forr,
<<
With a high impedance source, the Qe of the driver disappears and the unit resonance is only damped by its Qm which Qt equals now.
High impedance idea may be a valid idea far above the main resonance. Its advantages as well as the restoration of the damping of the main resonance by motion feedback has been dealt by Hawksford :
http://www.essex.ac.uk/ese/research...ve patent.pdf
>>
The patent you reference is interesting. It mentions another approach that is endorsed by Colloms too, negative resistance amp. This is another way that I find intriguing, but do not understand all the ramifications of.
<<
With a high impedance source, the Qe of the driver disappears and the unit resonance is only damped by its Qm which Qt equals now.
High impedance idea may be a valid idea far above the main resonance. Its advantages as well as the restoration of the damping of the main resonance by motion feedback has been dealt by Hawksford :
http://www.essex.ac.uk/ese/research...ve patent.pdf
>>
The patent you reference is interesting. It mentions another approach that is endorsed by Colloms too, negative resistance amp. This is another way that I find intriguing, but do not understand all the ramifications of.
Zenmasterbrian,
My main source of information was first the AES anthologies I bought.
I only know two articles dealing with current driven loudpseakers : by Hawskford and by Greiner & Travis, as already mentionned.
Negative resistance amplifiers driving loudspeakers is a much more covered subject. It's very close to velocity motion feeback (MFB) and seems to have been thought of as soon as in 1926 by Paul Voigt.
The negative resistance technique applied to loudspeakers can be considered to be the contrary of current driving them : the last one gives an infinite value to Qe; the first one, ultimately, nullifies it.
However, the negative resistance cannot be left alone if its absolute value approaches the value of the DC resistance of the voice coil. This is where Erik Stahl had a very clever idea : to make the driving amplifier having a particular impedance to change the apparent mechanic parameters by electrical means. There is a patent and an AES paper.
My main source of information was first the AES anthologies I bought.
I only know two articles dealing with current driven loudpseakers : by Hawskford and by Greiner & Travis, as already mentionned.
Negative resistance amplifiers driving loudspeakers is a much more covered subject. It's very close to velocity motion feeback (MFB) and seems to have been thought of as soon as in 1926 by Paul Voigt.
The negative resistance technique applied to loudspeakers can be considered to be the contrary of current driving them : the last one gives an infinite value to Qe; the first one, ultimately, nullifies it.
However, the negative resistance cannot be left alone if its absolute value approaches the value of the DC resistance of the voice coil. This is where Erik Stahl had a very clever idea : to make the driving amplifier having a particular impedance to change the apparent mechanic parameters by electrical means. There is a patent and an AES paper.
I think all these options are worth exploring. All are simple in concept. None of them are simple when all the design factors and trade offs have to be considered.
These AES anthologies sound real good.
forr, if you don't mind my asking, are you involved professionally in something that draws upon such expertise?
These AES anthologies sound real good.
forr, if you don't mind my asking, are you involved professionally in something that draws upon such expertise?
I have the Martin Colloms book in front of me. His discussion of current drive starts on page 264.
He references something I don't know any easy way to get.
Mills and Hawksford, 'Amplifiers for current driven loudspeakers', J. Audio Engng Soc., 37, No. 10 (1989)
It looks like to make this work you need also a second voice coil to use velocity feed back. This velocity feed back is used to obtain the good damping that you would get with a low impedance voltage source amp.
So there are two feed backs. One is via the small resistor in the speaker return leg. This is what makes it a transconductance amp. Colloms says this is difficult to realize.
Then the second is via this second voice coil, a velocity feedback.
He says that the distortion performance of even a voltage source amp can be improved this way.
I would like to see articles that discuss this, and how to do it.
I am not aware of drivers made for this. In my mind, this is all more of a subwoofer issue.
The total system performance, considering the driver, the accuracy of this velocity feedback, and the electronics would have to be considered.
I would like to see books, paper, and examples.
I'd even like to be able to do it in Class D.
He references something I don't know any easy way to get.
Mills and Hawksford, 'Amplifiers for current driven loudspeakers', J. Audio Engng Soc., 37, No. 10 (1989)
It looks like to make this work you need also a second voice coil to use velocity feed back. This velocity feed back is used to obtain the good damping that you would get with a low impedance voltage source amp.
So there are two feed backs. One is via the small resistor in the speaker return leg. This is what makes it a transconductance amp. Colloms says this is difficult to realize.
Then the second is via this second voice coil, a velocity feedback.
He says that the distortion performance of even a voltage source amp can be improved this way.
I would like to see articles that discuss this, and how to do it.
I am not aware of drivers made for this. In my mind, this is all more of a subwoofer issue.
The total system performance, considering the driver, the accuracy of this velocity feedback, and the electronics would have to be considered.
I would like to see books, paper, and examples.
I'd even like to be able to do it in Class D.
The article should be in the folder with Malcom's publications, at least it was. If not, drop me a line, I downloaded it from there long ago.
You fill find the math there. I am not convinced that a second, collinear VC is the way to go. Also, I think he has gone overboard with his amplifier topology.
I am the first to agree that converting a voltage feedback amp to a VCCS (voltage controlled current source) has its drawbacks, but have not yet seen a sensible topology that does a better job.
You fill find the math there. I am not convinced that a second, collinear VC is the way to go. Also, I think he has gone overboard with his amplifier topology.
I am the first to agree that converting a voltage feedback amp to a VCCS (voltage controlled current source) has its drawbacks, but have not yet seen a sensible topology that does a better job.
http://www.ultrasound-hifi.com/Us_wh_1/main_frame_en.html
These threads have moved so fast, and I lose these links if I don't repost them. I need to spend real time with them.
These threads have moved so fast, and I lose these links if I don't repost them. I need to spend real time with them.
http://sound.westhost.com/project56.htm
Want to continue this thread, starting with the above.
Cross posted at Audio Explorations:
http://groups.google.com/group/audioex_amps_atob?lnk=li&hl=en&
Want to continue this thread, starting with the above.
Cross posted at Audio Explorations:
http://groups.google.com/group/audioex_amps_atob?lnk=li&hl=en&
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.