So there we have it. The choice is to follow the electrical textbooks, which tell you about how components and circuits work (based on physics and maths) so you can gain understanding, or to follow recipes from audio gurus who have either never read the textbooks or failed to understand them and show this by designing circuits which cannot work properly because they contain basic errors.
It is as though we are being asked to admire a new car, when we notice that the exhaust is hanging off and about to part company from the engine and one of the wheels is missing. We are told by the proud designer that the loose exhaust brackets are part of an advanced engine tuning system and the missing wheel helps the car turn corners. Meanwhile, to improve the acceleration the car has significant amounts of 'go-faster' stripes added all over the bodywork and blue LEDs underneath which are said to keep the road warm and reduce the risk of ice. We decline the opportunity to read the thoughts of the Masters of which this car designer is the Apprentice.
It is as though we are being asked to admire a new car, when we notice that the exhaust is hanging off and about to part company from the engine and one of the wheels is missing. We are told by the proud designer that the loose exhaust brackets are part of an advanced engine tuning system and the missing wheel helps the car turn corners. Meanwhile, to improve the acceleration the car has significant amounts of 'go-faster' stripes added all over the bodywork and blue LEDs underneath which are said to keep the road warm and reduce the risk of ice. We decline the opportunity to read the thoughts of the Masters of which this car designer is the Apprentice.
The other section with 0.48H is even worse. Mr Drlowmu fails to understand that an undamped or weekly damped resonance in the audio range will be finally damped by the load. Unfortunately such load is the 2A3 which should NOT be supposed to do that in order to give an amplified copy of the input signal. There is no free lunch. Who cares what the capacitors do if the amplifier is not linear anymore? He likes distorted sound obviously. Nothing agaist it. What I don't really like is the attitude. He doesn't know for example that many times when I built protoptypes I made power supplies this way becasue that's what I had on the bench. Then when I was satified with the amplifying circuit I put in place a proper supply I could hear the difference. I know how this poor LSES supply works and sounds. It's just RUBBISH.....in a proper HiFi system!
He probably can't even spot it by listening because he has serious issues elsewhere....
Last edited:
I am not seeing a fuse or fuses, and it looks like there is no ground wire in the power cord. Maybe it is just my eyes.
Win W5JAG
Win W5JAG
Member
Joined 2009
Paid Member
I've read about people's impressions of Dennis Frakers amplifiers from people who had 'no dog in the hunt' and had heard them at an audio show - their response was wholly positive about the subjective performance in comparison with other good amplifiers. I don't doubt that Dennis knows how to design and build a very good sounding amplifier. SET amplifiers are not known for technical excellence but for producing great satisfaction.
I've never met you Jeff, but you are in amongst engineers on this forum and I fear there is going to be some of the usual 'oil and water' issues when you try to mix your subjective experiences into this along with a burning desire to rescue 'us' engineers from our dogmatic adherence to the laws of nature 😀
I am willing to give the benefit of doubt to some of these ideas, as soon as I get time to build and experiment - I keep getting distracted by other projects !
I've never met you Jeff, but you are in amongst engineers on this forum and I fear there is going to be some of the usual 'oil and water' issues when you try to mix your subjective experiences into this along with a burning desire to rescue 'us' engineers from our dogmatic adherence to the laws of nature 😀
I am willing to give the benefit of doubt to some of these ideas, as soon as I get time to build and experiment - I keep getting distracted by other projects !
I wish someone would just build one of these amps and compare to a conventional amp, rather than theorizing endlessly. I don't have the time now, though I did build a PP 2A3 with a power supply inspired by Jeff/ Dennis thoughts many years ago. I don't have any complaints.
Hello Jeff,
I find your reply a bit strange, as you have zero information on my background, experience, etc., so to make such bizarre assumptions about me is somewhat absurd. You're clueless to know anything beyond my forum presence and yet quickly jump to conclusions which are meaningless.
You easily discount solid, proven technical design and engineering practices that have been proven through time. You also seem to imply that the 2A3 has some special operational features/requirements that force a specific power supply configuration that only 2-3 people on the planet actually grasp and understand, while convincing yourself that the rest of the world is daft, ignorant and incapable of understanding what you simply gloss over in "marketecture" and vague descriptions at best. Yet, you think you be taken seriously by everyone and nobody is really qualified to challenge whatever it is you're trying to push.
As nothing in your posts or writing to date have any factual data, such as some simple measurements, i.e., S/N ratio, bandwidth, distortion, etc., it's difficult to take you seriously where you offer nothing beyond self praise for your own equipment while at the same time make it a point to cleary state that virtually every other SET on the planet is junk and useless... how many SET amps have you actually heard to feel qualified to make such a statement? You want people on the forum to read compilations (I'm guessing these to be white papers, unless those are also old school and as such of no value to your new world order) on the LSES power supply, which you claim will enlighten, elevate and separate them from the unwashed masses. So just post them here or provide a link where they can be downloaded. How hard is that?
I'll ask, but I don't think I'll get an answer, but, can you provide any actual performance data on your amplifier(s)? Have you given any thought to my simple proposed test for how well your power supply design performs under real conditions? You want everyone to believe you, and yet all you really offer is to have us drink the "kool-aid" and follow. I don't think it's going to work out that way. Still, I'll be happy to read your compilations if you post them and/or provide a link... and of course Merry Christmas and have a Healthy and Happy Holiday season.
Regards, KM
If all were honest due to the imperfection of all of the devices we use to make our amplifiers, no matter how fine your understanding of the basics of electronic design, they all end up being effects boxes.
The engineer's main task is to build a product that turns on when you ask it to and does not burn your house down. How the product sounds is the purview of a very select group.
AS one can tell from this group of forums the vast majority of the engineers are more concerned with just about everything but how the component sounds while playing music. To refine it further: playing music that is not produced by other amplified boxes of speakers but actual instruments.
The guys in the effects boxes business usually come up with odd sounds just by trying things. Their concern is anything but fidelity to the original.
Akin to Medical Doctors who have a real idea of what people should eat to avoid disease as opposed to just remaining alive.
When one complains about another's moral indignation they might reflect on their own indignation, moral and otherwise.
The engineer's main task is to build a product that turns on when you ask it to and does not burn your house down. How the product sounds is the purview of a very select group.
AS one can tell from this group of forums the vast majority of the engineers are more concerned with just about everything but how the component sounds while playing music. To refine it further: playing music that is not produced by other amplified boxes of speakers but actual instruments.
The guys in the effects boxes business usually come up with odd sounds just by trying things. Their concern is anything but fidelity to the original.
Akin to Medical Doctors who have a real idea of what people should eat to avoid disease as opposed to just remaining alive.
When one complains about another's moral indignation they might reflect on their own indignation, moral and otherwise.
Hi, fuse is buried beneath the chassis lip, at the bottom of the amp. As is a CL-90 thermistor for start up purposes. The second AC switch on the chassis is an " Attended Listening " switch, which shorts-out the thermistor and the line fuse, so that the amp plays even more to my personal liking on my ALTEC movie speakers. Quite audible. Its not a for-production amp, and, being a DIYer, I can have what ever I want .
The line cord is only two conductor, correct. Two Fulton lengths ( 114 1/4 inches ) of Fulton Brown Speaker Wire.
Thanks for looking so carefully.
Jeff
Hi KM,
I stand 100% behind what I say.
I am no computer expert.
The Compilations from EEs, Messers Hasquin and Swenson, are a large amount of material to post. Since my computer skills are limited, I ask people to kindly email me, and I promptly return their email with the Compilations.
Feel free to do it that way, yourself. It will be an interesting read, I guarantee that. It LOOKS at a power supply from a different perspective, that what people are typically taught. The best thing of all is , it works, and it sounds great .
Merry Christmas.
Jeff Medwin
Most engineers try to ensure that their circuits don't sound at all, but merely pass on a bigger version of the input signal. Merely requiring that the item does not burn down the house is easier, as an empty box satisfies that requirement. So engineers are very concerned about how their item sounds when playing music, and do their best to eliminate it so listeners just hear the original recording (as far as rooms, speakers etc. will allow).rickmcinnis said:
Member
Joined 2009
Paid Member
You speak for yourself. I engineer my DIY products to meet a set of goals, which usually go beyond 'merely passing on a bigger signal' and from reading the presses so do a lot of others.
Perhaps there's a perception problem here. In general, a power amplifier should simply "amplify" the signal fed to it. That implies that the amplifier does not add any audible signature of it's own, meaning it's a faithful reproduction of the original signal (as much as is feasible of course). If your design goals are not to achieve this, then in the simplest of terms, you are "NOT" designing a power amplifier. What you are designing is some level of effects processing that also provides some degree of amplification. This is perfectly fine, just as long as you don't use industry standard terminology and twist it around to serve some warped personal requirement or perception. Just call it what it is, how hard can that be?
Case in point relative to this thread. By industry definition, a power supply has but a simple function. That is to provide a stable and clean voltage supply to the circuit it powers, nothing more. This implies it is not frequency sensitive, it's output doesn't waver at certain frequencies. It also doesn't sag or develop any noise and/or harmonics due to dynamic load changes. In short, it's a "power supply". Again, if this is not your design goal, then don't refer to it a power supply, call it what you designed it to be. Again, how hard is that??
Getting to the bigger picture, the rest of the amplifier should have similar goals. In short, signal/noise is paramount, period. Flat frequency response for at least 10 octaves should be expected/realized. Dynamic linearity should be intact throughout it's entire output range. Finally, distortions of any kind should be minimized as best as possible. This includes, traditional THD, IM levels as well as phase shift. Once again, if this is not your design goal, then don't refer to it as an amplifier. Call it what is it, basically it's a non-linear, phase producing, non-flat, frequency-dependent effects processor.
As an example, back in 2007, I did a clean sheet SET design. It took about 9 months (mostly spare time) to do the initial design, prototype it, finalizie the design, then engineer it into a product that could be produced into a working pair of amplifiers.
The amplifier uses all triode circuitry with a 45DHT. The design topology is a two stage, direct-coupled input/driver based on a 5814a. This provides ~43dB of voltage gain with sufficient slew rate and voltage to deliver more than 150 volts pk-pk and flat beyond 50KHz (driving the 45 grid). The driver stage is capacitive-coupled to the 45 triode, which uses an AC filament with fixed DC balance and adjustable AC balance. The power supply is a somewhat traditional Pi-filter driven by a 5V4G rectifier. With a simple RC decoupling to the input stage. Driver stage is on the same supply point as the 45 output stage.
Final working specifications are:
Rated output power: 2.0 watts - clipping at 2.25 watts
Frequency response: 25Hz - 50KHz within +0/-1 dB at 1-watt output
Signal/Noise: 80dB minimum noise (unweighted) referenced to 1-watt output (I have a few pairs that achieve close to 90dB S/N)
Measured THD at 1-watt: less than 0.4% at 1KHz, rising slightly at frequencies below 100Hz and decreasing a bit at frequencies above 10KHz.
In short, the amplifiers did exceed my original design specifications a bit, but I also swapped over 50 different 5814a tubes through the input/driver stage and over 60 different 45 triodes in the output stage to ensure the design was both stable and consistent. Point being, if you're going to design, engineer and build a product of any sort, having a realistic set of design goals and specifications are a basic requirement. Tossing something together via trial and error and listening to each variation whatever it is, doesn't qualify as design work... it's a hack, plain and simple.
Good luck to all as you endeavor to enjoy your hobby in the audio world. It's great to have a hobby you can enjoy. It's fine to do whatever you like as well. Just do the forum members a simple favor: Try and accurately describe what is it you're trying to design and build. Don't use standard industry terms to describe something you think qualifies, especially if you know it doesn't. It's that simple!
Best Regards to all,
KM
Case in point relative to this thread. By industry definition, a power supply has but a simple function. That is to provide a stable and clean voltage supply to the circuit it powers, nothing more. This implies it is not frequency sensitive, it's output doesn't waver at certain frequencies. It also doesn't sag or develop any noise and/or harmonics due to dynamic load changes. In short, it's a "power supply". Again, if this is not your design goal, then don't refer to it a power supply, call it what you designed it to be. Again, how hard is that??
Getting to the bigger picture, the rest of the amplifier should have similar goals. In short, signal/noise is paramount, period. Flat frequency response for at least 10 octaves should be expected/realized. Dynamic linearity should be intact throughout it's entire output range. Finally, distortions of any kind should be minimized as best as possible. This includes, traditional THD, IM levels as well as phase shift. Once again, if this is not your design goal, then don't refer to it as an amplifier. Call it what is it, basically it's a non-linear, phase producing, non-flat, frequency-dependent effects processor.
As an example, back in 2007, I did a clean sheet SET design. It took about 9 months (mostly spare time) to do the initial design, prototype it, finalizie the design, then engineer it into a product that could be produced into a working pair of amplifiers.
The amplifier uses all triode circuitry with a 45DHT. The design topology is a two stage, direct-coupled input/driver based on a 5814a. This provides ~43dB of voltage gain with sufficient slew rate and voltage to deliver more than 150 volts pk-pk and flat beyond 50KHz (driving the 45 grid). The driver stage is capacitive-coupled to the 45 triode, which uses an AC filament with fixed DC balance and adjustable AC balance. The power supply is a somewhat traditional Pi-filter driven by a 5V4G rectifier. With a simple RC decoupling to the input stage. Driver stage is on the same supply point as the 45 output stage.
Final working specifications are:
Rated output power: 2.0 watts - clipping at 2.25 watts
Frequency response: 25Hz - 50KHz within +0/-1 dB at 1-watt output
Signal/Noise: 80dB minimum noise (unweighted) referenced to 1-watt output (I have a few pairs that achieve close to 90dB S/N)
Measured THD at 1-watt: less than 0.4% at 1KHz, rising slightly at frequencies below 100Hz and decreasing a bit at frequencies above 10KHz.
In short, the amplifiers did exceed my original design specifications a bit, but I also swapped over 50 different 5814a tubes through the input/driver stage and over 60 different 45 triodes in the output stage to ensure the design was both stable and consistent. Point being, if you're going to design, engineer and build a product of any sort, having a realistic set of design goals and specifications are a basic requirement. Tossing something together via trial and error and listening to each variation whatever it is, doesn't qualify as design work... it's a hack, plain and simple.
Good luck to all as you endeavor to enjoy your hobby in the audio world. It's great to have a hobby you can enjoy. It's fine to do whatever you like as well. Just do the forum members a simple favor: Try and accurately describe what is it you're trying to design and build. Don't use standard industry terms to describe something you think qualifies, especially if you know it doesn't. It's that simple!
Best Regards to all,
KM
Member
Joined 2009
Paid Member
In the audio industry the term amlifier does not appear, from the marketing information I see, to have the meaning I think you believe it does.
Okay, care to post some references? Also, what do you consider marketing information?
Regards, KM
It's a waste of time kmaier. The rule is there are no rules! 😀
This thread shows clearly why SET amplifiers have become the best example of effect box while properly made and used they aren't.
This thread shows clearly why SET amplifiers have become the best example of effect box while properly made and used they aren't.
Dear KM,
Your 45 amp, and its design, does not impress me............ one iota.
I built two 45 amps, two stage, SET, in 2015 and 2016 and either one of them will, from your description, easily CREME your amp on music playback.
You don't have a clue !! Are you in Europe by any chance???
You made several FATAL and NON RECOVERABLE errors with your design.
(1) You describe your power supply as a pi filter. Was that as in L1/C1/L2/C2 ? If so, were any of the Ls going to the Finals over 20 Ohms in DCR. ?? They probably were !!
If so, "F minus" is your grade on the DHT power supply. That I learned in 1982, chokes need to be 20 Ohms or less. Its 2016, time for you to learn that.
(2) You choice of a driver tube, for a two stage SET was ALL WRONG. Medium mu , no good.
It needs to be a mu of 70 as the bare minimum, not 19 or 20, which will only sound OKish at best - when pushed. I prefer mu of 100 tubes, best of all, in a two stage 45 amp. A mu of 100 driver is a) more lively sounding, b) more FUN to hear, c) has a greater sense of " presence" and d) is always better-sounding when the amp is played at lower volume levels. One must be CAREFUL in implementing a high gain tube, to get it correct. Takes thought !!
(3) as SOON as you elected to capacitor-couple to the grid of the Type 45 tube, you TOTALLY "BLEW" any possibility for any REAL fidelity in your amp !!!
You can NOT cap-couple a two stage amp, into the DHT !!! Ever, and get the best result. The best result comes ONLY one way - it is to DIRECT COUPLE the driver tube to the grid of the finals, with about three runs of paralleled 18 AWG pure silver wire.
(4) Unless the 45 tube was an EML made tube ( with nice symmetrical cathode structures ) your choice of a 45 as the Finals tube was also bogus !!
I have not had the opportunity to hear the EML Type 45, but I discovered, as of September 2016, for SURE - one can not build a satisfactory amp from a NOS Type 45 tube. Sure, I hear the tube's nice top end, and I hear its surprisingly good bass response, but Jeff also hears BAD things in a Type 45, that no one seems to acknowledge, or want to even talk about.
KM, there is a shallow and very broad suck-out, in the Type 45, in the entire midrange of the tube, when it plays on high efficiency speakers ( ALTEC - alnico 515B, 802s ) . The entire midrange on the piano is NON LINERALY depressed, very broad, but shallow. The 45 CAN't do justice to a piano. No way. Its the NOS Type 45, with its stinking " M " or "W" non linear filament structure, that causes this lack of true and accurate fidelity. EMLs don't have this poor filament design, and are likely OK.
In October 2016, I ripped down both of my Type 45 NOS amps, and harvested some of their parts, for use in the amp you saw pictured at the AA asylum.
I am selling off all my 45 NOS, used 45 tubes, on eBay, to those who don't know any better, and will buy them. Since my enlightenment and lesson-learned, ( about how poorly NOS Type 45s play music is new to me, a few months old ), I thought it might be good to post my findings, for others to consider, and learn from.
If you have a NOS 45, borrow an EML and give a listen to it. I would guess, it would likely improve fidelity.
I don't give a hoot about how your amp measures into a static load. It won't play music as it should, it never could from the start, it never will, and it would NEVER satisfy me one iota. You are lost KM, get with it.
Jeff Medwin
Last edited:
Not surprisingly, nothing you've posted to date impresses me in the least either. You have zero facts to back up anything, and you probably don't own a single piece of test equipment, nor would you know how to use one if you did. Yet, you rave about paralled caps, multiple silver wire runs, black-magic design philosophies and puke on everything that's not yours. Then you parade around like the self proclaimed Messiah of Audio. You should lay off the Kool-Aid, seriously. You spend high sums of $$ for silver wire (I do as well), yet you use Hammond power transformers and chokes, which are really bottom basement junk components. Simply put, you're delusional!
You're telling me about EML 45 tubes being the way to go, yet state you've not heard them! I actually have a pair. Bottom line, they're NOT real 45 triodes. They don't measure the same, are huge and ring like crazy from any mechanical excitation. While testing them, I picked up a trumpet, played a few notes from 10 feet away (around 3 meters if you're in Europe) and they went off the chart! The internal structure is the cause, the structure has zero damping, so they are really lively. Needless to say, you will certainly get a "sound" from them, but not quite what you expect... then again, striving for anything but accurate sound seems your calling.
You're also stating my amp as being two-stage... it's not, it's a 3-stage amp (input/driver/final). I've also done numerous 2-stage direct-coupled amps, probably before your time, both with 45 and 2A3 tubes. If you actually know how to design a proper amp, the 45 triode is perfectly fine and doesn't suffer from the noted afflictions of your design. So to make such bogus claims that your amp would creme mine is utter nonsense. Once again, it's clear you exist in your own little world of make believe and go into a rant every time somebody won't agree you, grow up! Good luck in your delusional world.... check please!
Regards, KM
You're telling me about EML 45 tubes being the way to go, yet state you've not heard them! I actually have a pair. Bottom line, they're NOT real 45 triodes. They don't measure the same, are huge and ring like crazy from any mechanical excitation. While testing them, I picked up a trumpet, played a few notes from 10 feet away (around 3 meters if you're in Europe) and they went off the chart! The internal structure is the cause, the structure has zero damping, so they are really lively. Needless to say, you will certainly get a "sound" from them, but not quite what you expect... then again, striving for anything but accurate sound seems your calling.
You're also stating my amp as being two-stage... it's not, it's a 3-stage amp (input/driver/final). I've also done numerous 2-stage direct-coupled amps, probably before your time, both with 45 and 2A3 tubes. If you actually know how to design a proper amp, the 45 triode is perfectly fine and doesn't suffer from the noted afflictions of your design. So to make such bogus claims that your amp would creme mine is utter nonsense. Once again, it's clear you exist in your own little world of make believe and go into a rant every time somebody won't agree you, grow up! Good luck in your delusional world.... check please!
Regards, KM
Last edited by a moderator:
- Status
- Not open for further replies.