Best buffer for a chipamp?

Status
Not open for further replies.
Hi all,

I am planning to make an unity gain stage to use for my chipamps (LM3886, LM4780 and LM3875).
Whats the best sounding buffer design?
Should I use a discrete design (i.e. Pedja Rogics JFET buffer)?
Or may I use just an OPA134 in unity gain? (I dont know if it sounds good at unity gain)
Or do you advice a better alternative?

Thanks in advance for your replies.
 
I can't decide between Pedja's buffer and AD797 in unity gain - they are both extremely good - and if you want to drive low Z headphones too, or very long cables - go for JISBOS.
 

Attachments

  • ad797 buffer.gif
    ad797 buffer.gif
    4.7 KB · Views: 1,330
Finally I have decided to make Pedja's.. And finished in one weekend 😀
An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

Now its ready for departured! But I have a problem with the power supply. Which is;
http://www.diyaudio.com/forums/showthread.php?s=&threadid=124548
I havent got any replies from Power Supply Design guys until now. May you say something? Because I am growing impatient to try my JFET buffered gainclone...
 
AndrewT said:
get a 20-0, 20-0Vac transformer and you won't need a regulator/s

But Pedja advices regulated supply in his web site...

Nuuk said:
Or replace the 1083 regs with another pair of 317/337 to see if that cures the problem.

Hmm... With 30v supplies, one LM3875 sink more than 3.5A, in this case, do you think 317/337 will enough, especially if you take consider my speakers sensivity is just 86dB!
 
Hmm... With 30v supplies, one LM3875 sink more than 3.5A, in this case, do you think 317/337 will enough, especially if you take consider my speakers sensivity is just 86dB!

I'm confused Ozgur! Are you using the same supply for the buffer and your amp?

As Andrew says, the ideal solution is to use a lower voltage traffo for your buffer supply. I used an 18v one for mine, followed by two stages of regulation so each stage drops about 5 volts which doesn't stress the regulators, or require heatsinks. Full details
here . 😉
 
I guess, the problem is using the same PSU for buffer stage.. Youre right..
Because a lot of people use that (the first stage of PSU) design for chipamps.. However using LM337 just after LT1083 may cause that problem. So I will remove the 317/337 regulators and test it again...
I hope it will work.
 
Status
Not open for further replies.