Hello.
I was thinking that I would love to build an amplifier of my own design. The problem, however, is that it doesn't seem like there's much use in designing a new amplifier when there are already so many other great ones out there.
So I was thinking to myself, wondering why people still keep building more amplifiers.
Naturally, they do it because it's fun. I understand that perfectly. Those who remember two years ago will attest that I still loved playing with the simulator even though I had little idea how anything really worked.
However, I still have little idea what in an amplifier makes it desirable, other than low component count, low distortion, high efficiency, and really neat design.
So, asking those who design amplifiers, I would like to know something. If I were to design an amplifier on which to stamp my own signature, what all would I be aiming for? What makes an amp great?
- keantoken
I was thinking that I would love to build an amplifier of my own design. The problem, however, is that it doesn't seem like there's much use in designing a new amplifier when there are already so many other great ones out there.
So I was thinking to myself, wondering why people still keep building more amplifiers.
Naturally, they do it because it's fun. I understand that perfectly. Those who remember two years ago will attest that I still loved playing with the simulator even though I had little idea how anything really worked.
However, I still have little idea what in an amplifier makes it desirable, other than low component count, low distortion, high efficiency, and really neat design.
So, asking those who design amplifiers, I would like to know something. If I were to design an amplifier on which to stamp my own signature, what all would I be aiming for? What makes an amp great?
- keantoken
Hi, keantoken!
I do not design amplifiers; however if you start to plough through Self, Slone, Duncan, the articles of Nelson Pass, Walt Jung, Borbely and others you start to get your own ideas how an amp topology should look like. (electronics and electrodynamics knowledge greatly help you with this).
Of course that is not the same as building all famous amps and listen to them yourself. That's also a nice way to make your own mind up (maybe a bit expensive, though).
The only advice I can give you is to find out for yourself! Asking other people for advice on such a war topic gives you more different replies than users registered here 😀
Have fun, Hannes
I do not design amplifiers; however if you start to plough through Self, Slone, Duncan, the articles of Nelson Pass, Walt Jung, Borbely and others you start to get your own ideas how an amp topology should look like. (electronics and electrodynamics knowledge greatly help you with this).
Of course that is not the same as building all famous amps and listen to them yourself. That's also a nice way to make your own mind up (maybe a bit expensive, though).
The only advice I can give you is to find out for yourself! Asking other people for advice on such a war topic gives you more different replies than users registered here 😀
Have fun, Hannes
Why don't you listen, and read reviews of, amps with known topology, transitor types etc, and try to see some kind of read thread, and then start to design a "new" amp with a type of topology and transistors you find interesting.
keantoken said:
It seems some amp designers approach this as a purely intellectual exercise, in the same way they would a crossword puzzle. I don't understand them. As the amp is ultimately only a tool in the reproduction of music, the concept of the design should reflect the designer's attitude towards sound.
Single ended? Negative feedback? Desired distortion spectrum? Clipping behavior?
If you don't know your own preferences, why would you start a new design? Just to kill some time over the weekend? To start a discussion with other similarly bored designers? To publish a magazine article?
Thanks, Hann.
I've been searching names here for about an hour and haven't gotten anything but books. I can't order books. Or kits. I don't have $$.
Are there any online schematics or resources?
I suppose searching these forums would be a must, since there are 'official' amplifier threads and such.
I'm mainly interested in the schematics and theory involved. Doug Self's site was perfect for me because it's all circuit and theory, and what he says he says so darn well.
Can you point out some quick links?
Thanks,
- keantoken
I've been searching names here for about an hour and haven't gotten anything but books. I can't order books. Or kits. I don't have $$.
Are there any online schematics or resources?
I suppose searching these forums would be a must, since there are 'official' amplifier threads and such.
I'm mainly interested in the schematics and theory involved. Doug Self's site was perfect for me because it's all circuit and theory, and what he says he says so darn well.
Can you point out some quick links?
Thanks,
- keantoken
Sure!
Read all of Nelson Pass' Zen articles and his legacy projects at
http://www.passdiy.com/
Erno Borbelys articles are here
http://www.borbelyaudio.com/special_articles.asp
When a bit advanced, you may find some interesting reading from Walt Jung here (especially his articles on regulators are a must)
http://waltjung.org/Library.html
Further advanced you can read Bob Cordells AES article on his Mosfet amp which features error correction, rarely seen unfortunately
http://www.cordellaudio.com/poweramp/
If you're into opamps, Walt Jungs Op Amp Applications Handbook might give you ideas
http://www.analog.com/library/analogDialogue/archives/39-05/op_amp_applications_handbook.html
and of course Ron Mancinis 'opamps for everyone'
http://focus.ti.com/lit/an/slod006b/slod006b.pdf
I guess this should keep you occupied 😀
Have fun, Hannes
Read all of Nelson Pass' Zen articles and his legacy projects at
http://www.passdiy.com/
Erno Borbelys articles are here
http://www.borbelyaudio.com/special_articles.asp
When a bit advanced, you may find some interesting reading from Walt Jung here (especially his articles on regulators are a must)
http://waltjung.org/Library.html
Further advanced you can read Bob Cordells AES article on his Mosfet amp which features error correction, rarely seen unfortunately
http://www.cordellaudio.com/poweramp/
If you're into opamps, Walt Jungs Op Amp Applications Handbook might give you ideas
http://www.analog.com/library/analogDialogue/archives/39-05/op_amp_applications_handbook.html
and of course Ron Mancinis 'opamps for everyone'
http://focus.ti.com/lit/an/slod006b/slod006b.pdf
I guess this should keep you occupied 😀
Have fun, Hannes
The thing about desgining and building an amplifier is that it's concrete evidence that you understood the theory and correctly interpreted the datasheets.
I build a lot of prototypes for a wide variety of functions. Be they ever so simple, I always get a kick when they work. Perhaps it's not very admirable that I feel the need for constant self-reassurance like this, but the understanding of a lot of this stuff felt hard-won in a lot of instances, and although I do not regard myself as stupid, I'll take any encouragement I can get.
The primary consideration in the design of any (linear) amplifier is that it should exactly replicate the input signal at higher power.
How well it works is, however, sometimes secondary to the fact that it works at all...
It's nice to be able to pick up an old tube, or a bunch of components and turn them into something useful.
Modular and discrete amplifiers of more than adequate performance for most purposes are widely available at economic prices. From that point of view there is little to be gained by designing an amplifier, although there are geniune advances taking place in digital and class D amplification.
w
I build a lot of prototypes for a wide variety of functions. Be they ever so simple, I always get a kick when they work. Perhaps it's not very admirable that I feel the need for constant self-reassurance like this, but the understanding of a lot of this stuff felt hard-won in a lot of instances, and although I do not regard myself as stupid, I'll take any encouragement I can get.
The primary consideration in the design of any (linear) amplifier is that it should exactly replicate the input signal at higher power.
How well it works is, however, sometimes secondary to the fact that it works at all...
It's nice to be able to pick up an old tube, or a bunch of components and turn them into something useful.
Modular and discrete amplifiers of more than adequate performance for most purposes are widely available at economic prices. From that point of view there is little to be gained by designing an amplifier, although there are geniune advances taking place in digital and class D amplification.
w
About 10 years ago I built a well know D. Self based kit amplifier. This measured well and sounded like a D Self amplifier is reported to sound like. This amp has served as a useful reference for subsequent "next amps" which really is a must if your objective is listening to music. If you already have an amplifier and are very familiar with the cct topology then that will also be fine.
A few pointers
1. It's often best to start with a scaled down version of your cct such as a preamp. This is cheaper to build and easier to protect which means less $$$ going up in smoke when something doesn't quite work out as planned. When you're happy with the design after mods etc. you simply up scale the voltage rails and supply currents as required.
2. Some test equipment is mandatory such as DVM, ac mV meter, 1 MHz Audio generator with a nice sqare wave and if you can buy or borrow then a 100 MHz CRO is really handy and in my view a must have. If you're aiming high then things can and usually do oscillate at RF frequencies. I've even had parasitics as high as 400 MHz.
3. The reason people build more than just one amplifier is usually because it is very much a learning curve and we all learn from our design shortcomings and open our eyes to new ideas such as those presented here in DIYA from well respected pro and amateur designers. The holy grail of amplifiers is a limit we can never achieve but nonetheless is one we all strive for.
This is your mission if you wish to choose it.
Good Luck
A few pointers
1. It's often best to start with a scaled down version of your cct such as a preamp. This is cheaper to build and easier to protect which means less $$$ going up in smoke when something doesn't quite work out as planned. When you're happy with the design after mods etc. you simply up scale the voltage rails and supply currents as required.
2. Some test equipment is mandatory such as DVM, ac mV meter, 1 MHz Audio generator with a nice sqare wave and if you can buy or borrow then a 100 MHz CRO is really handy and in my view a must have. If you're aiming high then things can and usually do oscillate at RF frequencies. I've even had parasitics as high as 400 MHz.
3. The reason people build more than just one amplifier is usually because it is very much a learning curve and we all learn from our design shortcomings and open our eyes to new ideas such as those presented here in DIYA from well respected pro and amateur designers. The holy grail of amplifiers is a limit we can never achieve but nonetheless is one we all strive for.
This is your mission if you wish to choose it.
Good Luck
Hello Keantoken.
I am an amplifier designer.
Of course, what you should be aiming for depends entirely on what you are trying to get out of this undertaking. What do you mean by "stamp my own signature"?
I am an amplifier designer.
I design them because there are very few out there that are great.
I only care about how faithfully the recorded music reaches my ears.
To me a great amp is one that has no sonic signature. It is an audibly invisible link in a chain. I do not feel qualified nor do I presume to add my own flavour to a musicians' creation.
Of course, what you should be aiming for depends entirely on what you are trying to get out of this undertaking. What do you mean by "stamp my own signature"?
traderbam said:
That's fine for a commercial design, but i like to build amps that add a little bit of "flavour" to the music, the exact amount to make me feel at home instead of feeling inside a white cold anaechoic chamber, anyways, we are in the sub 0.05% THD, so probably the same musician performing in a different room would sound even more different.
traderbam said:
Ah, the 'Mountain Climber' approach.
Very much a solo pursuit.
There are lots of amplifiers out there that are great for listening to music. Particularly the ones large enough to drive speakers for say, a party.
w
Let's keep our eyes on the prize.
keantoken said:
Speaking only for myself, a great amp must satisfy me, and no one else. And it must satisfy me on multiple levels.
First and foremost it has to sound good to me. And to that end I don't care about the intentions of anyone upstream. It gives me no pleasure to to know that I'm perfectly recreating what someone upstream may have intended if at the end of the day it doesn't sound good to me.
It must also satisfy me aesthetically. It has to look good. Even if it may sound good, I don't want something ugly sitting on the shelf staring back at me.
Finally, it must satisfy me philosophically. I rather like the elegance of simplicity and like to see this expressed both on the inside and the outside.
But this is just how I look at it. You may find it appealing or you may not. The only advice I would urge you to consider in answer to what you should be aiming for is that which satisfies you. Without any consideration as to what may satisfy someone else. And the only person who can ultimately determine that is you.
Good luck!
se
traderbam said:
That's sounds all well and good on paper, but in reality, it presents something of a conundrum.
It presumes that the entire playback system used to create the recording in the first place was essentially perfect and had no sonic signature of its own. While some would argue that this is rather easily achieved for electronics, not even the most hardened, objectivist "nay-sayer" would claim that this has been achieved when it comes to loudspeakers.
So given that, how could playing a recording back on a system which is for all intents and purposes "perfect" and having no sonic signature of its own, be faithful to what the person producing the recording had intended?
se
I agree...I really meant amplifiers that are great in terms of what I am looking for.wakibaki said:
Curiously, I am a mountaineer. But I don't climb solo...it is too dangerous.
I agree. There are lots of ways to skin a cat. I have mine. I know what you imply about some "so called" accurate amps sounding cold and perhaps a little sterile but my experience is that this is because they are not accurate. A transparent amp should let both the clarity and the warmth of the sun shine through, if you will.ionomolo said:
I would agree if the nature of the distortions in each link were similar. My personal experience is that they are not and that the unnatural, electronic distortions that power amps can create are particularly displeasing to my ears. Much more so than typical loudspeaker distortions, for instance. When I talked about "no sonic signature" that doesn't require no distortion at all, but no distortion would be ideal. My experience is that some types of distortion are more polluting than others.Steve Eddy said:
keantoken said:
I can relate, and I feel you CAN have a UNIQUE amp that sounds GREAT and does not color the sound, but plays it more accurately than any consumer-grade amps.
I design amps, and I'm not a professional, but a DIY guru that enjoys the hobby and the rewarding feeling of building something yourself that is superior to anything you can buy for the same cost.
What I aim for to make a great unique amp......
1. Cool Temperatures - most amps you buy run the heatsinks hot enough to burn you, however, I prefer a heatsink cool enough, so you can hold your hand on it, and also use fans whenever possible
2. Durability - Oversized heatsinks, transformers, and overrated transistors put in more parallel pairs than necessary for the design to last a long time under heavy load
3. Drive capability - great bass output without distortion, by using large capacitors and power supply components, and high gain output transistors - to have stronger drive than the avg amp
4. RMS power - build the amp with REAL RMS power, and not exaggerate the power output like consumer amps do, actually I prefer to under-estimate it so when the amp is built, there's more power than planned
5. Supporting circuits - Circuits other than the amplifier itself.....Do you want simple on/off, or a fancy electronic on/off circuit with speaker-on delay that also protects the amp and shuts off during a fault? Also your preamp and power for that applies here too.
6. Looks - Make a nice case that looks great and is functional for proper cooling. And also whether or not the heatsinks are on the inside or outside. Last Nice case I made was from gray-wood paneling outside, with wooden frame inside, and two fans on the back. Heatsinks all inside. Looks clean, looks good, and very functional 😎
Now these six things vary a LOT between amp builders, so I agree that you can always design new amps, and the amp design is more than just the schematic for the amplifier itself, it's the WHOLE complete project.
i like to design amps because i can test my ideas on what works well or what doesn't. do a lot of amp repairs, so i get a "from the trenches" view of various manufacturers good and bad ideas (since i repair them i see more of their bad ideas). i have an idea called the "universal amplifier". since no two people hear the same thing exactly the same way, people have different opinions of what an amplifier should sound like. if everybody did hear everything alike, there would be a "standard amplifier" amd everybody would be happy. my "universal amplifier" idea is that if an amplifier could be made to have some of it's operating parameters adjustable or presettable to match peoples preferences, then it might be a successful product. i don't want to "cheat" by using DSP (like some guitar amp manufacturers do). i want all of the solutions to the challenge to be analog. also many of the things that make amps sound different are very subtle but measureable. another thing i shoot for is reliability. i see a lot of "DUH!" mistakes in commercially made equipment, such as electrolytiv caps right next to heat sinks and high wattage resistors, 1 amp regulators run at 950mA, and other assorted snafus. also the mech engineers that design the physical part of the amp must think "no one will ever have to repair this" or they majored in chinese puzzle design. so my "universal amplifier" will also have simple and logical construction. it won't have the power supply or amp boards buried somewhere deep in the chassis.
keantoken said:
It depends what you want from an amp.
An audiophile will probably want something that has very low distortion.
A rock guitarist will be looking for a valve amp or a valve pre amp to give out lots of distortion to richen and fatten the sound.
I built a valve pre amp mixer that does both, it has 3 clean channels for my mobile disco and one dirty channel for my guitar.
- Status
- Not open for further replies.