Help me design a Class-D audio amplifier

Status
Not open for further replies.
H

hugo_nz

Member level 2
Joined
Jul 20, 2003
Messages
52
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
548
Hi everyone,
I am working on a Class-D audio amp for a design project. It takes an analogue input and uses no fancy custom IC's. I can't find much on this site about class-d - am i looking in the wrong place or what? I would have though someone would have done this before. I am using an LM566 to generate a triangle wave, a comparator, some lockout logic, some mosfet driver chips (such as the MAX626) and a power mosfet h-bridge. The switching frequecny will be about 100kHz or maybe less (its only for a subwoofer) and the total power to load will be 50-100W rms. Any one have any suggestions, comments, etc.
Cheers, hugo_nz
 

audio class d project

There was a recent topic on this. Here are a few suggestions. Test your circuit at low power supply voltages before using the full value. You will need a low pass filter before the speaker. It should go high impedance out of band.
 

c o m /d o w n l o ad_en.htm

So it turns out i need to do this thing on a single sided pcb - i.e no ground plane -- everthing i've been reading says a low impedance & inductance ground is vital for reliable performance. Any one know if this will be possible. (im new to this forum, and to switchmode stuff -- flatulent can you give a link to the 'recent discussion' cos i cant find it) thnx & cheers, hugo_nz
 

hip4080a projects

oh, just to remind everyone, "Filter Free" is a very easy and reasonably comprehensive way to design all sorts of filters. i used it to design a balanced 2nd order butterworth filter for this amp in a few minutes.
link: f*p://ftp.nuhertz.com/filter/FF3_1.exe
site: h**p://www.nuhertz.com/filter/
its Freeware - yay i dont need to fuk round finding a krack for it.
hugo_nz
 

class d grounding forum

I think that it would be nice if somebody can give us links (http) to the class D amplifier projects.

Thanks in advance
 

other discussion

**broken link removed**


The lack of a ground plane would make radiation problems for your circuit. Otherwise it will work. Remember that 74LS circuits were done with wire wrap at higher switching frequencies than this.
 

Do not download the file PWN_DAC_1.pdf but visit:
**broken link removed**
file is converted htm document.
 

I understand the advantages of using all N-channel devices in the Power stage (H-Bridge) - better efficiency etc... but how are you supposed to drive the high-side N-channel MOSFETs if the power supply voltage is higher than 20V which seems to be the maximum for all the reasonably priced and easily available high-side N-Channel driver chips (the ones with a charge pump) that i can find. Enlighten Me!
 

supertex

Try looking around the Supertex web site for application notes on this. I have a 7 year old paper catalog from them with application notes about high drivers that use capacitors, resistors, diodes, and transistors.
 

HI everyone,
thanks for all the help so far. i'm getting the circuit together now. FYI i found a chip for the mosfet driving, HIP4080A, by Intersil, that contains a comparator for PWMing, delay logic (to prevent shoot-throughs) and the full-bridge high and low side FET drivers (+charge pump and bootstrap). It runs up to 1Mhz. Any Australian/New Zealand readers know of anywhere to get one. They have them at RS, but RS are expensive (*sigh*). Thanks again, hugo_nz
 

Re: Class-D audio amp

U can use the LM4651 of National.

Is a 170W subwofer class-d amplifier.
With some change u can do a 20Hz-20Khz amplifier.

Work really good...
attention at all ground plane in the board and
value of inductance for output filter.

Bye!
 

Re: Class-D audio amp

The project is due in a couple of days. I still haven;t gotten the board working tho. here is my final report, as of right now. Any recomendations, and comments, even spelling mistakes appreciated. cheers
 

Re: Class-D audio amp

hi
it can help u
 

Re: Class-D audio amp

thanks j_explorer2 i had already read that doc. it is very useful, and the theory for the feedback loop in there is what i used to stabilise my loop.
 

defects

The major defect in these circuits is the imperfections in the triangle generator. They are slightly nonlinear and produce distortion. This distortion is also on the set of signals at the sampling frequency and can extend down into the baseband signals producing dissonant sounds. This is why the sampling frequency has to be so high which further affects the system because higher frequency triangle generators are more nonlinear than lower frequency ones.

This is why these amplifiers are only used for subwolfers and not the other frequency range. About 30 years ago the major hi if companies tried class D and could not make it work with low distortion for the rest of the audio range.
 

Re: Class-D audio amp

Can someone please tell me how ot get protel and pspice files exported to some format that is portable, i.e pdf or eps or even an image file. i can get the files out of psice as dxf (autocad)...
and i know how to get is as a pdf if you can get the print to file option, but neither has it. by the way this is Protel 98 and Pspice 8, cos they are on university computers.
 

Re: defects

 

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…