my first power amp

Nantuko

Member
2007-12-23 6:47 pm
well this will be my first thread...

i built this amp using Toshiba C5198, A1941 and one Phillips NE5532N.
it is a darington pair push pull construction with alot of noise so i was wondering if someone would like to give me some tips to make it better..

Schematic and Pictures:

[IMGHTTPDEAD]http://img90.imageshack.us/img90/8176/myampgs0.th.jpg[/IMGHTTPDEAD] [IMGHTTPDEAD]http://img88.imageshack.us/img88/9243/dsc00566mv7.th.jpg[/IMGHTTPDEAD] [IMGHTTPDEAD]http://img137.imageshack.us/img137/8141/dsc00570ks9.th.jpg[/IMGHTTPDEAD]
 
Hi Nantuko,

there are a number of problems with the circuit as drawn,
remove the link between Q1 Base and Emitter as this stops Q1 from functioning.

swap the emitter and collector terminals for Q3 and Q4 then remove the link that connects Q4 base and emitter. This should then look like a mirror image of the top pair.

The next problem is there is no bias voltage, this a small voltage between the base of Q2 and Q4 so that both pairs are nearly turning on. Without it there will be lots of cross over distortion.

I suggest you take a look at the following link which is a similar circuit. the main difference is this uses a CFP output pair not darlington. In this circuit the bias voltage is generated by the 2 diodes, if you use a similar scheme you will need 4 diodes to correctly bias the darlingtons.

http://sound.westhost.com/project76.htm

Sorry if thsi is confusing but it's rather late here.
 

Nantuko

Member
2007-12-23 6:47 pm
Symon said:
Hi Nantuko,

there are a number of problems with the circuit as drawn,
remove the link between Q1 Base and Emitter as this stops Q1 from functioning.

swap the emitter and collector terminals for Q3 and Q4 then remove the link that connects Q4 base and emitter. This should then look like a mirror image of the top pair.

The next problem is there is no bias voltage, this a small voltage between the base of Q2 and Q4 so that both pairs are nearly turning on. Without it there will be lots of cross over distortion.

I suggest you take a look at the following link which is a similar circuit. the main difference is this uses a CFP output pair not darlington. In this circuit the bias voltage is generated by the 2 diodes, if you use a similar scheme you will need 4 diodes to correctly bias the darlingtons.

http://sound.westhost.com/project76.htm

Sorry if thsi is confusing but it's rather late here.

ahh i see mirror... i have connected q4 wrong..

but i dont understand whats wrong with the bias. i dont need the diodes where i am using an opamp voltage follower to compensate for the 0.7v that the transistors use...

http://www.ecircuitcenter.com/Circuits/pushpull/pushpull.htm

the link between Q1 base and emitter resolves some of my problems.. it removes some noise.. dont know why..

if you look here:

http://www.ecircuitcenter.com/Circuits_Audio_Amp/Output_Stage/Output_Stage.htm

you see the resistors re1 and re2. that was my thought so first i conected 2 trimports where the link is now. and i trimmed them. things got better when they reached zero. but probably thats because i connected q4 wrong?


i general i was just wondering if it is better to use opamps or diodes for the bias. i dont see the draw back. if i am using diodes and a voltage divider. then the transistors is allmoast turned on right? but if the transistors are hot wouldent they need a lower voltage to be turned on. i think this is called thermal runnaway. this problem is solved by the voltage follower(opamp)?

is opamps a limit or some sort of bad thing when making amplifiers. do they result in bad sonic properties?

i'm looking to buy the burr brown OPA 134 in exchange with the ne5532. may this make things better?


thanks symon:D

@longthrow: yes it does work. nothing is burning anyway..:D