Are there any open source audio amplifier designs?

[lufihengr]

So I want to build a class D sub amp, but ideally I would use an already proven open source design. I tried to look for one, but I came up with not a whole lot.

While there are some designs available, they are usually anywhere from a few watts, to maybe 100W. I'm looking for something along the lines of 1kW+

Anyone have any knowledge about any projects?

[bob91343]

I would look into application notes for the devices you are considering.  Failing that, contact applications engineers at the manufacturers and see if they have suggestions.

[lufihengr]

Thanks

[Doctorandus_P]

For anything audio related, diyaudio.com is probably a better place to look around.

[dietert1]

You can search ebay for "iraud 1000W". Currently there are boards for 1000W at € 52 and for 2000W at about € 88,26. And you may need something like a SMPS800R (connexelectronic). Plus a speaker protection board.

Schematics of those amplifiers can be found in the web once you know what to look for. Iraud amplifiers are an old, yet good solution. The ready made boards may need some rework though, as they take the low voltage supplies from the high voltage rails, which isn't very nice.

Regards, Dieter

[radar_macgyver]

I used the TAS5630 a while back for a 600W amp, it was mostly a 'follow the eval board' design. The one trap I fell into was the output inductors - I used a commonly available surface-mount drum core, and they got extremely hot when running at full power. TI's application engineers mentioned that the output inductor selection is critical, with cores chosen to avoid an inductance slope with applied current.

[AaronD]

Old thread, by almost a year, but I just saw it now.  How's this?  (even older):
https://www.diyaudio.com/forums/class-d/224052-systemd_2kw-design-33.html#post3404577

Completely open-source 2.4kW half-bridge class-D.  Analog input, self-oscillating.

***WARNING!!!***  This amp requires several different DC supplies, the highest of which is easily enough to kill.  Do not attempt unless you can safely produce those voltages at their required power levels, and you're already familiar with electrical safety in general.  (this is not the project to learn that with)

[james_s]

IMHO if you're looking for >1kW you might be better off looking at more efficient speakers. Are you trying to outfit a large arena or something?

[AaronD]

IMHO if you're looking for >1kW you might be better off looking at more efficient speakers. Are you trying to outfit a large arena or something?

I'm sure it could be pared down.    Perhaps simply run at a lower voltage, knowing that it's well inside its SOA.

---

I agree that the need for a high-powered single amp channel is likely a sign that something else is wrong.  At least in the context of audio (re)production.

And yes, efficiency IS more important than power!  Just a 3dB increase in the dB/W/m number represents TWICE the sound for the same electrical power!  But very few people understand that.  Decades of consumer products speccing peak electrical input power (yeah, that's stupid) with a carefully selected test signal (not a valid test) has produced widespread expectations of ridiculous wattage for a given sound.

The boxes that we're all familiar with are only about as efficient as an incandescent light bulb.  There are far more efficient designs (mostly horns in some form or another), but they're not as well known and hardly ever sought out.  And they're a bit more sensitive to mass manufacturers cutting corners too, so if you've had one of those, it might poison your opinion as well.

---

I guess if you've done all of that research, and put an efficient driver that actually meets its specs, in an efficient enclosure that is made RIGHT, and you still want an array of them that all do the same thing, then you could series/parallel them onto a single massive amp like this.  One (minor) advantage of something this size is that you can rectify a 120VAC power input directly to the main power rails...if you can also GUARANTEE (no excuses, ever, I don't care why) that no one will touch the speaker leads either, and that you also use an audio transformer at the input that can block the mains voltage.  Good ones of those are expensive too.  If it ends up being worth that, it might still be useful as-is for a massive powered subwoofer.  (external connections are IEC and XLR, nothing else)

[Ground_Loop]

This one from a GA Tech professor has been popular

http://leachlegacy.ece.gatech.edu/lowtim/

[james_s]

This one from a GA Tech professor has been popular

http://leachlegacy.ece.gatech.edu/lowtim/

I built one of those years ago and it's a great amp, but it isn't class D and it isn't anywhere near 1kW. It's more of a hi-fi amp than a sub amp.

[floobydust]

I have a Double-Barrelled build aka Leach SuperAmp that is still working fine, it's a very good power amplifier among the best I've heard. Strange it's appreciated in price so much, acres of heatsinks and TO-3's, 1500VA power transformer just freak people out. It scares me a bit running at +/-85VDC rails. No Class D around here.
I have a bunch of the Low TIM's yet to finish building for the PC and AV rear channels. He had many many students building them at Georgia Tech. An SMD update is on my list too.
I have to laugh at modern AV receivers all maybe ~7 transistors and blow up working hard for a few songs, where he used 21 and it can do club duty.

Dr. Leach's design philosophy of lots of local feedback and little loop feedback really works well I find, but today THD is the Holy Grail so designs are the opposite.