Headphone Amp Redesign?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I designed a very basic headphone amp for a class last fall, and decided that it would be fun (not to mention useful) to build. So I made a trip to radio shack (i know, i know) to pick up some parts to throw together a bipolar supply and a couple small prototype boards (op amp, bjt's, and vishay/dale resistors from digikey). I put the whole thing together in a nice wooden box, and did my best to shield it (separating the PS and audio circuitry) with aluminum foil. And, wouldn't ya know, I'm very impressed with how it sounds. It measures very well both in simulation and on a spectrum analyzer, flat within a couple hundredths of db to several hundred kHz with low distortion.

But, as we all well know, things tend to escalate.

My original idea was to simply lay out the circuit on PCB, to make it a tad more durable; not to mention better looking. And so I had some more control over how the traces get routed. But then I started thinking, "well why not try to improve it if i'm going to eat the cost of a couple PCBs anyway?" So I've decided to put a Jung regulator in place of my current supply, which is not unlike this one here: http://www.generalguitargadgets.com/pdf/ggg_bipolar_ps.pdf

I'm pretty sure that alone will make a difference, but then I got to thinking about my actual circuit. I managed to make it without any capacitors in the signal path, but the DC offset is insignificant enough to not blow up my headphones. Not sure how much of a good idea that is, but it would be nice to not have to deal with the phase shift and LF attenuation involved with a signal cap. So I guess I'm looking for a suggestion about that. Also, my biasing scheme for the output transistors - would maybe some kind of current source be a better idea than just simple resistors?

Anyway, I attached a schematic of 1 channel - Vpos is 15V and Vneg is -15V. And RL=300 ohm headphones (HD650). Just looking for some suggestions on how to do this right. Anything would help - thanks a lot!

-Bob
 

Attachments

  • hpamp.gif
    hpamp.gif
    36.4 KB · Views: 402
Quick improvement - add a capacitor between the bases of the transistors to improve bias regulation.

Your amp is pretty much a tried and tested design. I've seen it a few times. One that springs to mind is Rod Elliots P113

I would probably use LM317/337 based regulators for better noise performance, rather than the 7815/7915 pair.. but that's just choice.. it's probably just fine.
 
Haha, that looks pretty familiar! Interesting idea about the capacitors on the OTs, I guess those are just to keep the AC signal from modulating the bias current?

He seems to have a bunch of extra parts on his design, compared with mine - this may sound pretty basic, but what exactly are R1L, R2L, and C2L for? Also R7L and R8L? Are they just some sort of DC protection?
 
Anyone? I know they're probably pretty simple questions...I'm just trying to figure out what R1L, R2L, C2L, R7L, and R8L do exactly in this circuit:
An externally hosted image should be here but it was not working when we last tested it.


I'd appreciate any help - and in return for your help, a joke:

A mechanical engineer, an electrical engineer and a software engineer are traveling in an old Fiat 500 when all of the sudden the car backfires and comes to a halt.
The mechanical engineer says "Ah! It's probably a problem with the valves, or the piston!"
The electrical engineer says "Nonsense! It's most probably a problem with the spark plugs or the battery!"
The software engineer says "How about we all get out of the car, and get back in again?"
 
Disabled Account
Joined 2006
R1L forms a simple high-pass filter with C1L, and it also provides a path for input current to flow for the positive input of the Op-amp.

R2L helps balance the impedances as perceived by the Op-amp at it's two input terminals.

C2L prevents DC from flowing to / from ground via resistor R3 (one of the two feedback defining resistors), hence contributing to zero DC output offset. I've not done the maths but it is also likely to set the low frequency response of the amp. If U1A has very low offset figures it may be possible to delete C2L and still get a low DC level at the output.

R7L and R8L... Consider spending a half-hour here...
http://www.st-andrews.ac.uk/~jcgl/Scots_Guide/audio/Analog.html


Also try the site where that circuit came from. Here is their op-amp intro... http://sound.westhost.com/dwopa.htm


(Both links checked and live at time of posting)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.