Experience is the key to reading any measurement. I assume you were baiting J.C. as the one he was showing clearly show lots of splatter. The related harmonics don't stand out the way the far off ones do. Get far enough away and -100 isn't really good enough. But full power graphs only show the amp isn't yet clipping. Distortion near crossover is quite hidden by high level measurements and you do most home listening at very low power levels.
I see John's point, you also notice the first has much less f1-f2 component at 1kHz probably also due to the "gobs of feedback". Measurements of first watt performance are probably more telling. I've been playing with double twin-tee passive notches, one could probably do 19k/20k IMD with a sound card. In simulation I could get -80db notches at both (pretty touchy trim in practice). One could average and use long FFT's and get way down in the noise floor.
In the text, they say that the 1 watt residual is noise-dominated. So the crossover song and dance is just a deliberate distraction.
Jan: There's a graph showing some higher order distortion products at under -100dB with 800W flowing. Clearly a major concern.
The measurements descriptions indicated that the residual at lower powers was just noise. The only things the amp lacked were a logo from someone on the Team, a fanciful story, and a royalty check for John.
No , look deeper ...
As you can hear 30 db into the noise floor (Depending on how you define noise) saying it can't be measured with the equipment they were using then isn't really informative.
Then there is the issue of what the noise floor is! Op-amps of the day were not exactly quiet.
But then what would would I expect from a guy that still uses vacuum tubes!
No that's not a game I play. I clearly see the 1 kHz diff tone in the JC-1 graph which is completely absent on the Crown graph. Also the sidebands near the 19+20kHz fundamentals were larger on the JC-1 graph than on the Crown graph.
Please tell me what I missed? The grass in each seems around -110dB but the ones I mentioned clearly rise above that.
Jan
You like to forget about the fundamental, a recurring theme.
30dB below a noise floor of...? Another recurring theme.
As you can hear 30 db into the noise floor (Depending on how you define noise) saying it can't be measured with the equipment they were using then isn't really informative.
Then there is the issue of what the noise floor is! Op-amps of the day were not exactly quiet.
But then what would would I expect from a guy that still uses vacuum tubes!
Shhhh, vinyl and RR is the problem 10W@10thd is good stuff ..
On the Crown I see the spikes at virtually every 1,000 hertz harmonic. These are unmusical, the spurs on the other's spurs would be masked by the music or even perceived as more musical.
No distortion would be best for many, some like bits of low order harmonic distortion, only one guy here likes lots of high order stuff and even non harmonic stuff. But as he is the gardener, it is okay.
The high NFB , two dim sound dats what ...
Then :
"Fig.7 shows that, at a slightly higher power output—1184W peak into 4 ohms and still prior to clearly visible signs of clipping in the waveform—the distortion spectrum, while still quite low (less than 0.01% at 1kHz) is beginning to get messy. But it may be academic; you're never going to put thios much power at 19kHz and 20kHz into any loudspeaker load and have a tweeter that will live to tell about it."
Couple that with its kneeing and sudden transition and you get an amplifier that trends towards being harsh at anywhere near being driven. 4 ohm nominal speakers with 45-60 deg phase angles will show case this allday long . Experience has shown that trending is more or at the very least as important as amount.
Last edited:
Ok I see that. So we're back to what is more audible: the 1 kHz standing out in the JC or the 'harmonic grass' below -110 on the Crown.
Or the TIM/PIM around the sidebands on the JC or the same stuff but lower/less wide on the Crown?
But you cannot really say how either sounds based on the graphs anyway.
Jan
If the JC Dont show any trending the sound will be more consistent to the ear ...
Yes, that's what the onset of clipping looks like in just about any well-designed amp. Back it off a few watts (to a mere 1100 watts!) and you end up with the very clean results shown a few posts back.
Well the IM measurement shows a clear 1 kHz spike above anything else. That's almost a text book definition of IMD?
Jan
@Sy,
Nope , its where he started measuring from , guarantee its does the same from 1-10, 10-100, etc , you dismiss because you cant tell at what amount can the ear detect the change. I'm suggesting when listening it is less sensitive to the amount than a change in the amount .
look at the last graph posted ..
Nope , its where he started measuring from , guarantee its does the same from 1-10, 10-100, etc , you dismiss because you cant tell at what amount can the ear detect the change. I'm suggesting when listening it is less sensitive to the amount than a change in the amount .
look at the last graph posted ..
Last edited:
Same scales, increased power, and the text supports that. See figure 8- this is taken partway up the inflection between 900W (900W!) and clipping. The power here is also at the single channel clipping point (see Table 1).
This is one amazingly good amplifier. At these levels, the Parasound would be reduced to a smoking cinder.
edit: The graph you referenced is under different conditions- parallel mono. Try apples to apples, it's more accurate.
- Status
- Not open for further replies.