I'm new to the forum but I've have some experience working with micro chips such as PIC16F887. I got to thinking what kind of sound quality could be achieved using a micro chip to convert low level audio to digital then create a pwm signal to drive an high powered h-bridge to achieve approximately 1500 watts peak power into 4 ohms for subwoofer duty.
Anyone have any experience with something similar or is this something that will not have satisfactory results?
Thanks in advance!
Anyone have any experience with something similar or is this something that will not have satisfactory results?
Thanks in advance!
Last edited:
A similar project was published on Hack a Day: 70 watt amp uses an ATtiny
It uses the other type of MCU but the idea is same. It drives the mosfets directly. For higher power it's not enough, I guess you could add some mosfet drivers. No idea about the sound quality, it has no feedback too.
It uses the other type of MCU but the idea is same. It drives the mosfets directly. For higher power it's not enough, I guess you could add some mosfet drivers. No idea about the sound quality, it has no feedback too.
Thanks for the link. That looks about like what I would want to try and build except with a bit more power. Is the no feedback thing good or bad?
I don't think open loop is ever really considered a good thing. Here's a thread that touches on the topic....
http://www.diyaudio.com/forums/class-d/85489-all-digital-pwm-open-loop-withfeedback.html
http://www.diyaudio.com/forums/class-d/85489-all-digital-pwm-open-loop-withfeedback.html
Ok so it is probably a lot easier to not use a microchip to create the pwm signal to feed the output. Next question... which produces better sound a single push-pull output or a balanced output from an h-bridge?
An H bridge gives twice the voltage swing for the same rails, so if you're interested in high power then it might be worth thinking about in order to keep the rail voltages low - that way you can use cheaper and/or better performing MOSFETs and make smaller explosions. Assuming you don't mind building twice the circuitary that is. Can't comment about differences in sound quality though.
My concern would be whether you can get a high enough switching frequency and enough resolution with typical uC clock speeds. As a quick back-of-an-envelope guestimate, A 50kHz switching speed divided up into 256 steps (as in 8-bit audio) means that the clock doing the timing needs to be at 12Mhz. That's the clock rate, not the oscilator that you put into the uC which will need to be at least twice that.
Maybe the PIC has special facilities for accutate timing though (I've never used one), or maybe there are cleaverer modulation schemes that relax the timing demands?
My concern would be whether you can get a high enough switching frequency and enough resolution with typical uC clock speeds. As a quick back-of-an-envelope guestimate, A 50kHz switching speed divided up into 256 steps (as in 8-bit audio) means that the clock doing the timing needs to be at 12Mhz. That's the clock rate, not the oscilator that you put into the uC which will need to be at least twice that.
Maybe the PIC has special facilities for accutate timing though (I've never used one), or maybe there are cleaverer modulation schemes that relax the timing demands?
That said, of course if you're driving a sub the switching speed can be lots lower, if you can build a good enough filter to keep it out of the speaker (bleed from 5kHz switching us going to get anoying pretty damned fast).
After some readings and looking at some chip specs I believe it would be possible to go the microchip route but at the expense of writing some assembly code and messing with some extra circuitry for no real accomplishments... I think its probably best to go with a more traditional proven route. Thank you for the input.
I just got done breadboarding 70 watt amp uses an ATtiny. The sound is not bad.
I had to substitute IRFZ34 and IR9Z34. Couldn't find a reasonable price for the others. They are running hot using a 12vdc supply. I assume this is because the RDS is lower than the IRFZ14. Adjusting the 12v supply down to 10v allows it to run cool with no heat sinks.
The sound is not bad. The only major snag I hit was programming the fuse bits on the attiny85. You have to make sure to switch the clock to PLL. Cost me a little time because it wasn't spelled out and I'm not an Atmel expert.
I had to substitute IRFZ34 and IR9Z34. Couldn't find a reasonable price for the others. They are running hot using a 12vdc supply. I assume this is because the RDS is lower than the IRFZ14. Adjusting the 12v supply down to 10v allows it to run cool with no heat sinks.
The sound is not bad. The only major snag I hit was programming the fuse bits on the attiny85. You have to make sure to switch the clock to PLL. Cost me a little time because it wasn't spelled out and I'm not an Atmel expert.
PIC16F887 (8-bit) processor a has limited instruction set , consider investigating Microchip`s dsPIC processor line 16bit processor.. which allows you to leverage all math operators which you definitely need to use when performing control loop dsp feedback algorithms used in class d
which is extremely hard. good luck exploring this area.
Analog electronics still provides simple to use building blocks with fairly forgiving design results compared to digital dsp algorithms.
I think eva can help in this area shes involved in dsp processing for switching converters. show us your work eva 😀
which is extremely hard. good luck exploring this area.
Analog electronics still provides simple to use building blocks with fairly forgiving design results compared to digital dsp algorithms.
I think eva can help in this area shes involved in dsp processing for switching converters. show us your work eva 😀
Last edited:
- Status
- Not open for further replies.
- Home
- Amplifiers
- Class D
- pic microcontroller pwm for H bridge