It helps to select best class power amplifier (AB- 40-50 watts)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all:

I am new to the forum and want to introduce myself, I am a computer engineer and electronics technician. Professionally I work in computer engineering but I always wanted to build an amplifier.

I do not question the quality of the sound of tube amps and transistor stages in Class A. But I would like to mount a class AB stage (if I remember correctly have better performance). I do not like this:
Amp with power consumption 180W, but only 25W RMS (only 14% of energy is intended to sound)

Do you know any good scheme that can build a stage of 40-50W (RMS) class AB or low bias current (little heat in the final transistors)?

If someone would kindly give me some schema and / or advice to start with the project would be very grateful.

Thank you very much in advance.
-
José Manuel.

PS: Excuse me for my English (google traslator).
PD2: My living room has a size of 8x4 m
 
Last edited by a moderator:
Administrator
Joined 2007
Paid Member
Hi José, I altered your title to english... hope I got it right :)

I have to say I agree with you on the Class A comments and while there are many many designs on the forum I'm going to unashamedly point you to a Lateral FET design of mine that does all you ask. Low quiescent current and fantastic sound quality.

http://www.diyaudio.com/forums/solid-state/119151-my-mosfet-amplifier-designed-music.html

Also don't overlook the many "chip amps" based on the LM3886 and variants... plenty of designs in the chipamps forum.
 
Molly thank you very much. I do not like amplifiers based on "chip", I prefer minimalist designs based transistors. I believe that the ability to alter the audio signal is lower.

I will study your design. I look forward to your help in order to resolve the questions you might find me.

Do other designs or suggestions?

Thank you very much to all.
 
Administrator
Joined 2007
Paid Member
Well there are many designs, some offering outstanding technical performance such as the "blameless" class B" types and its not really fair of me to criticise any. All I can say is that sonically there are huge differences in amplifier designs and that for me, the lateral FET design with its single ended input stage has never failed to please on audition. I've built many amps over the years and bought a few too... and this one wins hands down every time.
 
There are so many designs here that it is impossible to pick just one.

Among the class AB designs there are some with EC (error correction) including Hawksford type, and also some that try by various means to prevent output transistor turn-off on alternating cycles... not to mention some that are very refined in both the bipolar as well as the jfet/bipolar, jfet/mosfet and jfet/bipolar/mosfet variety.

In fact, perhaps too many choices to make a simple decision.

Guess you will have to build up several! :D

_-_-bear
 
There's something I have clear Lateral FET/MOSTFET amplifiers have impressed me.

Mooly you design is spectacular, i have a couple of questions:

1) Why use the TL071 op amp?
2) What speaker protector (soft-start) can use?.
3) What is the latest version of the schema / parts list?
4) What is the latest version of the PCB?

Bear: What is your favorite amp?
catalin: He knew that topology, do you have any design?

Thank you very much to all.
 
Administrator
Joined 2007
Paid Member
There's something I have clear Lateral FET/MOSTFET amplifiers have impressed me.

Mooly you design is spectacular, i have a couple of questions:

1) Why use the TL071 op amp?
2) What speaker protector (soft-start) can use?.
3) What is the latest version of the schema / parts list?
4) What is the latest version of the PCB?

Bear: What is your favorite amp?
catalin: He knew that topology, do you have any design?

Thank you very much to all.

Thanks for the kind words...

1. The TL071 is ideally suited to a DC sevo. A FET input is madatory for such an application. The TL071 also allows the use of a single negative rail for its supply, something many opamps wouldn't cope with (and they would need a positive rail as well). Remember the opamp is not in the "audio path" as such. It just provides the correct DC bias for the input stage and gives a true 0.00 volt DC offset at the output.

2. Thats a good question. What is vital is that there is a reliable switch on delay of around 4 seconds to allow the servo to stabilise. That can be achieved very easily with a simple circuit. I believe there are many such circuits available as kits etc from the usual sources that incorporate DC offset protection as well.

I've also become a huge fan of solid state relays using HEXFETS's for speaker switching. I think there are some links and details in the thread but if you can't find them its here (read the whole thread though:))

http://www.diyaudio.com/forums/solid-state/191449-output-relays-14.html#post2660474

3/4. The circuit is as in the first post. There never was a bill of materials as such. Alex's PCB designs are first rate (when I made the amp it was a one off hand drawn special... no PC's) and I wouldn't envisage any issues using those.
 
Hey lopezm0,

I don't have a *favorite* amp at the moment.
A few years back i would have chosen my own Symphony No.1 amp (look at my webpages) but something has gone south in my own unit, and I have to bring it back to spec to make that statement presently... :D

Other than that, I have not listened to ANY of the DIY designs that have popped up here in the last 5 years, and there are SO MANY that you could do nothing but build DIY amps!! There are some incredibly well designed amps now, taking into account subtleties that quite frankly were virtually obscure in years past. One would never think that there are so many approaches to solving the various issues and problems internal to an amp design! Quite frankly it is incredibly diverse and wonderful at the same time.

I would not know where to start!!

Having said all that, I can say that on *my speakers* which are wide range horns (<300Hz -->~14khz) that a specific implementation of the simple F5 amp was hands down the best thing I have ever heard on them! Compared them to all sorts of other amps including some well done Western Electric 300B amps, and there is/was no contest. Dunno if they would be king on any other speakers though... the 109dB/1w sensitivity is a big plus running an amp with this low a power (it is pure class A).

_-_-bear
 
Hi Molly:

I have generated a new schema and bill of materials. Please, Could you check if schema and bill of materials are correct?

I have some doubts about these components:

C5 => ? / 25V
C8 => ?
C9 => ?/250V
C12 => ?/250V
C13 => ?/250V
R8 => ?
R9 => ¿Exist?

All resistors except R23, R24, R25, R26 are 1/4W ?

What is the power of R23, R24 and R25?

¿How can i construct L1?, what is the section of wire?, how many laps?

Should I roll over R26?
 

Attachments

  • Molly_PowerAmp.pdf
    42 KB · Views: 155
  • Molly_PowerAmp.txt
    1.4 KB · Views: 56
Administrator
Joined 2007
Paid Member
José,
I'll look at your list and diagram shortly.

The coil is non critical and tbh is only needed for deliberate testing into capacitive loads and even then the 0.22 output resistor provides sufficient isolation to prevent stability problems. If you want to include it then here's a picture.
 

Attachments

  • Coil.jpg
    Coil.jpg
    106.6 KB · Views: 286
Administrator
Joined 2007
Paid Member
Here we go,

C5 is 10uf 25 volt

C8 is 0.1uf 50 volt or higher polyester/film type (not a ceramic)

C9/12 and 13 are 0.1 160 volt or higher polyester/film type.

R8 is determined on test. I don't like presets as they can be a source of problems, particularly when a few millamps is flowing through them. To determine the value initially you can use a 470 ohm preset tagged in place of R8. You start with the preset on minimum resistance (a short) and increase until 100 milliamps is flowing in the FET's. That is easily determined by measuring the volt drop across either of the 0.22 ohms and setting for around 22 millivolts DC.

R9 Not present :)

All resistors except R23, R24, R25, R26 are 1/4W ? R27, the 10K can usefully be a 0.5 watt depending on supply voltage.

R23 and R24 were small low inductance "ceramic" coated ones. 3 watt or higher.

R25 is best as a 2 watt carbon film (or metal) so as to be totally non inductive at high frequencies.

Not shown on the circuit is a 10 ohm 1 or 2 watt carbon or metal film ACROSS the coil. It's just to damp the coil.

The coil we've covered. It's very non critical (if you want to include it at all).

The opamp MUST be a genuine TL071 type. I think the D notation is for an SMD part but any standard 8 pin DIL package device is fine.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.