0.4 is also too low.
I would go for 2mA, as a starting point.
And - as JMFahey just mentioned - 10nF screws up the whole idea.
You have to remove it.
You will have the best quality (especially at higher swing) if you use CCS (dynamic load) at the OpAmp's output.
Tangentsoft gives lots of information and enough to get you experimenting.
https://tangentsoft.net/audio/opamp-bias.html
from tangentsoft
if you go for method 3 then the two jFETs should be different types.
select a low gm high Idss jFET for Q2
and select a high gm low Idss jFET for Q1
If you want around 3mA of current drain, then Idss of Q1 should be around 4mA. LSK170A, or similar
Then Q2 Idss should be at least double that i.e. >8mA BF244C, or similar.
Using this combination you may find that Rs = zero ohms.
Of the three methods described by Tangentsoft, the resistor method is least likely to give good results.
if you go for method 3 then the two jFETs should be different types.
select a low gm high Idss jFET for Q2
and select a high gm low Idss jFET for Q1
If you want around 3mA of current drain, then Idss of Q1 should be around 4mA. LSK170A, or similar
Then Q2 Idss should be at least double that i.e. >8mA BF244C, or similar.
Using this combination you may find that Rs = zero ohms.
Last edited:
There is another very easy way to bias an opamp into class A and not have to worry about any PSRR degradation - see attached diagram. DIYaudio forum member jcx showed this a few years ago and I used it on my Symphony preamp. If you are looking for the ultimate in quality, its best to also buffer the opamp as shown. You can then drive loads up to 2x the peak buffer standing current with virtually no loading effect on the opamp at all.
Separately, I used a class A/buffer design and measured it on an AP about 5 or 6 years go. The distortion was at the AP limit at around 1.5 ppm - there's a write up on my website with the measurements (look for 'class A Buffer').
Separately, I used a class A/buffer design and measured it on an AP about 5 or 6 years go. The distortion was at the AP limit at around 1.5 ppm - there's a write up on my website with the measurements (look for 'class A Buffer').
Great stuff Andrew! Looked at the article. CCS'ed current sink + CCS'ed buffer = perfect combination 😎
Thanks for sharing.
Thank you very much Mr. Bonsai,
I will try to understand your schematic.
It is a little difficult for me but I will try.
I will try to understand your schematic.
It is a little difficult for me but I will try.
Rclassa determines the current in Class A the Opamp deliveres... So, basically Ohms law... For each mA it's about 650 (Ohms)/ x (mA) (assuming Vbe = 0,65V )
The other two resistors determine the amplification of the total circuit. Let's assume the most left resistor is R1 and the other R2 : (R1+R2) / R1 = amplification
Last edited:
- Status
- Not open for further replies.