Headphone amp on perfboard: layout critics welcome

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2008
Paid Member
Since I'm having a hard time getting any layout done for months and years, I suddenly changed my mind to go with perfboard. I've got some lying around and this way I can simply get going with it and finally build something!

So I sat down yesterday and created a 100mil-spaced layout. I wanted it to be rater space-efficient and not widely spread over a giant, mostly empty board. I also didn't want to have much wire links in there, so I decided to leave the power supply lines unrouted till the end and finally wire them directly to the PSU.
Trying to obey the rules of proper grounding, I ended up with a power ground (dark blue trace) and a signal ground (light blue trace). Please have a look at the attached picture. The red square is the point that connects directly to PSU ground. I'm not sure however whether I can safely connect both grounds together on-board in the area encircled in red or if it's mandatory to run a second wire to the PSU from the red X.
Probably it's best to ultimately test it out, but I'd still like to hear some suggestions beforehand. Do you think that my overall thoughts are OK so far or did I even do something terribly wrong?

In case you might wonder: the output transistors are supposed to be mounted to the case, lying flat beneath the board.

Any comments are greatly appreciated :).
 

Attachments

  • perf_01.png
    perf_01.png
    126.5 KB · Views: 323
Here's the schematic.

Just off the cuff... you might be offending the symmetry gods here. :) I would be expecting a LED, or two 1N4148s in series, on either side of the feed. One going to each transistor base. It should work but the IC4B op amp output would settle at a Vbe drop off center. I'm no discrete wizard though, better get some more opinions.
 

Attachments

  • Capture 1.png
    Capture 1.png
    43.4 KB · Views: 268
Last edited:
the op amp output isn't reliably biased Class A with symmetric opposed CCS of the same value

I wouldn't use both CCS

the schematic is pretty generic - could be expected to find similar circuits 30 years ago

some in audio do make a fetish of brand or design by a "big name" - but audio design, EE is much deeper in history than most impressed by the audio ad copy may appreciate



but of course the 1st question hasn't been addressed - what headphone(s) - they vary way more than loudspeakers in I,V, power required - to the point that some amps are really bad matches to some headphones
 
for those headphones you can just use small signal Q, don't really need the BD139/140 - 20 mA will drive the 770 80 Ohms to 120 dB SPL

for deep Class A output stage even t0-92 would work if you drop your supply V by 1/2 - the 880 250 Ohm only needs 2.8 V to reach 120 dB SPL


I also don't know where you're going with all the inverting, the inverting input low Z input makes it poor match to a volume pot

the "always invert" recommendation really only applies when you try to get to single digit ppm distortion levels - and you won't with those op amps
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.