about ccs of Leach Amp
 

Leach Amp v4.5 uses two zener and some resistors as tail of differential input circuit .

i take the tail as ccs , in normal amplifier , tail ( ccs) is consisted of a transisitor and diode or LED


so the question is which one is better?

I would prefer using LED(s) to set the voltage for a CCS transistor.
Like the arrangement you mention.

What I know, LED produces less noise than zenerdiode.
There are also other factors like temperature stability to consider in some applications.

Generally we see zeners used more frequently as current sources in older amplifers.

A third method is to use JFET and a resistor.
This is the most simple way to make a CCS, constant current source.

Another method often used is two bipolar transistors.
One transistor holds constant voltage across the emitter resistor of the other, while limiting base current.
This will give a CCS of Vbe/Re.

He is using the current through a resistor as the current source, with the zener providing a stable voltage reference so that the current is constant. His page explains why he did not use "active" current sources.

He has thought about alternatives but found this to best solution
mainly for 2 reasons.

Hi lineup,

Sorry, I find Leach's explanation, or excuse, for his simple cascode rather baseless.

2 BJT current sources using low noise small signal BJTs are not noisy. Zeners are.

My SKA and 300W both use 2BJT CCS's because of the superior PSRR afforded by them. Any they have NO turn on/off thump.

More of a throwaway explanation that doesn't survive scrutiny.

I tried both alternatives on my 300W amplifier (always keeping an open mind) and the sim PSRR result is here -

http://www.diyaudio.com/forums/showt...678#post790678

He has included an RC filter in the early stage supply lines to improve things.

Cheers,
Greg

Yes, amplifierguru
this is what I have learned, too.

I am always surprised to see zeners in audio amplifiers.
I have seen Nelson Pass using zeners in some earlier amps.
Do not know if he still.

I would never consider put zener diodes into where noise should be avoided.

Hi Lineup,

My experience has been that simple resistive bias with bypass capacitors gives the best audition results, THD and noise (all)compared to zeners for cascode bias.

I used this on the Magnet 300F which was rebadged and won TAS Golden Ear.


Cheers,
greg

Reliability is a second grudge for me.
Already have 6 Leach amplifier boards layed out for the Zeners, but thinking of replacing those by a string of LEDs.

Re: about ccs of Leach Amp
 

imho it is more of convience than anything, imagine stringing what? 20 led's in series to get 40 volts! so a zener in this case is more practical.

where a transitor is used with a current source or sink, an led is more practical imho.

Earlier versions did not have the zeners. IIRC, they were added at Ver 4.4 along with some other value tweaks along the way to the current version. The amp will run fine without them, and seems to fit Guru's preference for resistive biasing of cascodes.

If you eliminate the zeners R13/14 change, since they no longer have to pass the zener current. From the Leach page I saved a while ago: R13 = R14 = (V - 38.2)/5.42, where V is the power supply voltage.

Sorry Jacco, this gives you another option to try ;)