Anyone interested in PCB's for OPA627/BUF634 based headphone amp?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Which opamp, instead of OPA627/OPA637, I could used in "class A" as preamp for BUF634 circuit ?
.
 

Attachments

  • buf634_class a.gif
    buf634_class a.gif
    12.2 KB · Views: 222
Last edited:
Careful with the 637! It isn't stable with low gain.

Any fet input single opamp will work but Pavel also recommends the OPA134 or OPA132 (similar, one is just better specs)

The OPA132/4 and dual OPA2132/4 are good, solid, inexpensive, DIY-friendly opamps.

PS I missed the 'class A' part, that may involve a bit more investigation ...
 
Maxhawk - just curious, how much background hiss does your design get with the input grounded and the volume all the way up? I'm building something similar in another thread with an OPA627 + NJM4556. I didn't know about this old thread. Looks like a good design! I wanted to add the Jan Meier switchable crossfeed from the old Headwize post but couldn't figure out a way to get it stuffed into the feedback loop without messing up the DC parameters.
 
Maxhawk - just curious, how much background hiss does your design get with the input grounded and the volume all the way up? I'm building something similar in another thread with an OPA627 + NJM4556. I didn't know about this old thread. Looks like a good design! I wanted to add the Jan Meier switchable crossfeed from the old Headwize post but couldn't figure out a way to get it stuffed into the feedback loop without messing up the DC parameters.

I'll try it within the next couple of days and get back to you.
 
6 years ago? wow if hes still got one i'd be just a little surprised. I really wish people would read the date of the last post before bumping a thread

Yes, true 6 years is a long time and it does seem a bit silly ...


But we get to use this:

:zombie:

Zombie Thread!!1!! OMG Zombies!!!1!

And sometimes 'accidents' end up bringing together a bunch of seemingly unrelated projects. Sort of crazy out-of-the-box type thinking!

For example, I'm putting together an OPA627/634 buffer circuit and will now follow this thread ... might not have found it otherwize ...
 
I would use LME49710 instead of opa627 if you want another op-amp. In my active crossovers I originally had opa132PAs and they were fine but when I replaced them with LME offset almost completely disapeared (no caps in the circuit in series apart from HP).

If it was to be a standalone HPAmp taking signal from outputs not controlled by volume control then one could place a small value pot between the input and the output op-amps.

cheers,
 
Maxhawk - just curious, how much background hiss does your design get with the input grounded and the volume all the way up? I'm building something similar in another thread with an OPA627 + NJM4556. I didn't know about this old thread. Looks like a good design! I wanted to add the Jan Meier switchable crossfeed from the old Headwize post but couldn't figure out a way to get it stuffed into the feedback loop without messing up the DC parameters.

Just tried it using my Shure SE530 buds (noise isolating) and the only thing I get is some AC hum. I get zero hiss.

I suppose the hum means my AC/DC power supply could use some choke filtering as right now it's just a bunch of parallel 4700uF caps.
 
Just tried it using my Shure SE530 buds (noise isolating) and the only thing I get is some AC hum. I get zero hiss.

I suppose the hum means my AC/DC power supply could use some choke filtering as right now it's just a bunch of parallel 4700uF caps.

That is good news, thanks. Nearest I could determine from the OPA627 specs it seemed that the noise should be very low in the audio band.

Yeah, I noticed that power supply rejection ratio was one failing of the 627 from the data sheet. PSRR starts dropping rapidly at 100 Hz and is down to just -60dB at 20Khz. Vs. around -100dB for things like the LME49990 and LME49720. So easy fix, like you say the chip will require a very clean set of power supply rails. I'm going a little nutz with the PS and feeding a LT1963A/LT3015 with a CLC filter. But even just a 1R or 5R resistor between the rectifier and the 4700uF bank would probably do the job. 5R with two 4700's in parallel should put the corner frequency of the LPF somewhere around 3Hz. I'll bet (voltage) regulation is much less important with the 627 than the clean rails.
 
Last edited:
I never used LME in such configuration and I don't think it is really necessary but you can try starting with low current biasig. You must remember that that to keep the op-amp in class A, it has to be passing more current all the time than the load would take on its own but only just enough. Op-amp must also be capable of delivering significantly more current to the load than strictly required if you want to add bias.
cheers,
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.