This whole damping argument is vastly overblown, - usually by the same parties selling the amplifiers in the first place.
You can of course do the other way, - my own solution why?
It's well known current feedback techniques DO NOT WORK below 100hz.
ie.
Multiple output windings of the OPT actually used as part of a crossover network.
One of the wiring harnesses used for the LF speakers (total impedance 22 ohms, OPT winding resistance approx 1.1ohm).
Totally linear frequency response as a result 35hz> and none of the claimed problems with speaker resonances.
Another harness is used on a different set of windings to feed the mid range array, totally unconnected with the LF winding apart from through the transformer iron.
I have never seen this technique used elsewhere.
The ability to handle large LF signals (eg, around 20-30hz) simultaneously with large HF frequencies without any form of modulation of the latter (to my mind) was pretty staggering.
"The ability to handle large LF signals (eg, around 20-30hz) simultaneously with large HF frequencies without any form of modulation of the latter (to my mind) was pretty staggering."
I get this with my Carfrae Little Big horns from a slightly different method
Full range Lowther DX4 (Backloaded) Is driven by the bridged ACA with no crossover or hardware in between the amp and the driver at all. A signal taken from the speaker cables feeds an active subwoofer in each speaker with the cut of set at the LF the horn gets down to. To quote one review:
", Little Big Horn’s active sub means it does its own thing in the bass region, and the system amp doesn’t need to supply any serious current below 80Hz".
The cutoff for the sub is actually closer to 50hz than 80hz. The active subs in my reproductions are 100 watt (8ohms) Hypex DS2 plate amps.
It works well for me at the moment and swapping out a much loved Push pull triode amp for a pair of ACA has proved a great move.
I get this with my Carfrae Little Big horns from a slightly different method
Full range Lowther DX4 (Backloaded) Is driven by the bridged ACA with no crossover or hardware in between the amp and the driver at all. A signal taken from the speaker cables feeds an active subwoofer in each speaker with the cut of set at the LF the horn gets down to. To quote one review:
", Little Big Horn’s active sub means it does its own thing in the bass region, and the system amp doesn’t need to supply any serious current below 80Hz".
The cutoff for the sub is actually closer to 50hz than 80hz. The active subs in my reproductions are 100 watt (8ohms) Hypex DS2 plate amps.
It works well for me at the moment and swapping out a much loved Push pull triode amp for a pair of ACA has proved a great move.
I assume you are talking about using a current source drive under 100 Hertz.
You can use it under 100Hz for sure. For example, a sealed box with a Q=0.703 and -3dB at box frequency 40 Hertz. Now use an LCR network tuned to 40 Hertz and if you get the current EQ's flat, flat impedance, flat current phase angle, then you can drive the sealed box and the Q will still be 0.707 at all times. The response will always be -3dB at 40 Hertz.
Is this still current drive if using a current source as an amplifier? I don't know, people can have different views about that.
But let us have a bit of fun with the subject, what is we make the LCR a bit more aggressive and reduce the R to the point that the response at 40 Hertz -5dB and you will have a Bessel sealed box alignment. If you go further and get it down to -6dB, you will have a "Critical Q" alignment. Mind you, the impedance will have a noticeable dip around 40 Hertz.
And the beauty is that the amplifier doesn't matter. It can be tube, it can be SS, it can even be a pure current source with infinite source impedance. But I would likely stay at flat impedance and Q = 0.707 is a locked-in 2nd order Butterworth.
This largely closes down discussion which is one of the reasons some people got frustrated and gave up on the blowtorch thread. I recall DF96 (RIP) was very patient and attempted to explain repeatedly where you were going wrong by making assumptions about the relevance of the current measurements at the amplifier output. I suggest you take his comments and those of others seriously, believe it or not they were trying to help.
Joe have you ever pushed a car? At first it is hard. Is it even moving? Yes it is, but slowly.. Then soon enough it is rolling along and you can ease up on the pressure.
This is inductive reactance (assuming you run around later and push it the other way). The pressure and velocity are out of phase, yet connected intimately.
P.S. I want a cut if it makes any money 😉
This is inductive reactance (assuming you run around later and push it the other way). The pressure and velocity are out of phase, yet connected intimately.
P.S. I want a cut if it makes any money 😉
Thanks dave, I'm beginning to understand. Appreciate your replies, and others' also.
The requirement of low Z amp with non-sagging power supplies - and then the 1K series resistor loses me again. I fully understand what the 1K does. However it seems I could put that in series with the 50 Ohm output of my function generator, end up with 1.050K - and be in a barely different measuring situation than the low Z non sagging supplies power amp with 1K in series.
It's beyond me why the HiZ method of measuring T/S doesnt just use the 50 Ohm output of nearly all standard function generators as the standard, but - oh well. It is what it is.
I'm happy to see some coherence in understanding (a component of...) what tube amplifiers do, regarding having "a sound", which is toward OPs initial post.
Interesting to read how the well regarded ACAs have a similar out Z to a SET amplifier. Interesting to know you could measure T/S using the source Z of the intended amplifier, or family of amplifiers - and be successful using parameters derived that way in the design of a speaker enclosure. Interesting to see that actually reflected across enclosure design examples.
I'm glad to read amplifiers with the old "damping factor" control do exist and can be designed using today's chips - I'm all for one more knob to turn - with an understanding of what that does. I would think that'd be a "standard feature" of all modern designs - an output impedance potentiometer control ranging from 0.001 to 16, linear from 2 - 16 Ohms. For those that dont understand, the OEM could just put a cap on it "as adjusted from the factory".
Time for me to put a cap on it and go test my router diameter setting for a 16.625" cutout.
The requirement of low Z amp with non-sagging power supplies - and then the 1K series resistor loses me again. I fully understand what the 1K does. However it seems I could put that in series with the 50 Ohm output of my function generator, end up with 1.050K - and be in a barely different measuring situation than the low Z non sagging supplies power amp with 1K in series.
It's beyond me why the HiZ method of measuring T/S doesnt just use the 50 Ohm output of nearly all standard function generators as the standard, but - oh well. It is what it is.
I'm happy to see some coherence in understanding (a component of...) what tube amplifiers do, regarding having "a sound", which is toward OPs initial post.
Interesting to read how the well regarded ACAs have a similar out Z to a SET amplifier. Interesting to know you could measure T/S using the source Z of the intended amplifier, or family of amplifiers - and be successful using parameters derived that way in the design of a speaker enclosure. Interesting to see that actually reflected across enclosure design examples.
I'm glad to read amplifiers with the old "damping factor" control do exist and can be designed using today's chips - I'm all for one more knob to turn - with an understanding of what that does. I would think that'd be a "standard feature" of all modern designs - an output impedance potentiometer control ranging from 0.001 to 16, linear from 2 - 16 Ohms. For those that dont understand, the OEM could just put a cap on it "as adjusted from the factory".
Time for me to put a cap on it and go test my router diameter setting for a 16.625" cutout.
Last edited:
Scott, you know who you are dealing with. The few that do this online will never acknowledge, admit or receptive to it because of their commitment to the commercial interest.
some intresting articles from Audio Review.
Pilotaggio in corrente - AudioReview
Pilotaggio in corrente 2 - AudioReview
Pilotaggio in corrente altoparlanti
Pilotaggio in corrente altoparlanti
Pilotaggio in corrente - AudioReview
Pilotaggio in corrente 2 - AudioReview
Pilotaggio in corrente altoparlanti
Pilotaggio in corrente altoparlanti
NO you can't.
ESA explains it very clearly why in his book, and in detail if you care to ask him.
It was one of the main reasons I totally lost interest.
Carry on looking for unicorns.
George gives us a clear idea of how damping factors (ie sort of 1/Rout as measured at the amplifier is exponential in nature, and above a certain point, more is moot.
dave
It can be done. You just have to kill the resonant peak. Joe does it in Elsinore, A low Qms woofer in an aperiodic box can do it as well. And it has been done historically.
It only takes 1 example to show that your 1st staement is incorrect.
Unfortuneately in a world where over 99% of systems are speciafically designed with the assumption of a voltage amps suitable drivers that are currently available are rare.
dave
The simple way to do it comes with its own compromises, Daniel has been working on one that doesn’t but it requires an mpu to implement the FB. ie compromise of not simple.
dave
It only takes 99 999 examples out of 100 000 to prove statistically your one so called exception is fatally flawed rubbish.
I have no time for IFs or BUTs, if it doesn't work, it doesn't work, and I don't follow the bizarre logic of trying to prove it might, when I don't need to, with a perfectly adequate solution that waits for glue, wood cutting and some decent drivers.
I've done that, and I consider the rest to be BS.
Are you saying drivers cannot be made to work under 100hz with current drive or that existing drivers cannot work under 100hz?
The First Rule of Engineering:
"For any given engineering problem, there are at least 100 solutions, of which at least 3 solutions will actually work, but only if they are implemented properly"
"For any given engineering problem, there are at least 100 solutions, of which at least 3 solutions will actually work, but only if they are implemented properly"
Sorry that you take that attitude, I see the above as an attempt by YOU to shut down the conversation. Why? What exactly is it that you are trying to do? This is an OPEN discussion that anybody can join in. As for he, no, sadly he was never patient and not even kind. I am sad to hear the news, I wish no evil to anyone.
I take ALL constructive criticism as valuable. Alas, some just criticise based more on where their ego is and NOT on substance.
SUBSTANCE IS EVERYTHING!
.
- Status
- Not open for further replies.