Poweramp Needs Help Will Not Get Louder

[prorms]

I can not get this amp to crack up the power is ther and all sound like is 40watts I just useing a portable singal generator my transformer get up to 70volts and 4.5 A with the a va of 100 I can add a lot more amps when I do get a little stonger in volume I mean that it will IT WILL CARRY FUTHER BUT NOT LOUDER ,
ANY PARTS YOU THINK ADDING OR CHANGE BE REALY APPREACTED
TRANSISTOR ARE TIP147 & TIP142 DARLINGTON TRANSISTORS
MJE15034 AND MJE15035

[MikeB]

Hi, it's a wonder that this amp plays anything... You messed up all biasings, you can't connect the vas transistor via a cap (c13) to the diffamp output, you are feeding the voltagesupply to the diodes for the ccs (d5/d2) through a 1k, that's way to small, 70v will cause 5watts to dissipate in the 1k, try 47k. Also, in my experience, c1 can be a disaster, and NEVER connect anything to an asymetric inputstage without couplingcap ! What is c4 good for ?
R9 looks shorted ? Also, i don't see how the outputstage protection should properly work, it needs to measure the current in the outputstage but is only connected to the output itself. Also your inputstage is not balanced in any way... And you missed the zobel at output...

Sorry, i can't recommend any single parts added to get this amp work, it needs to be redesigned. Where did you get this schematic ?

Mike

[prorms]

I redesing the power AMP to what was said and it realy made differents on the simulator and breadboard much louder and clearer.the parts that i change are


c13 remove
c4 removed
r5 to 47k
c1 removed
c2 to 100uf
r10 to 1m
c3 to 100pf

add+++
zobel c30 100uf
r30 100

my bais section is not good at all what changes should i make

[Tekko]

That looks like a messed up version of the 1982-1983 Elektor amp.

[MikeB]

Hi sean, that's partially better, but still lot of things wrong.
Let's start with the diffamp.
- You really need an input cap, like 4.7uf or 10uf between r9 and input. (if elyt + to input, - to r9)
- The resistors to the bases need to be identical, i suggest 10k for r10 & r7, to keep gain identical you have to increase r8 to 200ohms.

Biasing, i did not mean the vbe multiplier to set bias in outputstage, you have to carefully select the value for r3 to get balanced operation.
Both transistors in the diffamp need to have each half of the current supplied into the diffamp. This means the voltage dropped across r3 needs to match the biasing for q6. You supply (1.2-0.6)/470 = 1.2ma into the diffamp, means 0.6ma across r3 need to set correct bias for q3.
You supply (1.2-0.6)/100 = 6ma into the vas. For q3 to supply the same current, you need a voltage of 0.006*47 + 0.6 = 0.88v. To get 0.88v with 0.6ma you need 0.88 / 0.0006 = ~1.5k. Okay set r3 & r4 = 1.5k. (might need finetuning like 1.6k)

Why did you remove the miller-cap ? Set a 100pF from b to c for q6. (or 47pF)

You should have a 500ohm or 1k resistor to the base of q5, otherwise you risk that your amp does not recover from negative clipping and stays stucked to -v, destroying your speaker. (i did pay with one speaker for this experience)

Skip the outputprotection, it does not work that way... I can't help you here, i have no experience with them.

A Zobel is a RC (typically 10ohm+100nf in series) from output to ground.

btw, you have to reverse c2, voltage at q2 base is negative.
A single outputpair might be insufficient at these powerlevels.

Heads up, my first circuits were a disaster...

Mike

[prorms]

c4 4.7uf added
r10 and r7 change to 10k
r3 to 1.5k
r12 added
r16 added 1
l1 added 0.05uh
q4 changed to bd139

[Duo]

As an early project, why not start with something like the Project 03 from Rod Elliot?

P03 is of about the same level of complexity but it's a proven schematic. This design of his works very well and will exceed your expectations.

Project 03Amplifier


Perhaps try it out and see what you get. You'll easily get more power out of it than the one you're trying right now. And it will really perform.


Also, make sure to read his article there. It's full of excellent information.

[MikeB]

Hi sean, you got r12 wrong, it needs to be in front of q5 base, between the diodes and the base. You are still missing the miller cap ?

Duo's advice is good, for a start use a proven design and lower voltage...

Mike