So I'm planning on building a remote controlled headphone amplifier soon and I have a few questions.
I decided to use the PGA2310 for volume control and figured the gain from the chip is enough for headphone use. Should I put op-amp buffers in the input AND output? If so whats a good op-amp for that, one that's not super expensive like I see often used (although I'd only need a single dual chip).
Second concern is with the power supply. The PGA(and any op-amp buffers) will need a bipolar supply (around +/-15V or so I guess, havent looked into it much) and the PGA's digital control and a microcontroller will need +5V. I plan on using a wall transformer and rectifying and filtering myself. I figure two regulators could regulate the +/- 15V.
The problem is then with the +5V. Dropping from ~15V (or whatever I'll need) to +5V is alot of wasted power as heat. It'll also put a larger current draw on the positive rail which might cause the rails to become asymmetrical. Lastly is something I have no idea about but I've read that its not good to power analog and digital components from the same supply.
Taking all of this in account whats the best and cheapest solution for me. ATM the design is barebones but I'll def have an IR receiver for the remote. I might add an LCD or 7-seg display as well as some other things maybe. The circuit will also contain a CD4052 multiplexer. Microcontroller will be an AVR atmega since I have one laying around already.
I think thats all I have for pre-design concerns, if you have any thing else I should look at for or suggestions I'd be more than happy to hear them. Its my first real project (well actually a complete redesign of a previous trial). I'm an EE student so I'm eager to learn more as well. Thanks for any input and answers you guys will provide.
The reason to have a buffer on the input is so you don't have the volume control affecting the source with varying impedance when your output stage has the buffer in a global feedback loop with the gain stage, a la the National Reference Design for the LME49600 with the LME49720. Otherwise the gain stage would act as the input buffer followed by the pot, then an ouput buffer. Global feedback loops incorporate the gain stage with the output buffer, so you need an input buffer before the volume pot.
The reason to have a buffer on the output is for amperage delivery. The input buffer doesn't need to have extra amp delivery like the output, so your only criteria for the input role are: stability at unity gain, good performance in a local feedback loop, and cheap.
If you're running a simple design with the gain stage in its own local feedback loop, then an input buffer isn't needed, a la the Objective2, because any opamp acts as an input buffer whether you want it to or not.
Now, the output buffer has some different shoes to fill: unity gain stable, stability at low output impedance, short circuit protected at low output impedance, delivery of at least 60mA, and cheap. The reason for that 60mA number is because some dynamic headphones and IEMs can pull that much, certainly more than most opamps deliver (25-35mA), so this matters. If you're using planar magnetic headphones then you'll need a lot more amperage than my 60mA figure.
So you only have a few options and they're all expensive except the one that's spec'd in the O2, the NJM4556A.
I'll let someone else work on the power supply. Those things are my weak spot.
Okay makes sense, I'll start looking into it.
My main concern is the power supply since without that I can't even start experimenting properly outside of my school's lab with benchtop supplies.
The only thing I can think of is having two separate transformers but that creates alot of problems in the cost and space departments. Not to mention adding unwanted complexities. Wallwart keeps everything simple.
Anybody have a nice simple yet very informative post about that ^_^
|All times are GMT. The time now is 05:18 PM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright ©1999-2015 diyAudio