Author Topic: why is class A audio amp saturating?  (Read 3538 times)

0 Members and 1 Guest are viewing this topic.

Offline d4n13lTopic starter

  • Regular Contributor
  • *
  • Posts: 71
  • Country: mx
why is class A audio amp saturating?
« on: August 13, 2019, 07:47:52 pm »
Hi,

Before you say it, I know class A audio amplifier are not efficient, I just decided to make the circuit to see how they work, I want to test the different kinds. Also I'm learning about how op amps work.

I made the circuirt you see in the attrachment and I'm using a small radio as input, however for some reason the output is saturated on the positive side of the wave and can't figure out why. I can hear a clean audio from the speaker but the volume is also low.

Any help will be appreciated.
« Last Edit: August 13, 2019, 07:52:08 pm by d4n13l »
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: why is class A audio amp saturating?
« Reply #1 on: August 13, 2019, 09:11:22 pm »
First, calling your circuit a "class A" amplifier is a bit over the top.

Bur to your circuit: your resistances are waaay too high for en LM358. Divide the values by 20 and you'll probably get an acceptable result.

 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: why is class A audio amp saturating?
« Reply #2 on: August 13, 2019, 09:22:02 pm »
Surely class A :)

That being said, running constant current through the speaker is undesirable. Firstly, it dissipates a few watts of constant power in the speaker, make sure not to exceed its power rating if it's small. Secondly, it pushes it away from its neutral position, presumably increasing distortion.

As for the limited output - does the opamp run on 9V too? LM358 is limited in output to some 1.5~2V below its positive supply, add 0.7V difference between Q1 base and emitter and you get hard clipping at 6.6V.
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 3382
  • Country: gb
Re: why is class A audio amp saturating?
« Reply #3 on: August 13, 2019, 09:33:24 pm »
Are you powering this with a little 9v battery or a decent power supply?  The LM358 output will likely only swing to about 1.5v below the rail, plus the Vbe drop from the transistor so you are already down to about 6.9v maximum at the speaker.  If the power rail is sagging (e.g. from a small battery, long cables etc.) then it's not a stretch to think you'll only see the 5.6v pkpk that you are getting.  Use your scope to look at the op-amp output and the supply rail. 

Benta's comments about the bias resistor values is also worth heeding, the LM358 is an older bipolar op-amp so input bias currents are something you have to consider.  With no audio input what is the voltage at speaker?  You want it to be at half the rail voltage, but chances are it won't be.

Finally you might know this already, but speakers do not like having large DC biases applied to them like this circuit does.  It causes high power dissipation in the voice coil and biases the cone away from it's mid-point which will introduction distortion. 
 

Offline d4n13lTopic starter

  • Regular Contributor
  • *
  • Posts: 71
  • Country: mx
Re: why is class A audio amp saturating?
« Reply #4 on: August 13, 2019, 10:10:40 pm »
I'm using a 0.5W speaker which should be fine here I think. I'm using a power supply, not batteries, and I'm also feeding 9V to the op amp.

I didn't know the op amp can't go all the way to the supplied voltage, so I guess that's the problem here.

I checked the voltage without input and it is higher than it should at ~5.8V. So I take it this shift from mid-point is caused mostly by the DC bias in the speaker? But that's the problem with this class of amplifier right? the speaker is receiving DC all the time, nothing to do about it. I want to try a push-pull next
 

Offline hfleming

  • Regular Contributor
  • *
  • Posts: 52
  • Country: no
Re: why is class A audio amp saturating?
« Reply #5 on: August 13, 2019, 11:38:39 pm »
You can also use a low-impedance transformernto drive the speaker, or just resistor in place of the speakor, and drive the speaker via a large capacitor. Am I right in assuming this is just a learning excercise, as a 2N3055 is a bit of an overkill and has a very low gain, so that will further load the op-amp. For what you are doing, even a 2N2222 will work.
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 2138
Re: why is class A audio amp saturating?
« Reply #6 on: August 14, 2019, 12:28:10 am »
At idling you nominally have 4.5V across the 8\$\Omega\$ speaker - I make that 562mA, and 2.5W. 2N3055 are not noted for their current gain (20-70 IIRC), your LM358 is going to struggle to drive the required base current.
 

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2005
  • Country: us
Re: why is class A audio amp saturating?
« Reply #7 on: August 14, 2019, 01:11:28 am »
At idling you nominally have 4.5V across the 8\$\Omega\$ speaker - I make that 562mA, and 2.5W. 2N3055 are not noted for their current gain (20-70 IIRC), your LM358 is going to struggle to drive the required base current.

It's worse than that.  An 8-Ohm speaker is typically about 6 Ohms DC resistance, making the current roughly 750mA.   The Hfe of a 2N3055 at 4V VCE and 750mA is about 100, and the 358 opamp should be able to drive that amount of current into the transistor base, so at quiescent conditions the circuit should sort of work (ignoring the bad practice of driving this DC current through the speaker).  But once you try to drive that transistor more positive then the opamp just can't do it.  It requires more voltage difference between the positive supply and the opamp output.  Thus the clipping.
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 

Offline Audioguru again

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ca
Re: why is class A audio amp saturating?
« Reply #8 on: August 14, 2019, 01:23:30 am »
The datasheet for an LM358 opamp shows that its inputs are PNP and have a maximum input bias current of 0.5uA. This current in your 1M bias resistor causes 0.5V and the input to be at (9V/2) + 0.5V= 5.0V. The DC negative feedback causes the output at the speaker to also be at 5.0V. But the opamp output must be 0.75V higher than the speaker voltage due to the base-emitter of the transistor so the idling output of the opamp is 5.0V + 0.75V= 5.75V. If the opamp has a lower input bias current then the idling output of the opamp can be 5.3V.

The datasheet of the LM358 shows that its maximum output at fairly low current is about 1.4V less than the supply voltage so it will be 7.6V. Then the positive output swing is only 7.6V - 5.75V= 1.85V. But the negative output swing is 5.0V. That is why there is clipping at the top of the waveform. It has nothing to do with the continuous current in the speaker.

Your poor little 0.5W speaker dissipates DC power at (5V squared)/8 ohms= 3.13W!   
 
The following users thanked this post: d4n13l

Offline d4n13lTopic starter

  • Regular Contributor
  • *
  • Posts: 71
  • Country: mx
Re: why is class A audio amp saturating?
« Reply #9 on: August 14, 2019, 06:52:14 am »
Yes, this is just for learning.. I actually tried a 2N2222 instead of the 2N3055 but although it was within current limit it got very hot, I don't know if the transistor would be able to dissipate it so I switched it.

---

Audioguru, thanks a lot for that explanation, I learned yet again new things about op amps. Also I totally forgot to account the dc bias  when I calculated the power for the speaker, however I didn't noticed it getting hot, probably because I only did brief tests. I switched to a bigger one I have that can handle 25W. I adjusted the schematic a bit a now it works much better although it is still a little off the mid point but I think I can move on now.
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 3382
  • Country: gb
Re: why is class A audio amp saturating?
« Reply #10 on: August 14, 2019, 11:50:08 am »
Audioguru, thanks a lot for that explanation, I learned yet again new things about op amps. Also I totally forgot to account the dc bias  when I calculated the power for the speaker, however I didn't noticed it getting hot, probably because I only did brief tests. I switched to a bigger one I have that can handle 25W. I adjusted the schematic a bit a now it works much better although it is still a little off the mid point but I think I can move on now.

Why aren't you heeding the advice given for the resistor values?  Why have you changed R2 but not R3?

Reduce the feedback resistor values by at least factor of ten, 10k input resistor, 100k feedback.  Reduce the rail splitter resistors R2 and R3 to ~10k each.  The intention is to put this point at half the supply voltage so the resistors need to be equal.
 

Offline d4n13lTopic starter

  • Regular Contributor
  • *
  • Posts: 71
  • Country: mx
Re: why is class A audio amp saturating?
« Reply #11 on: August 14, 2019, 03:27:53 pm »
Hi mikerj,

I think I did heed the advice, as audioguru and you said the op amp won't go all the way to the supply voltage and I have to account also for the voltage drop in the transistor, so I don't want the rail splitter resistor to be the same anymore. As for the resistor values I didn't change them as you mention simply because it didn't affect the output I think, I'm doing this just an exercise, it's not a circuit I'm going to keep.

Anyway I did try the changes you mention (fig 1) and on fig 2 you can see the output with the changes I made. I couldn't get to the mid point because I don't have appropiate resistor values but close enough I think.
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 3382
  • Country: gb
Re: why is class A audio amp saturating?
« Reply #12 on: August 14, 2019, 05:46:48 pm »
Hi mikerj,

I think I did heed the advice, as audioguru and you said the op amp won't go all the way to the supply voltage and I have to account also for the voltage drop in the transistor, so I don't want the rail splitter resistor to be the same anymore.

I take your point about centering the operating point in the useable output range of the op-amp, but the Vbe voltage drop through the transistor will be taken care of with negative feedback.  The reason your output bias is not what your were expecting is because your feedback and rail splitter resistor values are all too large.  The old LM358 draws significant bias currents on it's inputs, and it is these currents flowing through your large value resistors that cause this offset error.
 

Offline d4n13lTopic starter

  • Regular Contributor
  • *
  • Posts: 71
  • Country: mx
Re: why is class A audio amp saturating?
« Reply #13 on: August 14, 2019, 08:15:46 pm »
ok, got it, thanks. I'm getting used to do the analysis as how the component behave in the real world instead of their ideal behavior.
 

Offline cvanc

  • Frequent Contributor
  • **
  • Posts: 675
  • Country: us
Re: why is class A audio amp saturating?
« Reply #14 on: August 14, 2019, 08:29:00 pm »
Shouldn't the speaker be capacitively coupled to the amp output?
 

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2005
  • Country: us
Re: why is class A audio amp saturating?
« Reply #15 on: August 14, 2019, 08:41:04 pm »
Shouldn't the speaker be capacitively coupled to the amp output?

In the real world yes, but then you would be using a push-pull output stage.  The OP's circuit is an attempt to avoid that.  A practical class-A would probably use an output transformer driving the speaker.
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: why is class A audio amp saturating?
« Reply #16 on: August 15, 2019, 05:22:19 am »
Transformer won't help, the primary would be a short at DC.

In a more proper single ended class A stage, the speaker would get a series capacitor and additional power resistor or active current sink would go from the emitter to ground to provide negative drive.
 

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2005
  • Country: us
Re: why is class A audio amp saturating?
« Reply #17 on: August 15, 2019, 03:34:55 pm »
Transformer won't help, the primary would be a short at DC.
Well you wouldn't use the same driver design.  There would be feedback to set the transformer current, and probably voltage feedback to control the AC gain.  This is how the old-fashioned class-A amps were designed.  Some still are, but they usually use vacuum tubes.

In a more proper single ended class A stage, the speaker would get a series capacitor and additional power resistor or active current sink would go from the emitter to ground to provide negative drive.
That's the resistor (class A) or push-pull (class AB) design I mentioned.  And yes, in a Class-A design a resistor could be replaced with an active current sink, but at that point you already have the makings for a well-performing class-AB stage.  I suppose we could argue about distortion...
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 

Offline Audioguru again

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ca
Re: why is class A audio amp saturating?
« Reply #18 on: August 15, 2019, 04:01:57 pm »
I think the class-A amplifier produced a low loudness in the speaker because the very high DC current in the speaker caused its coil and cone to be pushed hard against one side of its magnet structure.
Then the circuit does not amplify anything, it simply heats everything.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20363
  • Country: gb
  • 0999
Re: why is class A audio amp saturating?
« Reply #19 on: August 15, 2019, 09:12:28 pm »
I think I should highlight the point made about biasing, as I don't think everyone has picked up on it. In this case, biasing the DC point at half the supply voltage is undesirable. The output should be biased at half the maximum voltage swing of the amplifier output. If the maximum positive output swing of the amplifier is 6.6V (1.5V loss for the op-amp and 0.9V for the transistor, as emitter followers drop a lot more than 0.7V, at high currents) and the negative is 0V, then bias point needs to be half way between these voltages, which is 3.3V.

Using a pull down resistor will reduce the power dissipated by the speaker, but also efficiency of the circuit. If the resistor is the same value as the speaker's impedance, then the maximum power delivered to the speaker will be reduced by a factor of four, because it will drop half the voltage as P = V2/R. Using a lower value resistor will drastically increase the static power dissipation and using a higher value will reduce the maximum power output even more. Replacing the resistor with a constant current sink, but it would require an additional transistor and it will be cheaper to just use a class AB push-pull topology. Exercise for the original poster: work out the optimal current for the sink, given a voltage swing of 6V peak-to-peak (the current sink will prevent the output voltage reaching 0V) and an 8Ω speaker.
 

Offline d4n13lTopic starter

  • Regular Contributor
  • *
  • Posts: 71
  • Country: mx
Re: why is class A audio amp saturating?
« Reply #20 on: August 16, 2019, 11:56:47 pm »
Hi,

Sorry I don't have much free time so I moved to more practical amps, I made a class AB that is simple and works great. My only remaining questions are:

are class A amps still used today? if so, could you give examples of where are preferable and is just because people prefere to waste lots of power in exchange for the low distortion or there is some other reason?
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: why is class A audio amp saturating?
« Reply #21 on: August 17, 2019, 05:51:01 am »
Mostly still used when people want to burn lots of power in exchange for lower distortion.

Or to get acceptable performance from oddball 3-transistor topologies like below :)
http://www.firstwatt.com/prod.html
« Last Edit: August 17, 2019, 05:53:20 am by magic »
 

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2005
  • Country: us
Re: why is class A audio amp saturating?
« Reply #22 on: August 17, 2019, 06:29:55 am »
Mostly still used when people want to burn lots of power in exchange for lower distortion.

Or different distortion. Some guitar amps use Class-A (tube / transformer output stage) because when overdriven the harmonics are mostly even harmonics (due to asymmetrical clipping).  This sounds "sweeter" than the odd-harmonic symmetrical clipping of a push-pull circuit. 
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 

Offline Calvin

  • Regular Contributor
  • *
  • Posts: 179
  • Country: de
    • Calvin´s audio page
Re: why is class A audio amp saturating?
« Reply #23 on: August 17, 2019, 06:56:26 am »
Hi,

almost all discrete built lowlevel and linelevel audio (pre-)amps are biased in class-A.
Its just that the number of class-A biased power amps is rather small and mostly restricted to the socalled Highend scene, where the emotional experience counts not figures and numbers.
It can still be found quite often among tube amps, i.e. singleended Triode amps with typically less than 10W of output power.
Especially those simple SE Triodes (some even without global feedback) can perform on a level other amps aren´t capable of .... it may not be perfect in a HiFi sense, but can be a more envolving emotional experience.
They´re often combined with off mainstream speaker systems like efficient fullrangers or electrostats.
They are certainly specialty amps for a niche of users, no mass-market devices.

jau
Calvin





..... it builds character!
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20363
  • Country: gb
  • 0999
Re: why is class A audio amp saturating?
« Reply #24 on: August 17, 2019, 03:29:53 pm »
Hi,

Sorry I don't have much free time so I moved to more practical amps, I made a class AB that is simple and works great. My only remaining questions are:

are class A amps still used today? if so, could you give examples of where are preferable and is just because people prefere to waste lots of power in exchange for the low distortion or there is some other reason?
The only application of class A is for low power voltage gain stages, for example all op-amp input stages are class A.

Class A power amplifiers are of no practical use, but ome people like them because they believe they sound better or are interested in old designs.

For completeness here's an LTSpice simulation of the current sink, with the component values optimised.
 
The following users thanked this post: bmdaly


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf