2 x 15W rms Audio Amp for small speakers

[maouna]

i have 2 2 way speakers for laptop and i want to build a discrete amplifier for these using tip41-tip42, 2n5551-2n5401,bc549-550. any schematic in mind?

 

[Bigun]

I'm not familiar with all those devices, but you could look at something like the DX amplifier with 27V power rails.

Greg's Web Site

[I did something similar here: http://www.diyaudio.com/forums/solid-state/140461-tgm-amplifier.html ]

 

[tinitus]

isn't 15watt too low for classAB anyway ?

 

[knutn]

Maybe you should try the JLH Class A amp?

 

[maouna]

class A will create much heat and there will be the need for big heatsinks.i dont want to use ic's.I would like something with descrete components.

 

[krokkenoster]

15 watt amp

isn't 15watt too low for classAB anyway ?

Tinitus Iremember the days when a 10 watt amp could be heard for two miles from the source on a farm! Fifty years ago we were happy with 5 watts/channel class A with a EL84 output tube. 15 watts with two in push pull fed a large Railway station with sound. So if he need it for his office or home with his computer then I think 15 watts class a/b will be ample! THIS IS WHY A LOT OF US HAVE TINNITUS AND ARE HARD OF HEARING? (pun intended)

 

[jaycee]

You can do it with a pretty standard Lin/Blameless type topology.. powered from 18V rails or so. Use your TIP41/42 as the output transistors, and 2N5401/2N5551 for everything else.

 

[destroyer X]

This is really interesting....Class A amplifiers seems to play louder

It is non sense, but 10 watts in class A plays very loud.

Please folks, i cannot understand because i am not addicted nor skilled with the stuff, why this happens.

I have perceived that when i have increased the Dx Super A to 700mA per rail quiescent current.... i found it louder.... really louder than another amplifier playing almost same level.... say.... watching in the waveform it looks the same..but the Class A was louder while listening.

Thid does not make sense to me.... can you explain...please

regards,

Carlos

 

[lineup]

You can do it with a pretty standard Lin/Blameless type topology.. powered from 18V rails or so. Use your TIP41/42 as the output transistors, and 2N5401/2N5551 for everything else.

This is a blameless using those transistors you mention.
The power supply should use a transformer 2x15VAC, which will give like 20-21 VDC rails.
I have tested with SPICE. It should work.

 

[maouna]

I think tha the ideal for me would be jean hiraga's ''le monstre'' class A 8 watt,as described in other threads here in the forum and here JEAN HIRAGA - Le Monstre Amplifier PCB ,on monday i m going to have two speakers 7 watt rms/4ohm rated each like the ones in the picture. I dont know if the result willl be remarkable but maybe it worths giving it a try...

 

[Ian Finch]

class A will create much heat and there will be the need for big heatsinks........

Are you sure about what you want?

 

[tinitus]

I think the ideal for me would be jean hiraga's ''le monstre'' .......

may I ask, based on what ?
reading old threads ?
not there is anything wrong in that, but...

though, I see one advantage of the small 8watt
it doesnt need much of anything

btw, if those are your speakers, they look like they are powered speakers

 

[destroyer X]

I have tried Hiraga once

And i like it.... not hundred percent but i like it...a good try.

regards,

Carlos

 

[jaycee]

a big problem with le monstre and indeed any amp that has fets in it's input, is getting the fets... they are practically unobtanium unless you are willing to go to the ends of the earth for them, or gamble on ebay

 

[maouna]

yes the speakers are powered with internal amplifier but i want to add an external amp

 

[bobodioulasso]

Build some good speakers first.
The ones you show deserves no more than the chip amps inside.

 

[Bigun]

Carlos - that's an interesting observation because Class A no-feedback tube amplifiers are often reported to sound louder and it is usually explained by their soft clipping but I never heard of a SS amp with feedback being accused of sounding louder before.

A Class AB 'blameless' is also my vote - like the one I made and linked in my post above. It was designed for low voltage rails. At really low voltage rails things need a bit of care because you start to lose some headroom and the LTP benefits from a current source like Lineup shows rather than the resistor I used.

 

[Ian Finch]

My experience with recent small speakers is that they are now very inefficient, like 80-84 dB/W/1m, though this is hidden by their being fitted with dedicated amplifiers and not being required to disclose sensitivity.

Some small "professional" powered speakers with 5" woofers seem to require two 40W chips in each box. Because they need to trade off so much mid/treble sensitivity for bass, frankly, I don't think you'll quite get the power you may need there. Your 15W plan seems more promising. Just aim to do whatever an LM1875 chip does in discrete format and you should be on target for a wide range of small commercial boxes.

 

[maouna]

I should aim at something like classic 3 stage topology like Bigun and Lineup mentioned before.This is my design but then comes in mind some more questions like:should i use current mirror in LTP? or a buffer in VAS? or some kind of cascode?or Sziklai output due to lower quiscent current than EF output? or they are to much for such a simple application?

 

[maouna]

Sorry,values are not clear enough.Here's a new image.