Hi, Chris
So say those who haven' seen the book. I think it's also much telling that even the staff of an academic journal are unable to find any other justification for the status quo than the status quo itself.
If it is not allowed to say that technology A is much better than technology B, even when the evidence is provided, how can there ever be necessary progress? With such a mindset, I wonder if we had ever made it out of the caves. (This is nothing personal.)
There is a great wall of misinformation and wishful thinking regarding the concept of damping, 'damping factor', and 'controlling' the diaphragm by voltage; and I only cannot see how this deep-rooted barrier of misconceptions and consequent prejudices about current-drive could ever be passed or broken without some degree of frankness. I feel this message must not and cannot be watered down to be comfortable for all, to 'just another way to drive a speaker'. Doing so would be dishonesty.
There is also a note in the preface: "Existing faults are brought out at times even quite vigorously. However, the issues treated are universal, and nobody has reason to react to them personally although recognizing one's own misconceptions may sometimes produce inconvenience."
anatech said:
So say those who haven' seen the book. I think it's also much telling that even the staff of an academic journal are unable to find any other justification for the status quo than the status quo itself.
If it is not allowed to say that technology A is much better than technology B, even when the evidence is provided, how can there ever be necessary progress? With such a mindset, I wonder if we had ever made it out of the caves. (This is nothing personal.)
There is a great wall of misinformation and wishful thinking regarding the concept of damping, 'damping factor', and 'controlling' the diaphragm by voltage; and I only cannot see how this deep-rooted barrier of misconceptions and consequent prejudices about current-drive could ever be passed or broken without some degree of frankness. I feel this message must not and cannot be watered down to be comfortable for all, to 'just another way to drive a speaker'. Doing so would be dishonesty.
There is also a note in the preface: "Existing faults are brought out at times even quite vigorously. However, the issues treated are universal, and nobody has reason to react to them personally although recognizing one's own misconceptions may sometimes produce inconvenience."
"esp" site - variable amp impedance
Hi,
this new approach to the old ways are not so easy.
A curious article that looks at the current <-> voltage drive discussion from a slightly different point of view is found on the Rod Elliot site "esp".
(Haven't figured out how to add the relevant link but .. )
"esp" website -> main index -> Design Guidelines -> Variable Impedance Amplifiers (last article)
Curiously, on the next section below, in the index, is an article about Seris Vs Parallel Crossover Networks (again, last article) - rather apprpriate, IMO.
My F3 (First Watt) current gain amp uses 12dB Series Xovers at 150Hz with high efficiency 12" bass in Reflex 5.25 cu ft chamber and a high efficiency 8" full ranger in a 1 cu ft spherical chamber and I'm achieving quite definite improvements as I understand more of the techniques outlined in Esa'a book.
The change in the way you look at (the traditional approach to) amp/speakers is not so easy as I imagined it to be, unfortunately - getting set in my ways (= old!)
I'm a bit dissapointed in the reported refusal to publish the advertisement for your book, Esa, in what is said to be the "audio authority journal" because in the editor's view," industry" doesn't want it! Perhaps, the editor meant ....?
Hi,
this new approach to the old ways are not so easy.
A curious article that looks at the current <-> voltage drive discussion from a slightly different point of view is found on the Rod Elliot site "esp".
(Haven't figured out how to add the relevant link but .. )
"esp" website -> main index -> Design Guidelines -> Variable Impedance Amplifiers (last article)
Curiously, on the next section below, in the index, is an article about Seris Vs Parallel Crossover Networks (again, last article) - rather apprpriate, IMO.
My F3 (First Watt) current gain amp uses 12dB Series Xovers at 150Hz with high efficiency 12" bass in Reflex 5.25 cu ft chamber and a high efficiency 8" full ranger in a 1 cu ft spherical chamber and I'm achieving quite definite improvements as I understand more of the techniques outlined in Esa'a book.
The change in the way you look at (the traditional approach to) amp/speakers is not so easy as I imagined it to be, unfortunately - getting set in my ways (= old!)
I'm a bit dissapointed in the reported refusal to publish the advertisement for your book, Esa, in what is said to be the "audio authority journal" because in the editor's view," industry" doesn't want it! Perhaps, the editor meant ....?
I have earlier discussed one of Elliot's articles in post #10 ("Effects Of Source Impedance on Loudspeakers"). They seem to reflect quite typically the general imagery that the EMFs somehow become 'suppressed' by low source impedance and this EMF suppression then somehow magically keeps the cone in 'control', even up to high frequencies. In this article ("Variable Amplifier Impedance") he then concludes that the current drive system should never be used in practice because he claims it to be prone to radio interference.jameshillj said:
That's odd as during my many years of experience with the topology, I have never observed signs of such interference or heard this be a problem; nor is there any reason that a properly designed and compensated circuit would behave so.
While it may be interesting and educative to experiment with variable amplifier impedance, there is no reason to strive to equalize the power consumed by the speaker or to otherwise stay in any half-way solutions when we can as well pursue the whole solution.
Just to avoid any confusions: The Pass F3 seems to be quite a conventional amp in terms of output impedance. F1 was the model that comes closest to current drive with its 80 ohm impedance.
Is there perhaps something that needs more clarification?
Fascinating thread !
As i was making my way through it, i suddenly remembered something that really excited me when i first discovered it in a Hi-Fi magazine in the 1970'2, which you may find of interest !
Audio Pro ACE-Bass invented by Karl Erik Ståhl from Sweden.
Quote -
Negative resistance
The solution is to serially connect the amplifier output with a negative resistance. But you can't buy that at your nearest electronics parts shop. It can only be achieved by positive feedback in the amplifier's output stage.
There are several sensitive factors that can easily cause the system to self-oscillate. However, correctly trimmed, you can achieve a system that exactly equalizes all the resistances that are in series with the drive units' properties. In this way, you can also reach and affect the loudspeaker's mechanical attributes electrically.
You can also choose to let this influence decrease with increasing frequency, where it's no longer needed.
By using this feature fully, Audio Pro maintains that you can decrease the speaker's physical volume by 90%, compared to a conventional bass reflex box!
Audio Pro - Sound of Scandinavia
Here's his patent from 1978 Loudspeaker lower bass response using negative resistance and impedance loading - Patent 4118600
We also have a thread on here about it STAHL - 'Acebass' Synthesis of Loudspeaker Mechanical Parameters by Electrical Means http://www.diyaudio.com/forums/subw...r-mechanical-parameters-electrical-means.html
The Variable Amplifier Impedance Variable Amplifier Impedance is one i've seen before and is similar, but not the same as ACE-Bass. Non the less it is worthy of inclusion for discussion and examination in this area.
As i was making my way through it, i suddenly remembered something that really excited me when i first discovered it in a Hi-Fi magazine in the 1970'2, which you may find of interest !
Audio Pro ACE-Bass invented by Karl Erik Ståhl from Sweden.
Quote -
Negative resistance
The solution is to serially connect the amplifier output with a negative resistance. But you can't buy that at your nearest electronics parts shop. It can only be achieved by positive feedback in the amplifier's output stage.
There are several sensitive factors that can easily cause the system to self-oscillate. However, correctly trimmed, you can achieve a system that exactly equalizes all the resistances that are in series with the drive units' properties. In this way, you can also reach and affect the loudspeaker's mechanical attributes electrically.
You can also choose to let this influence decrease with increasing frequency, where it's no longer needed.
By using this feature fully, Audio Pro maintains that you can decrease the speaker's physical volume by 90%, compared to a conventional bass reflex box!
Audio Pro - Sound of Scandinavia
Here's his patent from 1978 Loudspeaker lower bass response using negative resistance and impedance loading - Patent 4118600
We also have a thread on here about it STAHL - 'Acebass' Synthesis of Loudspeaker Mechanical Parameters by Electrical Means http://www.diyaudio.com/forums/subw...r-mechanical-parameters-electrical-means.html
The Variable Amplifier Impedance Variable Amplifier Impedance is one i've seen before and is similar, but not the same as ACE-Bass. Non the less it is worthy of inclusion for discussion and examination in this area.
Ah Esa,
(I've been looking for "clarification" for years now!!)
No, it's just a lot of that accumulated prejudice and misconceptions that resist the acceptance of new ideas - takes a bit of time to assimilate - will perservere. (Reminds me of my initial reaction to some of Ed de Lima's rather contraversial ideas about amp/speaker integration some years ago [AudioPax])
Yes, the F1 is pretty good (marvellous sound, IMO) but prefer the dynamics of the F3 and the balance of the speakers, particularly the control and "speed" of the 12" bass driver - rather surprising that.
(I've been looking for "clarification" for years now!!)
No, it's just a lot of that accumulated prejudice and misconceptions that resist the acceptance of new ideas - takes a bit of time to assimilate - will perservere. (Reminds me of my initial reaction to some of Ed de Lima's rather contraversial ideas about amp/speaker integration some years ago [AudioPax])
Yes, the F1 is pretty good (marvellous sound, IMO) but prefer the dynamics of the F3 and the balance of the speakers, particularly the control and "speed" of the 12" bass driver - rather surprising that.
Hi,
This is quite an old hat. The earliest applications (if deliberatly or not) where in the late 1930's german studio monitoring systems using Eckmiller coaxial drivers. They used Pentode driven Amplifiers without NFB (and hence effectively current drive) and speakers designed to damp mechanical resonances mechanically, rather than through the amplifiers. These systems where evaluated quite critically directly against complex real sounds and did surprisingly well.
If one considers how much of modern audiophile concerns these systems addressed well it is almost unbelievable that not much else came out of it.
As long as Amplifier designers design amplifiers to drive resistors and speaker designers find it cheaper to damp mechanical resonances electrically and ignore distortions nothing will change.
Sadly there is a heavy real and "moral" investment into the traditional technology and it is not a small change to move to drive-units that self-damp mechanical resonances. So just like many other technologies where the worst but cheapest option has won and carried the day so voltage driven speakers will stay.
We will get the usual two stupid twin brothers of "what will we do tomorrow" answered by "what we did yesterday" as a result (as seen in the JAES part of this affair) together their ugly & vicious sister "get them before they get us", as a result of this muddy bog of sullen inertia of so-called common sense.
As for the usual nay sayers and pseudo-skeptics who will rubbish everything they fail to immediatly understand, the problem actually boils down to a few very simple and self evident facts coverd in really basic electrical theory.
1) The force applied to the diaphragm of the speaker is proportional to the current in the voice coil.
2) The current in the voice coil is proportional to the voltage applied and the impedance of the driver.
3) The impedance of a coil that is wound around a solid steel core is severely non-linear (cubic function), promoting mainly odd order harmonics with rising current and frequency. That is actually why mains transformers are made from thin laminations, not with much easier to make solid steel cores and why high frequency coils use ferrite cores and not steel cores, not even laminated ones.
4) With the impedance to which the (linear) voltage is applied being non-linear, the current must be nonlinear, hence the force moving the diaphragm is nonlinear (distorted).
So, the whole issue is quite simple, easy and has been discussed before.
Reducing this distortion component can be attained in many ways.
If we increase driver efficiency we reduce the current levels and due to cubic relationship between current and distortion doubling efficiency quarters distortion. So moving from a (say) 87dB/W/m system to a 97dB system will approximately reduce this distortion component tenfold. Also distortion from several other sources will be reduced in this case by reducing the diaphragm travel at a given SPL and frequency.
The downside? We need to give up LF extension or make a very large speaker.
Or we can use drivers that reduce the inductive component of the voice coil impedance (copper sleeve or similar), however this tends to reduce efficiency and hence is not that great a choice, but commonly found in high performance drivers as it is easily applied in current technologies.
Current drive is another possible tool. I rather like it as it also cancels power compression as well. Yet this would require a driver that is build with a Qm = desired Qt and with a symetrical and linear damping system. Again, such a system is not particulary difficult, but requires a chang in thinking. Last time I discussed the needed design features with a designer of drivers I drew blank looks, as saig guy never thought the subject through.
I used current drive for midrange and treble in a experimental speaker in the 80's, which also used a type of motional feedback for the Bass. But such a system needs to be fully active and is hard to design, even in this day and age.
With most drivers having very low mechanical losses (high Qm >> 2 once inside a cabinet) and significant sample variations, using current drive and flattening the result using EQ (e.g. Linkwitz Transform) is not an option I'd like to try in series production, though it may be doable in a single DIY design.
In such a case current drive would almost mandatory require also motional feedback and that opens another can of worms. Tannoy used such a system in a commercial (and commercially very unsuccessful) subwoofer a while back.
Seas used to (may still have) drivers with a low Qm (< 2) and generally linear mechanical systems in their range, these may be a good choice for the experimenter, as a Qt of (say) 2 can be handled okay using standard equalisation techniques and current drive is trivial to implement using the all popular chip amps.
I think there was a discussion of this before here, where my evil twin suggested a more specific experimental setup using Seas drivers.
Ciao T
PS, before implementing Motional feedback using an electret microphone capsule in the 80's I tried "negative impedance" which really just reduces to motional feedback using the back emf as position signal (the fallacy of doing this should be painfully obvious), plus it was very instable due to the thermal changes of voicecoil resistance.
For now I have given up on all these schemes, I prefer using other methodes.
This is quite an old hat. The earliest applications (if deliberatly or not) where in the late 1930's german studio monitoring systems using Eckmiller coaxial drivers. They used Pentode driven Amplifiers without NFB (and hence effectively current drive) and speakers designed to damp mechanical resonances mechanically, rather than through the amplifiers. These systems where evaluated quite critically directly against complex real sounds and did surprisingly well.
If one considers how much of modern audiophile concerns these systems addressed well it is almost unbelievable that not much else came out of it.
As long as Amplifier designers design amplifiers to drive resistors and speaker designers find it cheaper to damp mechanical resonances electrically and ignore distortions nothing will change.
Sadly there is a heavy real and "moral" investment into the traditional technology and it is not a small change to move to drive-units that self-damp mechanical resonances. So just like many other technologies where the worst but cheapest option has won and carried the day so voltage driven speakers will stay.
We will get the usual two stupid twin brothers of "what will we do tomorrow" answered by "what we did yesterday" as a result (as seen in the JAES part of this affair) together their ugly & vicious sister "get them before they get us", as a result of this muddy bog of sullen inertia of so-called common sense.
As for the usual nay sayers and pseudo-skeptics who will rubbish everything they fail to immediatly understand, the problem actually boils down to a few very simple and self evident facts coverd in really basic electrical theory.
1) The force applied to the diaphragm of the speaker is proportional to the current in the voice coil.
2) The current in the voice coil is proportional to the voltage applied and the impedance of the driver.
3) The impedance of a coil that is wound around a solid steel core is severely non-linear (cubic function), promoting mainly odd order harmonics with rising current and frequency. That is actually why mains transformers are made from thin laminations, not with much easier to make solid steel cores and why high frequency coils use ferrite cores and not steel cores, not even laminated ones.
4) With the impedance to which the (linear) voltage is applied being non-linear, the current must be nonlinear, hence the force moving the diaphragm is nonlinear (distorted).
So, the whole issue is quite simple, easy and has been discussed before.
Reducing this distortion component can be attained in many ways.
If we increase driver efficiency we reduce the current levels and due to cubic relationship between current and distortion doubling efficiency quarters distortion. So moving from a (say) 87dB/W/m system to a 97dB system will approximately reduce this distortion component tenfold. Also distortion from several other sources will be reduced in this case by reducing the diaphragm travel at a given SPL and frequency.
The downside? We need to give up LF extension or make a very large speaker.
Or we can use drivers that reduce the inductive component of the voice coil impedance (copper sleeve or similar), however this tends to reduce efficiency and hence is not that great a choice, but commonly found in high performance drivers as it is easily applied in current technologies.
Current drive is another possible tool. I rather like it as it also cancels power compression as well. Yet this would require a driver that is build with a Qm = desired Qt and with a symetrical and linear damping system. Again, such a system is not particulary difficult, but requires a chang in thinking. Last time I discussed the needed design features with a designer of drivers I drew blank looks, as saig guy never thought the subject through.
I used current drive for midrange and treble in a experimental speaker in the 80's, which also used a type of motional feedback for the Bass. But such a system needs to be fully active and is hard to design, even in this day and age.
With most drivers having very low mechanical losses (high Qm >> 2 once inside a cabinet) and significant sample variations, using current drive and flattening the result using EQ (e.g. Linkwitz Transform) is not an option I'd like to try in series production, though it may be doable in a single DIY design.
In such a case current drive would almost mandatory require also motional feedback and that opens another can of worms. Tannoy used such a system in a commercial (and commercially very unsuccessful) subwoofer a while back.
Seas used to (may still have) drivers with a low Qm (< 2) and generally linear mechanical systems in their range, these may be a good choice for the experimenter, as a Qt of (say) 2 can be handled okay using standard equalisation techniques and current drive is trivial to implement using the all popular chip amps.
I think there was a discussion of this before here, where my evil twin suggested a more specific experimental setup using Seas drivers.
Ciao T
PS, before implementing Motional feedback using an electret microphone capsule in the 80's I tried "negative impedance" which really just reduces to motional feedback using the back emf as position signal (the fallacy of doing this should be painfully obvious), plus it was very instable due to the thermal changes of voicecoil resistance.
For now I have given up on all these schemes, I prefer using other methodes.
Last edited:
Let me rewrite this ad for you...
An optimal operation method reducing distortion and interference effects
As they say, "the medium is of the message" and your medium is seriously flawed. I, for one, would have been scared away from your book by that ad.
Current driving of loudspeakers:
An optimal operation method reducing distortion and interference effects
Why has a whole field of loudspeakers operations been left unexplored ? A frank exposure of the limits of voltage drive and a thorough, innovative guide to current drive, a potentially better way to feed electrodynamic loudspeakers.
As they say, "the medium is of the message" and your medium is seriously flawed. I, for one, would have been scared away from your book by that ad.
It is the second sentance that I do not understand.
I would appreciate if you try to make it easier for me (and harder for you
In a current drive system, power delivery is proportional to impedance. So it is possible to have a system where what are variations of FR (when viewed from voltage drive perspective) can be compensated for by complementary changes in impedance response (under current, or partial current drive)
The practical instances we see this in today are some single driver systems, that under voltage drive tend to have FR that droops at the top and bottom. The natural tendencies of a single driver impedance curve can be used to flatten the frequency response of the speaker system by increasing the output impedance of the amplifier -- one of the reason one often finds a no global feedback single ended amplifier driving FR speakers sounding so good.
dave
Here is an ongoing thread on a variable trans-impedance amplifier (aka TransAmp) which may be of interest to anyone wishing to explore the effects of output impedance.
Also let me point out that the Mark Audio Alpair 12 has a specified Qm of 1.45, which makes it usefully close to a low enuff Qm to thrive under current drive. We are currently in the midst of experiments to see how low we can get the actual system Q using various time tested tricks.
dave
Also let me point out that the Mark Audio Alpair 12 has a specified Qm of 1.45, which makes it usefully close to a low enuff Qm to thrive under current drive. We are currently in the midst of experiments to see how low we can get the actual system Q using various time tested tricks.
dave
I'm not saying that the concept in itself is very new. Rather, it is the knowledge of the serious flaws of voltage drive and the practical applications of current-drive that have been missing. However, I have to agree with you in most points.ThorstenL said:
Harmonic distortion due to the steel core is surely one of the issues that happen, but it is not the only one and not even the worst. If it were, I likely wouldn't have even bothered to set about this endeavor. There is also a host of other effects that I have summarized here.ThorstenL said:
ThorstenL said:
Unfortunately, increasing driver efficiency usually increases the nonlinear inductance too, so this aid may not materialize.
That's a bit pity as there is no real substitute for current drive, except perhaps with panels, which then have their own issues.ThorstenL said:
There is also a saying "Do not shoot the messenger". I can't help what the laws of nature in these issues are. Blame me only if I say something that does't hold true.00940 said:
Hi,
In traditional "single amplifier & passive multiway speaker" setups that is indeed so. But why limit ourselves so much? In the 80's when I experimented with this stuff myself using a fully active speaker it was quite possible to make comparisons that one can consider completely fair. At least for the Midrange and high frequency drivers, where current drive arguably has the greatest impact.
Forgive me for inquiring, but is this assertion the result of actually experimenting (and listening/measuring)? That is, is your view that speakers "will certainly sound ... worse when driven by a current source" based on actual experience? And if so, can you describe why reduced high order harmonics and reduced thermal compression in your experience sound worse?
Ciao T
In traditional "single amplifier & passive multiway speaker" setups that is indeed so. But why limit ourselves so much? In the 80's when I experimented with this stuff myself using a fully active speaker it was quite possible to make comparisons that one can consider completely fair. At least for the Midrange and high frequency drivers, where current drive arguably has the greatest impact.
Forgive me for inquiring, but is this assertion the result of actually experimenting (and listening/measuring)? That is, is your view that speakers "will certainly sound ... worse when driven by a current source" based on actual experience? And if so, can you describe why reduced high order harmonics and reduced thermal compression in your experience sound worse?
Ciao T
Hi ETM,
I think you missed the message that was intended. Perhaps because you are so use to defending yourself so often you can't see non-partial observers?
To make an effective case, you need to include others in your work, rather than challenge an entrenched industry. So why not go over what you see as the existing problems with your approach to driving a diaphragm (that differ from the established voltage drive method)? Going further on how to adjust a loudspeaker design to work with your ideas might be very helpful. An open discussion on the problems of voltage drive would also be helpful.
Dave, ETM,
I did read the materials that were referred to. It didn't add much to what I already know on this subject, because I am already familiar with it. The major shift in understanding what amplifiers have as limiting factors is pretty simple. Voltage amplifiers mostly have a current compliance issue, whereas a current amp will have a voltage compliance issue. Clipping a V amp involves running out of voltage swing, a C amp would run out of current as it's clipping mode - in a simplified view.
The end result of either way to drive a speaker is to create a linear displacement for a given signal level. The driver is a current operated device (never once disagreed here) mostly, so we only need some way to linearize the motion of the diaphragm w.r.t. the driving signal. Since the composite impedance of a loudspeaker is anything but flat, it would seem that the choice might be more along the lines of which loudspeaker characteristics are the easiest to live with, and accept that for now. That would determine how we go about applying a signal to the loudspeaker.
ETM,
You might find it interesting to study the Carver TFM-45 amplifier (or the solid state Silver 7) as these do sense current as well. They would represent a kind of hybrid approach to the problem. The control involved is labeled "damping" and has a large effect on amplifier gain as well. If nothing else, obtaining one of these amplifiers would allow you to experiment more freely with a combination of approaches. They are normally inexpensive to buy these days and allows you direct access to some current feedback.
I would be happy to maybe help you out in understanding how these amps work if needed. The added benefit is the amount of power available for experiments. They have a protection network that tends to limit damage to the amplifier, but at about 375 wrms per channel, the voice coil is often destroyed without any amplifier faults. That's if you can't contain the impulse to turn it up.
Best, Chris
I think you missed the message that was intended. Perhaps because you are so use to defending yourself so often you can't see non-partial observers?
I think what 00940 is trying to say was an expanded version of what I said earlier. I have to say that I really do agree with his point of view. The wording that you choose to use is combative. If you look again where your statements were reworded, you might be able to see that a similar message is sent, but the reworded statements invite those who you are trying to reach to learn more. The way you have talked so far is a challenge to everyone who doesn't completely agree with your point of view. That's no way to introduce a different idea, now is it?
Now this kind of comment is more palatable. You haven't closed any doors and are offering information. A response might have been to ask what the shortcomings are inherent in voltage drive (but not common to both methods).
This issue is common to both current drive and voltage drive. Therefore it is nulled out and shouldn't be addressed. You want to highlight the differences between both approaches.
This tends to sum up the situation that exists today. As mentioned before, early amplifiers were current source devices due to the high impedance they presented (or damping ability as normally defined, ie: a high output impedance). For whatever reasons, it was determined many years ago to use a voltage drive system in spite of the short comings. This was at a time when driver impedance was highly variable. There were no standards in use at the time, and when the first "normal" speaker loads were considered, they were 15 or 16 ohm. A 4 ohm load is a relatively recent change within the last 20 some-odd years. That means that early developers, and the entire industry, moved toward a voltage drive solution for a reason. I don't know what that reason is, but it's possible that newer technologies make it so that a current drive solution is easier to do well. I would suspect that these factors may be with the speaker drive unit itself, since the early amplifiers did tend to be current drive types anyway.
To make an effective case, you need to include others in your work, rather than challenge an entrenched industry. So why not go over what you see as the existing problems with your approach to driving a diaphragm (that differ from the established voltage drive method)? Going further on how to adjust a loudspeaker design to work with your ideas might be very helpful. An open discussion on the problems of voltage drive would also be helpful.
Dave, ETM,
I did read the materials that were referred to. It didn't add much to what I already know on this subject, because I am already familiar with it. The major shift in understanding what amplifiers have as limiting factors is pretty simple. Voltage amplifiers mostly have a current compliance issue, whereas a current amp will have a voltage compliance issue. Clipping a V amp involves running out of voltage swing, a C amp would run out of current as it's clipping mode - in a simplified view.
The end result of either way to drive a speaker is to create a linear displacement for a given signal level. The driver is a current operated device (never once disagreed here) mostly, so we only need some way to linearize the motion of the diaphragm w.r.t. the driving signal. Since the composite impedance of a loudspeaker is anything but flat, it would seem that the choice might be more along the lines of which loudspeaker characteristics are the easiest to live with, and accept that for now. That would determine how we go about applying a signal to the loudspeaker.
ETM,
You might find it interesting to study the Carver TFM-45 amplifier (or the solid state Silver 7) as these do sense current as well. They would represent a kind of hybrid approach to the problem. The control involved is labeled "damping" and has a large effect on amplifier gain as well. If nothing else, obtaining one of these amplifiers would allow you to experiment more freely with a combination of approaches. They are normally inexpensive to buy these days and allows you direct access to some current feedback.
I would be happy to maybe help you out in understanding how these amps work if needed. The added benefit is the amount of power available for experiments. They have a protection network that tends to limit damage to the amplifier, but at about 375 wrms per channel, the voice coil is often destroyed without any amplifier faults. That's if you can't contain the impulse to turn it up.
Best, Chris
Interesting topic, ETM. With limits imposed by conventional speaker technology, it might not be universally practical, but it IS an interesting alternative, and your arguments appear sound, to me. After all, we usually design motor drives with current. In fact, I used the 'modified Howland pump' topology, back in the late 60's for a motor drive amp. Thanks for the refreshing look at the problems and potential solutions to loudspeaker drive.
Hi Sy,
Quote:
Yes.
Let me be absolutely clear, you tested current on speakers under conditions where the frequency response changes around the resonance either do not matter or are corretcly compensated?
And you found the sonic and/or measured results inferior to voltage drive?
Oh come on now. You exactly wrote:
For clarity I omitted some of the noise from the quote.
First, in commercial production all Speaker drivers are designed for Voltage drive, so the omission of that part of the quote should need no explanation.
Second, I elected to render:
"sound quite different (and worse)"
as:
"sound ... worse"
Do you really claim that: "A speaker ... will certainly sound ... worse when driven by a current source" is an unfair rendering of the actual content of your original writing?
If so, where you actually NOT saying that current drive of a speaker (a common commercial one) will sound worse? As you clearly indicated the sound would be different, where you trying the sound would be better and took the wrong word?
Ciao T
Quote:
Yes.
Thorstenl said:
Let me be absolutely clear, you tested current on speakers under conditions where the frequency response changes around the resonance either do not matter or are corretcly compensated?
And you found the sonic and/or measured results inferior to voltage drive?
Thorstenl said:
Oh come on now. You exactly wrote:
For clarity I omitted some of the noise from the quote.
First, in commercial production all Speaker drivers are designed for Voltage drive, so the omission of that part of the quote should need no explanation.
Second, I elected to render:
"sound quite different (and worse)"
as:
"sound ... worse"
Do you really claim that: "A speaker ... will certainly sound ... worse when driven by a current source" is an unfair rendering of the actual content of your original writing?
If so, where you actually NOT saying that current drive of a speaker (a common commercial one) will sound worse? As you clearly indicated the sound would be different, where you trying the sound would be better and took the wrong word?
Ciao T
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.