Until such time as there are solid scientifically valid double blind listening tests that establish a correlation between the measured data and what people hear, nothing conclusive can be established with any of those tests.
They can thou be very useful in the hands of a designer that knows how to use them to ferret out issues with the system.
If Geddes' research is considered the THD are not much use.
dave
The measurements are multi tone and H2/3/4/5 Vs freq. Why ref to THD?
What do you suggest manufacturers use to accurately gauge peformance
of their drivers?
I believe 3rd chart down had THD
I suggest they use the metric that Toole & Olive came up with. It is the only metric that has had any correlation to what people hear. It does thou still yet need to be independently duplicated.
dave
Further:
When evaluating any measures (or listening tests for that matter) how do you set a reference... what you are really comparing is amp + speakers in each case.
A test that i feel could be useful (again no correlation exists) is a ring down test. You use a shaped 5 cycle burst (or similar, the one i am familiar with was developed by PEARL, but Linkwitz uses similar). You then look at the waterfall (time axis in periods), rotate it until you are looking into the time axis. The chart with the most even spacing between curves should have an edge. It should go down at least 40 dB.
dave
When evaluating any measures (or listening tests for that matter) how do you set a reference... what you are really comparing is amp + speakers in each case.
A test that i feel could be useful (again no correlation exists) is a ring down test. You use a shaped 5 cycle burst (or similar, the one i am familiar with was developed by PEARL, but Linkwitz uses similar). You then look at the waterfall (time axis in periods), rotate it until you are looking into the time axis. The chart with the most even spacing between curves should have an edge. It should go down at least 40 dB.
dave
I have to withdraw the last sentence above, as it has appeared that nothing can be really discontinued from Amazon.com. (They never cease taking in orders.) So, the older version is now made available at the same discounted price as before.ETM said:
http://www.amazon.com/Current-Driving-Loudspeakers-Fundamental-Reproduction-Technology/dp/1448695325/ref=sr_1_1?ie=UTF8&s=books&qid=1254919079&sr=1-1
Perhaps this lecture, that I quickly found, could help you to get some grip on the motional EMF.anatech said:
You can also try the following experiment: Take a speaker driver and attach an AC voltage meter at the terminals. Flick the cone with fingers, and you will notice that voltage is developed across the terminals though virtually no current flows through the meter and coil. Next, attach different resistors, greater than 100 ohms or so, across the terminals, and flick similarly. You will notice the voltage is quite the same as before and independent of the resistor value and hence independent of the current taken. You may also measure the current taken and discover it is quite inversely proportional to the resistance (including meter resistance). So, you shouldn't deny any more that in a current-driven system, where the amplifier presents high impedance, the EMF is not producing any current.
Have you read the first paragraph of that page: "The following is just a brief summary of the disastrous effects that voltage drive has on the sound quality of all electro-dynamic loudspeakers. A detailed analysis of the subject with measurements and illustrations is rendered in chapter 4 of the book."anatech said:
I have demonstrated how to measure the magnitude of the microphone effect and shown some easily reproducable results which indicate clearly that the microphone EMF due to cabinet noise forms a considerable part of the terminal voltage at the lower mid-frequencies. If you are saying these results are invalid, you'd better have some good proof to back up your claim.anatech said:
Damping of the transducer is synonymous with lowering the effective Q value. When the Q is settled to the desired value, by whatever means, the speaker is damped, and there is nothing further that can or needs to be done about it. We can, for example, manufacture a driver whose mechanical Q value is somewhere close to 1, and then we may further apply heavy cabinet stuffing by suitable material to lower the Q to the desired value, e.g. 0.7 or so. When this is done, the speaker behaves on current-drive exactly as a voltage driven speaker of the same Q value and resonant frequency because their frequency responses and hence their transient properties are the same. The damping can also be effected by active equalization with the same end result.anatech said:
As I have emphasized, for any given frequency response (amplitude & phase) there exists one and only one transient response (for any given input), and this holds also in practice, assuming driver nonlinearities are not yet coming to play.
Assuming the cone is reasonably rigid and the driver properly functional, there is no such thing as "uncontrolled movement". The driver's operation is, in the closed box case, completely defined by the second-order transfer function it follows; and every movement of the cone, at whatever transient signal, is fully predictable and can be simulated by the principles explained in the book. The moving mass, spring constant, damping constant, and force factor are all that are needed to exactly determine the acceleration, velocity, or position of the diaphragm at any instant.
So, when the Q value and the frequency response are fixed, there is nothing that can be done to the transient properties unless we chance the frequency response (amplitude and/or phase). It works automatically; we don't need to worry about any energy issues.
It is important to realize that damping and the Q value have effect only near the resonant frequency. Instead, at all other frequencies, from about 200 Hz up, any damping doesn't have any effect at all.
According to the popular notion, the cone position can be "controlled" by the voltage applied to the driver terminals, and this cone position then would translate into sound pressure. Both of these imaginations are false. There is not any electromechanical relationship in the driver's operation that would make the position of the cone follow the applied voltage at the usable frequencies of the driver. In fact, in the normal operation region, the cone's position is about in opposite phase with the applied signal, as follows directly from the properties of second-order systems. Secondly, the sound pressure generated by a driver does not follow the cone's position but its acceleration. This is also proved quite easily by linear system properties. In fact, if cone position really translated into sound pressure, then sound pressure would also roll off inversely proportional to the square of frequency, as cone position does.
Thus, as things are seen in the light of reality, rather than the popular imagery supported by the marketing departments, there is not any need to control the position to begin with; all we need to do is to provide the radiating surface with a pure acceleration, which in the mass controlled region is governed by the Newtonian law F=ma.
Chris, please be concise to keep this thread readable.
Hi Esa,
There is a need to bound any comments so they are not taken out of context. I believe that my posts are pretty readable, not that I enjoy a great deal of typing.
I'll give your test a try, but since the source is a low impedance generator, I wouldn't expect a 100 ohm resistor to make much of a difference to the voltage output. When testing drivers to determine the T/S parameters, I often had to perform the testing at night with the warehouse heaters turned off. Any loud noise generated enough output from the speaker to make measurements impossible. Yes, we are well aware that dynamic loudspeakers operate nicely as a microphone. No disagreement there.
What I do know is that a meter does require a small amount of current to be drawn form a circuit to operate. I also know that the higher the amount of current drawn from a coil moving across a magnetic field, the more force will be required to maintain the same velocity. The same holds true in reverse. The higher the current that flows through a coil located in a magnetic field, the higher the acceleration of that coil will be. If you place a voltage on that coil, it will not move until current actually flows through the coil. Static repulsion will not get you where you want to go.
The of magnetic flux around a coil is dependent on two things. The amount of current flowing through the conductors, and the number of turns the conductor that can re-enforce the magnetic flux linking together. This is all current stuff, not voltage stuff. I see the voltage as a side effect of the impedance in your speaker, connections and crossover. This is actually playing exactly in to what you are trying to prove to us.
A dynamic loudspeaker is a current operated device, I am agreeing with you here. From your perspective, you are promoting that the current in the driver is controlled and allow the voltage across the components be the uncontrolled variable. I'm fine with that. What we are currently doing is to apply a voltage across a network of crossover and drivers, the resulting current is the uncontrolled variable. Since the speaker is a current operated device, the relationship between the control signal and the motion of the cone is not directly controlled. The best we can say is that the motion of the cone varies with the applied voltage, but not directly. By using a current drive system, we are directly controlling the forces acting on the voice coil, and hence the forces acting directly on the speaker cone.
Thought experiment for you Esa. Imagine you have just excited the speaker with a large displacement (high current in the voice coil), what happens if you instantly reduce the current you are delivering to the speaker to zero? A true current source has infinitely high impedance. You now have a solenoid type winding, or inductor, that has a great deal of energy stored in it's magnetic field. The nature of an inductor is to resist changes in current flow. What happens in this situation?
Before telling me the situation will not occur, imagine someone runs a test and feeds in a square wave that is unipolar to the input of your transconductance amplifier, no capacitive coupling. From where I am sitting, I see a sudden application of a high current, followed by a complete cessation of that current.
Otherwise, it's somewhat like buying software from an unknown company. If the product is not what it says it is, you're out the purchase price. Very Microsoft-ish if you ask me.
Driving a system with a current amplifier will radically change the system "Q" an also the frequency response. The transient response will be another thing again, I expect far more variability with this aspect of a speaker's attributes.
Acoustic power is determined by how much volume of air is displaced. Therefore, since the area of the cone is fixed, only displacement is the variable we can "control". To control displacement, we must control the position of the cone at any instant in time. If we don't, I imagine distortion will be the result.
Esa, can we concentrate on how the amplifier interacts with a speaker, and how the two systems differ? Presenting actual proof (or accessible references) to back up a statement at the time you introduce an idea will make this thread a great deal more readable.
Best, Chris
There is a need to bound any comments so they are not taken out of context. I believe that my posts are pretty readable, not that I enjoy a great deal of typing.
I do "have a grip" on how coils and magnetic fields interact, but thank you. It should be noted that I have finite time with which to play on the internet. I am currently reading the articles pointed out earlier as well. Once finished those, I'll have a better idea of your arguments with that material are. I will mention that I feel it is unwise to dismiss the work of earlier, accepted work without a clear explanation offered at the same time. So far you have not offered much in the way of information.
I'll give your test a try, but since the source is a low impedance generator, I wouldn't expect a 100 ohm resistor to make much of a difference to the voltage output. When testing drivers to determine the T/S parameters, I often had to perform the testing at night with the warehouse heaters turned off. Any loud noise generated enough output from the speaker to make measurements impossible. Yes, we are well aware that dynamic loudspeakers operate nicely as a microphone. No disagreement there.
That may depend a great deal on the situation. I'm personally not sure of this as most speaker systems use a passive crossover. There are plenty of parallel paths for current to move. The closest I have been to a current drive is the 1 K resistor in series with my oscillator that drives a speaker under test in my jig. I do not have access to a transconductance amplifier unless I build one. No time for that now.
Until I can perform an experiment to prove what is happening, I'm not about to blindly accept any statements. I'm not saying you are wrong, I am admitting that I don't know for sure, and that if I can see proof of this, I'll accept it.
What I do know is that a meter does require a small amount of current to be drawn form a circuit to operate. I also know that the higher the amount of current drawn from a coil moving across a magnetic field, the more force will be required to maintain the same velocity. The same holds true in reverse. The higher the current that flows through a coil located in a magnetic field, the higher the acceleration of that coil will be. If you place a voltage on that coil, it will not move until current actually flows through the coil. Static repulsion will not get you where you want to go.
The of magnetic flux around a coil is dependent on two things. The amount of current flowing through the conductors, and the number of turns the conductor that can re-enforce the magnetic flux linking together. This is all current stuff, not voltage stuff. I see the voltage as a side effect of the impedance in your speaker, connections and crossover. This is actually playing exactly in to what you are trying to prove to us.
A dynamic loudspeaker is a current operated device, I am agreeing with you here. From your perspective, you are promoting that the current in the driver is controlled and allow the voltage across the components be the uncontrolled variable. I'm fine with that. What we are currently doing is to apply a voltage across a network of crossover and drivers, the resulting current is the uncontrolled variable. Since the speaker is a current operated device, the relationship between the control signal and the motion of the cone is not directly controlled. The best we can say is that the motion of the cone varies with the applied voltage, but not directly. By using a current drive system, we are directly controlling the forces acting on the voice coil, and hence the forces acting directly on the speaker cone.
I have no complaints with that statement at all. However, if the amplifier must create a current to resist the motion of a voice coil, it then may appear to the voice coil in that instant to be a low impedance source. If this is the case, your current source appears to be a voltage source. I feel that the dynamic situation may not be well explained by thinking in static ways.
Thought experiment for you Esa. Imagine you have just excited the speaker with a large displacement (high current in the voice coil), what happens if you instantly reduce the current you are delivering to the speaker to zero? A true current source has infinitely high impedance. You now have a solenoid type winding, or inductor, that has a great deal of energy stored in it's magnetic field. The nature of an inductor is to resist changes in current flow. What happens in this situation?
Before telling me the situation will not occur, imagine someone runs a test and feeds in a square wave that is unipolar to the input of your transconductance amplifier, no capacitive coupling. From where I am sitting, I see a sudden application of a high current, followed by a complete cessation of that current.
Yes, I did read that. Do you know what I learned from this statement?
I learned that you are prone to exaggeration. Have you posted chapter 4 of your book? After all, you are referring to it as a source. It should be available so that anyone can immediately be able to evaluate your statement. If you'd rather not post chapter 4 on your site, then perhaps you could make your statements in the text on your web page.
Otherwise, it's somewhat like buying software from an unknown company. If the product is not what it says it is, you're out the purchase price. Very Microsoft-ish if you ask me.
Yup. That's what you've been saying .... without any proof.
Well, now you are talking about the characteristics of the driver, or load. This is true for both the current drive and voltage drive. Understand that a speaker in a box is a mass coupled to a restoring force through a compliant system. It has (predictably) a resonant frequency and will tend to move at that frequency whenever it is excited by any energy input. Any continued motion at that frequency is in fact - uncontrolled. You have to take action in order to arrest this motion.
I'm unsure about this. Your system "Q" is made up of an electrical component, and a mechanical component. There is energy in each component, and you are saying that disabling the electrical control, the mechanical motion is going to change and be properly damped some how? From designing more than a couple speaker systems, I know that the output resistance of the amplifier will possibly modify the "Q" of the system I designed. Speaker designs work (have similar response driven from different amplifiers) because most amplifiers have an output impedance similar in magnitude. In fact, many have an output impedance that is dominated by connection resistances, which works to provide a similar impedance from a wide variety of amplifiers.
Driving a system with a current amplifier will radically change the system "Q" an also the frequency response. The transient response will be another thing again, I expect far more variability with this aspect of a speaker's attributes.
I am not comfortable with this idea. I agree that damping and "Q" are most important around resonance, I'm not prepared to dismiss that these things are not affecting the speaker at other frequencies. I have no experience with this situation.
I'm pretty sure that it's accepted that a dynamic speaker responds to current through the voice coil. I've been very consistent in saying this. However, the cone of a speaker will move in response to an applied voltage. It's the exact value of current this creates in the voice coil that is uncertain.
Yep. Just like any other reactive component (like a capacitor or inductor). But you are describing the characteristics of a speaker, not of whatever is supplying the energy to the speaker.
Marketing has nothing to do with this concept. Marketing departments will not go anywhere near this idea unless they are selling a motional feedback system.
Acoustic power is determined by how much volume of air is displaced. Therefore, since the area of the cone is fixed, only displacement is the variable we can "control". To control displacement, we must control the position of the cone at any instant in time. If we don't, I imagine distortion will be the result.
Esa, can we concentrate on how the amplifier interacts with a speaker, and how the two systems differ? Presenting actual proof (or accessible references) to back up a statement at the time you introduce an idea will make this thread a great deal more readable.
Best, Chris
It is a book. Sound Reproduction, Floyd Toole. required reading.
Briefly... trying to cover it off briefly doesn't do it justice.
dave
You need to consider just a driver... a single speaker. The design of a speaker for current drive is necessarily different than when designing for voltage drive.
You are using the speaker impedance to do voltage to current conversion. It works perfectly (only) of the impedance of the speaker is purely resistive,
Fourier already did. This is the basis of every FFT based measuring system we use.
dave
Hi Dave,
That's what I have been saying all along. Also, I mentioned that the design for each would be different, probably in response to the OPs comment that people ought to take a risk and invest in current drive. You have no disagreement from me at all with these concepts.
Looking at a single driver, yes. You have an impedance with values that are not a constant. The impedance of a speaker depends on instantaneous VC temperature, and pole piece temperature (getting picky since someone else will if I don't!
). VC position will determine how much inductance the VC has as well. Then there will be all the mechanical effects coupled with back-emf. Got a slide rule?
Let's just say that it is a complex impedance and that we can give it some average numbers that are close enough for Rock and Roll. However, once you hang the speaker inside a crossover with other drivers, you now have to deal with a network. Just imagine the crossover was poor;y designed and suffers from inductor saturation and coupling between inductors. Yuk!
Now, as for your last quotation, I'm sure you are aware that I'm more than familiar with Fourier. But my comment was not aimed in that direction at all. My concern was the equality assumed similarities between two hypothetical drivers designed in different ways. Transient response is the more difficult parameter to control, as most acoustic techs and engineers know. The arguments seem to change between exactness and close enough at Esa's convenience. Since his position is that the current state of the art is "fatally flawed", and that he knows the true way to audio nirvana, he should be able to present a clear and convincing case. His statements, his responsibility. I am doing my best to consider his arguments as there is some truth to what he is saying.
-Chris
That's what I have been saying all along. Also, I mentioned that the design for each would be different, probably in response to the OPs comment that people ought to take a risk and invest in current drive. You have no disagreement from me at all with these concepts.
Looking at a single driver, yes. You have an impedance with values that are not a constant. The impedance of a speaker depends on instantaneous VC temperature, and pole piece temperature (getting picky since someone else will if I don't!
). VC position will determine how much inductance the VC has as well. Then there will be all the mechanical effects coupled with back-emf. Got a slide rule? Let's just say that it is a complex impedance and that we can give it some average numbers that are close enough for Rock and Roll. However, once you hang the speaker inside a crossover with other drivers, you now have to deal with a network. Just imagine the crossover was poor;y designed and suffers from inductor saturation and coupling between inductors. Yuk!
Now, as for your last quotation, I'm sure you are aware that I'm more than familiar with Fourier. But my comment was not aimed in that direction at all. My concern was the equality assumed similarities between two hypothetical drivers designed in different ways. Transient response is the more difficult parameter to control, as most acoustic techs and engineers know. The arguments seem to change between exactness and close enough at Esa's convenience. Since his position is that the current state of the art is "fatally flawed", and that he knows the true way to audio nirvana, he should be able to present a clear and convincing case. His statements, his responsibility. I am doing my best to consider his arguments as there is some truth to what he is saying.
-Chris
In a multiway designed for current drive, you need to exercise a completely different topology. What would be a parallel network in a voltage drive speaker becomes a series network in current drive and visa versa. One has the same travails, in a different way, from the typical voltage drive XO. Going active makes things so much easier (and more intimate). One can also mix current & voltage drive, so that one can more easily deal with woofers that have high Qm (ie made in world where voltage drive is assumed). My current drive loudspeaker will start with drivers having a Qm ~ 1.2, so should be able to readily get system Q <1
With current drive those become non-issues.
As Fourier showed there is a duality between the FR & the impulse. All Esa was going was stating that duality. No need for him to prove it.
dave
The impedance of a speaker depends on instantaneous VC temperature, and pole piece temperature (getting picky since someone else will if I don't!
With current drive those become non-issues.
As Fourier showed there is a duality between the FR & the impulse. All Esa was going was stating that duality. No need for him to prove it.
dave
Hi Dave,
Yes, I am well aware of the differences between voltage and current drive. Current drive is used in some non-audio applications. It's no mystery.
Remember, Esa is the one who is telling us that voltage drive is a horrible mistake, and that his current drive method cures all that plagues the current state of the art.
Dave, you have the book. Has Esa had a dynamic speaker designed and build for current drive? Has he confirmed that it does not suffer from other problems (ie; it is a superior way to drive a loudspeaker?). If this has not been done, it's all just so much hot air. In fact, if this method had true promise, there would be at least one audio manufacturer that would dive all over this in an instant. Think B&O, Naim or many small whacked out companies that only care to differentiate themselves. This technology possess no risk at all for them. They can even patent it and get a 20 + year free ride out of it.
-Chris
Yes, I am well aware of the differences between voltage and current drive. Current drive is used in some non-audio applications. It's no mystery.
Actually, I am going to disagree with you on this. The strength of the magnetic field affects the driver no matter how you are driving it. The same is true for VC position and the mechanical effects. These are things that don't care how you are moving the cone. They do matter, current drive is not a cure-all and does not fix all ills. This is the tone I was picking up on from Esa.
What he has to prove is that current drive does completely invoke the duality. To point at a theoretical or mathematical fact is all well and fine. Now prove that current drive does not suffer from real world effects, else it may be no better than what is being done right now.
Remember, Esa is the one who is telling us that voltage drive is a horrible mistake, and that his current drive method cures all that plagues the current state of the art.
Dave, you have the book. Has Esa had a dynamic speaker designed and build for current drive? Has he confirmed that it does not suffer from other problems (ie; it is a superior way to drive a loudspeaker?). If this has not been done, it's all just so much hot air. In fact, if this method had true promise, there would be at least one audio manufacturer that would dive all over this in an instant. Think B&O, Naim or many small whacked out companies that only care to differentiate themselves. This technology possess no risk at all for them. They can even patent it and get a 20 + year free ride out of it.
-Chris
I was referring to the compression from the voice coil heating up which is none of the above.
The duality applies no matter what the drive.
He outlines a couple designs. He provides sufficient information to encourage one to try it out.
He is very enthusiastic, you just need to let that slide. The goal of the book is to provide sufficient information for you to try it, and then you can decide.
Current drive brings its own issues, many due to trying to sort out issues brought by having to design with drivers made for voltage drive.
My interests go back to conversations i had with a Western Electric veteran who was appalled that people would use voltage drive, comments by Thorsten Loesch about the superiority of current drive, Hawksford's papers, Nelson's encouragement (the big right turn the industry made, and that maybe it was a wrong turn), experences with highish output impedance tube amps, and the variable transconductance amp we have here that is one of the best peices of solid state kit i've heard.
You should just buy the book and read it... Esa has nothing to prove ...
I see an analogy between multi-amped systems... (ie moving the XO to where we have a much more benign load), to current drive where the voltage to corrent conversion is moved to before the amplifier where we can have much more control instead of leaving it to the mercy of the speakers impedance.
The 1st pass thru the book indicates a solid presentation... i am now rereading the derivations & math bits to get a more thorough understanding.
dave
Hi Dave,
As I mentioned, in a perfect world - yes. In the real world, other factors come into play and the idea may have more problems than the existing method we are currently using.
My point is really simple here. Esa came in hot and heavy, claiming the existing technology was too flawed to listen to. My actions were designed to get him talking about his views (good for book sales), and perhaps learn something. So far I have not learned anything new beyond what I already know on the subject. My other concern is a request for book sales billed as the only responsible way to drive speakers. Given that most people are on a budget, buying a book that is a waste of money rubs me the wrong way. I did not say this book is worthless, what I did say is that the author has a responsibility to back up his claims for statements made in public. Simply saying that a question is dealt with in chapter "x" and interested parties should buy and read that section is not the way to go about things. If the ideas are sound, I'll support it.
Example. There is an author named "Weems" (I wish I could remember his name!), he has put out at least three books on speaker design that are quite good. He has built all the projects detailed in these books. I bought all three (one at a time) and got a great deal of use out of them.
I am asking questions and exchanging thoughts on the matter.
Sorry Dave, Esa has some questions he should answer. There are also some prototypes that need building and measuring. Only then is he able to support what he has said.
-Chris
That's yet another issue I pointed out earlier in the thread. We are both saying the same thing here. Note that I have pointed out several advantages with current drive. I believe my stance on this is pretty fair.
That's my point!
As I mentioned, in a perfect world - yes. In the real world, other factors come into play and the idea may have more problems than the existing method we are currently using.
My point is really simple here. Esa came in hot and heavy, claiming the existing technology was too flawed to listen to. My actions were designed to get him talking about his views (good for book sales), and perhaps learn something. So far I have not learned anything new beyond what I already know on the subject. My other concern is a request for book sales billed as the only responsible way to drive speakers. Given that most people are on a budget, buying a book that is a waste of money rubs me the wrong way. I did not say this book is worthless, what I did say is that the author has a responsibility to back up his claims for statements made in public. Simply saying that a question is dealt with in chapter "x" and interested parties should buy and read that section is not the way to go about things. If the ideas are sound, I'll support it.
Example. There is an author named "Weems" (I wish I could remember his name!), he has put out at least three books on speaker design that are quite good. He has built all the projects detailed in these books. I bought all three (one at a time) and got a great deal of use out of them.
All due respect Dave, but that is just not good enough. If he has not been able to prove any of this to himself empirically as yet, publishing the book is premature and borders on being irresponsible. Nelson builds before he recommends, as do most other people in the field. Esa is directing other peoples funds into an area that is not traveled yet. For any surprises, it's the experimenter that is on the hook, not Esa.
That's what I'm doing now.
I am asking questions and exchanging thoughts on the matter.
Yuk. They sound odd, and I have heard them in my own system. Different strokes I guess ....
Ahhhh, no! The book should not be recommended until the author has shown he has done his own reasonable amount of research. Right now, it looks like it's worthy of "paper status", not book status. If the author demonstrates that he understands the current state of the art, and his particular direction to explore, I'm more apt to believe the book has value. With such a strong viewpoint, he must be able to point to hard evidence and real experiments with the technology. He can't offer more than thought experiments right now. So, as much as I would like to believe him and support what he is doing, I can't.
An electronically crossed over system is completely different than a passive system. The electronic system allows us to control every aspect. Damping, slopes and phase. These can not be properly controlled with a passive crossover. The electronic crossover can not be said to have a more benign load. The impedances the electronic crossover sees is far less variable, and it doesn't even have a load to speak of. Totally different situation there.
Sorry Dave, Esa has some questions he should answer. There are also some prototypes that need building and measuring. Only then is he able to support what he has said.
-Chris
ETM,
This is an interesting topic concering the effects loudspeakers have on the amp. What effect would a pure ribbon speaker without a coil produce to the amplifier. This would seem to be a very good means to counter reverse emf.
You would of course have to design the amp to withstand the stress of .5 ohm loads so as not to incur problems from a transformer setup.
Do you think that if solid state amplifiers were designed before tube amps we might not be have this discussion. There is no doubt that early SS amp engineers were using tubes as a foundation for there endeavors. With more and more robust devices finding there way into audio we should see many new single stage current drive amps being developed.
Why did you name the thread with tube amps only. Do you not think all of the other products share the same fate?
Tad
This is an interesting topic concering the effects loudspeakers have on the amp. What effect would a pure ribbon speaker without a coil produce to the amplifier. This would seem to be a very good means to counter reverse emf.
You would of course have to design the amp to withstand the stress of .5 ohm loads so as not to incur problems from a transformer setup.
Do you think that if solid state amplifiers were designed before tube amps we might not be have this discussion. There is no doubt that early SS amp engineers were using tubes as a foundation for there endeavors. With more and more robust devices finding there way into audio we should see many new single stage current drive amps being developed.
Why did you name the thread with tube amps only. Do you not think all of the other products share the same fate?
Tad
Exactly right. In the very best systems, amplifier, speaker, and EQ have to be thought out as a system. No one "right" answer.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.