β18 Stereo Headphone Amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
HI
I found this project β18 and I wondering is this a good project worth of time and money ??
Total cost of parts for stereo will be around 30-35$ (AD8610 Dale Very good caps) plus power supply itp.

beta18_ampsch.png


1) Anyone Build this amp ?
2) Can I use J511 in place of CR470
3) What do You think about my pcb -dual mono

PCB_V.jpg


THX for all comments
 
Last edited:
what headphones? Z, sensitivity - what I, V will they need

why 17 A power Mosfet - has massive nonlinear parasitic C, 10x smaller die would still be overkill for all headphone's pwr, current requirements I know of

how much bias, what is the V stackup for crd, Vth
 
the 600 Ohms would be better driven by the op amp by itself - you don't want to lose the 3-5 V off both supplies from the crd, Mosfet Vth, the AD8610/20 has enough output current unbuffered

the 60 Ohhm 94 dB headphones could benefit from a buffered output - whether this circuit's output Vswing is high enough is arguable, lateral or low Vth Mosfet could make the difference if you don't want to use bjt buffer/follower
 
Last edited:
From the measurements published, it looks like he's losing about 5 V in MOSFET Vth in the case of a 33 ohm load, plus whatever minimum voltage swing the opamp has. So 4..5 Vrms normally is about it.

Still, that's decent enough in terms of maximum output. A Clip+ with its 800 mVrms - 14 dB less - just about passes the "Boléro Test" on HD420SLs when cranked all the way up. (Which consists in playing a recording of said Ravel piece in fairly quiet surroundings. It starts out very quietly, at about -55..-60 dBFS in this case, and that part has to be distinctly audible already.)

One could in principle adapt the circuit to BJTs, as GDS pinout is basically compatible with ECB. Not that much choice in good TO-220s though.

The more worrying part is the hum found, though that one may have been the result of an ordinary ground loop depending on how the amp was constructed.

There are a few things I'd change in the circuit. Preferably use a 10k pot instead of the 50k one. The 475 at the opamp's noninverting input can also be a 100..220 ohm part.

Re: the PCB:
That 1 µF MKP sure is wasting a lot of space. While specified in the circuit, it is of dubious benefit - the electrolytic alone can handle the audible range just fine.
In return, I don't see the 0.1 µF NP0 which was specified as a bypass cap near the opamp - that one is rather more important. (It can also be a film or ceramic X7R type.)
The two bypass caps (in parallel to the 4700µs) create rather narrow spots in the ground.
Oh, and... no power connectors?
 
Indeed, the bias adjustment is not the "long-term safe" variant (which would have the pot on the lower side, where bad contact can only cause bias to go down, not up). Not to mention that a 25-turn pot seems like overkill - one might use one for prototyping, but in a finished circuit it should be possible to narrow down the range enough for a regular pot + series R to be sufficient.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.