Jump to content
Electronics-Lab.com Community
Sign in to follow this  
Kevin Weddle

circuit distortion

Recommended Posts

IC devices have a low amount of noise according to data sheets. But is it a good indicator of signal distortion? Transfer characteristics show to be different depending on the transistor. So does this mean that transfer charateristics are negligible, even though they do operate on small areas of the transfer curve?

Share this post


Link to post
Share on other sites

Very old opamps like the 741 and LM324/LM358 have a high noise level.
They also have different but high distortions. The main distortion for the LM324/LM358 is crossover distortion because they are biased for low power. The bias current is so low that it causes crossover distortion. Their amount of distortion is not even spec'd.

Newer opamps like the OPA0134, OPA2134 and OPA4134 are designed to have extremely low distortion of only 0.00008%. The LM4562 has a distortion of only 0.00003%.

Of course the saturation voltage rises if the collector current rises unless the base current is also increased.

Share this post


Link to post
Share on other sites

For instance, two different IC amplifers that produce the same result, but have different specifications that overlap. Using a more modern opamp would have lower noise, but operating it near the end of it's specification creates more distortion. Frequency response is the only indicator of performance near the end of a devices specifications.

Share this post


Link to post
Share on other sites

For some reason my previous post was delete, maybe it was my mistake.

I've retrieved it from the bin.


The saturation voltage increases unless the base current increases?

The saturation voltage will always increase if the collector current is increased even if the base current is increased.

A transistor is saturated when an increase in base current no longer causes an increase in collector current.

What about transistor transfer charcteristics? I've read that FET'S have different transfer characteristics than bipolar transisitors.
MOSFETs are voltage controlled devices.

If a MOSFET saturates, it means that it'll draw a constant current, regardless of the voltage across the source and drain.

If a MOSFET is to be used in a switching application, saturation is a bad thing because the voltage across it will be high leading to a high power dissipation. To use a MOSFET as a switch it needs to be in the on state, i.e. when an increase in gate voltage does not cause an increase in current. There might be a more fancy name for this but I can't remember it. ;D

Share this post


Link to post
Share on other sites

For instance, two different IC amplifers that produce the same result, but have different specifications that overlap. Using a more modern opamp would have lower noise, but operating it near the end of it's specification creates more distortion. Frequency response is the only indicator of performance near the end of a devices specifications.


What do you mean?

Operating an op-amp with too higher voltage swing at too higher frequency will result in slew rate distortion which is nothing to do with the noise or distortion specifications.

Share this post


Link to post
Share on other sites

An opamp has extremely low distortion at low frequencies because its gain is extremely high which results in a lot of negative feedback.

A TL07x opamp has 0.003% distortion at 1kHz with a load of 2k ohms or more, a 30V supply and an output of 6V RMS. Its gain is 1.
Its open-loop gain at 1kHz is 3200 so it has a very high amount of negative feedback.

If the load is less than 2k ohms, if the frequency is higher than 1kHz or if the gain is higher than 1 then the distortion is higher.

Share this post


Link to post
Share on other sites

Analog IC's vary in design depending on their function. Different IC's can be used in circuits and provide the same function. You might use a simpler transistor circuit over an opamp for example. Although it's easier to just connect complicated IC's together, it still guess work.

Share this post


Link to post
Share on other sites

A transistor has a max open loop voltage gain of about 280 and distortion at full output of about 40%.
If it has negative feedback to reduce its gain and distortion then its gain is 3 and its distortion is 1%. Two transistors with bootstrapping have distortion of 0.015 at lw levels rising to anput 0.1% at high levels.

An opamp has an open loop gain of 200,000 or more at low frequencies. With negative feedback to reduce its gain and distortion its gain is 3 and its distortion is barely measureable (0.001% or less).

It is easy to bias an opamp or transistor so it works properly. Guesswork is not needed.

Share this post


Link to post
Share on other sites

I don't believe that the design of an opamp provides for that low of distortion. But it might. Maybe two or three transistors would not provide the gain, impedance, or current capabilities, but it should produce better results because of simplicity.

Share this post


Link to post
Share on other sites

The TL071 single, TL072 dual and TL074 quad audio opamps are inexpensive and have distortion of only 0.003%.

The OPA134 single, OPA2134 dual and OPA4134 quad audio opampscost a little more and have distortion of only 0.00008%.

Two or three transistors have much more distortion.

Share this post


Link to post
Share on other sites

I don't believe that the design of an opamp provides for that low of distortion. But it might. Maybe two or three transistors would not provide the gain, impedance, or current capabilities, but it should produce better results because of simplicity.
Look at the datasheets of the ICs listed by audioguru.

Build an amplifier with a couple of transistors and compare its performance (in terms of distortion) to the op-amp equivalent and see for yourself.

Study op-amp theory and negative feedback and you'll see that if an amplifier with a high gain and some distortion can become a low gain amplifier with virtually no distortion when lots of negative feedback is applied.

Share this post


Link to post
Share on other sites

A simple resistor and transistor signal amplifier produces maybe a gain of 50 or 100. An opamp provides high gain, high input impedance, and lower output impedance given the design. A single transistor with good voltage regulation has to be advantageous.

Share this post


Link to post
Share on other sites

A simple resistor and transistor signal amplifier produces maybe a gain of 50 or 100.

Yes, but its distortion is as much as 40%.

An opamp provides high gain, high input impedance, and lower output impedance given the design.

Yes, but since its open-loop gain at 1kHz is as much as 10,000 times, its distortion is extremely low.

A single transistor with good voltage regulation has to be advantageous.

Nope. The transistor produces severe distortion and the opamp doesn't.

Look at what a single transistor does to a sine-wave:

post-1706-1427914403083_thumb.png

Share this post


Link to post
Share on other sites

That amount of distortion must be due to the large output voltage swing. An opamp circuit is also capable of high distortion, but has very low distortion when operated within it's recommended but very narrow limits. Many amplfiers share some common design, but they are tailored for different applications.

I've seen designs which utilize single transistor amplifers cascaded together. But I think their main purpose was for filtering and not so much for amplification.

Share this post


Link to post
Share on other sites

The transistor has high distortion because it is controlled by current but the input is a low impedance voltage. Then its output is exponential. Also, the transistor is running wide open with no negative feedback. The distortion is about 3% when the transistor has lots of negative feedback but then of course its voltage gain is only 1 to 4.

An opamp with negative feedback has extremely low distortion until its output hits the power supply voltages which is not a narrow range.

Share this post


Link to post
Share on other sites

So you think more transistors must mean more distortion?

Here's an example of a circuit which proves your theory to be wrong.

In the first circuit there's three separate stages of voltage amplification, Q1 and Q2 just provide current amplification to drive R4, Q3 & Q4, Q6 & Q7, are Darlington pairs which have lots of gain, Q5 doesn't have much gain and just acts as an inverter.

R5 to R7 help bias the transistors and provide negative feedback, C3 is a phase compensation capacitor which prevents oscillation.

Notice how the output is a pretty clean and undistorted sinewave?

In the second circuit, Q5 to Q7 have been removed so the closed loop gain is much lower. The cross-over distortion can be clearly seen on the output.

This output stage is biased in class B which yields a low standby current but is very non-linear.

The first circuit has a very high open loop gain and lots of negative feedback which dramatically reduces the distortion.

The second circuit has a much lower open loop gain and much less negative feedback so the output remains very distorted.

Paste the text between the code tags into a text editor, save with a .asc extension, open it using LTSpice and see for yourself.

Version 4SHEET 1 880 680WIRE 272 -256 192 -256WIRE 416 -256 336 -256WIRE 192 -160 192 -256WIRE 192 -160 -176 -160WIRE 272 -160 192 -160WIRE 416 -160 416 -256WIRE 416 -160 352 -160WIRE 672 -160 416 -160WIRE 64 -80 -320 -80WIRE 208 -80 64 -80WIRE 464 -80 208 -80WIRE 592 -80 464 -80WIRE 464 -48 464 -80WIRE 64 -16 64 -80WIRE 208 -16 208 -80WIRE 592 16 592 -80WIRE -320 32 -320 -80WIRE 464 64 464 32WIRE 528 64 464 64WIRE -320 128 -320 112WIRE 592 176 592 112WIRE 672 176 672 -160WIRE 672 176 592 176WIRE 688 176 672 176WIRE 800 176 752 176WIRE 592 192 592 176WIRE 800 208 800 176WIRE 64 240 64 64WIRE 96 240 64 240WIRE 208 240 208 64WIRE 208 240 176 240WIRE 464 240 464 64WIRE 528 240 464 240WIRE 464 256 464 240WIRE 464 256 320 256WIRE 208 304 208 240WIRE 256 304 208 304WIRE 464 304 464 256WIRE 592 304 592 288WIRE 800 336 800 288WIRE 400 352 320 352WIRE 208 368 208 304WIRE 64 384 64 240WIRE 64 384 -80 384WIRE -496 400 -528 400WIRE -288 400 -432 400WIRE -176 400 -176 -160WIRE -176 400 -208 400WIRE 64 416 64 384WIRE 144 416 64 416WIRE 464 416 464 400WIRE -176 432 -176 400WIRE -144 432 -176 432WIRE 64 432 64 416WIRE -528 480 -528 400WIRE 0 480 -80 480WIRE -176 496 -176 432WIRE 64 560 64 528WIRE 208 560 208 464WIRE -528 576 -528 560WIRE -176 592 -176 576FLAG 464 416 0FLAG 592 304 0FLAG -320 128 0FLAG -528 576 0FLAG 800 336 0FLAG 208 560 0FLAG 64 560 0FLAG -176 592 0SYMBOL npn 528 16 R0SYMATTR InstName Q1SYMATTR Value BC547ASYMBOL pnp 528 288 M180WINDOW 0 60 32 Left 0WINDOW 3 59 68 Left 0SYMATTR InstName Q2SYMATTR Value BC557ASYMBOL Misc\\EuropeanResistor 448 -64 R0SYMATTR InstName R1SYMATTR Value 470RSYMBOL npn 400 304 R0SYMATTR InstName Q4SYMATTR Value BC547ASYMBOL voltage -320 16 R0WINDOW 123 0 0 Left 0WINDOW 39 0 0 Left 0SYMATTR InstName V1SYMATTR Value 10VSYMBOL cap -432 384 R90WINDOW 0 0 32 VBottom 0WINDOW 3 32 32 VTop 0SYMATTR InstName C1SYMATTR Value 220nSYMBOL voltage -528 464 R0WINDOW 123 0 0 Left 0WINDOW 39 0 0 Left 0SYMATTR InstName V2SYMATTR Value SINE(0 50mV 500)SYMBOL Misc\\EuropeanResistor 784 192 R0SYMATTR InstName R4SYMATTR Value 100RSYMBOL Misc\\EuropeanPolcap 688 192 R270WINDOW 0 32 32 VTop 0WINDOW 3 0 32 VBottom 0SYMATTR InstName C2SYMATTR Value 100













































































































Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Sign in to follow this  

×
×
  • Create New...