JLH1969 MOSFET Class A/AB amp

millwood

is it possible that you didn't have a decoupling capacitor/capacitors on the supply rail? a small but fast capacitor will do. try a 22uf electrolytic capacitor on the breadboard.

millwood

I did some testing but will post pictures tomorrow.

when the output is like 4vpp, I have about 10mvp ripple on my supply rail.

when the output is like 30vpp, I have about 60-80mvp ripple on my supply rail.

the transformer is a Plitron toroidal, rectified by a kpu8 bridge and filtered by two 2200uf nichicon capacitors. - nothing special.

scooby

See you after the test photos! Above waveform is not very good ah!

gogowatch

scooby,

The waveforms from my unregulated power was really bad indeed.  The waveforms from the notebook power supply are okay.

Cheers,

Biglee_163

单纯的整流滤波电路的纹波确实是如此的..要消灭这些纹波,你应该考虑使用CLC的方式.
究其原因,主要是你使用的JLH是单端线路,这种线路非常简单,但却存在一个非常致命的缺陷:其PSRR的能力非常低下!也就无法完全对抗电源中传输过来的纹波,不信你搞个差分的电路上去实验下,这点不稳压的电源纹波是可以有效抑制的.
使用开关电源,你的示波器中当然是看不到低频的纹波了,因为纹波的频谱已经转移到了较高的频段去了.而且真正好的开关电源不多,适合音频放大用的开关电源就更屈指可数了.当用了一段时间后你就会区别出开关电源(普通的)和老式模拟整流滤波电源的区别了.
为了甲类上模拟线形稳压电源不现实也非常不经济,因此还是建议楼上的兄弟用CLC的方式实验下.
有低频干扰的另一个原因还在于你的地线的排布是否合理.由于地线排布引起的瘟瘟声更是挥之不去的梦魇.

gogowatch

Biglee_163,

谢谢指导.  我会用CLC的方式实验一下的, 我用面包板的目的就是多实验呢!  但现在是上班时间, 不方便玩这些, 到下班时便可以做了.

我是正在努力学习中, 所以有很多问题,

1. 开关电源的高频纹波对整个系统将会有什麽影响?
2. 普通的开关电源为什麽不适合做音频电源?
3. D类放大器的高频纹波在输出端是不能避免的, 这样对声音有否影响呢?

Cheers,

漫步猫

请问，输出功率管换为IRFP240可以吗，元件数值需要改变吗？

gogowatch

漫步猫,

建议你先仿真一下, 然後才实验, 我都是用这方法去用手头上的元件的.

Cheers,

xiaolu36

我已经把1969小甲做为推动级，后面再接一独立的电流放大，这样既保留了小甲的音色，又无需使用大面积的散热片，夏天开它N小时都不怕。
在制作中把小甲的功放管换成中功率管，调小电流，末级不设反馈。
现在的问题是这样的接法会不会提高输出功率？独立的电流放大级不设反馈会不会阻尼系数变差？

millwood

biglee is right on about the poor psrr for SE amps.

differential input pairs actually suffer from similar problems, but only on the tail side.

however, if you use a current source there, you can greatly improve your psrr.

原帖由 gogowatch 于 2007-11-22 10:37 发表
Biglee_163,
1. 开关电源的高频纹波对整个系统将会有什麽影响?
3. D类放大器的高频纹波在输出端是不能避免的, 这样对声音有否影响呢?

shouldn't be a problem. just make sure that you use a fast metalized film / polyester cap on the board to decouple.

2. 普通的开关电源为什麽不适合做音频电源?

regulated power supplies, either smps or linear, are great for class A amps as they tend to draw a constant current from the power supply. ideal for regulated power supplies.

class AB/B amps, on the other hand, have very dynamic current draws based on the music - that's what makes them efficient. that means your power supply must have very low dynamic output impedance. that's difficult to achive for regulated power supplies as  they have opamp, or transformer / choke based and cannot produce tons of current on short notice.

millwood

原帖由 漫步猫 于 2007-11-22 10:40 发表
请问，输出功率管换为IRFP240可以吗，元件数值需要改变吗？

shouldn't be a problem. irfp240 (i am looking at a fairchild datasheet for the same part) has slightly higher Vgs threshold voltage so your idle current for the input and phase splitter transistors should be slightly bigger but nothing to be alarmed for.

millwood

原帖由 xiaolu36 于 2007-11-22 11:15 发表
我已经把1969小甲做为推动级，后面再接一独立的电流放大，这样既保留了小甲的音色，又无需使用大面积的散热片，夏天开它N小时都不怕。
在制作中把小甲的功放管换成中功率管，调小电流，末级不设反馈。
现在的 ...

that is actually a very innovative idea.

the jlh is essentially a common emitter amplifier on the positive cycle and a common collector amplifier on the negative cycle. so it has low output impdance on the positive cycle, and high output impedance on the negative cycle.

that's not a problem for the BJT version as the two transistors work in push/pull fashion during the entire cycle. neither is it a problem for the Class A mosfet version.

for the class ab mosfet, the output devices no longer work in push/pull so the impedance mismatch can be an issue and your idea will address that completely.

I would include the buffer in the feedback loop, however.

another way to try is to use a current transmission line as VAS, as is in the ths3001 or lm6361 case.

the ths3001 is a little bit too complicated to with discrete components but lm6361 is quite simple and completely doable with discrete transistors. you can also bias it in either entirely class A or class AB.

here is a suggested schematic, as well as the original one for your enjoyment

gogowatch

回复 #137 millwood 的帖子


hello millwood,

I do have a decoupling cap on the breadboard.  It is a 3300uF.

Picture 1 and 2 are the traces using 22uF as decoupling cap instead,

Picture 1: X=2ms/div
- Upper trace is base of the input transistor 10mv/div
- Lower trace is the output before capacitor 100mv/div

Picture 2: X=2ms/div
- Upper trace is base of the input transistor 10mv/div
- Lower trace is +ve rail 100mv/div

Idle current 500mA.  Hum problem is worsen than before.

Cheers,

millwood

looks like your supply ripple is about 600mvpp, 100hz.

so I simulated it a little bit, using V5 on the supply rail, 600mvpp, 100hz sine wave. Notice that the original input signal is turned off (Vampl of Vsin is  set to zero).

the 2nd chart is ripple at the base of the input transistor. it is about 3mvpp (16.153v - 16.156v).

the 3rd chart is ripple at the output capacitor. the output ripple is about 40mvpp (18.07v - 18.11v).

yours seems to be too high, vs. the simulation.

what value of r13, r14 and c5 are  you using?

gogowatch

hello millwood,

Thanks for your simulation.  

Value of r13, r14 and c5 are 10k, 9.7k and 4.7uF (3.5uF measured) respectively.  My simulation gave result similar to yours.  That's really puzzle me.  I have to do a thorough checking on my circuit and components.  

Cheers,

xiaolu36

原帖由 millwood 于 2007-11-22 20:38 发表


that is actually a very innovative idea.

the jlh is essentially a common emitter amplifier on the positive cycle and a common collector amplifier on the negative cycle. so it has low outpu ...

建议你打拼音 ，我的想法已经实现，声音确实保留了原来的音色，线路过段时间再发上来。

zhengfaming

楼上大虾，等你的线条好多天了！！！！！！！！

xiaolu36

末级的图来了，大家研究一下。

Adams_Leo

原帖由 millwood 于 2007-11-4 22:31 发表


I am Chinese too.

I wish I could have a more efficient way to input Chinese.

你可以在英文的操作系统里面输入中文，使用拼音输入，非常的简单，我现在用的也是英文的操作系统，但是输入的是中文，呵呵

aili

试译12＃：

By the way, R8 is a variable resistor and R2 can also be a variable resistor.

R8 is used to adjust idle current (adjusted so that the voltage drop on the RE resistors is about 25mv), and R2 used to adjust DC voltage on the output.


R8的是一个可调电阻，R2也可以是一个可调电阻器。

R8的作用，是用来调节偏置电流（调整后，使电压下降，大至约25mv ） ，R2用来调节直流电压输出。