You do that, not me
. And please also do not invite me for listening tests of plywood vs. oak cover. Thank you .
Loudsinks...
I am well familiar with heatsinks 'singing' according to power amplifier output level and loading, and differences according to heatsink construction and shape....also rubber bungs between fins intended to damp/prevent heatsink 'singing'.
Ditto transformers and other components also, but IME producing rather lower acoustic level and different spectrum (ie sound) compared to typical aluminium heatsinks.
My initial question is more to do with measurement data showing spectral and distortion harmonics differences imparted by otherwise identical steel or aluminium enclosures....anybody quantified this ?.
Thanks, Dan.
I am well familiar with heatsinks 'singing' according to power amplifier output level and loading, and differences according to heatsink construction and shape....also rubber bungs between fins intended to damp/prevent heatsink 'singing'.
Ditto transformers and other components also, but IME producing rather lower acoustic level and different spectrum (ie sound) compared to typical aluminium heatsinks.
My initial question is more to do with measurement data showing spectral and distortion harmonics differences imparted by otherwise identical steel or aluminium enclosures....anybody quantified this ?.
Thanks, Dan.
Yes, they concentrate only on dI/d(theta). Yet nowhere is a derivation attempted to discern the integrated resistance and subsequent IR drop over time. Of course, that was not their intent.
Nor did A. B. Dwight back in 1923. His paper is a virtual duplicate of the one being discussed. I am delighted to find that papers from the 20's are available via google and IEEE btw...some of the best stuff on this topic is from '58 back.
So your sticking point is that I said the word eddy currents as discussed within their document, you agree that they were talking about eddy currents, but now you are hung up on the fact that they did not specifically say the word eddy currents when they were talking about eddy currents and I pointed out they were talking about eddy currents????
Is that what your gonna hang your hat on??.
jn
ps..this discussion with you is nice. I would have it no other way, thank you for your time and effort..
What is the ratio of useless posts witch can be considered as noise contribution in this thread ?
This said, you were very near, indeed, to advice cones and choose of rare woods ;-)
I'm, like you, sure that the change of this amp's sound, depending from the cover's material, can be measured and explained. Or some noise from the PSUs, or some microphonic part in the amp, like a poor weld or poor contact somewhere or some mechanical element, playing a role of capacitance, and sensible to vibrations.
As habit, the bad way to solve this problem is the audiophile way: playing with material's quality, OMHO.
As said Chuck Norris, "When the wise man points at the moon, the fool looks at his finger".
Last edited:
My vote is 99.9% useles, 0.1% valuable. But even those 0.1% are sometimes damned interesting
+1But even those 0.1% are sometimes damned interesting
i've learned or my attention have been pointed on various interesting things i do not worried enough about.
Reason why we all stay trolling or shaking the coconut here, i believe.
Thanks to John, (among some others), even if i am far to agree all the time with him. The issues he raises are often interesting.
Last edited:
"To Bonsai's comment: well, the motivation is, "because it can be done", and secondarily, because it hasn't been done all that well yet. And beyond that, because other inductive sensors may be able to benefit from the various techniques brought to bear. And as John points out, there is no one-size-fits-all opamp for the purpose, that can manage the drive requirements and provide a good noise match."
I am a bit late with this response. Apologies.
As I said, once the stylus hits the vinyl . . .
I have other concerns, namely what happens when you drive an RIAA network from a class B opamp output stage.
I am a bit late with this response. Apologies.
As I said, once the stylus hits the vinyl . . .
I have other concerns, namely what happens when you drive an RIAA network from a class B opamp output stage.
It's amazing what you can discover when you connect a 200MHz bandwidth scope to an amplifier . . .
Last edited:
If you look at the suggested applications, they just might have a major customer with very deep pockets.
You may be in danger of making your view unfalsifiable. Any paper which does not explicitly explore time variation of the resistance you can dismiss as 'not their intent'. This could be all papers, as all people working on skin/proximity effect know that the resistance does not have time dependence.jneutron said:
Any decent EM textbook will show that the field solution for skin effect (the 'self' version of proximity effect) has geometry which depends on frequency alone, so there is no time variation of resistance - just a frequency variation of resistance. You would have to start by assuming standard EM theory is wrong in order to explore alleged time variation of the resistance. Then you would move from being a 'Maxwell solution denier' to being a 'Maxwell equation denier'.
Going back to your oft-repeated point that it is rate of change which induces eddy currents; consider Lenz's Law. This says that induced currents act to oppose the driving currents/fields/whatever. What will oppose an applied sinusoidal current? Another sinusoidal current, producing a sinusoidal field. The net effect of the applied current and the induced current is a change in the geometry of the current density - which we call skin/proximity effect. A change at twice the applied frequency would not oppose the driving current, so Lenz's Law will fail. I am trying another tack, as arguments based on
1. linearity of Maxwell's equations,
2. absence of frequency-doubling in RF/microwave circuits,
3. skin effect theory in EM textbooks,
4. proximity effect calculations in published papers discovered by you.
all seem to fail to convince you. I am running out of ideas. How do I convince someone who simply refuses to believe the simple truth?
Here are your approx slopes
(What are the intermediate peaks (blue line)?
Pavel, do you imply that by increasing the load, the slope increases?
You two have me(not to mention jn who has blocked me completely
Oneand I'll think of this while counting sheep
George
George,
What I see is that the harmonics from rectification roll off around 6 dB/octave (AKA 20 dB/decade).
The capacitor filter is also single pole so it too is 6 db/octave.
Some random noise is also present and it follows to some extent 1/f.
Now looking at the increasing sensitivity of the hearing mechanism, we are doing slightly better than breaking even.
If we were to add a theoretically perfect inductor to the basic power supply we would gain an advantage.
An inductor after the rectifier will reduce our power supply output voltage. They also are hard to find that work at low voltages and high currents.
So you went and looked at CRC filters. Is there a combination that will approximate the LC results?
You do that, not me
I knew that it was coming so I added that paragraph.
Nevertheless, you are always invited (not for wood comparisons)
Is it full moon today?
My first act of the day was to post on wooden tops.
Second one was to kill a fine speaker driver that I was modifying for 2 days.
Third one was to brake a Valvo EL84.
Six more hours till midnight.
George
I'm sorry. RIP. Old electricians have scars all over the soul.
Go for 'Solid' state devices
That is certainly not the case.
The point is, the specific reason for the paper under discussion (as well as others) was to derive an expression which allows engineers to use a generic factor to calculate the dissipation within a conductor of some geometry, this paper ends with a Rac/Rdc value.
At no time within this paper is any consideration given to when the dissipation occurs. It's all about time averaged numbers.
It isn't. I've been desperately trying to find the papers I read back in the 80's when I was building sub nanosecond TRR fixturing.
And as you well know, stating that a paper doesn't derive an entity certainly does not prove it doesn't exist..
Historically, that is incorrect. Unfortunately, I've no clue as to how to redo the equations with infinite conductivity.
Perhaps you could modify the paper's equations to demonstrate the impossibility? Or will you continue to state it doesn't exist because nobody's tried?
Nah, no point, it doesn't add anything to invoke lenz.
Within the bridge setup...If a square wave with a specific slew rate is driven into the bridge (for ease in discussion), what is the end terminal resistance value and voltage drop during the ramp up..then, what is the value during ramp down? How long after ramp will the resistor value be it's dc value?
Better to have Ed repeat with low resistance values. It should be simple enough to provide data one way or the other..
jn
- Status
- Not open for further replies.