Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Construction Tips Construction techniques and tips

Building a Class Amplifier from Scratch
Building a Class Amplifier from Scratch
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 22nd May 2020, 04:09 PM   #1
hpro is offline hpro  Switzerland
diyAudio Member
 
Join Date: Apr 2012
Location: Bern, Switzerland
Default Building a Class Amplifier from Scratch

Hi everyone here..
I'm new here.. see my introduction then you know..
Building a Class A Amplifier from scratch.

Answering following Questions before you start. Here I talk about YOUR IDEA, not a design someone else posted on the net, or you find a finished circuit in some Magazine or likewise.

Questions should be as follows
1. Why Class A
2. How much can you spent on it
3. Are you conform with what it takes to build an Amplifier, never mind the class?
4. What the specs of the amp should be?
5. do you have the necessary tools to build it.
6. Do you have the necessary Computer Design and Simulation programs?
7. How about your technical knowledge?
8. Do you know how Class A operates versa Class B & Class AB - not Talking about the rest of Amps.


To build a decent amp, you need to be able to simulate and go beyond the regular designs. Today many Class A are sold which are in fact just AB amps BIASED with A Bias.

For me there are two Kind of Class A.
1. Single End Class A = this is an Amplifier where one Transistor per Stage making all the Work. Meaning amplifying the whole 360 degrees of the signal.
2. Push Pull where one Transistor amplifies the Signal and the Second one is the Current Source which enables the "Signal Transistor" to Work freely

The difference between Class A and Class B, respectively Class AB is, that a Class A Amp does not break the signal in two pieces.. Upper and Lower Signal

Class A is the only Amp where Distortion is the at minimum, where heat doesn't matter like in other configuration regarding Signal amplification, this only when the amp is designed well and the Working point of the Transistor is exact in the Middle of the Working field of that Transistor.

That the Sound of it is more close to the Natural Sound what you hear everyday. that the Amp reproduces the sound without coloring it.

The disadvantage is that you loose about 80% of the power. So if you have a Class A Single End Amplifier and you have a power supply with 0 + 50 Volts then the max Power you can expect to hear out of the Speaker are about 17 - 20 Percent. this means that you will get an RMS Voltage at the output of 10 Volts in the best case and only 8.5Volts in the NORMAL CLASS A Amplification scenario. So if you use Single End, then you are lucky if you get this 10 volts out, this would mean that you have about 12.5 Watts RMS @8Ohm, not yet looking at the Distortion which might occur because of so many reason like Components, layout of the PCB, Supply Voltage, etc etc.

Knowing all this I'm still going for Class A. Because if you really seriously do your job and giving your Best, consult the big guys if you get stucked, then success will be on your side. Then the next part of the story is what you do about the HEAT of a CLASS A. As the Amp burns Power away more than its able to deliver to the speaker..


In the days to come I will place some Schematics, PCB layouts and Building "instructions" which are necessary if you want to build a Class A.

Building Class A Amps you need to be able to Jump over you Shadow in sometime building something where all the world would say: na this never works.. but if you believe in what you do then I'm Sure that success is on your side. This is a small example. If anyone likes to see how the Amp looks like inclusive Specs and everything, then just ask. Developed by me. There is one resistor R2 10Ohm, its 200Watt.. the rest is all regular. this config has an output of about 12.5 Watt @8Ohm linear 10 -100Khz with very low distortion below .01 Linearity is +- 0.2 db over full bandwidth..
BTW I'm open for everything and with everything I mean I also can take Critics and all what comes with it.. Just let me know

Thanks for reading
Regards Chris
http://www.hecco.ch/pictures/class_a... Amplifier.pdf

Last edited by hpro; 22nd May 2020 at 04:20 PM. Reason: Typos
  Reply With Quote
Old 22nd May 2020, 11:11 PM   #2
PRR is online now PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
Building a Class Amplifier from Scratch
> Developed by me.

Are you claiming "invention"?

The idea has been around since the 1960s. I've done it with a resistor-load, but efficiency is few-% at best, optimum supplies are not balanced, and linearity is poor. I can't find my notes on that.

With your opamp on +/-22V, 10r and 8r, you can't get 8.88Vpk 6.3Vrms before clipping.

> two Kind of Class A.

There are MANY kinds of class A audio amps.

D.Self argues convincingly that a properly biased AB BJT is lower THD than any other amplifier.
Attached Images
File Type: gif Tori-Amp-NPN.gif (5.2 KB, 242 views)
  Reply With Quote
Old 23rd May 2020, 10:16 AM   #3
hpro is offline hpro  Switzerland
diyAudio Member
 
Join Date: Apr 2012
Location: Bern, Switzerland
Default No

Quote:
Originally Posted by PRR View Post
> Developed by me.

Are you claiming "invention"?

The idea has been around since the 1960s. I've done it with a resistor-load, but efficiency is few-% at best, optimum supplies are not balanced, and linearity is poor. I can't find my notes on that.

With your opamp on +/-22V, 10r and 8r, you can't get 8.88Vpk 6.3Vrms before clipping.

> two Kind of Class A.

There are MANY kinds of class A audio amps.

D.Self argues convincingly that a properly biased AB BJT is lower THD than any other amplifier.



Hello

Thank you for your response.
No, I do not claim invention. BTW development and invention are two different things. At least in German, French Language and understanding and if this isn't like that in English then I'm sorry, my BAD.

I'm here to learn and give my knowledge further if this is OK with you. But sure I'm not here to discuss nonsense.

I would have a lot of Questions to you and the Circuit you presented is impressive. Now my build are not on the same level as yours.

I'm not satisfied when I get only 6.*** volts out of an Amplifier which I feed +-37 Volts or more into. And I will do anything to change that as long it is in my financial power.

First of I will download the Datasheet of the components I'm going to use.. and then read them through fully and after that I will use my Computer Program which is BTW National Instruments Circuit Design 14 or call it MultiSimm14 .
Then I will bring down my thoughts into that Program and calculate what is possible and what not. IT'S very hard to cheat that program and I'm for one far off from cheating in Circuits or on the results, because the only one I would cheat, is myself.

So far so clear I hope. I also know the rules of OHM. Also I know the CLASS A Rules and what can be done and what not. Also can read Resistor Colors. Reading a Scope whatever Brand and also Digital and Analog Multimeters.

Specs of Operational AMP OP37E & OP37GPZ ( Texas Instruments) I use, both of them (price per piece 5USD and more) Voltage P-P @22Volts Rail +- is 13.5 Volts. this is 9.5445 Volts RMS.


Let the Facts Talk:
http://www.hecco.ch/chris_hess_class_a_amps_2020.pdf
Enjoy weekend as I do..
anything else you might want to know sounds suspicious to you, Just ask.
Regards Chris Hess

Regards Chris Hess
  Reply With Quote
Old 23rd May 2020, 05:49 PM   #4
Mark Tillotson is offline Mark Tillotson
diyAudio Member
 
Join Date: May 2018
Location: Cambridge UK
+/-22V is really pushing the opamp supply, +/-19V is a more realistic range to choose to reliable long term operation.

There's not much point having the main supply much larger than the driver supply, all that does is waste more power and generate more heat.

The feedback network is rather high impedance, why not use 10k/1k and get less noise?

The amp isn't strictly class A because the NE5534 output stage is class B internally. You can bias it externally with several mA of pull-down current to force the output stage to class A. Arguably the output transistor already does this, but you'd need to check that's the case.

There are ways to bootstrap an opamp to track the signal and be a floating driver, which might be worth persuing for such a design to give more voltage swing and power. At higher power the opamp is going to struggle to deliver enough current unless a driver transistor is added.
  Reply With Quote
Old 23rd May 2020, 06:26 PM   #5
hpro is offline hpro  Switzerland
diyAudio Member
 
Join Date: Apr 2012
Location: Bern, Switzerland
Sorry I was in a hurry this morning when I read your comment.

Anyway as you see in the pictures it is possible to get as much as 12.25 volts RMS out if Rail is 22 Volt or more.. With more I mean going 10% over the maximum Rating of the IC.. it doesn't kill the IC nor does it any bad as I can see. So the Voltage I apply to OP37 as well as the NE5534 is 23.60Volts + and 23.60Volts -. Speaking for the OP37E gets a better sound but less Output Level as the NE5534.

I have build many amps in my past 30 years, just for fun not for Sale. Here at home in Switzerland, I still have a Forte Audio A100 - Threshold - Speaking about that amp then the designer is the same as First Watt or Pass Labs.. Nelson Pass.

I spend a lot of my personal Money to buy amps in these 28 Years which I lived in Asia. I had a Threshold and many others also some Valve amps.

But I wanted to build my own. So I'm still on it..

This what you see is just a part of a project I'm trying to achieve. Nothing more nothing left. Use Google to find Single End Class A and you will not find a Class a with a IC in front nor with dual voltage. I have build several Transistor Class A.. Show you one here..

Usually I build every two or three years a new Amplifier just to proof to myself that I still have the patience to sit down and calculate thinking about the design.etc etc. I'm almost 70 and so I have another 2 years before I will stop and return to my second home in Thailand, and will stay away from public, just enjoy the last years of my life,

Now I'm still have that drive to try something new to me. Anyway my goal is this project I'm working on to have a Class A with on or two Tubes at the input and then Single end Transistor output. without C-Coupling as fas as possible and of course no Coupling transformers..So this Class A I have showed are just to find out what way I need to go.. I do not intend to use Mosfets or Fets..
OK a few Pics will follow.. this is the PCB of Version 7 created sometime in September in 2017. its A single end Class A from Input to output. Idle or Bias is 2.5 amps at+- 48Volts
According my measurements its deliver 50 Watts@8 Ohm over a bandwith of 10 - 100K.
Input is similar to many amps in the open, single transistor. I use a PID to control DC at the output.

First pic is the PCB in two variations. Designed by ME
Following pics is the Amp itself and the Preamp American Made
http://www.hecco.ch/chris_hess_class_a_amps_2020_01.pdf
  Reply With Quote
Old 23rd May 2020, 07:23 PM   #6
hpro is offline hpro  Switzerland
diyAudio Member
 
Join Date: Apr 2012
Location: Bern, Switzerland
Default Yes - Your completely right.

[QUOTE=Mark Tillotson;6216096]+/-22V is really pushing the opamp supply, +/-19V is a more realistic range to choose to reliable long term operation.
There's not much point having the main supply much larger than the driver supply, all that does is waste more power and generate more heat.

The feedback network is rather high impedance, why not use 10k/1k and get less noise?


Thank Your for your Input. it's appreciated.


This is CLASS A Single End output, with no " CURRENT SOURCE TRANSISTOR"other rules apply.

Take at the best 20% of the Line voltage or Rail voltage and then the Amp Clips. It's about 17% in reality, but the sound is more lively than with using a Current Source Transistor. I have several Current Source Class A

37 + 37 : 17% = 14.8 only so the 15Volts RMS can be schieved.

We have a saying in Swiss, a Class A Amp which don't heat up isn't a class A. And here nobody touches it.. My Wife stays two meter away from the amp..

Another one I build was up to 140C.. nothing happen to it. it was running for hours like that. This was 7 years back as I build that one.. still running perfect till today. Rail voltage 48Volts +-

I have about 10 or 12 self made Class A here at home, and another 5 in Thailand in my second home.. I'm Swiss, Wife is Thai. Owing a Thai Restaurant here in Swiss. This is just for fun.

If you read my follow up this is part of a project of me. I like to achieve the impossible.This is only the start.
I just finished calculations for full transistor with single end wait I let you guys see soon.. Layout is finished already.

As always I use a PID (proportional-integral-derivative controller) to keep DC at the Output under control.. you can also call it Servo or whatever.
It doesn't matter how it is called, it matter that it keeps 0Volts DC at the Output.. Besides that it will run and work without the SERVO very well at least in multisim 14.

Yes about the Feedback Network I will try that, thanks for the input

About the IC-OP37 Texas Datasheet give max 22. the NE5534 isn't the problem.. It's the OP37 which just delivers a little bit to little for me.. I wish to have 15Volts RMS this would equal 26 Watts - 28 Watts depending on the Distortion level which should be achieved..You know, a good Amplifier deosn't need to be turnet up all the way, because it will sound excellent on low output level already.. I seldom listening loud.. protectiing my ears is worth much more..
Talking about distortion everything under .5% cant be heard by Human as Studies from Germany showed some times ago.. they even were talking about 2 %
If you are looking at the specs of First Watt F7 then you know what I'm talking about even Nelson Pass check this 2% mark at 4 ohms
So it actually doesn't matter if its between 0.001 and .1

Thanks again for your Input, and Input is welcome.. positive or negative.. doesn't matter, because important is to hear what other people say.. it's always some truth in it, it's always a Win win Situation.. I'm proud to be here..
  Reply With Quote
Old 23rd May 2020, 08:18 PM   #7
nigelwright7557 is offline nigelwright7557  United Kingdom
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
Here is a JLH 1996 class A amplifier i came across.
Didnt quite work when I built first one but works now.
Click the image to open in full size.
__________________
2020 versions of PCBCAD51, PCBCAD360 and PCBCAD720 out now >>> https://www.murtonpikesystems.co.uk
  Reply With Quote
Old 23rd May 2020, 10:34 PM   #8
hpro is offline hpro  Switzerland
diyAudio Member
 
Join Date: Apr 2012
Location: Bern, Switzerland
@nigelwright7557

Yes, thank you very much. Looks really good..

There ain't be problem with the Amplifier I Build. It works, all what I do now, is just perfecting it. If you read my last post then you see my goal I set myself. This Amp the Single end Class A with Op37 as input and driver, is only the first step to get the limits close up.
Building electronic devices I go always for the maximum, there is nothing like " OK that's good enough" this sentence is not used.

@ Mark Tillotson I changed the Value of the feedback resistors, the way you mention.

Outcome. NE5534 is now more Linear than before max output RMS of the Amp with 8ohm load is 11.20Volts up to about 70Khz starting at 10 Hz.
As for the OP37E and OP37GPZ the output still not reaches the "announced level" of Texas instruments" Now its not a big thing.. But because of your input about the Network, I found a error in the PCB layout. the RAIL to BASE of the MJL4281 is definateley to "Long and also to close to collector Rail" so that this Oscillate on the Base of the POWER OUTPUT Transistor. But only whit the OP37, this don't count for the NE5534. If I reduce the Impedance of the Feedback more that I already have, then I fear that I loose the spirit of the Amp, meaning that it will sound dry. so now I stop that one, because it maxed out already.

Tomorrow I will manufacturer the PCB with the transistor and then a second one where I use a Emitter switch to increase Voltage Gain by 100% this means that the OP37 then only needs to amplify 50 % of now, instead of 11.5 volts. 5 - 6 volts are enough.. but as I already wrote, the transistor PCB with Class A from input to output comes first.

Earls in the morning will get that PCB done. I hope that I can test tomorrow afternoon.
I let you know..
Thanks at everyone.. This will be a LOOOOOOOOOOOOOONG journey..

The difficulty will come when I start the TUBE PART.. I never ever have build a Tube..

I will place all Layouts and Circuits online here as soon I can be sure that in long

term they don't fail.
Guys it 0:30 Am in Swiss, time to sleep for an OLD man.
See ya Happy Sunday
Chris
  Reply With Quote
Old 23rd May 2020, 10:56 PM   #9
analog_sa is offline analog_sa  Europe
diyAudio Member
 
analog_sa's Avatar
 
Join Date: Aug 2002
Location: Cascais
Building a Class Amplifier from Scratch
Quote:
Originally Posted by hpro View Post
but the sound is more lively than with using a Current Source Transistor.

It is even more lively when instead of a resistor a choke load is used. Power and efficiency also go up. This choke will perhaps work well in your circuit

http://www.lundahl.se/wp-content/upl...heets/2733.pdf
  Reply With Quote
Old 25th May 2020, 02:22 AM   #10
PRR is online now PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
Building a Class Amplifier from Scratch
> instead of a resistor a choke load is used. Power and efficiency also go up.

Yes. Choke loaded efficiency "can" approach 50%. CCS-loaded, 25%. Resistor loaded, 8.5% (8.5786438...%).

But then it becomes a very different amplifier. Bias current can be very hard to control. We can't buy a perfect choke so there is some DC on the speaker.
  Reply With Quote

Reply


Building a Class Amplifier from ScratchHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Little help building Class D amplifier daKube Class D 2 15th February 2016 08:39 AM
Class D Amplifier -- building blocks Sch3mat1c Tubes / Valves 15 23rd July 2015 11:34 PM
Second attempt at Building my own amp from scratch Vostro Solid State 106 9th May 2014 01:21 PM
Building Class D amplifier with PIC16F877A RealForce Class D 2 22nd April 2010 10:10 PM


New To Site? Need Help?

All times are GMT. The time now is 05:16 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki