Help in Class D amplifier designing

[dileepkella85]

I have attached the circuit, I want to redesign this, as this circuit is on pdf circuit i am unable to know the values of capacitors, diodes and transistors
I bought all ICs and implemented on breadboard but unable to drive the MOSFETs, I have used 327 and 337 transistors which were available to me. I am getting the output till transistors but unable to get the output from MOSFETs.

Thank you.
Dileep

[ChocoHolic]

Hi Dileep,
welcome to your first thread !
In your schematic the P-Chanel MosFets will always remain OFF.
You are feeding the PWM-signal directly into the 12V supply, instead towards the gates.

Good Luck
Markus

[ChocoHolic]

Reasonable values for C18 and C19 would be 33pF..100pF to start with .
Reasonable values for C20 and C21 depend on your PWM frequency.
Which PWM frequency does your 555 circuit generate?
(No, I do not want to search the 555 data sheet and start calculations...)

Furtheron you should use a soft 12V supply, because this circuit might have some cross conduction during start up.

[dileepkella85]

Hi Markus!
Thanks for your reply and wishing....
On your first statement that Pchannel Mosfets are fed with 12V supply, this is what I was also thinking but that circuit is in working condition on PCB board.

PWM frequency is 40KHz, so accordingly what should be the value of C20 and C21?

I am using 12V powers supply which is available in Lab.

and what could be the solution of above 12V supply to P Channel Mosfets

[dileepkella85]

Hi but you did not discuss about the transistors? please give your thoughts about that what transistors should I use?

[ChocoHolic]

Originally Posted by dileepkella85

On your first statement that Pchannel Mosfets are fed with 12V supply, this is what I was also thinking but that circuit is in working condition on PCB board.

Does this mean you do have a working PCB?

If so - you should check on the working if C20 and C21 are really connected between the emitters of the bipolar transistors and +12V.
I doubt it. C20 and C21 should be connected between the bipolar transistors and the gates of the P-Chanel MosFets.
I do not doubt that the source pins of the P-chanel MosFets are connected to +12V. Feeding the source pins of the P-chanel MosFets with 12V is fine, but the PWM signals which are coming from the bipolars through C20 and C21 must be routed to the gates.

Bipolar transistors:
BC327 and BC337 should be fine.

[dileepkella85]

Hi, yes, you are right that it should be connected to Gate of the MOSFETs. I went through the PCB again and changed the schematic which I have attached again with the correction.

I have also attached one pic of PCB, for one question that there is R22 and R23 marking of resistors, but actually there is no resistor on the PCB at position of R22 and R23 and one pin is grounded, please have look at the picture, you will get my question clearly.

What is purpose of this R22 and R23? and how can I implement this thing on my breadboard

Thank you for your co-operation and good suggestions.

Regards,
Dileep

[81bas]

D1 and D2 most probably are Zener diodes for 6-7 volts...

[ChocoHolic]

You could use zener diodes for D1 and D2 as proposed by 81bas, but it is not necessary.
Basically you just need a normal diode like LL4148 or similar there.
Both diodes are acting together with C20 and C21 as a charge pump.
Note: The voltage across the 270 Ohms will be a PWM rectangle with DC component. Means you will have a DC portion through both 270 Ohms.
But caps cannot conduct DC ! Without diodes resistors (Edit: R18 &R19) would see a pure AC rectangle. Peak to peak voltage of the rectangle would remain the same, but no DC component. The trick with the diodes is to clamp the unnecessary polarity of the drive signal (no need to have the MosFet more off than OFF) and at the same time allowing a certain current peak to refill that portion of the charge of the caps, which had been consumed by the 270 Ohms during ON time. So you can get instead of a +/- 2..3V signal a reasonable Ugs signal of -0.7V / +4...5V
I would not implement R22 and R23.
Probably they had been intended to allow a certain DC bias, which would reduce the necessary current peak for recharging the caps - but they would also bring the danger that the OFF state of P-Chanel MosFets becomes less reliable. Probably reliable values for R22 and R23 would bring not much improvement regarding the charging and peak and I guess that's why they are not in place.
Don't bother about the recharging peaks. We are not talking about troublesome 50A peaks...

[ChocoHolic]

More specific:
- There are two charge pumps, one pump for each P-chanel MosFet.

Detail correction:
- Ugs +0.7 / -4..-5V