Hiss probelm on JLH class A

[rchung]

HI All,

I recently built the JLH class A (High power version on Geoff's website with dual current source,plus and minue power supply rail, rectifier+47000uF cap power supply) and have a lot of probelm with hums and hissing noise. I was careful on how to route the return current already,star gournd was used and I was making sure the noisy current return does not disturb the feedback and signal ground.

Finally,, I need to resort to a big ferrite based choke to separate the feedback and signal ground to the rest of the ground. The remaining issue is the hiss noise, I just cannot get rid of it!

Have anybody experienced similar issue? I am using this as a head phone amp at work so the hiss is annoying. Driving a speaker is less problematic with the hiss.

I should also point out that I am NOT paralleling the output NPN power transistor, just using one set. Both output NPN has a emitter feedback of 0.15 Ohm on them and a sense resistor of 0.47 ohm on the colllector on one of them. (That is not on Geoff's schematic, I just added it there as an experiment). In addition, I need to add a zobel network of 8.2 Ohm and 0.47 uF to stabilize the system

I also tried to use the preamp section of Sansui AU-317 to buffer the input. That helped some but a little hiss is still there.

Any suggestion from this forum ?



Thank you all in advanced.


RUssell

 

[sreten]

The standard 100R/8R heaphone attenuator will reduce hiss.
The amplifier will need more input, and the amp may clip.

sreten.

 

[rchung]

Do you mean a voltage divider in the output ? I looked at the scope on the output . It is running no more than 200 mV peak to peak right now when driving the headphone at work. Floor Speaker is at a much higher level. Thanks. I'll try that out.

Russell

 

[sreten]

Yes exactly that, noise should drop 23dB.

sreten.

 

[millwood]

I suspect that the amp is oscillating.

I personally don't like those mods with active loads as it, in my view, defeats the purpose of this little simple amp.

as a headphone amp, you probably don't need that much power so maybe the original jlh will work better for you.

 

[Nelson Pass]

Is the problem on only one channel or both? (key question)

There are more alternatives: noisy transistor, noisy resistor,
bad connection and so on. After you've attentuated the input,
if that didn't do it, start looking here also.

It's possible, but the JLH, (like Zens, shameless plug) are not
prone to oscillation.

 

[mwh-eng]

Not familiar with your amp, but a hiss noise (pink noise) can be make by reverse biasing a base emitter junction enough to reach breakdown, and then amplifying this noise source. This is probably not the problem. Without an input signal check all your DC voltages to see if they are where they should be. Ground the input, if this is okay for this amp. Check for this hiss. If you have a good quality scope/probe, see if you can start at the output, finding the noise and work back to the front of the amp. Depending on the probe capacitance you may not get very far before the probe affects the amp/noise. This probe capacitance may stop or reduce the noise; however, this effect may help you locate the stage or transistor causing the noise.

Lots of luck

 

[blu_glo]

The star earth should contain just PSU grounds, from amp, psb caps and transformer. The input ground must go from input phono connector to ground near input components on the pcb. What I mean is I do not recommend following this earthing method too literally (I've seen it done) and taken input ground directly to the star earth!

 

[scafas]

problem

I solved the problem by completely separating supply (transformer and stabilizer) .