a DSP drivin "digital" amps cannot surpass the performance of a tri-state class BD system. another benefit, the carrier is effectively suppressed even before the output filter -- which is what allows generous amounts of GNF. a switching amp that employs global NF and rests at a physical ground when idle (not an averaged "ground" as seen thru an inductor) is the closest thing we have yet to a traditional linear amp. in every performance aspect (except the extremely high efficiency of course). damping factors can be easily in excess of 200 and s/n ratios well over 100 with no DSP at all.
i met with the author of the patent (james strictland) and had a very lenthy conversation about it (as well as a listening test =). it's nothing like 99.9% of the switcher designs out there as unimpressive as the patent description may seem. viewing the patent without the diagrams is pointless -- make sure you can see them all.
i met with the author of the patent (james strictland) and had a very lenthy conversation about it (as well as a listening test =). it's nothing like 99.9% of the switcher designs out there as unimpressive as the patent description may seem. viewing the patent without the diagrams is pointless -- make sure you can see them all.
Re: Damping Factor
Hi, Randy,
I think you are right here regarding damping factor.
If you would really like to see low damping factor in Class D look at: http://www.piak.nl/Mueta_analog_pwm_amplifier.pdf. I can give you also patent number http://l2.espacenet.com/espacenet/bnsviewer?CY=gb&LG=en&DB=EPD&PN=WO0042702&ID=WO+++0042702A1+I+ .
My limited expirience is that although good damping factor is technically superior, it may not sound best. In one of recent projects I was building 3 phase motor drive for high speed motor. I used output filters and two halfbridges running in oposite phase (class BD) per leg since motor was running on 16kHz 3 phase voltage, so the whole thing was actually 3 channel class D amplifier. I found it amusing to connect soundblaster as source and listen to some music from output of motor drive. I used dummy resistor as load and divided output voltage to drive headphones. The best sound was at first step when i drove output stages open loop from triangle modulator. Adding feedback across filter stage only increased noise, there was no improvement in sound quality.
So now I will try open loop design and use digital modulator, since it is impossible to do analog modulator with specs like TAS5015.
There is a relatively easy way to lower output impedance of TAS5015 design and that is by use of PEDEC (see my previous post)
Regards, Jaka
Hi, Randy,
I think you are right here regarding damping factor.
If you would really like to see low damping factor in Class D look at: http://www.piak.nl/Mueta_analog_pwm_amplifier.pdf. I can give you also patent number http://l2.espacenet.com/espacenet/bnsviewer?CY=gb&LG=en&DB=EPD&PN=WO0042702&ID=WO+++0042702A1+I+ .
My limited expirience is that although good damping factor is technically superior, it may not sound best. In one of recent projects I was building 3 phase motor drive for high speed motor. I used output filters and two halfbridges running in oposite phase (class BD) per leg since motor was running on 16kHz 3 phase voltage, so the whole thing was actually 3 channel class D amplifier. I found it amusing to connect soundblaster as source and listen to some music from output of motor drive. I used dummy resistor as load and divided output voltage to drive headphones. The best sound was at first step when i drove output stages open loop from triangle modulator. Adding feedback across filter stage only increased noise, there was no improvement in sound quality.
So now I will try open loop design and use digital modulator, since it is impossible to do analog modulator with specs like TAS5015.
There is a relatively easy way to lower output impedance of TAS5015 design and that is by use of PEDEC (see my previous post)
Regards, Jaka
i owuld like to say there are many different embodiments of class BD systems. there are diff. H-bridges, sub. H-bridges... ect... the one I speak of is a single ended output with three possible states. this is possible by employing two in-phase half bridges, one between +v and gnd, and the other between gnd and -v. you then switch between them at TWICE the sampling frequency to create the tri-state operation on a single ended output.
when there is little or no signal condition, you will switch to each bridge while it is in a ground state, providing a true ground at the output with the absense of a carrier even before the output filtering.
when there is little or no signal condition, you will switch to each bridge while it is in a ground state, providing a true ground at the output with the absense of a carrier even before the output filtering.
I admit that a BD topology would facilitate the feedback. But it is still possible to take feedback from a class AD output stage without any problems. If you are worried about the carrier take an output filter with zeroes and thake the feedback from after the filter. In this case you can even take care of the filter's rising output impedance and its nonlinear behaviour (inductor cores) as well.
The oldest patent for this approach I know is the following, it was later refined and licenced to Peavey (DECA series amps):US4178556
The main advantages of BD output stages are IMO reduced RFI and less supply pumping problems.
Regards
Charles
The oldest patent for this approach I know is the following, it was later refined and licenced to Peavey (DECA series amps):US4178556
The main advantages of BD output stages are IMO reduced RFI and less supply pumping problems.
Regards
Charles
Re: Re: Damping Factor
OK, I'll admit that I was wrong to use the term 'damping factor' to begin with. I guess I probably shouldn't have tried to continue associating amplifier characteristics to what I personally (and perhaps no one else) associated with the 'spirit of the meaning' of damping factor. I'll be the first to agree that's it's important to assign very specific meaning to technical terms.
I'm sorry about that.
However, I was trying to be clear (and objective) in how what I was describing was different from 'damping factor'. I tried my best to explain my reasoning, rather than just making statements.
I do appreciate the opportunity to discuss things here. I find that my own understanding of things is enhanced when I try to explain them the others. It also helps to expose the details of my thoughts to feedback on where they might be wrong.
I'll agree that instability rarely happens on the fundamental of the original signal. Negative feedback does cause lots of small harmonic instabilities and ringing in many amplifier designs, much of it from the introduction of back-EMF. Some amplifier designs do a better job of reducing or perhaps even eliminating feedback-induced artifacts in the audio band.
One way of reducing feedback-induced artifacts is to reduce or eliminate feedback. It's seems to me that this is a fairly popular approach. It's effectiveness can be evaluated objectively by measurements when driving an actual speaker (as opposed to a dummy resistor), and it also seems to commonly be supported by subjective listening tests.
Of course if a feedback circuit is designed well enough so that it has the bandwidth to completely control the output signal throughout the entire audio range, but yet with absolute stability, that amp (regardless of class) will excel at controlling the output waveform (what I was referring to as 'the spirit of the meaning' of damping factor).
In my own admittedly biased opinion, this has not totally been achieved. I base this opinion on the observation that all of the best sounding amps that I've heard (subjectively to my ears) were low-feedback designs.
I fully expect that sometime in the future, an amp design with feedback will appear that will excel at controlling the output waveform and also have better sound (and I'm guessing that it will also be based on a switching output).
In the meantime, I ask that my other comments on output waveform control be considered along with the notion that negative feedback in its present state of the art isn't fully effective without undesirable side-effects.
I would welcome any information you or others may have regarding implementation of improved negative feedback techniques. As I mentioned before, I believe in the long term that this will turn out to be a better approach to high-end audio.
I don't believe this is true.
For a given level of output, the effective impedance of a switching transistor is less than a steady-state transistor operating in its linear range.
I'm not able to whip out the math to prove this, but I offer the following evidence:
A class B amp will have an efficiency of 50%.
(Class A and class AB amps have worse efficiencies, but that's due to power being wasted during the time current isn't being delivered to the speakers. So the class B 50% figure should be used for my comparison.)
A switching amp will have a typical efficiency exceeding 90%.
The effective impedance of the output stage is directly proportional to the inverse of its efficiency.
(I believe this is a good argument for directly connecting each speaker driver to an individual amp and eliminating the crossover. Of course this applies to any type of amp.)
True, but we're talking 20uH in the case of the equibit design. The resistive component of the inductors is probably more significant. The effects of this filter reactance are a major loss component in the amp's efficiency, and are including in the 90% figure.
I was originally trying to explain it in terms of being a desirable trade off (reduced damping factor vs. undesirable side-effects of negative feedback). Hopefully I've clarified how I was framing this.
Given the premise that negative feedback has undesirable consequences on the amp's output (I recognize this is debatable), I think that the rest of my explanation is scientifically reasonable.
I would gladly accept further discussion on the merits of individual points of this topic. And I certainly don't want to portray myself as a high-level authority on this subject. But I would appreciate not having a wholesale dismissal of what I'm writing.
Regards,
Brian.
R. McAnally said:
OK, I'll admit that I was wrong to use the term 'damping factor' to begin with. I guess I probably shouldn't have tried to continue associating amplifier characteristics to what I personally (and perhaps no one else) associated with the 'spirit of the meaning' of damping factor. I'll be the first to agree that's it's important to assign very specific meaning to technical terms.
I'm sorry about that.
However, I was trying to be clear (and objective) in how what I was describing was different from 'damping factor'. I tried my best to explain my reasoning, rather than just making statements.
I do appreciate the opportunity to discuss things here. I find that my own understanding of things is enhanced when I try to explain them the others. It also helps to expose the details of my thoughts to feedback on where they might be wrong.
I'll agree that instability rarely happens on the fundamental of the original signal. Negative feedback does cause lots of small harmonic instabilities and ringing in many amplifier designs, much of it from the introduction of back-EMF. Some amplifier designs do a better job of reducing or perhaps even eliminating feedback-induced artifacts in the audio band.
One way of reducing feedback-induced artifacts is to reduce or eliminate feedback. It's seems to me that this is a fairly popular approach. It's effectiveness can be evaluated objectively by measurements when driving an actual speaker (as opposed to a dummy resistor), and it also seems to commonly be supported by subjective listening tests.
Of course if a feedback circuit is designed well enough so that it has the bandwidth to completely control the output signal throughout the entire audio range, but yet with absolute stability, that amp (regardless of class) will excel at controlling the output waveform (what I was referring to as 'the spirit of the meaning' of damping factor).
In my own admittedly biased opinion, this has not totally been achieved. I base this opinion on the observation that all of the best sounding amps that I've heard (subjectively to my ears) were low-feedback designs.
I fully expect that sometime in the future, an amp design with feedback will appear that will excel at controlling the output waveform and also have better sound (and I'm guessing that it will also be based on a switching output).
In the meantime, I ask that my other comments on output waveform control be considered along with the notion that negative feedback in its present state of the art isn't fully effective without undesirable side-effects.
I would welcome any information you or others may have regarding implementation of improved negative feedback techniques. As I mentioned before, I believe in the long term that this will turn out to be a better approach to high-end audio.
I don't believe this is true.
For a given level of output, the effective impedance of a switching transistor is less than a steady-state transistor operating in its linear range.
I'm not able to whip out the math to prove this, but I offer the following evidence:
A class B amp will have an efficiency of 50%.
(Class A and class AB amps have worse efficiencies, but that's due to power being wasted during the time current isn't being delivered to the speakers. So the class B 50% figure should be used for my comparison.)
A switching amp will have a typical efficiency exceeding 90%.
The effective impedance of the output stage is directly proportional to the inverse of its efficiency.
(I believe this is a good argument for directly connecting each speaker driver to an individual amp and eliminating the crossover. Of course this applies to any type of amp.)
True, but we're talking 20uH in the case of the equibit design. The resistive component of the inductors is probably more significant. The effects of this filter reactance are a major loss component in the amp's efficiency, and are including in the 90% figure.
I was originally trying to explain it in terms of being a desirable trade off (reduced damping factor vs. undesirable side-effects of negative feedback). Hopefully I've clarified how I was framing this.
Given the premise that negative feedback has undesirable consequences on the amp's output (I recognize this is debatable), I think that the rest of my explanation is scientifically reasonable.
I would gladly accept further discussion on the merits of individual points of this topic. And I certainly don't want to portray myself as a high-level authority on this subject. But I would appreciate not having a wholesale dismissal of what I'm writing.
Regards,
Brian.
Re: Re: Re: Damping Factor
Thanks for the thoughtful reply, and sorry if I came across to strongly. Thanks for clearing up those few things. Theres only one thing left -- i am going to have to think hard about this one, and thanks for giving my brain a workout today =).
Explain to me this: my class D sub amp, with 90% efficiency, has a higher output impedance (lower DF) than my class B full range amp operating at 60% efficiency (DF of over 200).
According to your theorem, the lower efficiency should imply a higher output impedance, which isn't the case. =) The class B amplifier generates more heat due to the V*I (P=V*I)characteristics of the output stage, since the transistors are "always on", but operating in their linear region. This implies in no way there is added resistance seen by the load when power is lost as heat in the output stage. The load could "care less" about the voltage drop the current must go through to reach it.
Of course, you could always look at in a different perspective... that P= I^2*R. Replace R with V/I and you get the same thing - P=V*I =). However, a BJT is not a variable resistor in practice. They are viewed as either a current controlled current source or more precicely a voltage controlled (Vbe) current source. if we viewed BJT's (or mosfets) as resistors depending on their Vce current/voltage there would be alot of braincells wasted =). Transistors are active devices which cannot be viewed as having a fixed resistance, in any mode of operation.
In a nutshell, efficiency has nothing to do with output impedance, and the nature of a class D output does not inherently create a lower output impedance.
note: I made a mistake earlier by stating that in a lineardesign the BJT's might be "half way" on. They are in fact always on, and simply operating at varying voltages. there is no such thing as a "half way on " BJT. sorry...
Thanks for the thoughtful reply, and sorry if I came across to strongly. Thanks for clearing up those few things. Theres only one thing left -- i am going to have to think hard about this one, and thanks for giving my brain a workout today =).
Brian Brown said:
Explain to me this: my class D sub amp, with 90% efficiency, has a higher output impedance (lower DF) than my class B full range amp operating at 60% efficiency (DF of over 200).
According to your theorem, the lower efficiency should imply a higher output impedance, which isn't the case. =) The class B amplifier generates more heat due to the V*I (P=V*I)characteristics of the output stage, since the transistors are "always on", but operating in their linear region. This implies in no way there is added resistance seen by the load when power is lost as heat in the output stage. The load could "care less" about the voltage drop the current must go through to reach it.
Of course, you could always look at in a different perspective... that P= I^2*R. Replace R with V/I and you get the same thing - P=V*I =). However, a BJT is not a variable resistor in practice. They are viewed as either a current controlled current source or more precicely a voltage controlled (Vbe) current source. if we viewed BJT's (or mosfets) as resistors depending on their Vce current/voltage there would be alot of braincells wasted =). Transistors are active devices which cannot be viewed as having a fixed resistance, in any mode of operation.
In a nutshell, efficiency has nothing to do with output impedance, and the nature of a class D output does not inherently create a lower output impedance.
note: I made a mistake earlier by stating that in a lineardesign the BJT's might be "half way" on. They are in fact always on, and simply operating at varying voltages. there is no such thing as a "half way on " BJT. sorry...
Re: Efficiency vs. Output Impedance
I've been mulling over the issue of efficiency vs. output impedance and how it relates to switching output stages some more. I'm less sure about this than I originally thought I was.
I've come up with enough examples, some that support a relationship, and some that refute it, that I think it's necessary to come up with a better fundamental explanation.
There's a guy I know who is an authority on high power inverters. I'm sure he'll be able to give me a better explanation on this topic than I can come up with by myself. I won't be able to talk with him until after the new year. Hopefully at that time I'll be able to report back with a better answer.
Regards,
Brian.
I've been mulling over the issue of efficiency vs. output impedance and how it relates to switching output stages some more. I'm less sure about this than I originally thought I was.
I've come up with enough examples, some that support a relationship, and some that refute it, that I think it's necessary to come up with a better fundamental explanation.
There's a guy I know who is an authority on high power inverters. I'm sure he'll be able to give me a better explanation on this topic than I can come up with by myself. I won't be able to talk with him until after the new year. Hopefully at that time I'll be able to report back with a better answer.
Regards,
Brian.
halojoy - On "digital" amplifier
I was the first one to answer in this thread,
see this link:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=7754
There I said, I was not yet interested in making of
a "digital" amplifier. I am still NOT interested!
As I see it, the difference is only where in the chain
you put the DAC - Digital to Analogue Converter
Even so called digital amplifiers have analogue outputs
even if they a placed within the Loudspeakers.
This comes from the fact that Loudspeaker Coils have an Analogue input.
For present I have the DAC in my CD-player.
And it works good enough for me!
-----------------------------------------------------------------
After reading this article, I can see where it all is going.
Into the bright future .......
- we will soon belong to the dying species -
- we will be no longer - it is just a matter of time -
at least if these Prophets are correct in telling the future .....
Now telling the future is never easy, for sure!
------------------------------------------------------------------
Quote is from an Audio Magazines Archive:
Amp Designs - from "Audio Media" Archives
It was written in March 2001:
"For this article I sought the impressions of a number of power amplifier designers and product specialists. I wanted to find out not only what is happening in the field of amplifier design, but also what is on the horizon. As we’ll see throughout this article, there are some significant advancements in the technology for power amplification. The people I spoke with also represent a good cross section of the professional audio industry."
----------------------------------------------------------
halojoy - hope he will die in the coming future
- there is a good chance he will!
I was the first one to answer in this thread,
see this link:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=7754
There I said, I was not yet interested in making of
a "digital" amplifier. I am still NOT interested!
As I see it, the difference is only where in the chain
you put the DAC - Digital to Analogue Converter
Even so called digital amplifiers have analogue outputs
even if they a placed within the Loudspeakers.
This comes from the fact that Loudspeaker Coils have an Analogue input.
For present I have the DAC in my CD-player.
And it works good enough for me!
-----------------------------------------------------------------
After reading this article, I can see where it all is going.
Into the bright future .......
We, who love our Analogues, Turntables, Tube-amps, Pass-amps and all other Classical Amplifiers and Analogue Audio Equipment,
- we will soon belong to the dying species -
- we will be no longer - it is just a matter of time -
at least if these Prophets are correct in telling the future .....
Now telling the future is never easy, for sure!
------------------------------------------------------------------
Quote is from an Audio Magazines Archive:
Amp Designs - from "Audio Media" Archives It was written in March 2001:
"For this article I sought the impressions of a number of power amplifier designers and product specialists. I wanted to find out not only what is happening in the field of amplifier design, but also what is on the horizon. As we’ll see throughout this article, there are some significant advancements in the technology for power amplification. The people I spoke with also represent a good cross section of the professional audio industry."
----------------------------------------------------------
halojoy - hope he will die in the coming future
- there is a good chance he will!
Re: halojoy - On "digital" amplifier
Cheer up!
Don't you like the idea that there might be a new option that could ultimately lead to better sound? It's not going to make your existing system sound any worse.
Also, I don't think traditional analog linear amps and turntables are going to go extinct any time soon. IMO tonearms and cartridges are much better than they were when CDs first came out. Analog linear amps are a fairly mature technology, but I'm sure that we'll continue to see improvements in them as well.
My vinyl LP collection still far exceeds the number of digital recordings that I have. I still find the best of these recordings to be fantastic. I'm not about to give up the turntable that I custom-built twenty years ago or the amp that I first used with it (the Hitachi HEXFET output devices seemed like pretty radical new technology to me at the time). I still occasionally have an urge to update sections of the turntable, or possibly to even start from scratch on a new one incorporating more recent things that I've learned. But for now I'm focusing on other things (like my new digital amp). The most recent change I made to that system was fifteen years ago and it's served me very well. As much as I like to continue to make improvements to my system, I've found it to be a good thing to reach a point of balance, leave it alone, and then simply use it to enjoy music. I know that there's a lot of things I could do now to improve it, and I've myself heard a number of systems that I thought were better. But even after all these years, most people who hear it think that it's the best system they've heard.
I've found that if you're not constantly replacing and updating things, audio is one of the very best hobby investments there is. Compared to most other hobbies, the equipment is amazingly reliable. Once you've got it, the equipment takes very little money to maintain.
I felt that technology has advanced enough now that I was ready to take a crack at an entirely new system. The biggest opportunities I see for improvements are the availability of digital high-resolution multi-channel recordings, and digital crossovers, room correction, and cross-talk cancellation (note that I didn't mention digital amplifiers).
I see digital amplifiers more as a new type of building block, one that fits in well with the other advances above.
I also think that for a given amount of money (whether it's $100, $1000, or $10,000), digital amplifier technology holds the potential of achieving better sound quality than a 'traditional' line-level DAC and analog linear amplifier combination. The jury is still out on this point, and I want to emphasize again that a digital amp isn't required to gain the benefits of the other new technologies I've mentioned above.
Just as analog amps vary in quality, I'm sure we'll see digital amps both good and poor.
It seems pretty certain that most regular consumer products will wind up using digital amplifiers because of their greater efficiency, lower cost and smaller size. I suspect that these factors are of low importance to most of the people in this group, but these factors are a major driving force behind the development of this technology. I think the good news is that digital amplifier technology has already shown great potential for high quality audio (with TacT leading the way), and that it will be possible for us as DIYers to leverage the new devices coming out (such as the TI Equibit chipsets) for our own benefit.
So even if you're not interested in these things for yourself, I wouldn't worry about them.
Regards,
Brian.
halojoy said:
Cheer up!
Don't you like the idea that there might be a new option that could ultimately lead to better sound? It's not going to make your existing system sound any worse.
Also, I don't think traditional analog linear amps and turntables are going to go extinct any time soon. IMO tonearms and cartridges are much better than they were when CDs first came out. Analog linear amps are a fairly mature technology, but I'm sure that we'll continue to see improvements in them as well.
My vinyl LP collection still far exceeds the number of digital recordings that I have. I still find the best of these recordings to be fantastic. I'm not about to give up the turntable that I custom-built twenty years ago or the amp that I first used with it (the Hitachi HEXFET output devices seemed like pretty radical new technology to me at the time). I still occasionally have an urge to update sections of the turntable, or possibly to even start from scratch on a new one incorporating more recent things that I've learned. But for now I'm focusing on other things (like my new digital amp). The most recent change I made to that system was fifteen years ago and it's served me very well. As much as I like to continue to make improvements to my system, I've found it to be a good thing to reach a point of balance, leave it alone, and then simply use it to enjoy music. I know that there's a lot of things I could do now to improve it, and I've myself heard a number of systems that I thought were better. But even after all these years, most people who hear it think that it's the best system they've heard.
I've found that if you're not constantly replacing and updating things, audio is one of the very best hobby investments there is. Compared to most other hobbies, the equipment is amazingly reliable. Once you've got it, the equipment takes very little money to maintain.
I felt that technology has advanced enough now that I was ready to take a crack at an entirely new system. The biggest opportunities I see for improvements are the availability of digital high-resolution multi-channel recordings, and digital crossovers, room correction, and cross-talk cancellation (note that I didn't mention digital amplifiers).
I see digital amplifiers more as a new type of building block, one that fits in well with the other advances above.
I also think that for a given amount of money (whether it's $100, $1000, or $10,000), digital amplifier technology holds the potential of achieving better sound quality than a 'traditional' line-level DAC and analog linear amplifier combination. The jury is still out on this point, and I want to emphasize again that a digital amp isn't required to gain the benefits of the other new technologies I've mentioned above.
Just as analog amps vary in quality, I'm sure we'll see digital amps both good and poor.
It seems pretty certain that most regular consumer products will wind up using digital amplifiers because of their greater efficiency, lower cost and smaller size. I suspect that these factors are of low importance to most of the people in this group, but these factors are a major driving force behind the development of this technology. I think the good news is that digital amplifier technology has already shown great potential for high quality audio (with TacT leading the way), and that it will be possible for us as DIYers to leverage the new devices coming out (such as the TI Equibit chipsets) for our own benefit.
So even if you're not interested in these things for yourself, I wouldn't worry about them.
Regards,
Brian.
It is interesting to note that most delta-sigma DAC chips these days are in fact multi-bit delta-sigma which means that the DAC part is not one-bit like older bitstream DACs but may be 3, 4, 5 or possibly 6 bits. This is a bit discouraging for digital amps as they have to be one bit. If chip designers had to move to multi-bit delta-sigma to greater performance then I think the one-bit output stage of a digital amp will have to be quite an impressive bit of engineering to compete with the best of today's convetional DACs and linear amplifiers.
The technology I'm sure is coming along very well but you may find that 95% of the motivation is for lower cost, smaller size etc. to satisfy mass market suff rather than improved sound quality.
The technology I'm sure is coming along very well but you may find that 95% of the motivation is for lower cost, smaller size etc. to satisfy mass market suff rather than improved sound quality.
tas5015 shematics
Hello,
if anyone is interested here is preliminary schematics of TAS5015 control board. It is very simple, basically only optical input, DIR1703, AD1896 upsampling to 24bit 96khz and TAS5015 in master mode run from 100MHz oscillator. Output is SN65LVDS389 transmitter connected to 2 RJ45 jacks. Communication with power stage will use standard UTP cat5 network cables. I would first like to have working modulator circuit, then proceed with the power stage. I hope to have first pcb ready by the end of the february.
Any comments welcome!
Regards,
Jaka Racman
Hello,
if anyone is interested here is preliminary schematics of TAS5015 control board. It is very simple, basically only optical input, DIR1703, AD1896 upsampling to 24bit 96khz and TAS5015 in master mode run from 100MHz oscillator. Output is SN65LVDS389 transmitter connected to 2 RJ45 jacks. Communication with power stage will use standard UTP cat5 network cables. I would first like to have working modulator circuit, then proceed with the power stage. I hope to have first pcb ready by the end of the february.
Any comments welcome!
Regards,
Jaka Racman
Damping factor and efficiency
Amongst other things the Rdson of the MOSFETS used is an important factor regarding the efficiency of ANY type of switching amplifier.
The output impedance (and hence damping factor) of said switching amps is also heavily influenced by the Rdson of the MOSFETS. So far there is in fact some correlation between DF and efficiency.
The power supply and the output filter will also have significant influence on the amplifier's damping factor (the latter makes it even very frequency dependant !).
As soon as you are using NFB, the DF will be higher than that of the "bare" amplifier alone.
Since the Equibit topology is of the non NFB type you end up with quite a low DF. I once heard the TacT MillenniuM and I didn't have the impression that it lacks control !
My personal (and therefore subjective) view is that it isn't very important to have huge DF figures but rather reasonable ones, i.e. between 10 and 200. BUT the DF should stay the same for ANY output power AND also ANY phase-angle of load current. Thats where class A, generously dimensioned clas B/AB and some switching amps excel, compared to whimpy class AB amps with badly engineered VA protection circuitry.
Furthermore the DF should be monotonic with frequency, i.e. it may fall with rising frequency but should do so without bumps and dips etc.
Regards
Charles
Amongst other things the Rdson of the MOSFETS used is an important factor regarding the efficiency of ANY type of switching amplifier.
The output impedance (and hence damping factor) of said switching amps is also heavily influenced by the Rdson of the MOSFETS. So far there is in fact some correlation between DF and efficiency.
The power supply and the output filter will also have significant influence on the amplifier's damping factor (the latter makes it even very frequency dependant !).
As soon as you are using NFB, the DF will be higher than that of the "bare" amplifier alone.
Since the Equibit topology is of the non NFB type you end up with quite a low DF. I once heard the TacT MillenniuM and I didn't have the impression that it lacks control !
My personal (and therefore subjective) view is that it isn't very important to have huge DF figures but rather reasonable ones, i.e. between 10 and 200. BUT the DF should stay the same for ANY output power AND also ANY phase-angle of load current. Thats where class A, generously dimensioned clas B/AB and some switching amps excel, compared to whimpy class AB amps with badly engineered VA protection circuitry.
Furthermore the DF should be monotonic with frequency, i.e. it may fall with rising frequency but should do so without bumps and dips etc.
Regards
Charles
Tripath releases new digital input chip
Its been a while coming, but Tripath has finally released a digital input controller for its power output stages. I can't find the datasheets on the site yet, however as Tripath has a decent DIY following with chipsets relatively easy to use I'm sure a few of you following this thread would be interested to check this out. See www.tripath.com for more info.
SAN JOSE, California, January 8, 2003—Tripath Technology Inc. (Nasdaq: TRPH), creators of Digital Power Processing (DPP®) and Class-T™ Advanced 1-bit Digital Audio Amplifiers, announced today the availability of its TC6000 6-channel digital input Class-T™ controller.
The TC6000 offers impressive versatility for the designer through its ability to drive a wide range of power stages from less than 10 watts to more than 500 watts. Its 6-channel digital input and I2C volume control makes it ideally suitable and addresses the sweet spot for the powered DVD home theater systems, multi-channel AV receivers, automotive audio electronics and other multi-channel audio systems. Utilizing its proprietary Digital Power Processing (DPP®) technology, the TC6000 provides the highest fidelity listening experience available in a high efficiency switching amplifier. The TC6000 Class-T™ controller has a THD+N of less than 0.03% and offers both the high fidelity of Class-AB and the power efficiency of Class-D audio amplifiers.
Tripath's patented signal processing is combined with a switching mode approach that generates high fidelity sound with considerably lower heat dissipation than Class-AB amplifiers. The TC6000, when mated with power stages available from Tripath, achieves greater than 90% efficiency with a dynamic range of greater than 100dB. The controller comes in a single 80-pin QFP package.
Tripath's broad line of Class-T™ advanced 1-bit digital amplifiers, ranging from 10 watts to 2000 watts, are rapidly gaining market acceptance within high growth consumer electronics categories, such as powered DVD home theater systems, micro-component systems, set-top boxes, personal computers, aftermarket automotive electronics, LCD TVs and flat panel displays. These tightly integrated Class-T™ based designs set a new standard for high power, exceptional sound quality and high efficiency. Industry giants who have chosen Class-T as their digital amplifier technology include Aiwa, Apex Digital, Apple, Denon, Fujitsu, JVC, Panasonic, Sony and Motorola.
"Tripath continues to pave the road for cutting-edge technology in the digital audio market," said Dr. Adya Tripathi, Chairman, President, and CEO of Tripath. "We are very proud of the TC6000, our latest addition to the industry's broadest range of 1-bit digital audio amplifiers, and will continue to introduce innovative products with the highest efficiency and lowest distortion."
Specifications: TC6000
I2S compatible formats for digital audio input
Signal to noise ratio greater than 100dB
Efficiency greater than 90%
THD+N less than 0.03%
Up to 24-bit resolution (16, 18, 20, and 24 bit)
Input sampling rates from 32 kHz to 192 kHz
Standard 2-wire control interface
Digital volume control - 120 dB range - 1/2 dB step size in 1/8 dB increments - Zero crossing detection for click free transitions
80-pin QFP package
Applications -- multi-channel DVD home theater systems, multi-channel AV receivers and automotive audio
Its been a while coming, but Tripath has finally released a digital input controller for its power output stages. I can't find the datasheets on the site yet, however as Tripath has a decent DIY following with chipsets relatively easy to use I'm sure a few of you following this thread would be interested to check this out. See www.tripath.com for more info.
SAN JOSE, California, January 8, 2003—Tripath Technology Inc. (Nasdaq: TRPH), creators of Digital Power Processing (DPP®) and Class-T™ Advanced 1-bit Digital Audio Amplifiers, announced today the availability of its TC6000 6-channel digital input Class-T™ controller.
The TC6000 offers impressive versatility for the designer through its ability to drive a wide range of power stages from less than 10 watts to more than 500 watts. Its 6-channel digital input and I2C volume control makes it ideally suitable and addresses the sweet spot for the powered DVD home theater systems, multi-channel AV receivers, automotive audio electronics and other multi-channel audio systems. Utilizing its proprietary Digital Power Processing (DPP®) technology, the TC6000 provides the highest fidelity listening experience available in a high efficiency switching amplifier. The TC6000 Class-T™ controller has a THD+N of less than 0.03% and offers both the high fidelity of Class-AB and the power efficiency of Class-D audio amplifiers.
Tripath's patented signal processing is combined with a switching mode approach that generates high fidelity sound with considerably lower heat dissipation than Class-AB amplifiers. The TC6000, when mated with power stages available from Tripath, achieves greater than 90% efficiency with a dynamic range of greater than 100dB. The controller comes in a single 80-pin QFP package.
Tripath's broad line of Class-T™ advanced 1-bit digital amplifiers, ranging from 10 watts to 2000 watts, are rapidly gaining market acceptance within high growth consumer electronics categories, such as powered DVD home theater systems, micro-component systems, set-top boxes, personal computers, aftermarket automotive electronics, LCD TVs and flat panel displays. These tightly integrated Class-T™ based designs set a new standard for high power, exceptional sound quality and high efficiency. Industry giants who have chosen Class-T as their digital amplifier technology include Aiwa, Apex Digital, Apple, Denon, Fujitsu, JVC, Panasonic, Sony and Motorola.
"Tripath continues to pave the road for cutting-edge technology in the digital audio market," said Dr. Adya Tripathi, Chairman, President, and CEO of Tripath. "We are very proud of the TC6000, our latest addition to the industry's broadest range of 1-bit digital audio amplifiers, and will continue to introduce innovative products with the highest efficiency and lowest distortion."
Specifications: TC6000
I2S compatible formats for digital audio input
Signal to noise ratio greater than 100dB
Efficiency greater than 90%
THD+N less than 0.03%
Up to 24-bit resolution (16, 18, 20, and 24 bit)
Input sampling rates from 32 kHz to 192 kHz
Standard 2-wire control interface
Digital volume control - 120 dB range - 1/2 dB step size in 1/8 dB increments - Zero crossing detection for click free transitions
80-pin QFP package
Applications -- multi-channel DVD home theater systems, multi-channel AV receivers and automotive audio
The TC6000 datasheet will be available in about two weeks and will be posted on their website soon after. Shipping to select clients will begin in about a months time and then to the general public in a few months after that. I have already talked to the Apps Engineers about this one. It is meant to work with their existing drivers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.