I have many friends who like build old amplifier (around 80an) of class B with huge crossover distortion, low slew rate, and high THD. They think it is the best amplifier. I have other friends who like tube amplifier and hate solid state amplifier. They think tube amplifier is the best.
In fact, this noise. Regardless of size. We can't hear.
Just like bacteria are 10000 times larger than molecules. We can't see bacteria either.
But ad827 thd+n is really high. But a lot of people. Including merchants. Users. They will tell you that ad 827 sounds better than ne5532.
The reason is that ad827 is more expensive. Or ad827 SR is higher.
This is obviously wrong.
In fact, in the audio industry. Maybe 95% of the ideas are wrong.
No one can change.
I don't like imitating the design of manufacturers.
But engineers from famous manufacturers. Display is the most professional group in the audio industry.
We can refer to their design scheme. Study their ideas. This is still very beneficial.
For example, most audio manufacturers use njm4580d. Or ne5534.
No engineers use the ad827 in sound. This is the difference between major and amateur.
From my measurements on output stage crossover distortion:
Worst to best, in class AB
Disqualified:
Quasicomp
CFP
Darlington
Lateral FETs
Acceptable:
Vertical FETs with resistors
BJT EF2 with resistors
Good:
Vertical FET without resistors at ~200mA bias
Fantastic:
BJTs without resistors at ~100mA bias
In fact, the example you give is contradictory.
Your friend's amplifier may be just your imagination.
Because if your friend can't deliberately make an amplifier with poor performance.
If he doesn't think this is a low-performance amplifier. So how do you know that its performance is very low. Don't guess what others have. Unless it's your own thing.
Any amplifier can be good. Or bad.
It's like a transistor. TTA1943 TTC 5200。
Or njw1302 3281. We can find them on the $30 machine.
They can also be found on a $100000 machine. For example, mark Levinson. So we can't regard similar things as the same.
As someone said. Asians are all short. But Yao Ming is also very good in NBA.
Digital power amplifiers are also good. For example, iraudamp4 iraudamp5
Including arc ds225 dsi200 Of course, there are many very poor digital power amplifiers.
For example, many USB mini speakers. Low cost car amplifier.
Low cost car amplifier.
But take the best hiend amplifier as an example. Almost all of them are class ab. Machines with more than $100000 do not have class D
Last edited:
"Hi-end" is like fashion. It is not about quality.
I and my friend design an amplifier and he and his friends compared it with branded amplifier that cost more than 3x. All of them think my amplifier sound better. My friend asked them if any chance they will buy my amplifier. They answered "NO!" Because my amplifier's case look ugly and cheap.
Ah, another self proclaimed prophet declaring intellectual deficit in all others to understand his wisdom.
Ever heard of Liner Audio magazine?
Linear Audio
Believe me, even in my spare time. All performance indexes of class D amplifier. Including sound. Will not exceed class ab
In an amateur situation. We can make the smallest amplifier of thd. It is lm4702. Unfortunately, this chip is not easy to buy. This is a simple amplifier that anyone who has DIY for the first time can do well. Thd less than 0.00002%
I quite understand what you mean.
I thought the same as you 20 years ago.
But in fact, I also like the manufacturer's machines very much now. For example, class 460 sounds great. Made in Canada. Not ClassD. It is class. It's a brand.
Of course, we will try our best to design better products. But please don't always belittle those professional teams.
Last edited:
I haven't heard of it. In fact, my statement is a little exaggerated.
In fact, among the clients I know. 90% of them prefer class ab.
Or class A. Although I think class A is meaningless at all.
The best class D amplifier that I ever heard (at high-end audio exhibition several years ago) still worst than the best class AB and A that I ever heard.
I agree with your opinion. Although maybe your taste and my taste is different. But you are lack of understanding about relationship between measurement and sound perception.
Class D still improving. It huge improvement from 10 years ago. Now they use GaN mosfet which can switch in very high frequency. It is like when BJT power transistor have fT 4 Mhz at old day and now we have 30 Mhz and even 60 Mhz.
Point is that, at diyAudio,, there is a lot of members with vast knowledge and even larger practical design experience. They know all about subjective vs. objective performance of audio circuits and then they know even more. Look up who is Jan Didden.
Discussing with such members, using car wheels and bacteria size analogies instead of technical circuit properties, doesn’t leave impression that you really understand what is discussed about.
The different about amateur and professional engineer is:
1. Amateur analyse engineering product using statistic, like they did in social science.
2. Professional engineer analyse engineering product using physics.
1. Amateur analyse engineering product using statistic, like they did in social science.
2. Professional engineer analyse engineering product using physics.
In fact, I know what you think. Because I have met many people.
They think the same as you. Some of these ideas are correct. Some are incorrect.
Because I like to study these theories more deeply. I find that many of the theories you believe are actually incorrect.
For example, the frequency of class D is higher. Better performance. In fact, this is also incorrect.
For example, we can find it in iraudamp5. Its recommended PWM is 400k.
In fact, it can work at 900K But in fact, the same amplifier. We set PWM to 900K. Its performance is degraded. Instead of improving. Relative to the frequency of 400k.
Because the higher the frequency. Then MOSFET TRR TFF rises. The process of decline will be more.
This will increase the calorific value of MOSFET. Thd will also rise.
I found this phenomenon when designing L30D irs2092 irfb4227.
But a higher PWM frequency has one advantage. It can reduce the capacity of LC.
Lower the voltage of PWM output. But I don't think it will make much difference.
I recommend that the PWM working frequency not be too high.
Why should we agree with you? Most of what you say is wrong.
You are not honest, first you say, I used JR 250, then when I point out that it is class D, you say I don't use JR 250.
You still don't understand the difference between an analog and a digital amp even after it is explained.
You are ignorant on the latest class D developments like Purify and the latest from Hypex, which are demonstable better than class A.
Apparently you are suffering from Kruger-Dunner.
Come back when you are really interested in dialog and discussion instead of throwing up nonsense.
Jan
You are not honest, first you say, I used JR 250, then when I point out that it is class D, you say I don't use JR 250.
You still don't understand the difference between an analog and a digital amp even after it is explained.
You are ignorant on the latest class D developments like Purify and the latest from Hypex, which are demonstable better than class A.
Apparently you are suffering from Kruger-Dunner.
Come back when you are really interested in dialog and discussion instead of throwing up nonsense.
Jan
We have many simple methods. For example, instrument is my best teacher.
He won't cheat me. isn't it?
I can design 0.0015% class AB amplifier for $3, such as mx50se. And many people use it. No one said it sounded bad.
I just don't want to talk about my products. Because I don't think this kind of discussion will be fair.
Everyone doesn't like to talk about the shortcomings of their own things. Ha.
I have used many amplifiers. Far more than Jeff rowland250
Even Krell ksa50, ksa250 and kav400, I don't remember how many amplifiers I have.
I like to take them apart. Anatomy. Then sell it.
Is there a problem.
This does not mean that I use this machine now.
Class A is originally a commercial advertisement.
You should not believe that class a will be better.
Class A is originally an advertising word. It can only cheat those who don't know how to cook n
A really professional person. Class a B C D is not used to judge the amplifier.
I suggest you learn about self's class B ultra low THD
I have all the class D amplifiers you mentioned. For example, Jeff rowland B&O.
ICE POWER HYPEX UCD MCD IRS TA20XX。
I don't even remember what I designed. Because there are too many.
If you are interested, I can show you many photos.
I also disassembled Goldmund S330. In addition, I copied its PCB.
In addition, I also disassembled proceed amp2 There are too many machines. I can't remember too much for the time being.
Attachments
