Constant Current Pros and Cons

Status
Not open for further replies.
Mike,
you hold steadfastly to your mistaken beliefs. The PS caps will necessarily and unavoidably remain in the signal path whatever you do. For lowest distortion and highest ripple rejection, the governing voltages must be directly applied between gate and source. (Voltage reference to the source side, current source to the drain side).
Efficiency is the same, lousy in both cases. It is the price of linearity.
Med venlig hilsen
 
What you write is in contradiction to almost everyone else, and most of all i direct contradiction to those who have actually tried it...
Also simulation shows that there NO PSU modulation on the B version.. and plenty on the A version....

I fail to see that your explanation is of value.
 
Minor correction: The bias voltage should be referenced to the positive rail as well, as shown below.

p.s. *ahem* Feeding trolls generally doesn't end well.
 

Attachments

  • ckt2.JPG
    ckt2.JPG
    14.8 KB · Views: 154
Last edited:
godfrey,
now it`s working, but again, incorrect and inferior. Be sure if it was valuable, it would have been invented long before Ciuffoli did. Now he says his new balanced amp sounds much better. In that case, something is wrong with the power follower.
 
Hi Jan,
tell me very confidentially why would the signal bypass any accessible path? How are the (low impedance) AC signal grounds established? If the cap was not in the signal path, it would not effect the sound. Make a power supply with small capacitor bank and check the bass performance. Study appropriate Thévenin equivalent circuits. Obtain trustworthy information.
 
You lecture about things that simulators refuse to show and treat it as some basic common knowledge.. that we all should see and know....Yet you utter refuse to give explanations and tell us to go home and read....I think it's very poor attitude.. and does not do this community any good....

This is not about the importance of a good or bad power supply or supply ripple rejection...but weather the load driving current is modulated through the PSU or not...

In the first case it is, in the second it is not...or at least that is what simulation shows....
 
It is puzzling (to me) trying to understand what could make the corrected circuit B so inferior.
Is it just than one doesn't like having the drain as part of the signal loop for a source follower? But since the transistor sees the same current voltage in both cases any effect to Vgs is the same.
Might it be a dispute over an implementation flaw not at all evident for the simplified schematics being discussed?
Sure hope I'm led to the answer soon.

Thanks
-Antonio
 
Micke,
I can hardly convince you, so I`m saying read sources you trust.
Power supply components though their imperfections effect amplifying device function and linearity. As I stated earlier, you listen to the power supply modulated by the amplifier executing conversions, essentially DC-to-AC.
And don`t forget to delete the simulator.
😀
 
Hi Jan,
tell me very confidentially why would the signal bypass any accessible path? How are the (low impedance) AC signal grounds established? If the cap was not in the signal path, it would not effect the sound. Make a power supply with small capacitor bank and check the bass performance. Study appropriate Thévenin equivalent circuits. Obtain trustworthy information.

The power supply current drawn in fig B is 2A DC as set by the CCS. The FET and the load are in parallel (for AC). The FET varies it's current due to the input signal, so that when the FET current goes down by 1A, the load current MUST go up by 1A.
A way to look at it is that the FET determines which part of the constant 2A goes through the load, and the rest through the FET. But the result is that the power supply current is a constant 2A.

Edit: Thevenin is not much help in this case, but Norton is.

jan didden
 
There may be practical reasons for preffering a over b, one example:
Dont install any local Amp By-pass capacitors,
Move the power supply close to your listening position,
Hook a current probe on one of the power output lines,
Install a head-phone on the current amplifier

All in good fun
-Antonio
 
Status
Not open for further replies.