Hi Paulb
Thank you for your welcome, unfortunately I am once again leaving because of the mindless bullying from Millwood.
He hasn't a clue what he is on about, and it is not worth me making the effort to overcome the distraction.
Hi rod,
In view of your statment that I write rubbish, I have simulated your 101 amp, and found it wanting in exatly the same way that I imagined from seeing the component layout.
The loudspeaker damping is inductive - not resistive.
This was something that Peter Baxandall warned about in the '60/70s.
This amplifier will sound different to others with multiway dynamic loudspeakers because it will generate tweeter driving error potentials from mid-bass back emf. STATEMENT.
Thank you for your welcome, unfortunately I am once again leaving because of the mindless bullying from Millwood.
He hasn't a clue what he is on about, and it is not worth me making the effort to overcome the distraction.
Hi rod,
In view of your statment that I write rubbish, I have simulated your 101 amp, and found it wanting in exatly the same way that I imagined from seeing the component layout.
The loudspeaker damping is inductive - not resistive.
This was something that Peter Baxandall warned about in the '60/70s.
This amplifier will sound different to others with multiway dynamic loudspeakers because it will generate tweeter driving error potentials from mid-bass back emf. STATEMENT.
... and this applies to 99.9% of all commercial amps regardless of price. The impedance rises with increasing frequency, as there is less available feedback to keep it to miniscule levels.
In most cases, it will still be lower than the cable impedance at all frequencies of interest, and in all cases considerably lower than the impedances introduced by a passive crossover.
Graham, this is a non-issue as you well know, and your post is misleading by not quoting any values or frequencies.
Blatant rubbish. The error levels so generated are so low as to be negligible. The output impedance is < 25 milliohms at 10kHz! There are a great many amplifiers (and a gazillion cables) that can't come even close to that figure, and you know it (or should know it).
You criticise others for bagging you, yet bring it on yourself by mindless criticism with no figures quoted, no reference designs that 'solve' all the 'problems' you cite, and no contribution to the debate that helps to inform or enlighten.
Sorry squire, but you bring it upon yourself. This behaviour is much more typical of another forum, where the Black Knight* 'gurus' lambast anyone who has the temerity to expect proof.
* Black Knight - as in Monty Python. Denial to the extent that he wants to keep fighting after both arms and legs are severed.
While it is absolutely true that proof denies faith, the latter is flawed, and proof is either relatively easy (for effects that exist) or impossible (for effects that don't). So far, most of what you have spoken of here requires faith - a commodity that I find in short supply
Cheers, Rod
Graham Maynard said:
I thought you said that you had no access to a simulator.
As someone pointed out a while ago, amps are built to make sound, not simulated to make sound. so why don't you just build one P3A and measure it with your equipment and report back to us how it sounds.
Relying on a simulator to tell you how an amp sounds is always risky and usually inaccurate.
Very subtle
I would suggest always inaccurate. Simulators are good for checking instantaneous power dissipation, DC conditions and general behaviour of a circuit, but cannot be relied upon to determine real-world performance.I nearly forgot ...
Ultrasonic insect repellers have been discredited by just about every consumer magazine on the planet, but, even if it does work, it will cause immense discomfort to cats, dogs, birds, and the majority of other animals - and even very young children.
This is not an ecologically sound (pardon the pun
) idea, as should be obvious. Surely this cannot be the sum total of Mr. Maynard's contribution to DIY - are there any published reference designs that we can admire?Cheers, Rod
Graham Maynard said:
Hi, Graham:
I understand that there are two versions of the P101 amp from Rode, and Rode did not published component values nor devices used. so I was curious which of the two you simulated and what compoenet values / devices you used in your simulation.
Maybe what you found is more related to your simulator / simulation set-up / component or device selection than with the topology.
BTW, since you have the schematic / component value / devices, you must have bought RodE's PCB boards. so it should be easy to populate the boards with devices and let us know how it sounds.
We would always appreciate your input on sound quality.
This is an interesting thread. We have luminaries here, people who know exactly and precisely what they are talking about, we have mathematicians, and academics, but there is a dearth of comment connecting the topology with the way circuits sound.
Graham, who by any standard must be considered very smart, and who has published impressive designs in EW, on the net, and in this forum, has posited 'first cycle distortion'. He has even offered, at no obligation (or charge), a refinement of JLH's seminal Class A circuit. As one with a lousy math ability but deeply intuitive knowledge of Ohms and Kirchoff's laws, I have to admit I only partially understand what he is referring to, but I keep trying, and believe that since his writing is articulate and fluent it deserves genuine effort from all of us. One of the failings of the human animal is the tendency to deride what it cannot understand; this is, effectively, one of the most damaging traits of all since it is anti-intellectual. The effort should come from us, and now that he has explained himself we should grapple with our own understanding.
Rod has defended his design well. I know Rod, and I know his integrity and common sense, and all that he has said is sensible and correct AFAIK. I am confident his new mosfet amp sounds very, very good because I know a lot about this topology and the sonics it produces, though I cannot be definitive because I don't know his component values. I particularly endorse his comments on bootstraps, and would add that a properly designed bootstrap greatly adds to the subjective impression of deep, melodic bass. I would however take Graham to task on his 2-3 second time constant to avoid thumps; I have found bootstraps only need a 0.2-0.3s time constant to render any thump infrasonic, although this does move the voice coil to full excursion for about half a second at switch off.
Much of the thread is devoted to distortion in all its forms, and the more esoteric aspects of simulation. The point is well made that a simulation does not a listening make. Bravo, and bleeding obvious, I would have thought...... So would it not be more fruitful to identify topologies and operating points which result in superior sonics, and pare the subjectives down to simple observations like 'better bass, more impact', or 'fatter', or less prominent sibilance, or more body in the midrange? I realize most people's perceptions are different, but the marketplace, particularly for high end which must surely be our target here, is not concerned with dry, objective argument - only with the sonics. (Or perhaps the advertizing shows I'm wrong??)
I have just come to the end of a long development cycle for two of my own amplifier products. This has been focussed on improved sonics. there is a stock product, then a Level II upgrade, and soon a Level III. I don't believe there's any more in the basic design, but who knows? If I continue to read Graham's posts, I might find something else!! The improvements to the sonics are quite obvious to even a casual listener in AB testing on real music, and merely reflect different prices points for the market. This gives the consumer choice, and gradates that choice logically. Perhaps if more people did this sort of work the arguments about subtle technical points here might not be so incensing, and more people would connect the topology with the sound - cause and effect. For my part, I find Graham's comments interesting because he might well be giving technical reasons for many of the things I have found work better; this is invaluable for the practitioner, and a useful lesson for anyone trying to build a better amp.
In closing, the onus of proof is on each of us, not on Graham or Rod. They offer their insights, and explain them well. But we must open our minds, be polite, and bloody well listen, asking questions perhaps, unthreateningly, where we are unsure. Otherwise things quickly degenerate into name calling and the forum is a complete waste of time.......
Cheers,
Hugh
Graham, who by any standard must be considered very smart, and who has published impressive designs in EW, on the net, and in this forum, has posited 'first cycle distortion'. He has even offered, at no obligation (or charge), a refinement of JLH's seminal Class A circuit. As one with a lousy math ability but deeply intuitive knowledge of Ohms and Kirchoff's laws, I have to admit I only partially understand what he is referring to, but I keep trying, and believe that since his writing is articulate and fluent it deserves genuine effort from all of us. One of the failings of the human animal is the tendency to deride what it cannot understand; this is, effectively, one of the most damaging traits of all since it is anti-intellectual. The effort should come from us, and now that he has explained himself we should grapple with our own understanding.
Rod has defended his design well. I know Rod, and I know his integrity and common sense, and all that he has said is sensible and correct AFAIK. I am confident his new mosfet amp sounds very, very good because I know a lot about this topology and the sonics it produces, though I cannot be definitive because I don't know his component values. I particularly endorse his comments on bootstraps, and would add that a properly designed bootstrap greatly adds to the subjective impression of deep, melodic bass. I would however take Graham to task on his 2-3 second time constant to avoid thumps; I have found bootstraps only need a 0.2-0.3s time constant to render any thump infrasonic, although this does move the voice coil to full excursion for about half a second at switch off.
Much of the thread is devoted to distortion in all its forms, and the more esoteric aspects of simulation. The point is well made that a simulation does not a listening make. Bravo, and bleeding obvious, I would have thought...... So would it not be more fruitful to identify topologies and operating points which result in superior sonics, and pare the subjectives down to simple observations like 'better bass, more impact', or 'fatter', or less prominent sibilance, or more body in the midrange? I realize most people's perceptions are different, but the marketplace, particularly for high end which must surely be our target here, is not concerned with dry, objective argument - only with the sonics. (Or perhaps the advertizing shows I'm wrong??)
I have just come to the end of a long development cycle for two of my own amplifier products. This has been focussed on improved sonics. there is a stock product, then a Level II upgrade, and soon a Level III. I don't believe there's any more in the basic design, but who knows? If I continue to read Graham's posts, I might find something else!! The improvements to the sonics are quite obvious to even a casual listener in AB testing on real music, and merely reflect different prices points for the market. This gives the consumer choice, and gradates that choice logically. Perhaps if more people did this sort of work the arguments about subtle technical points here might not be so incensing, and more people would connect the topology with the sound - cause and effect. For my part, I find Graham's comments interesting because he might well be giving technical reasons for many of the things I have found work better; this is invaluable for the practitioner, and a useful lesson for anyone trying to build a better amp.
In closing, the onus of proof is on each of us, not on Graham or Rod. They offer their insights, and explain them well. But we must open our minds, be polite, and bloody well listen, asking questions perhaps, unthreateningly, where we are unsure. Otherwise things quickly degenerate into name calling and the forum is a complete waste of time.......
Cheers,
Hugh
The recent posts illustrate a couple fundamental problems with the "topology wars" business.
A- Even the best (you define what that is!) topology can be poorly implemented and the less than best can be well implemrnted.
B- All topologies involve a compromise - i.e., they give up one advantage in order to improve another quality. You just exchange one set of problems for another and your choice may be as simple as choosing the problems you are most adept at solving.
C- Not all trade-offs involve audible quailities, size, cost, and reliability are also considerations.
D- The specific application is relavant. As an example the fact that a particular design will sound highly offenseive when clipping may not be relavant in a case where the combination of power, speaker sensitivity, room size and lisening preference make the likelyhood of clipping near zero. The inverse cicumstance could also apply.
E- Judging a design just by the schematic is requires caution and caveats. I have read posts which state that "XYZ" won't work while in fact "XYZ" sits a few feet from me and is playing music quite nicely. Similarly the claim that "XYZ" will sound bad has ocassionaly met a similar fate.
F- I give much more weight to opinions from someone who has actually built what they opine about.
A- Even the best (you define what that is!) topology can be poorly implemented and the less than best can be well implemrnted.
B- All topologies involve a compromise - i.e., they give up one advantage in order to improve another quality. You just exchange one set of problems for another and your choice may be as simple as choosing the problems you are most adept at solving.
C- Not all trade-offs involve audible quailities, size, cost, and reliability are also considerations.
D- The specific application is relavant. As an example the fact that a particular design will sound highly offenseive when clipping may not be relavant in a case where the combination of power, speaker sensitivity, room size and lisening preference make the likelyhood of clipping near zero. The inverse cicumstance could also apply.
E- Judging a design just by the schematic is requires caution and caveats. I have read posts which state that "XYZ" won't work while in fact "XYZ" sits a few feet from me and is playing music quite nicely. Similarly the claim that "XYZ" will sound bad has ocassionaly met a similar fate.
F- I give much more weight to opinions from someone who has actually built what they opine about.
That must depend on one's point of view. In response to someone asking if anyone had built the amp, and how does it sound, GM wrote ...
As most people know, ESP is my sole source of income, and selling boards is the only way I can maintain my website and all the information thereon. Suggesting that someone should not build the amp because of an assumed 'flaw' in the design is an attack - on my design abilities, products and livelihood. From my perspective, GM's post was indeed an attack - polite though it may have been.
I do agree that the thread has gone off on various tangents for far too long, but there has been a lot of good information offered from many quarters - not all of it is directly related to the original topic, but information can be valuable regardless of how, when or where it is obtained.
First Cycle Distortion
At this time (and at the risk of appearing facetious), it would appear to be time to look at some of the claims made for this mysterious distortion mechanism, which has evaded detection by all other than GM.
What actually constitutes a 'first cycle'? Is it any of the following. Is it the first cycle ...
1 - of signal an amplifier ever reproduces
2 - of signal during a listening session
3 - of a musical passage
4 - of a bar
5 - randomly selected from all other cycles of signal
What determines whether it is a cycle that may be 'distorted'? Amplitude? There are low levels of noise present at all times, so what is the threshold level, and how is it determined?
Frequency? Are some frequencies more likely to suffer FCD than others? Again, what is the threshold and how is it determined?
I have seen other info on this forum where FFT was discussed in relation to FCD, but that is an inappropriate way to measure a single cycle - a subtraction method is more appropriate, where the input and output of the amp are subtracted from each other with an external circuit. After a complete null is obtained on the signal, any remaining signal is distortion.
Although it may be difficult to quantify, this method will at least show that the distortion exists. As discussed earlier, I have detailed two ways to achieve this, and other suggestions have been put forward as well.
This is just the kind of information that would allow others to duplicate the tests, and know what to look for. By omitting it, the idea is relegated to the apocrypha* basket. This may or may not be appropriate, but without more information, duplicated test results, facts and figures, we will never know.
(* Writings or statements of questionable authorship or authenticity - from dictionary.com)
Attacks
GM has (to some degree) invited the attacks that followed because he has not provided any verifiable facts. While these are abhored at some other forum sites, DIYAudio readers seem to prefer information that is backed up with data from real tests. This was sadly lacking in GM's posts, and he was asked on several occasions for facts and figures to prove his point.
Please don't see this as a further attack - that is not my intention here. I seek only to clarify the situation, and explain (in as far as this is possible) my reasons and rationale. Naturally, I cannot speak for others, but the majority of so-called attacks on GM were simply asking for proof (IMO at least - I could also be wrong, but I prefer to consider people to be of good intent until they prove themselves otherwise).
And yes, the insect repellant idea was off the wall, but no-one was to know that this was not the original purpose, until Graham explained how it transmogrified from a stray 'dog egg' preventer to an insect repellent. Fortunately, this has been clarified, and I for one apologise to Graham for any unintended offense.
Cheers, Rod
I couldn't design an amp if my life depended on it.
Thankfully Rod's site allows me to have some insite into this fascinating field. I'd hate to think the amount of hours he has put into this site for the DIY community and it is one of the best laid out sites I've seen for researching amps etc. Everyone should be thanking him for his work and from what I've read, most do appreciate what he does.
And yes, I have just bought this PCB and know that it will sound good (depending on my assembly skills) as Rod is not going to put a crappy design up as one bad design has a detrimental impact for ages. I have the utmost confidence in his designs and hopefully my assembly.
Well done Rod and keep up the good work and I'm sure this is the sentiment of most DIYers.
Thankfully Rod's site allows me to have some insite into this fascinating field. I'd hate to think the amount of hours he has put into this site for the DIY community and it is one of the best laid out sites I've seen for researching amps etc. Everyone should be thanking him for his work and from what I've read, most do appreciate what he does.
And yes, I have just bought this PCB and know that it will sound good (depending on my assembly skills) as Rod is not going to put a crappy design up as one bad design has a detrimental impact for ages. I have the utmost confidence in his designs and hopefully my assembly.
Well done Rod and keep up the good work and I'm sure this is the sentiment of most DIYers.
to all
Board,
For the record I was poking fun at GM when I posted that link to the insect repellant circuit. One of my many faults is a lack of tolerance for people who criticize others (seemingly)without compassion, humility or reason. For that I apologize to GM and the board.
Let us not forget there is no civilization without civility.
...and a little humor doesn't hurt
M
Board,
For the record I was poking fun at GM when I posted that link to the insect repellant circuit. One of my many faults is a lack of tolerance for people who criticize others (seemingly)without compassion, humility or reason. For that I apologize to GM and the board.
Let us not forget there is no civilization without civility.
...and a little humor doesn't hurt
M
Graham Maynard said:
I am very curious about this. Are you saying, Graham, that to reduce this "first cycle distortion", we should take out R10 / R11? Those are gate stoppers necessary for mosfets!
I guess those hitachi, RCA and IRF engineers are wrong in suggesting gate stoppers for their devices.
I also think Curl recommends the use of some sort of gate stoppers on fast BJT devices.
Graham Maynard said:
Graham, you later stated that you simulated the P101. With values unknown, how did you simulate the circuit?
Re: Hi !
Moderator hat on
samod, you have just commited a crime , threadjacking. Please start a new thread if your post has little or nothing to do with the thread topic. The post is moved to a new thread.
The thread and your post is here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=32746
samod said:
Moderator hat on samod, you have just commited a crime
, threadjacking. Please start a new thread if your post has little or nothing to do with the thread topic. The post is moved to a new thread.The thread and your post is here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=32746
yep
Peranders,
yes, tolerance is something I work on every day, that's all I can do.
Some may consider giving GM a free pass after belittling (call it what you want) Rod E's Mosfet design tolerance, I disagree.
And I plainly stated in my first response my words were my opinions. GM posted his first response as if it were Gospel.
Nuff said, I need to get back to work.
Enjoy your day,
M
Peranders,
yes, tolerance is something I work on every day, that's all I can do.
Some may consider giving GM a free pass after belittling (call it what you want) Rod E's Mosfet design tolerance, I disagree.
And I plainly stated in my first response my words were my opinions. GM posted his first response as if it were Gospel.
Nuff said, I need to get back to work.
Enjoy your day,
M
We, moderators can't be everywhere, but I think it's good to think (hard) how your post will be recieved. GM is fast on judging, I agree. If he have had Rod's amp and disected it into pieces it would have been a different ballgame.
First I thought Hugh's AKSA 50 and 100 was a piece of crap but I have learned that as long as I haven't listened to it, it will be untested with unknown capabilities. Crap maybe.
maybe 
maybe.This goes for all (me too): Try to separate opinions from things you actually know (through observations). To have a schematics and discuss that is only one part. The other part is the real thing. A good example is the Gainclone, the same schematics but with different performance due to real life things.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.