My DIY UcD amp from the junk box

[acid_k2]

Hi...
I'm proud to present my first UcD amp circuit.
All components come from the junk box (!) Yes, no money for this amplifier, no smd, no ic, no gold plated connectors, no strange japanese transistors.... on two homemade printed boards: a pcb for modulator, a pcb for power stage, 68X35 mm each.
It's perfectly running at 20Vdc (single rail supply, output cap!) on a 2 ohm load, delivering a very very clean, bright and powerful sound.
I've built a second power stage, and I'm planning to modify the modulator to drive two power board in bridged mode (so I can get away this horrible output cap!)
It's based on classd4sure schematics (threads: UCD Based Fully Differential Full Bridge 450W RMS Amp ; My DIY UCD ) (thanks to Chris, Bruno,etc. ), and on the big know-how of the people in this forum (thanks to all)!
Some data:
power Mosfets: IRF540
transistors in modulator: BC557 and BC237
transistors in power stage: all BC337 + 2SD478 (from the deepest and darkest place of my junk box) for stabilized voltage in gate drivers
all diodes: 1N4148, no schottky diodes!
output filter: 28 turns of 0.6mm2 enamel wire, on a ferrite bead... (about 22 microHenry), and 1microF cap
final output cap: a 30 years old, 1000 microF german electrolytic... last attempt to use it in a circuit!
This is the most simple and funny amplifier I have built in 15 years...

[gni]

Very good use of old parts. . .better to be heard than seen in a garbage heap!

[me262]

Hi,are from Italy,right?
Can you post some schematics?
Thanks.

[wwenze]

Wonder how it measures...

Might gain some insights on the design and part quality of commercial amps.

[stewin]

i too made a ucd using 1n4148 without schottky diodes, it worked but i don't know the disadvantage . not because of ignorance but lack of bat64 or in5819 i used class d from ru scheme BTW exellent schema.
what is the advantage of bc547/bc557 over 2n5551/2n5401 and 1n4148 over schottky diodes
Pictures and circuits of my DIY class D and other projects. ?

[acid_k2]

Hi,
this thread is about 4 years old, so forgive my vague replies.
Well, the basic schematic comes from the patented Philips design UM10155, with some modifications (given by the other discussions at the links in post *1)
http://www.nxp.com/documents/user_manual/UM10155.pdf

I haven't measured that, but the amp is still working "almost" correctly inside a subwoofer enclosure, the sound is clear and the motion of the speaker is good an well controlled.
A lot of simulation was done in 2004 with LTspice by a lot of people of the forum, demonstrating good performance, with an overall low THD. I assume the amp has - in reality- a lot worse THD and frequency response than simulation, but it doesn't matter much to me, considering it's a very low cost amp running a rude subwoofer.

There is no advantage of bc547/bc557 over 2n5551/2n5401, but it's what I had in hand, coming from the junk box... as the 1N4148, a relatively fast diode .... but definitely worse than a Schottky diode!

[stewin]

pls can you post your schematic . and what diodes can be better than 1n 4148 or a Schottky diode which is not smd package (or 1n58xx)

[acid_k2]

Schottky diodes have generally lower forward voltage drop than silicon diodes (about 0.4V vs 0.7V). IN4148 is a small silicon diode quite similar to a typical small Schottky diode. It's fast and with low forward voltage drop (at low current). In that application, it's enough good, but a lot more common and cheaper than a Schottky diode like BAT46 (non SMD).

I had used the schematic from member RX5, taken from thread: my 1st ever D-amp, WORKING!!!.

Link at the schematic is no longer available, but I suppose it's not a copyrighted material, so I post it (credit to RX5) :

[stewin]

it can operate up to how many vltg +/- ? are the irf540 comfortable or do they heat .

i have done a schematic based on class d from ru the 540 were heating even at +/- 12vlts .i thought it was the rds problem . i changed and used irfb23n20 and still the same problem thats when i thought my signal diodes were the issue . i replaced a 220r resistor with a 440r one on the negative side of the amp and it solved the problem . but i do not have theory proof or calculations to back me up thats why i was so puzzled. recently i used +/- 56vlts it works fine . till a speaker burns and shorts the outputs which destroys them.

[stewin]

clear view