Class D amplifier using NE555 Timer

Hey everyone,

I designed a H-Bridge Class D Amplifier to only amplifier some specific single , meaning it won't be used to amplify audio signal . the target signals are : voltage sags, voltage swell ..... etc ,

well i generated the PWM for my class D amplifier using NE555 , the Frequency of the PWM is 14KHZ then send it to gate drive which is IR2109 with 540ns dead time then to the first half bridge which consists of two MOSFET "IRF840" , i took the same PWM signal generated from the NE555 Time and invert it using 4049 then send it to the second gate driver which is also IR2109 then send it to the second half bridge ,

The two half bridge form i H-bridge .the bridge is fed with 10V for testing.
, finally i used butterworth filter of 1.4KHZ as cutt-off frequency here come the problem :

The output signal is small and does not seem to be sine wave in its low part also whenever i want to increase the bridge DC source the output signal will have a dc off-set and satueate at its top also to have a good PWM wave i was forced to add DC off-set to input sine
here are the circuit design using proteus and the output signal after the implimetation of the circuit:
https://imageshack.us/photo/my-images/27/classdamplifierusingne5.jpg/

the output signal :

https://imageshack.us/photo/my-images/560/outputsignal.png/

please not that is i increase the DC off-set of the input signal the output signal amplitude will increase but saturate in its bottom meaning the bottom of this sine wave will be square wave
i would like to note that the DC offset of the input sine is mandatory to have a PWM generated from the NE555 timer
secon, i increase the input sine wave amplitude the output signal amplitude increases but the problem of saturation appears .
Please help me because i ve no idea what the problem is !!!???

https://imageshack.us/photo/my-images...erusingne5.jpg
404 Not Found

https://imageshack.us/photo/my-images...tputsignal.png
404 Not Found

broken links

sorry i re-uploaded the images

I don't think that a single 555 can ever generate linear PWM. The transfer characteristic will be more or less exponentially and never achieve 100% modulation depth.

Hi all

I don't think that a single 555 can ever generate linear PWM. The transfer characteristic will be more or less exponentially and never achieve 100% modulation depth.

Click to expand...

Hit to the point !
555 is not good to create SPWM ! it can be used to generate the square wave and then taking an integral and then comparing the signal with that and then a good linear SPWM which is so trustful !

555 is such a non linear device !

Best Wishes
Goldsmith

goldsmith said:

Hi all

Hit to the point !
555 is not good to create SPWM ! it can be used to generate the square wave and then taking an integral and then comparing the signal with that and then a good linear SPWM which is so trustful !

555 is such a non linear device !

Best Wishes
Goldsmith

Click to expand...

Hi;

i want just to point that 555 timer uses Delta Modulation technique to generate the PWM signal and from the implementation and simulation PWM output i saw that it can be used in class D amplifier without any problem here is the simulation resulat

https://imageshack.us/photo/my-images/543/pwmsignal.png/

and here is the PWM signal taken from the scope :

https://www.youtube.com/watch?v=ycv60D4wi6g&feature=youtu.be

the PWM signal frequency si 14kHZ

O.k., a LM555 pulse width modulator will be fairly linear over a control voltage range of about 1.5 to 10V with 12V supply voltage. You might experience additional non-linerarity from incorrect operation of bootstrap driver (bootstrap capacitor too small).

FvM said:

O.k., a LM555 pulse width modulator will be fairly linear over a control voltage range of about 1.5 to 10V with 12V supply voltage. You might experience additional non-linerarity from incorrect operation of bootstrap driver (bootstrap capacitor too small).

Click to expand...

well for the output of the H-bridge i didn't see any nonlinearity it was a beautiful PWM signal ranging from +10V to -10V and actually this is what matters meaning that the H-bridge is giving a good signal !!

I checked the output of the drivers and the signals are perfect meaning there is non problem with the bootstrap capacitor.

Did you check bootstrap operation at high duty cycle?

Anyway, it was just a guess. To determine if the NE555 is still in linear range, you have to check the absolute CV pin level in relation to VCC. Did you?

There are ways to get a 555 to act as a PWM modulator (using some external components), but simply modulating the CV pin is a very poor way to do it. Besides the severe nonlinearity in D, you'll also be modulating frequency quite a bit.

A more straightforward way to make a linear PWM modulator is to create a nice linear ramp waveform using an op amp oscillator, then feeding that and your audio signal to a comparator. Look at the top of page 9 of this document: **broken link removed**

mtwieg said:

There are ways to get a 555 to act as a PWM modulator (using some external components), but simply modulating the CV pin is a very poor way to do it. Besides the severe nonlinearity in D, you'll also be modulating frequency quite a bit.

A more straightforward way to make a linear PWM modulator is to create a nice linear ramp waveform using an op amp oscillator, then feeding that and your audio signal to a comparator. Look at the top of page 9 of this document: **broken link removed**

Click to expand...

i know this method but the only thing that made me to choose the NE555 time is because in my country i didn't find a good comparator that could compare a triangular wave of 10KHZ with a sine wave of 50HZ in fact when i did the design on simulation i used this method to generate the PWM signal .

- - - Updated - - -

FvM said:

Did you check bootstrap operation at high duty cycle?

Anyway, it was just a guess. To determine if the NE555 is still in linear range, you have to check the absolute CV pin level in relation to VCC. Did you?

Click to expand...

For me when i saw that the H-bridge has a nice out PWM signal it's clear that everything before this bridge is working well , and i checked the 4 gating signal of the 4 MOSFETs everything was normal , and for your second question FvM i adjusted the Dc-offset and the amplitude of the sine wave to have a good and linear PWM signal

artasse said:

i know this method but the only thing that made me to choose the NE555 time is because in my country i didn't find a good comparator that could compare a triangular wave of 10KHZ with a sine wave of 50HZ in fact when i did the design on simulation i used this method to generate the PWM signal .

Click to expand...

Ordinary op amps and comparators should be able to do 10KHz no problem. A TL082 and a LM339 will do it no problem, and those parts should be available anywhere.

mtwieg said:

Ordinary op amps and comparators should be able to do 10KHz no problem. A TL082 and a LM339 will do it no problem, and those parts should be available anywhere.

Click to expand...

i tried LM339 it didn't compare quiet well , the frequency of the PWM that i want is 14Khz or higher

I've made PWM modulators with the 339, and it certainly can work at 14KHz. It has a rated delay of 1us, which should get you a duty cycle range of at least 2%-98%, which is plenty for SPWM.

If you really want something faster, the LM311 shouldn't be hard to find.

hi everyone ;

i would like to thank everyone for helping me with this problem i ve found the solution which to used comparator to compare sinewave and triangular wave and also changing the inductor to toroidal inductor and the output signal was quiet good .

hi ;
well the output signal of my class D amplifier is show below i would like to have your feedback knowing that the input sine wave is 10V peak to peak and the Vcc of the bridge is 18.75V .

**broken link removed**

Hey,

Have a look at http://www.elektor-projects.com/project/555-class-d-power-amplifier-130144-i.13004.html where i "linearize" a 555.
Working fine, good sound!

grt,
Frederik