Our audio world is currently heavily biased towards a voltage driven world (so it is hard to find stuff that fits into a not-voltage drive system), but voltage drive has some serious issues that current drive solves... the big ones being using the speaker's impedance as the reference resistance in the voltage to current conversion, and back EMF.
It is true that with a current amp the impedance curve of the loudspeaker is reflected in the FR curve, so one needs to have a flat curve (damping in the speaker and crossovers that aren't as evil as most -- or no XO at all), or one that is the inverse of the speaker's FR when driven by a voltage amp. With a loudspeaker suited for current drive one can achieve greater DDR.
The easy way out is to multiamp -- use a voltage amp in the bass where is is sometimes really hard to kill the system resonance, and current amps elsewhere where one can limit the bandwidth to the flat part of the impedance curve.
It can be really eye-opening to hear what happens when you play with the output impedance control of a variable tranimpedance amplifier. Every speaker seems to have an optimum setting. Particularily if you have speakers that are sympathetic with higher output impedance amplifiers.
oon, i encourage you to push forward, all too often the nay-sayers are blindered by the current lopsided prediliction for voltage drive... it may make it easier for the marketers to keep things simplier for the vast majority of customers who could care less about how it really sounds (steam-rolling some very important options for those that do) .
One has to always consider that the speaker & the amp have to be considered as a system. By making commerical speakera and commerical amps more or less suited to voltage source, any amp or any speaker can be mixed and (sorta) have something that is close to a system. Leaves out some very important options for those that want to achieve all that is possible.
A pivotal pointin my audio education was talking with an old Western Electric engineer who couldn't understand why anyone would build an amplifier that did not have output impedance of the amp = to the loudspeaker impedance ... makes one start to think.
dave
It is true that with a current amp the impedance curve of the loudspeaker is reflected in the FR curve, so one needs to have a flat curve (damping in the speaker and crossovers that aren't as evil as most -- or no XO at all), or one that is the inverse of the speaker's FR when driven by a voltage amp. With a loudspeaker suited for current drive one can achieve greater DDR.
The easy way out is to multiamp -- use a voltage amp in the bass where is is sometimes really hard to kill the system resonance, and current amps elsewhere where one can limit the bandwidth to the flat part of the impedance curve.
It can be really eye-opening to hear what happens when you play with the output impedance control of a variable tranimpedance amplifier. Every speaker seems to have an optimum setting. Particularily if you have speakers that are sympathetic with higher output impedance amplifiers.
oon, i encourage you to push forward, all too often the nay-sayers are blindered by the current lopsided prediliction for voltage drive... it may make it easier for the marketers to keep things simplier for the vast majority of customers who could care less about how it really sounds (steam-rolling some very important options for those that do) .
One has to always consider that the speaker & the amp have to be considered as a system. By making commerical speakera and commerical amps more or less suited to voltage source, any amp or any speaker can be mixed and (sorta) have something that is close to a system. Leaves out some very important options for those that want to achieve all that is possible.
A pivotal pointin my audio education was talking with an old Western Electric engineer who couldn't understand why anyone would build an amplifier that did not have output impedance of the amp = to the loudspeaker impedance ... makes one start to think.
dave
OK, now I'm confused. I thought you were already building an F2J. Where does that fit into the picture?
Anyway to answer your question: none of the above. I'd go for a design with variable output impedance. How can you resist a knob on the front of the amp that let's you change the output impedance while it's playing? Wonderful for experimentation. Anything else ties you down unnecessarily, IMO.
Dave, wasn't there a thread somewhere describing an amp of this kind?
P.S. Here's a nice paper by Malcolm Hawksford about current drive: DISTORTION REDUCTION IN MOVING-COIL LOUDSPEAKER SYSTEMS USING CURRENT-DRIVE TECHNOLOGY
Lots more of Hawksford's published work can be found here: Malcolm Hawksford - Publications
There are a few of them. One of them is specifically about the amp(s) the boy grnius next door built.
The best one so far (that i've heard) started out life as a driver amplifier for the electron gun in a CRT.
This is a good one too. http://www.diyaudio.com/forums/vend...vealed-much-more.html?highlight=current+drive
I recommend the book.
dave
Hi Godfrey,
The f2j is unfortunately my castle in the air at the moment. The other factor is my woofers are not 100db/w drivers, a lot less than that, so I am looking for a current source chipamp to complement it... now I realise, that prob current source is probably not the best thing if a majority of the bandwidth is around fs...
Oon
Oon
Sent from my MB860 using Tapatalk
The f2j is unfortunately my castle in the air at the moment. The other factor is my woofers are not 100db/w drivers, a lot less than that, so I am looking for a current source chipamp to complement it... now I realise, that prob current source is probably not the best thing if a majority of the bandwidth is around fs...
Oon
Oon
Sent from my MB860 using Tapatalk
Hi Dave,
Thanks for the lengthy reply. I am inclined to take your advice. Current amp for the full range and voltage amp for the bass. I am very inclined to in the direction that Klaus suggested too. A resistor and a coil in series to tune the output impedance. I prob should have posted this somewhere else, if I guessed where this thread is going. I thought that using a chipamp is an easy way of getting a current source. But the technique that Klaus described is very intriguing because I could just buy a preowned AV amp with 7 channels of 100w to do it.
In fact I was thinking about using the coil as a bass booster... simply treat it as a low pass filter at somewhere near fs, and I could achieve the 6 db / octave boost...
Oon
Sent from my MB860 using Tapatalk
Sent from my MB860 using Tapatalk
Thanks for the lengthy reply. I am inclined to take your advice. Current amp for the full range and voltage amp for the bass. I am very inclined to in the direction that Klaus suggested too. A resistor and a coil in series to tune the output impedance. I prob should have posted this somewhere else, if I guessed where this thread is going. I thought that using a chipamp is an easy way of getting a current source. But the technique that Klaus described is very intriguing because I could just buy a preowned AV amp with 7 channels of 100w to do it.
In fact I was thinking about using the coil as a bass booster... simply treat it as a low pass filter at somewhere near fs, and I could achieve the 6 db / octave boost...
Oon
Sent from my MB860 using Tapatalk
Sent from my MB860 using Tapatalk
Please see the thread; A headphone Transconductance amp for a change! so as to find a true current amp.
- Status
- Not open for further replies.