A friend of mine did a little work on my Bedini 150/150 mill and he measured 10% THD at normal listening levels.
After all these years it is still cleanest most dynamic amp I've had in my system.
So.... 10% sound very high when I see .01% for amps in their literature. And i just can't hear any distortion. What gives?
After all these years it is still cleanest most dynamic amp I've had in my system.
So.... 10% sound very high when I see .01% for amps in their literature. And i just can't hear any distortion. What gives?
Well, are you comparing it to other amps with 0.01% when you listen? Or do you just sit back and listen?
I think this question depends on how the 10% THD is generated. If it's from the amplifier clipping, and you cannot hear that, then you are wasting a lot of money on audio gear. If it's from 2nd harmonic non-linearity, well, that's way more than enough to be very audible.
I suppose it's possible that you have a tin ear, but what is more likely is that you simply like the sound of 10% 2nd harmonic. It is certain that an amplifier with 0.01% distortion would sound very different. It is not certain that you would like that sound better.
I think this question depends on how the 10% THD is generated. If it's from the amplifier clipping, and you cannot hear that, then you are wasting a lot of money on audio gear. If it's from 2nd harmonic non-linearity, well, that's way more than enough to be very audible.
I suppose it's possible that you have a tin ear, but what is more likely is that you simply like the sound of 10% 2nd harmonic. It is certain that an amplifier with 0.01% distortion would sound very different. It is not certain that you would like that sound better.
I know how 10% THD sounds to me.. it sounds like those horrible little mini/microsystems that were so prevalent. I'd rather hear silence
Solid state amp would probably be making something looking more than a little squared off at 10% distortion.
I would question the measurement technique and hardware. Or it's broken
I seem to recall working on a 150 about 25 years ago and it easily met its specifications.
I would question the measurement technique and hardware. Or it's broken
I seem to recall working on a 150 about 25 years ago and it easily met its specifications.
If the distortion is only even order harmonics, you hear it slightly bright. If you test it with a voice+bass as Fever of Peggy Lee, you will hear her voice modulated as if reflected by a rotating fan.
If the distortion is odd order but of opposite phase, you also hear it clean but too bright and enhanced dynamic.
If the distortion is odd order but of opposite phase, you also hear it clean but too bright and enhanced dynamic.
I long ago read somewhere the absolute human limit of perceiving THD was at 0.5%... That said, individuals who swear they can hear THD levels far below that & can hear 0.25 Db level changes...are just blowing smoke.
Test these individuals on a controlled double-blind basis, & they wind up failing...invariably they will insist the test was rigged & conceived improperly.
---------------------------------------------------------------------------Rick.....
Test these individuals on a controlled double-blind basis, & they wind up failing...invariably they will insist the test was rigged & conceived improperly.
---------------------------------------------------------------------------Rick.....
Pure Tones, 10% 2nd Harmonic, 10% 3rd Harmonic
Here's a limiting case, with tones.
All three tones are 1 kHz.
Pure is without distortion.
thd10pctsecond has 10% 2nd harmonic distortion
thd10pctthird has 10% 3rd harmonic distortion
Each one sounds like a different vowel being said...
each one lasts 5 seconds.
Download them, and play them and listen.
Here's a limiting case, with tones.
All three tones are 1 kHz.
Pure is without distortion.
thd10pctsecond has 10% 2nd harmonic distortion
thd10pctthird has 10% 3rd harmonic distortion
Each one sounds like a different vowel being said...
each one lasts 5 seconds.
Download them, and play them and listen.
Thanks djoffe! The effects are clear and understandable. On the other hand, our emotional interpretation and response to music makes it a lot more complex and harder to assess the multiple and changing distortions and decide whether they are even significant, perhaps spoil our listening pleasure or we just become accustomed to the effects and would miss them in a low distortion audio system.
I think many audiophiles do prefer moderate levels of low order distortion. When you look at typical tube amplifier figures, you could be excused for thinking that the extra distortion must be because it contributes desirable sound qualities. That may be counter-intuitive but some folk pay lots more for it than others do for very low distortion audio systems. It seems money talks where science fails to impress.
I think many audiophiles do prefer moderate levels of low order distortion. When you look at typical tube amplifier figures, you could be excused for thinking that the extra distortion must be because it contributes desirable sound qualities. That may be counter-intuitive but some folk pay lots more for it than others do for very low distortion audio systems. It seems money talks where science fails to impress.
I have met some audio and video people who could reliably state a monitor was 1 ft-lambert high or low
without another monitor to compare. I also met an audio operator on a major network show who sets his
audio by ear and the meters always agree with his ears so don't count out ability. Some people ARE that
good.
G²
I could reliably hear the 1% HD rated dynakit ST70 (good e-caps, newish rectifier & output tubes) versus .03% HD rated Peavey CS800s. ST70 was noticably fuzzier. Top octave solo Steinway piano had a little buzz. Tinkly bells had a little ssshhh. This was only audible with low harmonic distortion Peavey SP2-XT speakers. *****ier speakers with their own 10% HD, the amp distortion would be buried in the speaker distortion. My hearing is still good to 14 khz tested at my last job in 2008. Your results may vary, particularly with more hearing damage than me.
Last edited:
10% amp THD is very noticable and not something you enjoy so something is not right, even if it's "nice" THD it will significantly muddy/smear the sound.
Update. The amp in question is a Bedini 150/150 mkII which is a feed forward feedback use NPN on the positive and also NPN on the negative unlike negative feedback that use NPN/PNP neg and pos.
today I put in a rebuilt Eagle II amp and in comparison the Bedini sound horrible.
today I put in a rebuilt Eagle II amp and in comparison the Bedini sound horrible.
I had a Bedini home for a couple of days back in the 80's. ( or sometime like that, it was a long time ago) Took it back. To me it was no match for my then Plus 55 Sanyo VFET. That was the days of the Boulder Sound Gallery "amp of the week" club. Listened to a lot of bad amps. BUT...
Now on the measurements. I would look into the methods. I seriously doubt it exceeds .3% within normal use. I was just playing with RightMark and REW. My Focusrite on RightMark and similar results on a loopback in TrueRTA measures about .0008% THD. But REW said 1.7%. If your amp really is 10%, there is something very wrong with it.
I have noticed, ultra clean amps like some of the new class D, sound a bit lifeless to me. I am not sure if it is a lack of distortion we are accustomed to, or what if an AB amp, I would say insufficient power supply storage. I do know in speakers, when someone first hears a relatively low distortion speaker, then find them a bit dull, but they "reprogram" pretty quick. I was first tempted to get one and see if I reprogramed on it too, but I still find my own ( modified Hafler 120) to be very nice. Nothing wrong, nothing extra I am aware of. They do happen to measure .003% at 5W 20 to 20 and have three times the banks as was original. ( among some changes in IPS and VAS topology)
If our brain is involved, it gets complicated. We learn what to expect and without an alternative experience, come to prefer what we know. That is unless you are the admin of ASR forum and then anyone who does not believe a single THD+N measurement is the total end all of everything and all DACS and amps sound the same, then you are banned for sharing an experience. It is very simple there, you drink only their color of Koo-Aid. Kind of like a religious cult. Where I got my engineering education, an observation that does not fit the theory is open for discussion to see if anyone else had the same experience. One can learn from others experiences. Step one of a scientific process. As an example, the 10% does not make sense, so instead of declaring you are crazy to ever have liked it, or to declare the friend working on it is an incompetent idiot, I offer the friendlier suggestion something is amiss that should be looked into. Do you happen to have the schematic? Might be fun to toss into LTSpice and see how it models.
Now on the measurements. I would look into the methods. I seriously doubt it exceeds .3% within normal use. I was just playing with RightMark and REW. My Focusrite on RightMark and similar results on a loopback in TrueRTA measures about .0008% THD. But REW said 1.7%. If your amp really is 10%, there is something very wrong with it.
I have noticed, ultra clean amps like some of the new class D, sound a bit lifeless to me. I am not sure if it is a lack of distortion we are accustomed to, or what if an AB amp, I would say insufficient power supply storage. I do know in speakers, when someone first hears a relatively low distortion speaker, then find them a bit dull, but they "reprogram" pretty quick. I was first tempted to get one and see if I reprogramed on it too, but I still find my own ( modified Hafler 120) to be very nice. Nothing wrong, nothing extra I am aware of. They do happen to measure .003% at 5W 20 to 20 and have three times the banks as was original. ( among some changes in IPS and VAS topology)
If our brain is involved, it gets complicated. We learn what to expect and without an alternative experience, come to prefer what we know. That is unless you are the admin of ASR forum and then anyone who does not believe a single THD+N measurement is the total end all of everything and all DACS and amps sound the same, then you are banned for sharing an experience. It is very simple there, you drink only their color of Koo-Aid. Kind of like a religious cult. Where I got my engineering education, an observation that does not fit the theory is open for discussion to see if anyone else had the same experience. One can learn from others experiences. Step one of a scientific process. As an example, the 10% does not make sense, so instead of declaring you are crazy to ever have liked it, or to declare the friend working on it is an incompetent idiot, I offer the friendlier suggestion something is amiss that should be looked into. Do you happen to have the schematic? Might be fun to toss into LTSpice and see how it models.
10% THD will be plainly audible on a single 1 KHz test tone. It will sound different if it is all 2H as compared to all 3H, but an amp making 10% THD will also be making a considerable amount of IMD, and that will sound pretty bad on complex music.
I would say that the measurement is in error if you like the sound on multiple different musical selections.
I would say that the measurement is in error if you like the sound on multiple different musical selections.
I had my late friends Bedini 100/100 for a few weeks and thought it was one of the best SS amp I ever heard . He didn't want to sell. It was warm , bold and relaxed. Those amps sell in no time after they appear on used market.