Yes, that's a common excuse when the data don't agree with the preconception.
So my question: For devices which use junction isolation, is scr action at inputs an issue anymore? And, do designers consider the tub junction profile to tailor or speed up the junction's return to reverse bias? TFR, as it were.
I asked here because an input pair can suffer this as well if the input pair is overwhelmed by slew, so I figgered it was kinda on topic.
jn
IME you avoid super-integration in the same tub at times to take care of the most obvious cases. Operation after outside the rails fault conditions are difficult to characterize except on a case by case basis. We would generally only worry about permanent latch-up, and layout rules seem to make this problem pretty rare now. Never came across what you called TFR.
Yes I accept that the track wasn't perfect, although it was in fact a demonstration track from the early days of CD. Perhaps the really upfront nature of the recording was thought to impress and make equipment sound good back then.
That always surprises me, that when you really listen intently, just what minute differences can be discerned. The tests of a few months ago proved that, tiring, yes it can be, listening again, again, again trying to winkle that tell tale giveaway.
We rarely are talking about the obvious here except those that claim a single substitution of a Vishay resistor for an ordinary RN55 is so obvious you "put your hand over your ears in 30 sec". So you then have that group of designers claim sighted testing is always OK.
Thanks Waly. I tried listening to these myself before posting and found I could pretty reliably ID them apart, but as to which was preferable... not always easy.
Dan, that nails it for me. I've been there many times, a tweak or whatever and then find that its only through long term listening that all isn't as it seems.
Frank generally does really well on these kind of tests, I've seen that before on the earlier runs. More than that, he knows what kind of sound he likes and I think would choose equipment based solely on its sonic merit rather than a spec sheet. I'm pretty much the same tbh. Its no good kidding yourself, if you don't like the sound it doesn't matter what the specs say.
I think another trial would be good at some point... a "quieter" and altogether more subtle track and perhaps a change of active device... I know just the one
It can be. I mean, it is like everything, all depends on us. Blind or not is just a question of honesty, suspicion, care and... work.
This said, what do you mean by OK ? Some defect here, compensating a defect in an opposite direction ( even in your listening (or measuring) tools there and you're mistaken ? Blind or not ? Yes.
On my side, i use blind when i'm unsure, thinking i'm loosing my time for insignificant details, and I'd better listen to music or watch a movie, instead ;-)
And about hearing ability, what can-we pretend about high treble qualities, after 50 years old ?
That is the problem with listener 'preferences'. Sometimes, a circuit with MORE distortion can sound more 'alive' or frequency extended.
This happened with Mark Levinson and me, 40+ years ago. His system was these big electrostatic speakers driven a Phase Linear 700 power amp. He and I both agreed that the system sounded balanced and OK, but Mark's mother kept complaining that it bothered her, so we looked for a problem. I changed the quiescent current slightly to remove much of the crossover distortion and the system went completely out of balance. Mark asked me "What did you do with the high frequency response?" I said "nothing" but the highs had gone away. Months later when I visited, Mark added an electrostatic tweeter to restore the balance. This sometimes is necessary.
Personally I suspect that some speakers might like a little added distortion, at least over the short term, but perhaps long term, add listening fatigue.
This happened with Mark Levinson and me, 40+ years ago. His system was these big electrostatic speakers driven a Phase Linear 700 power amp. He and I both agreed that the system sounded balanced and OK, but Mark's mother kept complaining that it bothered her, so we looked for a problem. I changed the quiescent current slightly to remove much of the crossover distortion and the system went completely out of balance. Mark asked me "What did you do with the high frequency response?" I said "nothing" but the highs had gone away. Months later when I visited, Mark added an electrostatic tweeter to restore the balance. This sometimes is necessary.
Personally I suspect that some speakers might like a little added distortion, at least over the short term, but perhaps long term, add listening fatigue.
Last edited:
Car audio, sounds big and brash but is probably 99% distortion, listen to a neutral system and it can sound thin to some and quiet, you don't realise how much sound energy is often put out with a decent system, till you try shouting to the wife in the kitchen for another coffee. I do find that low sound quality systems can initially sound great, but after a few minutes they start to get on my nerves and listening fatigue sets in...like telly's straight out of the box, bright high contrast and everybody has a nice sun tan looks impressive initially then after a few minutes your eyeballs melt.
I agree.
The primitive calcs were for steady state conditions (thermal equilibrium) with a small temp rise. To the best of my knowledge, under these conditions, thermal resistance (Rth) is a constant.
Thermal impulse calc, where thermal impedance (Zth) concept should be applied, is for knowledgeable guys.
Thanks for confirming.
You mean 700mA [(75mA * 8=600mA)+100mA]
And jacco is right and always serious, don’t be fooled by his clown emoticon!
George
Jacco, a gift (reimportation
)http://www.nxp.com/wcm_documents/models/bipolar-models/mextram/bctm04rth.pdf
Attachments
A pretty similar thing happened to me when I started to crank up the bias current of my HK 680 integrated amp. As adjusted in the factory and noted in the sevice manual, its dual pair of output devieces (per channel) should be biased at 85 mA each. Just for the hell of it, I started cranking it up in first 15 mA do 100 mA, and then in 10 mA steps. Upon adding each increment, I gave muyself a whole day to listen to it using a battery of source material.
Heat sinking was not a problem, it has one darn big heat sinf fin, but I mean BIG. It got a little hotter, but nothing to worry about so long as it was not covered by anything, nor was anything placed on top of it.
At 140 mA, I noticed that the sound had lost a little focus, as if it became blurred. Dropped it back to 130 mA, and it was back to normal. Left it there for the last 12 years. About once a year, I just checked for my DC offset, and registered a 3 mV change from the nominal last time I looked, about a year ago.
This corresponds very cosely with my general experience with class AB amps. They mostly profit by adding some bias current, but sooner or later you reach a point when it no longer improves, but starts to actually degrade the sound. The only way then is to go for a complete, full class A operation, not really possible because of the required number of devices and incredible heat sinks one would required.
Last edited:
I shall reveal the actual file differences later today. You may be surprised. But I hope you don't change your minds.
I'm looking forward to that because I have no idea what Pano is going to show on this.
Oh yeah, Professor Holst lane.
Natlab used to be there, or so I heard on the history channel.
In case I get caught with my pants down, in flagrante, it enables a response that it was just a bit of humor to cheer people up : did you not notice the emoticon ? (it was recommended by my sollicitor, Frank)
- Status
- Not open for further replies.