Author Topic: Simple Audio Amplifier  (Read 2299 times)

0 Members and 1 Guest are viewing this topic.

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Simple Audio Amplifier
« on: August 22, 2019, 07:42:38 am »
Hi Guys,

I'm trying to learn how to design simple discrete audio amplifiers. I've come up with the attached schematic, based heavily off Self's blameless amp but with a few calculations made by myself. Can anyone offer any feedback or see any glaring issues with it before I order the components and give it a go?

Design Specs:

* 16W into 8 ohm
* Single Channel
* SMD components

My Calculations:

* Supply current is 2A RMS
* Voltage approx 12V RMS
* Planned supply (+VDC and -VDC in the schematic) is +/- 20V
* Long tail pair input stage - current source is a Wildar, aiming for approx 600uA
* Cascode VAS, simple current mirror source, aiming for approx 100mA
* I will be building this direct to PCB, running power from my benchtop supply (this is a learning circuit, not a practical one)
* My big concern for this being SMD is the thermal capacity of the output transistors - initial plan is to use a big ground plane with lots of vias to see how much cooling I can get that way, but I will have a range of heatsinks (SMD and custom) to try and see how it goes in practical use


Thanks very much for any feedback you have! :)
« Last Edit: August 22, 2019, 07:56:21 am by braddrew0 »
 

Offline Calvin

  • Regular Contributor
  • *
  • Posts: 179
  • Country: de
    • Calvin´s audio page
Re: Simple Audio Amplifier
« Reply #1 on: August 22, 2019, 08:26:26 am »
Hi,

the schematic shows several issues.
The most prominent one is the wrong ´polarity´ of the VAs stage (Q10).
The Collector potential of Q1 would have to be positive, the Base potential of Q10 is negative.
R2 should just be a fixed R of ~100R to 1k .... often it isn´t needed anyway.
A value of 1k is often used as R-part of an RC-lowpass filter with C in a range of 220pF to 1nF.
In that case place R2 ´in front of´, i.e to the left of C2.
Instead of two current mirrors for the long-tailed pair CCS use simple 1-Transistor/1LED current sources with transistor and LED thermally coupled (the CM design/dimensioning appears questionable in itself)
There´s no Miller compensation cap around Q10, nor any other compensation arrangement.
There´s also no bandwidth limiting cap parallel to the feedback R11 ... typical a few dozen pF.
Then there´s no resistor from the feedback point to gnd ... just C3.
You have to add a resistor in series to C3, the value depending on the required gain (Av=1+R11/Rx).
With R11 beeing 22k  Rx would probabely fall between 470R and 1k (typical gains ~30dB).
Finally You didn´t draw a V+ connection to the output transistors.
So back to the design board  ;)

jauu
Calvin
..... it builds character!
 
The following users thanked this post: braddrew0

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #2 on: August 22, 2019, 09:09:48 am »
To be honest I expected more to be wrong so this is a good outcome for me :)

* I'll pull out the trim resistor R2 - in my head, this was going to be volume control. I'm planning to test by plugging into the headphone output of my computer and I wasn't sure if that output would be too high for the amp.

* By 'polarity', do you mean Q10 should be a PNP? The design is based on VAS configuration D on the self article (http://www.douglas-self.com/ampins/dipa/dipa.htm) - that has both of the transistors as NPN, although what you're saying makes sense :)

* Likewise for compensation of Q10 - is C4 sufficient? (Again that came from the same Self configuration)

* I'll switch out the current mirror :)

* I totally missed the voltage divider on the feedback loop! I'll add that in, along with a cap

* Will also add the V+ line... that will help a lot :)


Thanks a bunch for your help!
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Simple Audio Amplifier
« Reply #3 on: August 22, 2019, 10:15:09 am »
This thread has several similar schematics you can look at:
https://www.eevblog.com/forum/projects/classic-40w-audio-amp-with-discrete-components/
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 
The following users thanked this post: braddrew0

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: Simple Audio Amplifier
« Reply #4 on: August 22, 2019, 12:08:09 pm »
......
* By 'polarity', do you mean Q10 should be a PNP? The design is based on VAS configuration D on the self article (http://www.douglas-self.com/ampins/dipa/dipa.htm) - that has both of the transistors as NPN, although what you're saying makes sense :)

* Likewise for compensation of Q10 - is C4 sufficient? (Again that came from the same Self configuration)


Q10 should be PNP, or the input differential stage should be PNP.  Usually the input stage and VAS stage use opposite polarity transistors. As the VAS stage tends to be more critical with speed, the preferred version is often PNP differential and NPN VAS stage.

A value of 47 µF for C4 is way to large. The usual order of magnitude is more like 100 pF-1 nF.
 
The following users thanked this post: braddrew0

Offline magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: Simple Audio Amplifier
« Reply #5 on: August 22, 2019, 02:55:24 pm »
Yes, turn the whole input stage upside down.

I don't see the alleged cascode over VAS anywhere on your schematic.

C4 certainly isn't what Self uses. You mixed up VAS compensation with bypassing the bias spreader. See Self's figure 33: C3 is compensation around TR12/TR4 Darlington VAS, C4 stabilizes voltage across bias spreader made of TR13 and surrounding circuitry.

BTW, schematics are easier to see (and more likely to be seen) when uploaded as image. At least when it's something that fits on one page, like here.

And there is something wrong with R11/C3 feedback network. I don't think you copied Self exactly here ;)
 

Offline Audioguru again

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ca
Re: Simple Audio Amplifier
« Reply #6 on: August 22, 2019, 06:19:14 pm »
And there is something wrong with R11/C3 feedback network. I don't think you copied Self exactly here ;)
I noticed that C4 completely removed audio negative feedback. Then this amplifier will have lots of "overdrive" producing lots of "fuzz" as an electric geetar amplifier (then it should use vacuum tubes).
 

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2005
  • Country: us
Re: Simple Audio Amplifier
« Reply #7 on: August 22, 2019, 06:26:38 pm »
Are Q2 and Q4 supposed to be a current mirror or something? They are shut off by the base-emitter connection and will do nothing.
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #8 on: August 23, 2019, 06:10:21 am »
Thanks everyone for the help! I've flipped the input stage, fixed the feedback loop and removed the two current sources to replace with a much simpler LED stabilised version. I was trying to avoid an exact copy of the Self amp (which is why the input and VAS didn't match, and why it had odd current sources) but I'll give this one a go and see how it works.

Thanks again! :)
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: Simple Audio Amplifier
« Reply #9 on: August 23, 2019, 08:49:13 am »
The general concept is OK and the circuit might actually work this way. There are just details that may need adjustment.

The simple current source usually only uses 1 LED. The TC of the LED and the transistor can about compensate with only 1 LED.
The current for the input stage is chosen rather high - this gives quite high an input current. A more usual current level is at 0.5-1 mA. So R4 should be more like 1-2 K. For a small DC offset it may help if R1 and R10 are the same value.

The MMBTA05 transistors are not such a good choice: they are relatively slow and low power. Q6,Q8,Q9 and Q10 may run rather hot. R9 and Q10 may need a slightly larger form factor. Less current (larger R14) could reduce the power for Q6 and Q8.

R16,R17 are relatively low value for the small transistors and 8 Ohms load.

R13 is a little low, so that one may not be able to set the standing current high enough.
 
The following users thanked this post: braddrew0

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #10 on: August 23, 2019, 09:12:57 am »
Thank you, these are exactly the kind of things I wanted to learn from this :)

Can you recommend a suitable SMD transistor instead of the MMBTA05/55? I found these through a Digikey search for ~60V (to give a buffer on the +/- 20V), minimum 500mA and SMD (most were SOT-23). Should I search for something that can handle higher current?
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: Simple Audio Amplifier
« Reply #11 on: August 23, 2019, 10:00:41 am »
For Q9, Q10 I would consider a slightly larger form factor, like SOT223 or SOT89. So something like BCP56/53 or similar.
These transistors see some 6 mA from R15 and maybe up to 20 mA from the base current.  This may be still just inside the MMBTA05 specs, but at the limit the transistor would be really hot inside and this has a negative effect on thermal stability. The limiting value is also valid for quite some copper area.
The limiting factor is not the current rating, but more power rating, possibly even the SOA.

60 V voltage rating is good, though other transistors (except Q6,Q8,Q9,Q10) could get away with a lower voltage rating.
 
The following users thanked this post: braddrew0

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #12 on: August 23, 2019, 08:49:12 pm »
OK, I've updated R4 to 1k (from 100R), R14 to 470R (from 100R), R13 to 470R (from 100R) and the output transistors to 0R47 (from 0R22). Back to one LED as well.

Q9 is now BCP56 and Q10 BCP53. Is there any benefit to running BCP56/53 on the other transistors in the audio path (Q1, Q5, Q8) or even for just all of them except Q11/Q12?

Thanks again :)
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: Simple Audio Amplifier
« Reply #13 on: August 23, 2019, 09:15:33 pm »
Using larger transistors where not needed also has a downside: they have more internal capacitance and thus tend to be slower, especially when used at a relatively low current. So it's better to have the small transistors for Q1-Q8  (Q7 is not critical).

There can be a slight advantage with large transistors as the noise might be lower, though often not specified. Noise would be an important factor for Q1 and Q5.
For Q7 the larger transistor has a chance to get a lower U_BE and thus more temperature effect, which can be a good thing - it's preferable to have the standing current to go down with temperature and not up. Chances are the difference is small.
 
The following users thanked this post: braddrew0

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #14 on: August 23, 2019, 10:03:43 pm »
Thanks, that's a great explanation! I'll check back in when it's made :)
 

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #15 on: September 19, 2019, 08:43:06 am »
Small update - it works! :D

Thanks to those who helped me with this project. You might notice from the DC readout that I have about 44mV DC offset... this seems quite high, is anyone able to shed light on why it's likely this much and/or potential fixes?

Lessons learnt for me:

* Drawing a max of about 500mA per channel from my benchtop supply (so ~1A total into 8 ohm = 8 watts) the speaker is about as loud as I'd ever want to have it...
* Running at this power setting for about 15 minutes, the parts didn't even feel warm...
* Don't use DPAK SMD power resistors without an oven! Seriously - these were near impossible for me to hand solder, even with a bucket of flux. Also I misread the pinout from the spec sheet - the two 'legs' of the DPAK are the resistor legs, the big pad is just for cooling. Next time will be through hole for both the inline and filter resistors (ie the high power ones).

Thanks again! :)
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15157
  • Country: de
Re: Simple Audio Amplifier
« Reply #16 on: September 19, 2019, 08:59:41 am »
The DC offset can have different origins, like a mismatch in R2 and R5 (emitter resistors at the input). Q1 to Q5 and Q2 to Q4 VBE matching can also be a factor, good for a few 10 mV. Another point is that normally R10 and R1 should be the same value to avoid an offset from the input bias current (increasing R1 to 22 K should be easy). Input Bias current should be on the order of 1-3 µA so a 12 K mismatch would cause some 10-35 mV of offset.

A point to fix adjust the offset would be at R2/R5 or R3/R6.  So something like an extra resistor in parallel to R3 or R6 should be able to adjust the offset.
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: Simple Audio Amplifier
« Reply #17 on: September 19, 2019, 01:46:37 pm »
Another potential source of offset and maybe some ppm-level nonlinearity is the difference (if any) between base currents of Q4+Q2 and Q8.

Collector currents of Q2 and Q4 are equal. Q5 supplies base current to Q4 and Q2 in addition to Q4 collector current, Q1 supplies base current to Q8 in addition to Q2 collector current. Therefore Q1/Q5 currents may become unbalanced if Q4+Q2 and Q8 base currents aren't equal and a permanent offset voltage will develop to produce that imbalance.

Self failed to notice this problem despite all his obsession with current mirrors ;)
Supposedly the latest 6th edition has been corrected.
« Last Edit: September 19, 2019, 01:56:00 pm by magic »
 

Offline braddrew0Topic starter

  • Regular Contributor
  • *
  • Posts: 79
  • Country: au
Re: Simple Audio Amplifier
« Reply #18 on: September 20, 2019, 06:56:43 am »
That's great info, thanks! I did change R1 to match the 22k of R10, but I didn't use any form of precision resistors - they're all 1% 0805. I'll have a go at trying to balance R2/R5 and R3/R6.

With regards to Q2/4 and Q8 - I actually changed out Q8 to a BC850 (with Q1 as a BC860) and left Q2/Q4 as MMBTA05. Should these transistors be matched?

Thanks again!
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: Simple Audio Amplifier
« Reply #19 on: September 20, 2019, 08:15:22 am »
No, only Q2 with Q4.

Ib(Q2)+Ib(Q4) is supposed to be close to Ib(Q8). You will never get it perfect in this input stage topology because transistors aren't exactly identical, even from one production batch. You can get close by design and calculations, knowing that Ib=Ic/β and Ic of each transistor is something you design and control while β is known approximately from the datasheet.

But things need to be put in perspective. Transconductance of your input stage is probably at least a few mA/V, which is a few µA/mV. So unless current imbalance exceeds a few µA, offset won't exceed 1mV. Okay, it seems I just proved it's a nothingburger unless you did something crazy like running the VAS on huge collector current using a very low-beta transistor.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf