go ahead and try out your circuit.
 
Hi Tiroth,

Your figures are pretty typical. It is hard to reduce the distortion further for lack of gain overhead.

As Ashok mentioned higher power amps have lower distortion ratings. If you do a search on tube amps in Google or your favorite search engine, you will find that the manufacturers of high powered tube amps have rated their high power amps at 0.05% and lower. But the power is 300 and more watts. That with transformers.

I have heard solid state amps with upwards of 10% distortion at full power and they sound OK. My own design, before NFB, had a rating of 4%, and to me sounded superb. Of course, NFB brought it down to 0.01%, and still sounds as good.

So, as another said, put it together, give it a listen, and tweak as desired. Because even a "golden ears" can barely hear the distortion of a pure sine wave of 1%. For music it is impossible to hear.

My additional 2 cents.
Gabe
 
Distortion, levels and types need to be considered.

When looking at amplifier distortion it is important to look at the type of distortion produced. Even order products are generally less disturbing while the odd order products are always more noticeable and thus more objectionable. Also out of band intermod caused by high harmonic levels of the primary signals causes in-band beat notes to be produced. There is also the issue of phase cancellation when various harmonics add or subtract from each other amplitudes in the wrong relationships. This causes the original harmonic structure of the input signal to be significantly modified.

Distortion measurements can provide a rough guide of the performance of a amplifier and it can be deduced that amplifiers with no negative feedback will likely have distortion figures that are likely between 1 and 5 percent at close to maximum output levels. The same amplifier with moderate feedback will likely run around .1 to .5 percent. If the amplifier has enough gain and the negative feedback is increased substantially the distortion figures may reduce to .01 to .05 percent. The output impedance of the amplifier will decrease with each increase in feedback and also there will likely be some changes in phase cancellation levels of the harmonic structure versus output amplitude.

Single tone distortion measurements are truly hard to analyze subjectively. It used to be a rough guide that Hi-Fidelity sound amplification required a 1 percent or less distortion level when using single tone harmonic distortion tests. These days with lots of amplifier gain available before the application of negative feedback while maintaining stability can produce distortion figures that are hundreds of times lower. Many say that the negative feedback levels required to produce such low distortion figures have undesirable side effects that are hard to measure but can be heard when reproducing music. One of the primary differences between most solid sate circuits and old tube technology is gain. The solid state units have in general much more gain and thus run at much higher levels of negative feedback.

Perhaps this is a little off the subject but perhaps worth considering when looking at distortion figures.

John Fassotte
Alaskan Audio
 
With a Class AB PP amp, you have some crossover distortion where the waveforms are "put back together" or something. To minimise this in mine (PP 6v6) I used a small amount of NFB.

With virtually no NFB there is a little bit of noticeable distortion at mid-volume, but when you adjust the level (I use a 10k variable resistor in series with a 4k7 resistor in the loop) you can pick a level that sounds cleanest. The amp sounds very nice now, accurate and clean but still with warmth.

The added bonus with NFB is that it removes any last traces of hum and noise, the amp is dead silent now. :D