G.Kleinschmidt said:
Hi Glen,
These are good points. I would say first that someone who wants to build a low-end amplifier need not bother with MOSFETs. The same probably goes for poorly-designed MOSFET amplifiers. And, indeed, a lot of people have demonstrated that it is quite possible to design a poor MOSFET amplifier.
I would also say that the example I cited was what I would consider a bit of a worst-case design, in that a MOSFET amplifier with only 20 dB of NFB at 20 kHz and "only" 200 mA of bias without EC is toward the low end of performance in general. I would worry about distortion due to transconductance droop in the crossover region in such a design before I'd worry about 0.5 uH of additional effective output inductance. A more decent non-EC design would have 30 dB or more NFB at 20 kHz.
Cheers,
Bob
Bob Cordell said:
OK, but a lot of people do advocate the use of low NFB. Low NFB in a low power amplifier (<~70W rms) with an output stage with an intrinsically high output impedance can quite easilly produce an amplifier with an equivalent output inductance in the uH range.
This is especially so in designs which drive a single pair of MOSFET's from a hot VAS. To match BJT performance in this regard, a driver stage and a decent bias current is pretty much mandatory.
Cheers,
Glen
G.Kleinschmidt said:
I can't disagree, but I think I can can build a better high-end MOSFET amplifier than a high-end BJT amplifier, even at 70 Watts. If I use EC, it will dissipate about the same; If I use no EC, it will dissipate more, and run a little hotter at idle. But if you have to design it to be able to do 1/3 power continuously, the higher idle bias won't really change the amount of heatsinking needed.
If I get a chance, I'll measure the effective output inductance of a hafler DH-200 that I have.
Cheers,
Bob
Bob Cordell said:
Hi Bob
Why do you say that? How about if you used EC with a high fT bioplar output?
What does the DH-200 have inside?
Cheers,
Glen
G.Kleinschmidt said:
Hi Glen,
Here's a link to the DH-200 Manual. It's over 8MB. Output stage uses the original Hitachi lateral MOSFETs.
andy_c said:
OK, thanks. That looks a lot like a power amp kit marketed here in Oz in the 80's under the 'Dynaco' brand name.
Cheers,
Glen
G.Kleinschmidt said:
The Dynaco company was founded by David Hafler many years ago ('50s maybe?), so that makes sense. After Dynaco went under, he founded Hafler Corp. I thought the original Dynaco company had already gone under by the time the '80s rolled around, but maybe there was some different arrangement in Oz.
Andy,
same story overhere, in the 80s you could either buy a complete Hafler amp or a Dynaco kit overhere.
Maybe a similar setup as the distinction between Dodge and Chrysler vehicles.
same story overhere, in the 80s you could either buy a complete Hafler amp or a Dynaco kit overhere.
Maybe a similar setup as the distinction between Dodge and Chrysler vehicles.
G.Kleinschmidt said:
EC with a high-ft bipolar output is fine also.
The DH-200 is a modest Lateral MOSFET design that is about 100 wpc and certainly cannot be said to be a sophisticated or expensive design.
Bob
sophisticated amplifiers
Real sophistication starts at thd<1ppm, not at 0.02% or 0.25% in case of tubes.
Real sophistication starts at thd<1ppm, not at 0.02% or 0.25% in case of tubes.
Re: sophisticated amplifiers
I would not go quite that far, especially in the case of solid state, and in view of the fact that single-number values can be misleading.
First of all, if you are talking about 1 kHz THD, a value of 1 ppm is pretty meaningless. If you are talking about THD out to 20 kHz, then I think you are getting reasonably sophisticated at anything less than about 0.002% from 1 watt to full power in a 200 kHz measurement bandwidth.
As far as 19+20 kHz twin tone IM, you are starting to get sophisticated when all of the IM products are down by more than 100 dB at full power, with preferably fifth and higher order products down by more than 110 dB.
It is not just a numbers race; it is important that the higher order products be down. A design with second and thirds down 90 dB and all the rest below 100 dB, and continuing to decrease as order increases, is a very good design, in my opinion.
As many have said, it is especially important that high order products are well down at low powers. However, it needs to be recognized that these can be a challenge to measure in light of noise, so that spectral analysis techniques must be used.
Cheers,
Bob
estuart said:
I would not go quite that far, especially in the case of solid state, and in view of the fact that single-number values can be misleading.
First of all, if you are talking about 1 kHz THD, a value of 1 ppm is pretty meaningless. If you are talking about THD out to 20 kHz, then I think you are getting reasonably sophisticated at anything less than about 0.002% from 1 watt to full power in a 200 kHz measurement bandwidth.
As far as 19+20 kHz twin tone IM, you are starting to get sophisticated when all of the IM products are down by more than 100 dB at full power, with preferably fifth and higher order products down by more than 110 dB.
It is not just a numbers race; it is important that the higher order products be down. A design with second and thirds down 90 dB and all the rest below 100 dB, and continuing to decrease as order increases, is a very good design, in my opinion.
As many have said, it is especially important that high order products are well down at low powers. However, it needs to be recognized that these can be a challenge to measure in light of noise, so that spectral analysis techniques must be used.
Cheers,
Bob
Hi Bob,
Of course I wasn't talking about a thd at 1kHz, let alone at 50 Hz or so. I meant at any frequency up to 20kHz and at any power level. You should know me better. For the rest of your comment, I fully agree with you.
Apart from this, you should read my previous post in conjunction with the preceding one.
Cheers, Edmond.
Of course I wasn't talking about a thd at 1kHz, let alone at 50 Hz or so. I meant at any frequency up to 20kHz and at any power level. You should know me better. For the rest of your comment, I fully agree with you.
Apart from this, you should read my previous post in conjunction with the preceding one.
Cheers, Edmond.
sophisticated amplifiers
BTW, Bob, there is a difference between sophisticated amplifiers and real sophisticated amplifiers. 😉
Let's say 20dB less THD and IMD, or, if you like, anything equal or better than Halcro.
Cheers. Edmond
BTW, Bob, there is a difference between sophisticated amplifiers and real sophisticated amplifiers. 😉
Let's say 20dB less THD and IMD, or, if you like, anything equal or better than Halcro.
Cheers. Edmond
Re: Re: sophisticated amplifiers
Hi Bob.
I don’t think that we need to get all that sophisticated to achieve these numbers.
For instance, in a low power amp we can cheat a bit by using class A. I’ve been fiddling with 1,10, & 20kHz notch filters based on some fancy opamps and as far as I’m able to resolve, the THD-20 of my amp here…..
http://users.picknowl.com.au/~glenk/12W.HTM
…..is <0.001% THD at 20kHz and rated Pout into 4R.
Don’t forget, D. Self did ~0.003% THD-20 with 2MHz fT BJT’s in an quasi-complementary class A amp with standard miller compensation! I’d like to see someone replicate those results with icky MOSFET’s in the output stage!
Cheers,
Glen
Bob Cordell said:
Hi Bob.
I don’t think that we need to get all that sophisticated to achieve these numbers.
For instance, in a low power amp we can cheat a bit by using class A. I’ve been fiddling with 1,10, & 20kHz notch filters based on some fancy opamps and as far as I’m able to resolve, the THD-20 of my amp here…..
http://users.picknowl.com.au/~glenk/12W.HTM
…..is <0.001% THD at 20kHz and rated Pout into 4R.
Don’t forget, D. Self did ~0.003% THD-20 with 2MHz fT BJT’s in an quasi-complementary class A amp with standard miller compensation! I’d like to see someone replicate those results with icky MOSFET’s in the output stage!
Cheers,
Glen
sophisticated amplifiers and simulation
Hi Bod,
Some second thoughts on sophisticated amplifiers and simulation.
First, maybe you felt offended by me and my (more or less arbitrarily chosen ) definition of "sophisticated'. It was certainly not my intention to categorize your design as non-sophisticated or less sophisticated. It definitely is sophisticated. No, I'm not "post flattering" my words, as I can prove here:
http://www.diyaudio.com/forums/showthread.php?s=&postid=1213620&highlight=#post1213620 post #1964
Second, regarding "Simulations don't count", it also depends on what is claimed. If somebody says "distortion is not affected by the quiescence current of the O/P stage", do you have any objections if a simulation shows that he is totally wrong?
Also, it depends on the complexity. No one would argue about a simulated response of a simple R-C filter (unless it's outrageously dimensioned, like 0.33mOhm and 68kF).
Also, it depends on proved "prior art". Now, I'm getting more serious. Let's look, for example, at my claim that "Transitional Miller Compensation" (TMC, see picture) markedly reduce the distortion. Quite a few people (not you Bob) are not convinced, because I never build it. The point is that 99.999% of a TMC amp has already been built and tested by others, that is, D. Self has measured that a two pole compensation does improve thd and E. M. Cherry has demonstrated a comparable improvement, by including the output stage inside the Miller loop. TMC is just a subtle variant on these two techniques and combines the advantages of both of them without the disadvantages. So, there is no need questioning the validity of my claim, which is only based on a simulation, and I have no intention to build the thing, just to satisfy the unbelievers.
Cheers, Edmond (not Edmund)
Hi Bod,
Some second thoughts on sophisticated amplifiers and simulation.
First, maybe you felt offended by me and my (more or less arbitrarily chosen ) definition of "sophisticated'. It was certainly not my intention to categorize your design as non-sophisticated or less sophisticated. It definitely is sophisticated. No, I'm not "post flattering" my words, as I can prove here:
http://www.diyaudio.com/forums/showthread.php?s=&postid=1213620&highlight=#post1213620 post #1964
Second, regarding "Simulations don't count", it also depends on what is claimed. If somebody says "distortion is not affected by the quiescence current of the O/P stage", do you have any objections if a simulation shows that he is totally wrong?
Also, it depends on the complexity. No one would argue about a simulated response of a simple R-C filter (unless it's outrageously dimensioned, like 0.33mOhm and 68kF).
Also, it depends on proved "prior art". Now, I'm getting more serious. Let's look, for example, at my claim that "Transitional Miller Compensation" (TMC, see picture) markedly reduce the distortion. Quite a few people (not you Bob) are not convinced, because I never build it. The point is that 99.999% of a TMC amp has already been built and tested by others, that is, D. Self has measured that a two pole compensation does improve thd and E. M. Cherry has demonstrated a comparable improvement, by including the output stage inside the Miller loop. TMC is just a subtle variant on these two techniques and combines the advantages of both of them without the disadvantages. So, there is no need questioning the validity of my claim, which is only based on a simulation, and I have no intention to build the thing, just to satisfy the unbelievers.
Cheers, Edmond (not Edmund)
Attachments
estuart said:
Edmond,
yes, it is quite easy to get THD of 0.002% at 1kHz, but it is not very easy to get it at 20kHz or 10kHz.
I have usually wondered why J.A. has made most of his spectral analysis at 50Hz, sometimes in seldom cases at 1kHz. I assume that the answer is quite clear - not to show what the systems under test are able to create in frequency domain ... 😀
Re: Re: Re: sophisticated amplifiers
Hi Edmund,
Just tell us a little bit about it (again, I apologize if you provided this information before and I missed it, or worse, forgot it).
Is the output stage MOSFET or BJT? If BJT, what ft devices?
Does it use EC in the output stage?
Is it Class AB, and what is the idle bias current?
If you simulated my amplifier and got 7 ppm THD-20, and I actually measured 6 ppm, that is very impressive!
If Halcro was coming anywhere near meeting its spec, virtually nothing would show on the AP and virtually nothing would show on the twin-tone spectral analysis. But things did show very clearly in both tests. Of course, there still could be a residual issue of distortion created by the particular dummy load connections, but this behavior was consistent across at least four reviews. Moreover, I saw at least one other review where the distortion appeared to be lower than that of the Halcro (I think it might have been the Boulder), so this suggests it is not the test equipment. But I do agree, this is a bit speculative. Nevertheless, bear in mind that even the measured performance of the Halcro is very, very good.
Cheers,
Bob
estuart said:
Hi Bob,
Yes, probably you did. In my post#1552, I was just a little bit joking, as somebody was referring to "sophisticated" amplifiers, which shows quite a lot of distortion. Therefore I stated: "Real sophistication starts at thd<1ppm, not at 0.02% or 0.25% in case of tubes."
Sure, you definition is adequate, but also 25 years in the future? 🙂
6ppm? NO, 7 ppm, according my simulation. 🙂
Maybe Candy is lying, but I never trust these measurements performed by such magazines. Besides, as you said, it's difficult to check the performance of a Halcro by means of an AP, for example. They should consult you to get reliable numbers.
Built? No, I'm retired and don't have the facilities any longer to build such amp.But I have designed amp's on paper with a thd (at 20kHz) around 20ppb. If some body give me the money and facilities to build such thing, I should really love it.
Well, it depends on the tools and models, and of course who is doing the simulation, it's really an art, just as building the real thing, to understand the power and limitations of a simulator.
Anyhow, my latest design shows, a thd (20kHz) of 20ppb. Maybe my sim is one order of magnitude wrong, that still leaves a 0.2ppm of distortion.
Cheers, Edmond.
edit: if you are really interested in my design, I'll send you the schematic (confidentially, of course).
Hi Edmund,
Just tell us a little bit about it (again, I apologize if you provided this information before and I missed it, or worse, forgot it).
Is the output stage MOSFET or BJT? If BJT, what ft devices?
Does it use EC in the output stage?
Is it Class AB, and what is the idle bias current?
If you simulated my amplifier and got 7 ppm THD-20, and I actually measured 6 ppm, that is very impressive!
If Halcro was coming anywhere near meeting its spec, virtually nothing would show on the AP and virtually nothing would show on the twin-tone spectral analysis. But things did show very clearly in both tests. Of course, there still could be a residual issue of distortion created by the particular dummy load connections, but this behavior was consistent across at least four reviews. Moreover, I saw at least one other review where the distortion appeared to be lower than that of the Halcro (I think it might have been the Boulder), so this suggests it is not the test equipment. But I do agree, this is a bit speculative. Nevertheless, bear in mind that even the measured performance of the Halcro is very, very good.
Cheers,
Bob