Vbe multiplier output stage design troubleshooting.

Thread Starter

truesavagemf

Joined May 26, 2023
5
I am designing a Vbe multiplier output stage with two power amplifiers below is my circuit attached. It is part of a three stage 2W audio amplifier and i am really struggling to understand the design and operation of it. Is using R5 the correct way to establish your Ibias. When I put a capacitor before thr load rsistor it just filters everything and no signal is passed. I just want to know where my problem lies.

Capture.PNG
 

Ian0

Joined Aug 7, 2020
10,277
R5 should ideally be a constant current source. The Vbe multiplier works best if the current through it is kept reasonably constant. I would suggest that R3 and R4 should be at least 20 times larger, so that the majority of the current in the driver stage goes through the Vbe multiplier transistor, not through its biassing resistors.
If your simulation doesn't work, check your voltage sources. They seem to be negative.
 

Thread Starter

truesavagemf

Joined May 26, 2023
5
R5 should ideally be a constant current source. The Vbe multiplier works best if the current through it is kept reasonably constant. I would suggest that R3 and R4 should be at least 20 times larger, so that the majority of the current in the driver stage goes through the Vbe multiplier transistor, not through its biassing resistors.
If your simulation doesn't work, check your voltage sources. They seem to be negative.
They are not negative they just needed to be rotated the top one is Vcc=+7,075 and the bottom one is -7,075. When i change the resistors it clips the top part of the wave when looking at VBB. Even when increasing the bias current.
 
Last edited:

Thread Starter

truesavagemf

Joined May 26, 2023
5
R5 should ideally be a constant current source. The Vbe multiplier works best if the current through it is kept reasonably constant. I would suggest that R3 and R4 should be at least 20 times larger, so that the majority of the current in the driver stage goes through the Vbe multiplier transistor, not through its biassing resistors.
If your simulation doesn't work, check your voltage sources. They seem to be negative.
Is the 2N2222 a good resistor to use in the vbe multiplier?
 

DickCappels

Joined Aug 21, 2008
10,246
Yes, the 2N2222 should be fine as a Vbe multiplier, though I wonder why you have so much current going through it (See Ian0's post @2)

V1 supplies negative voltage with respect to ground, V2 supplies positive. You need to turn around or swap them before your circuit can work.

If you put negative on the collector of an NPN transistor the collector will no longer be reverse biased and the TIP31 cannot operate as a transistor.

Check it out:
1716190505193.png
 

Ian0

Joined Aug 7, 2020
10,277
They are not negative they just needed to be rotated the top one is Vcc=+7,075 and the bottom one is -7,075. When i change the resistors it clips the top part of the wave when looking at VBB. Even when increasing the bias current.
Better to turn it round than to have it showing a confusing value.
As SPICE’s voltage sources usually show + and - on the symbol if yiu use them with negative values it can only lead to confusion!
 

Ian0

Joined Aug 7, 2020
10,277
Is the 2N2222 a good resistor to use in the vbe multiplier?
Yes, but almost any transistor can be used, NPN or PNP.
However, it is common practice to use a TO126 or TO220 packaged transistor, as it needs to sense the temperature of the output transistors, and one with a hole through it can more easily be attached to the heatsink.
(But if your amplifier is only 2W, perhaps that won’t concern you, but make a note of it for when you make a bigger amplifier)
 

Thread Starter

truesavagemf

Joined May 26, 2023
5
Yes, but almost any transistor can be used, NPN or PNP.
However, it is common practice to use a TO126 or TO220 packaged transistor, as it needs to sense the temperature of the output transistors, and one with a hole through it can more easily be attached to the heatsink.
(But if your amplifier is only 2W, perhaps that won’t concern you, but make a note of it for when you make a bigger amplifier)
@DickCappels I have altered my schematic the problem I have now is that it cuts off the top of the voltage waveform and there is still no voltage over the 8ohm output if a cap is added is there any other obvious problems and what causes the cutoff is it Q1?
See below the cut VBB voltage waveforme
Capture2.PNGCapture.PNG
 

Thread Starter

truesavagemf

Joined May 26, 2023
5
@DickCappels I have altered my schematic the problem I have now is that it cuts off the top of the voltage waveform and there is still no voltage over the 8ohm output if a cap is added is there any other obvious problems and what causes the cutoff is it Q1?
See below the cut VBB voltage waveforme
View attachment 322764View attachment 322765
Could adding resistors next to TIP31C and TIP32C leading to the load resistor(R2) be a solution?
 

Ian0

Joined Aug 7, 2020
10,277
I’m assuming that this is just the output stage, and you will be adding an input stage once you have the bias correct.
If so, then when you complete the feedback loop, it will sort out the DC conditions.
 

DickCappels

Joined Aug 21, 2008
10,246
It looks like you don’t have enough base current for your NPN.

what is the base of the NPN doing as the clipping is going on.

Off hand I would say with currents in this range (0.5 to 1 amp?) and probably not much drive through you 1uf capacitor, you might be better making the output transistors Darlington pairs.
 

Ian0

Joined Aug 7, 2020
10,277
In its present state, there is nothing setting the DC operating conditions, and we don’t know exactly which nodes have been used for the plot.
It could simply be clipping, and have drifted well away from the centre.
As it’s a bipolar power supply, remove C2. It’s. 20kHz HIGHpass filter, so it’s not doing much good in an audio amplifier, unless it’s audio for cats.
 
Top