DIYEFI.org Forum

KiCAD schematic has been updated as per the previous post.
https://github.com/DeuceEFI/Amplifier/c ... dc5d9e066d

Thanks! :-) Back to me for more progress, then :-/

Did some clearancing for the heatsink tonight. One more piece to cut/bend and i can look at fixing it down. Same goes for caps and bridges. It mostly sits where I want it now, so pretty soon I'll know how the boards need to look too.

OK, heatsink is in place, though not bolted down, yet.

Boards should be wide, not long. up to 100mm wide, up to 50mm from heatsink to edge, smaller in either dimension == even better. But not at expense of crappy routing/placement.

A hole and empty bit in each heatsink corner for mounting brackets, please :-)

Like this, roughly:



Fred.

Power in and speaker out should attach near heatsink somewhere and input signal at the back of the board away from heatsink. I think I talked about that in an earlier post? Feel free to seek clarification and/or to tell me to eff off :-)

Heatsink clearanced for tape deck mock parts, and mounted using three well aimed screws. The whole thing is stiffer as a result (ties back plastic to front steel).

Next up, power supply stuff, or maybe the TODO. I'll check that now and then edit this post.

viewtopic.php?f=37&t=1966&p=30972#p30972

Updated!

Fred.

Power socket/switch/fuse holder installed. viewtopic.php?f=37&t=1966&p=30972#p30972

Will glue it in once it's wired up, and/or recess the clips so that they work. It's such a nice tight (hand cut!) fit, though, it barely needs it.

Pics later.

Fred

Only cap mounting and power distribution buses to go! viewtopic.php?f=37&t=1966&p=30972#p30972 Then it's down to the PCBs and IO wiring and fascia looks etc.

Time to get some progress made on the speakers now! :-)

Fred.

Speakers are now semi usable, so I'm back to amp progress, but today, it's a different type ;-)

I've hijacked the amplifier module from those shitty Hama CS-482 PC speakers and am going to improve it a bit. Here's the baseline, or should I say bass-line, at 100 Hertz:



And at 50 Hertz:



Notice how the lower frequency signal has a much higher peak to peak? That's because it has a recovery time in between and gets recharged. You can watch the single-ended rail voltage sag during louder sections of music. Tomorrow I'll add 22000uF to the 2200uF that's in there, and retest with the scope. I bet it's louder, and possibly a bit clearer too.

Fred.

Just quickly doing the math, looking at more than double the power output (4W vs 2W), under some circumstances. Should also give a more definite "too loud" point, rather than gradual muddying depending on signal material. Eg right now a transient following a bass note could be peaking at 2W but able to go to 4W if not preceded by a bass note. This will make it always be 4W, which is of course nothing compared to the ~50W it's going to be fed fairly soon :-)