Hi all!
I am deeply inspired by the book “small signal audio design” and am designing a preamp myself.
What keeps me scratching my head is the amount of people saying you should not load a 5532 lower than 600ohms…
15V / 600ohm = 25mA
5V /200ohm = 25mA
Do those situations yield the same distortion or is there something I am missing?
What if I need a 5532 that will never need to put out more then 5V??
If loading distortion comes from the current flown, I was also wondering what the ideal output current is for a 5532 distortion wise. As in the book he is comparing one and two 5532’s driving 220ohm at 5V.
The distortion is lowered with 2 5532’s in parallel due to the lower current per 5532 right?
Then why can’t I find a graph that puts 5532 output current vs distortion? That would be really practical in determining feedback resistors etc.
Thanks for your time!
I am deeply inspired by the book “small signal audio design” and am designing a preamp myself.
What keeps me scratching my head is the amount of people saying you should not load a 5532 lower than 600ohms…
15V / 600ohm = 25mA
5V /200ohm = 25mA
Do those situations yield the same distortion or is there something I am missing?
What if I need a 5532 that will never need to put out more then 5V??
If loading distortion comes from the current flown, I was also wondering what the ideal output current is for a 5532 distortion wise. As in the book he is comparing one and two 5532’s driving 220ohm at 5V.
The distortion is lowered with 2 5532’s in parallel due to the lower current per 5532 right?
Then why can’t I find a graph that puts 5532 output current vs distortion? That would be really practical in determining feedback resistors etc.
Thanks for your time!
I believe most of Doug Self's perspective comes from analog mixing desk design, where 10V rms isn't that unusual, so distortion at a large or full swing is commonly quoted by him.
All amps distort more at high level (that's a mathematical inevitability for harmonic distortion) and also with lower load impedances (due to output stage non-linear impedances becoming a larger proportion the output circuit), for opamps there is a load level where this output stage distortion starts to dominate, and this sets the minimum load impedance without distortion increase (ie keeping to the datasheet headline-value for distortion).
I guess as a rough rule of thumb the lower the open-loop output impedance of an opamp, the more immune it should be to distortion due to heavy loads.
And of course eventually the output current rises beyond the opamp's ability to source or sink, giving a rapid increase in distortion - that is related only to the current of course.
In general you should be able to mitigate distortion due to heavy load by keeping signal level low, but I don't think its as simple as just being current related.
All amps distort more at high level (that's a mathematical inevitability for harmonic distortion) and also with lower load impedances (due to output stage non-linear impedances becoming a larger proportion the output circuit), for opamps there is a load level where this output stage distortion starts to dominate, and this sets the minimum load impedance without distortion increase (ie keeping to the datasheet headline-value for distortion).
I guess as a rough rule of thumb the lower the open-loop output impedance of an opamp, the more immune it should be to distortion due to heavy loads.
And of course eventually the output current rises beyond the opamp's ability to source or sink, giving a rapid increase in distortion - that is related only to the current of course.
In general you should be able to mitigate distortion due to heavy load by keeping signal level low, but I don't think its as simple as just being current related.
If you dig into the data sheet, you see the output swing saturates from 2 to 3 volts from the rails, 4 to 6 Volts less than the total supply. So operating a 5532 from 5V is not recomended. If you want to use a 5V supply, you need a "rail-to-rail" op-amp, not a 5532. Operating a 5532 from +/-5V, a total of 10V will clip at about +/-2V.
Zero. Any current drawn exposes the non-linear output stage.
I see that the NE5532 emitter resistors are 15 ohms. That implies that the output current should not be greater than ~1mA.
Ed
Erm, if you don't draw any current, you have no signal. These has to be energy transfer!
Also output-biasing can increase the output current and reduce distortion, so its not that simple.
Disagree with both. The output voltage can be measured with essentially no current drawn.
The output biasing is built-in. The transistors are sliding along an exponential function of current versus voltage. For minimum distortion, the current needs to be held constant.
Ed
Thank you! I think that makes more sense to me. I will do more research on the non linear impedances / output stage of an opamp!
Cheers
jorisds2 - Please read my article on Class AB biasing. It is written in the context of power amplifiers but applies to all class AB amplifiers.
The main distortion mechanism is the exponential relationship between current and voltage (Ebers-Moll). BJTs have other distortion mechanisms (Hfe variation with Ic and the non-linear part of the Early effect), but these are small compared to the exponential.
Ed
The main distortion mechanism is the exponential relationship between current and voltage (Ebers-Moll). BJTs have other distortion mechanisms (Hfe variation with Ic and the non-linear part of the Early effect), but these are small compared to the exponential.
Ed