Hi everyone,
I'm designing an amplifier circuit that would have two LME49600 as buffer for each channel.
I wanted to know how I should do my virtual ground, I use a single power supply of 12V and I want +6V/-6V?
Assuming that I can provide 250mA twice for the left and right channel, it makes me a total of 500mA.
Is a virtual ground-based TLE2426 followed by a buffer BUF634 would be sufficient?
Thanks for reading and helping me
I'm designing an amplifier circuit that would have two LME49600 as buffer for each channel.
I wanted to know how I should do my virtual ground, I use a single power supply of 12V and I want +6V/-6V?
Assuming that I can provide 250mA twice for the left and right channel, it makes me a total of 500mA.
Is a virtual ground-based TLE2426 followed by a buffer BUF634 would be sufficient?
Thanks for reading and helping me
Congratulations, you have found out why single-supply operation is a bit of a mess.
A virtual ground would have to be low-impedance and low-noise. That would take at least a 5532 class or lower-noise opamp with a voltage divider and buffer electrolytic, plus a hefty buffer inside the feedback loop (where all of them belong) to support the maximum of 500 mA.
As +/-6 V isn't exactly very exciting, an inverter to generate -12V with a sufficient current capability would be useful here. But being a switching power thingy, that again would be somewhat messy.
However you look at it, a simple oldfashioned split supply just saves you a major headache. If you're not skimping on electronics, don't do it for the power supply either.
A virtual ground would have to be low-impedance and low-noise. That would take at least a 5532 class or lower-noise opamp with a voltage divider and buffer electrolytic, plus a hefty buffer inside the feedback loop (where all of them belong) to support the maximum of 500 mA.
As +/-6 V isn't exactly very exciting, an inverter to generate -12V with a sufficient current capability would be useful here. But being a switching power thingy, that again would be somewhat messy.
However you look at it, a simple oldfashioned split supply just saves you a major headache. If you're not skimping on electronics, don't do it for the power supply either.
Just curious, why two per channel? Are you thinking of putting them in parallel? I ask that as someone who has wanted to do the same thing for quite a while until opc here
http://www.diyaudio.com/forums/head...-performance-headphone-amplifier-pcbs-16.html
managed to convince me a few months ago that not only would it not achieve anything with regards to powering headphones, it could actually be slightly detrimental with some balancing currents moving between chips. Which headphones are you planning on using them with? I have to admit that I've yet to find a pair that needs more than 250mA.
FWIW the LME49600 and BUF634 are virtually the same chip by way of the specs. If you really are stuck with a single supply then good old fashioned input/output coupling caps (3.3uF-or-so film on the input 1000uF or so on the output) and setting the internal DC bias point at +6V may be your best bet. I think the caps would mess around with the signal less than a virtual ground.
Take a look at this on VGs...
NwAvGuy: Virtual Grounds & 3 Channel Amps
49600 virtual ground
Headphones just need 10's of milliwatts. Perhaps you are intending to drive speakers? Virtual grounds are only used in some portables where sound quality must be compromised for minimum drain on a single battery.
Hi everyone,
I'm designing an amplifier circuit that would have two LME49600 as buffer for each channel.
I wanted to know how I should do my virtual ground, I use a single power supply of 12V and I want +6V/-6V?
Assuming that I can provide 250mA twice for the left and right channel, it makes me a total of 500mA.
Is a virtual ground-based TLE2426 followed by a buffer BUF634 would be sufficient?
Thanks for reading and helping me
Headphones just need 10's of milliwatts. Perhaps you are intending to drive speakers? Virtual grounds are only used in some portables where sound quality must be compromised for minimum drain on a single battery.
Yes, I know... But it's a portable project driven by a single 12V supply. I need to do the job with that.Sorry it's my mistake my circuit had 2 LME49600 buffer one pair channel
Thanks for the info, I will use good capacitors
if you must do the virtual ground thing, the lt1010 is a much better choice thats capable of the current needed all by itself and will drive it at 0v. where is the 12v coming from? more than one cell? just use a split battery supply and build a bipolar regulator circuit off that; then using 'real' ground. virtual ground is more trouble than its worth.
as has been mentioned, the lme49600/10 chips are basically the same thing as buf634, except the national chip would more readily drive its max 250mA due to a better dissipation package, so asking a buf634 to support twice its capability isn't going to work
as has been mentioned, the lme49600/10 chips are basically the same thing as buf634, except the national chip would more readily drive its max 250mA due to a better dissipation package, so asking a buf634 to support twice its capability isn't going to work
portable amp with single voltage
Do you need to comply with audiophile buzzwords. If not, consider the lowly Headbanger project. I did one just for fun, and it worked quite well. It uses the old LM386 found in modems and old PC's. Distortion has been minimized but it still has plenty of drive.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.