What? 😀
“Information is not knowledge. Knowledge is not wisdom. Wisdom is not truth. Truth is not beauty. Beauty is not love. Love is not music. Music is THE BEST.” ― Frank Zappa.
Sorry, to be correct work done is V x I x t.
Current x voltage is rate of work done, better known as power.
All conjecture is good publicity, right? Have you looked at the Temporal Coherence website...?
The distortion was seen on the horn loaded box, is what I read. Horn loaded boxes have a tendency to be over-dampened from feedback.
When it was switched to BR suddenly the damping became of value. The better the damping (motion-feedback) the more accurate a BR box will get. As volume rises, excursion tries to go all out, but it must get the correct resonance/hz to raise SPL. So the more accurate the resonances/hz, the higher the actual SPL from the off phase air passing through the port that builds on the driver's front facing pressure. Without the feedback it will just get sloppy (distortion) and that can decrease/not increase SPL when you don't have matching hz from port and driver front.
A horn does not work the same way. The problem here is that the driver needs to push air through a high pressure zone and let the tuning of the horn provide gain as it goes to low pressure, so excursion is more paramount than accuracy for SPL (opposite of BR? sorta). As the amp senses variation and tries to apply more e-brake to fix it, you ultimately just get a withered away bass that comes out sloppier than if you hadn't tried to fix it with feedback when fighting so much mechanical damping. This also has to do with horns generally having an assortment of irregularities, different little resonance peaks all over; and there is often a good amount of mechanical damping between the nature of a horn, and things like stuffing to reduce reflections. So you just kind of live with their variations and praise the magic of high to low pressure gain. That is why a no-feedback amp can put out a lot of bass on horn loaded single driver setups, since it ignores everything and pushes that excursion. Obviously none of this applies to horn shapes for tweeters like on the M2.
I thought the motion-feedback and Q15/16 increased damping capability some? I'm not a "master" like JC, so I guess I'll just stop there. I'm sure everyone will be in love with my insomniac-tired blubber about BR vs horn that cannot meet anyones standards.
Last edited:
From looking a the Marsh schematic, the current feedback is in phase with the voltage feedback. Given the resistor values the amplifier has mostly voltage feedback but as current goes up it starts to reduce the output voltage. So, it has some soft current limiting effect. Speaker power is limited somewhat at frequencies where it's impedance is very low. And of course, the limiting effect is in phase with speaker current.
Last edited:
That is interesting. I would have assumed the exact opposite (for an efficient motor) and something vaguely opposite (for an inefficient motor like a speaker). The voltage across Re simply creates heat. Have I missed something?Joe Rasmussen said:
Nevermind everything from before. I guess Richard is only refering to a BR cab?
In which case the most curious thing about JC's suggestion is what is the point of increasing current capability throughout? It would seem none. And now Mark is suggesting it limits it... which is confusing why his schematic says increasing it...
The driver is fairly large, and will have a more directional back pressure that may act more like a feedback loop/resonance generator. Perhaps this is problematic in that the damping can excite further distortion something like Keantoken suggest, or it simply climbs harmonic scales. Where as if you decrease feedback and simply dump current into the non-adjusting frequency, ala the motion feedback, you get a happy zone between what comes out of the port and off the front of the driver match so that spl increases and THD is present but not awful.
I suspect Mark or others probably will solve this...
In which case the most curious thing about JC's suggestion is what is the point of increasing current capability throughout? It would seem none. And now Mark is suggesting it limits it... which is confusing why his schematic says increasing it...
The driver is fairly large, and will have a more directional back pressure that may act more like a feedback loop/resonance generator. Perhaps this is problematic in that the damping can excite further distortion something like Keantoken suggest, or it simply climbs harmonic scales. Where as if you decrease feedback and simply dump current into the non-adjusting frequency, ala the motion feedback, you get a happy zone between what comes out of the port and off the front of the driver match so that spl increases and THD is present but not awful.
I suspect Mark or others probably will solve this...
Last edited:
Re, if it is the only resistance in the model, would have to include mechanical friction and air loading type things. Or, is there another resistor in the model for those? Ultimately, they all make heat one way or another. Out of phase voltage and current = no work done .
Here is how it looked to me: Suppose you have some input voltage from a preamp, say 1v. The amp and feedback network will need to put 1v at the inverting input to keep the inputs in balance. If current feedback is in phase with voltage, then the current feedback signal contributes a little to bringing the inverting input up to 1v. That means the amp output voltage can be a little lower and the input stage still in balance.
Depending on whether I got the phase of the wiring right, and to the extent the speaker current is in phase with voltage (speaker looks resistive) then that seems about right. If the speaker current was way out of phase then it could require the amp output voltage to raise. Depends.
Of course, it is 4:30 in the morning here, and I need to get some more sleep. So, I could have it wrong.
Last edited:
What a strange thing to say, why would I be irritated and is talking about possible speaker solutions that bad? Maybe the word solution is not really apt, maybe inquiry is much better?
You are allowed to think aloud - sometimes works for me too. 😀
I dunno, maybe you might find this an interesting read: Power Factor - The Reality
Didn't see anything new there, wonder why the title said 'the reality'? As though it were commonly misunderstood?
I agree. The negative current feedback increases the output Z of the amplifier. As opposed to positive current feedback like one of the simple Pass designs uses to try to make up for its problematically high output Z.
Joe I am still trying to understand this. You showed two overlaid graphs of SPL vs frequency, one red one green. One showed higher output than the other, but higher by varying amounts. What are these? Did you measure SPL at two different drive voltages? Did this driver exhibit level-dependent changes in frequency response?
Hi Joe,
-Chris
Well, no. That conclusion is a jump too far. They heard a change, but whether that is good for the speaker or not cannot be assumed out of the blue. If you are saying that they thought the speaker sounded better with some series resistance, that's another thing that I can't argue with.
I think that discussions are always fine as long as they remain inside the boundaries set by the rules of the forum. If it degenerates into verbal wars we would probably have to shut it down. Otherwise, it's fine to have that discussion. Probably best in it's own thread.
-Chris
To be general V and I have relative phase and the real part of the product is the power.
Apologies to Joe I should have said maybe he is disappointed that more folks are not paying attention to this way of designing speakers. You could eliminate all the wrong and/or confusing stuff and underneath there is a valid design process. There is no "breaking down" of Kirchhoff or any of the other nonsense.
The key is to understand is that the difference between the two are entirely down to the back-EMF and this test can be done physically and also using a modeling program like SoundEasy. It reveals that only the voltage that appears across the Re of the driver's voice coil is part of the force that creates the final dB-SPL (F = Bl*i) and the two graphs shows the relative difference under current versus voltage drive. The back-EMF part of the impedance is not part of the force, but rather acts like a voltage source, an opposing force to the current. It impedes current and that is what makes it an impedance.
While the current is what the device (a driver is a current device) actually responds to, it is the current times voltage (the power dissipated in the voice coil) that determines the dB-SPL. The fact that SoundEasy can do this modeling shows it to be based on actual currently understood maths and also, it is not difficult to do it physically, use a large value series resistor (50 Ohm or more) to simulate a current source and it behaves exactly like that.
Hope that helps, but I assure you this is no voodoo science, although the history of this thread tells me that... oh well. Negative comment is something I have come to expect.
Agreed, and also trying to be nice about it. No ill will intended towards Joe, who seems like a good guy.
It is just that to the extent speakers are improved by the addition of a network, it is completely explainable in conventional engineering terms. There is no need to develop novel ways of thinking about it or talking about it, and nothing to be gained by it.
However, it looks like over a long period of time and many arguments, Joe is strongly determined to stick with doing things his way. Too bad, he isn't going to learn a better, thoroughly vetted and proven way, and I'm not going to learn his way. Guess we won't be talking much about how speakers work.
- Status
- Not open for further replies.