Running op-amp in class-A

[Skorpio]

I have seen some connect a current source or a resistor to the supply from the output of an op-amp. This should be to run it in class-A?

Does it work, and how? I thought all low-level amps were in class-A?

I'm using a AD825 as follower (+1 gain), can I try with a resistor to the supply without problems? Witch supply should I use (Vcc or Vee) and witch resistor/current level? My supplies are +/-15V

Thank you....

 

[subwo1]

Here are some earlier threads for starters: One thread,

the search results page:http://www.diyaudio.com/forums/sear...d=641671&sortby=lastpost&sortorder=descending

 

[Bruno Putzeys]

Does it work, and how? I thought all low-level amps were in class-A?

I'm using a AD825 as follower (+1 gain), can I try with a resistor to the supply without problems? Witch supply should I use (Vcc or Vee) and witch resistor/current level? My supplies are +/-15V

Most good op amps are biased in class a up to a few milliamps (most of the quoted supply current). In normal use you'll rarely see an opamp come out of class A.

Drawing a bias current higher than this out of the output will cut off one of the output transistors, and use only the other output transistor (single-ended class A).

You can try this trick with about 10mA of bias current or so. Check the data sheet of the op amp to see which rail has the worst PSRR (ie which one has the miller compensated gain stage on it). Tie the current source between that rail and the output.

Using a resistor instead of a real current source adversely affects distortion and PSRR. Do use a real current source.

Note that top notch op amps tend to have such good output stages that they gain little by this modification.

 

[Nelson Pass]

I find that a 10K to V- is a good general value for audio
circuits. The V- rail is preferred because in most op amps
we would rather work the N type output device rather than
the P or the quasi-complementary "P", and 10K is high
enough not to significantly load the output and low enough
to give you SE Class A bias with most audio type loads.

 

[Morello]

A AD825 connected for unity gain do already have below -100dB THD or so. In fact, you can get more distortion with a current source at the output If you want class A, design a discrete OP-amp. It's not that difficult.

 

[Skorpio]

Discrete MOS-FET buffer

I am using the AD825 as follower.

How could I build a single stage MOS-FET follower?

Output from the DAC is voltage and with an off-set DC of 1.5VDC.

I am running through 3.3uF cap and 1M to ground and then to AD825...(+/-15V)

How shall I connect e.g. IRF620 as an discrete follower?

 

[needtubes]

For a MOSFET follower you might try the Szekeres amp. If that doesn't fit your needs, check out some other projects in the HeadWize project list.

 

[millwood]

Re: Discrete MOS-FET buffer

How could I build a single stage MOS-FET follower?

I built this as a pre-amp (gain about 2) but with a slight modification (moving the output pickup to the source of the mosfet) it also works as a follower as well.

the advantage is that it is very independent of supply voltage and transistors used (I have tried it with different mosfets / bjts).

it simulated to about 0.016 - 0.017% thd 1k - 20k, and has flat frequency response through mghz range.

 

[Skorpio]

Thank you, I will try this discrete follower

Also thanks for the interesting links on headamps