JLH1969 MOSFET Class A/AB amp

qjdj

这个现象是正常的。
如果你想改善，可以在输出电容后加一个1k欧的电阻到地，然后用喇叭保护电路来延时接通扬声器，让这个电阻先给输出电容充电。

海胆

不错，作个记号

htwlhu

我有喇叭保护电路的，如果在输出电容对地加一个1K的电阻，那这个1K的电阻不就成了负载电阻了，还有没其他好的办法呢，其他朋友用HOOD 1969有这个问题吗？

qjdj

肯定都有。
你看看后续的1969版本，jlh1996就已经加了这个电阻，对声音没有影响的，因为1k欧和8欧相比，相差很大。

htwlhu

我照你这个方法做了，一个声道好了，但另为一个声道开机还是有直流，但持久时间很短，只有6秒左右，是哪里出了问题呢，喇叭保护电路用的是UPC1237，最多可以延迟几秒呢

qjdj

不会是你的电容有漏电吧。
没有遇到这种情况，如果你想解决，可以仔细对比两个声道的参数。
upc1237的延迟接通可以调很大，你查查pdf吧，原理就是检查某个脚的电平，而这个脚电平的升高是电源通过一个电阻给电容充电得到的。可以通过加大电阻和加大电容来增加延迟时间。

htwlhu

板上的电容我都换成新的了,现在的问题是,为什么功放板在6秒后输出正常,为什么保护电路还不工作,是不是保护电路出了问题呢,一般说来功放板输出正常了保护电路就会接通

millwood

the original jlh1969's design is a little bit flaw when it comes to turn-on thump.

when the amp is first powered on, the 100u capacitor is charged up by the rail via the 39k resistor. this takes a long time (100u+39k -> time constant of 4 seconds), and holds point C low, which in turn holds the base of the input transistor low, which in turn holds point B low. with point B low, the current going through the upper output device, I2, is extremely high, thus the turn-on thump.

in my mosfet design, that can be in excess of 10 amp.

the solution is to shorten that time constant. I typically use 10k + 4.7u (a time constant of 0.05 second).

please note that too short of a time constant will cause high a/c hum. so it is a balancing act.

HIFI2008

按照millwood 老师给的电路图，我在EWB里面仿真了一下，想测测静态电流，却发现射极电阻上的电压好小？

上图，各位老师看看，我在哪里犯了低级错误了？

millwood

hifi2008, R1/R2 is really a pot. you adjust it to make sure that the output voltage is about 1/2 - no need to make it precisely that.

then you adjust R8 to make sure that the voltage drop on R13/R14 is what you wanted for idle current purposes. In this case, your amp is working in Class C.

the process may be iterative, as changing R8 will impact the output voltage and changing R1/R2 will also impact idle current.

HIFI2008

原来是这样，我还以为是我的仿真失误了

感谢millwood 老师的指点

对于反向器的选择，我看前面的帖子，millwood 老师有提到说用mje15030，但这个比较贵啊，BD139、BD669我这里又不方便买到正品～～

请问millwood 老师，对于反向器的选择，hFE有什么要求吗？？是不是越大越好？？

打算用安美森的MJE243，不知道合不合适？（hFE40~180，Cob 50pF，100V/4A）

millwood

the 243 should work just fine. it should have maximum current of at least 1a, enough Vce and reasonable beta. from the stats you quoted, it should be just fine.

drb888

millwood 老师，我打算用2SK851做输出管，相对IRF530而言，哪个会更适合你在贴里发表的电路？

DarkHorse

K851的功率是很大,但输入电容也很大(记得好像有2000多PF),相对而言IRF530的输入电容小些,高频会好些,我建议用IRF530好(实际你可以都试试,取你认为好的管子)!

millwood

drb888:

I don't know anything about 2sk851. I have used many mosfets, including irf540 and fqa25n50, among others. the only mosfet that I couldn't get it to work with is fdp3682, due to thermal instability.

if 2sk851 is like any other mosfet, it should work. if the Ciss is much larger than 3000pf, I would reduce the gate stoppers to like 220ohm or 110ohm, or increasing the idle current on the phase splitter.

hope it helps.

drb888

谢谢millwood 老师！
DarkHorse大侠，请问你作的“基于ALEX的JLH-MOS上修改来的电路”实验结果如何？

ch639827608

May I speak English?
I made a headphone amplifier using JLH circuit, it sounds quite good,but after I tested it ,I found this circuit has
inevitable defects.
1 low slew rate, less then 1v/us
2 low PSRR
3 high THD,about 2% at full scale output
4 low efficiency

How can these been solved?

让我来翻译：

我可以用英语来说吗？
我用JLH线路制作了一个耳放，它声音很好，但我测试它之后，我发现这线路有不可避免的缺点。
1，        低的转换速率，小于1V/us
2，        低PSRR
3，        高THD，满功率输出时大约2%
4，        低效率

这些问题怎么解决？

HIFI2008

这个电路需要用什么样的电源哈？？

貌似对电源的要求非常高啊

qjdj

这个电路我用过普通整流滤波电源，也用过crc整流电源，现在用20v的笔记本电脑开关电源。都没有电流声。

HIFI2008

开关电源用在上面，声音好吗？？

不会有干扰吗？

普通整流滤波、CRC整流、CLC整流、开关电源，哪种比较好？？