Recurring head amp problems - help please!

New member here.

I've built 2 headphone amps thus far, the CMOY and SOHA, both from the Headwize site which is no longer taking posts from what I can tell.  In both cases, the output is poor in my Grado SR60s (32 ohm, difficult load).  I've done lots of re-checking of layout and wiring and found only a few errors, all fixed.   I'm using an opamp (741) based virtual ground for the PS, with 220k splitter and 220uF across the battery.

Problems are similar on both amps.  Sound is sibilant and clipped even at low levels.  There is also both buzz and hum, though I'm not too concerned about the hum yet. Hum and buzz seem to be absent with no signal.  I've used the phones with portable CDP, computer and iPod with no problems, so it doesn't seem to be a problem with them.

Troubleshooting thus far:  for CMOY, tested voltage inputs for the opamp (OPA2137 for breadboarding) assuming maybe there was some DC on the output.  Getting 4.95v (+ and -) on both pins whether the phones are plugged in or not (no signal).  So I'm thinking no DC.  Plus I put a 330uF electrolytic cap on the output (per Chu Moy's article) and that had no effect on any of the distortions and noise.  I've done little troubleshooting on the SOHA because it's more complex.  

If anyone has ideas where to look next, I'd appreciate the input. 

Thanks,

Carl

Thanks.  I'll double check the inputs and outputs on the CMOY.  Opamps are what I had on hand, and only for breadboarding.  I'll use OPA2134 for the final build but also play with rollling them.  For now, there's some wire crimping, alligator clips and the like, so bad connections are definitely a possibility which I will explore more.

It does seem odd that the output problems are so similar.  The SOHA doesn't use a virtual ground yet the distortion and noise are quite similar so it would seem something else is at hand.  BTW, SOHA uses AC rectified and smoothed while the CMOY is a 9v alkaline.  I'll try to get back to it later today and do some more looking.  I'm guessing it has to do with the Grado's load, but I haven't looked at the SOHA schematic in a while to see if it should be able to handle the Grados.

Carl

PS, would Tangentsoft's TLE2426 virtual ground (http://tangentsoft.net/elec/vgrounds.html under the "Someting (sic) a little different" heading near the bottom of the page) be a better virtual ground?  Or should I just use Chu Moy's ideas?  The TLE is an interesting idea and would seem to be immune to the high current draw of the phones.

Hmmm.  The gain issue sounds plausible, as much of the problem sounds like that.  I'll look at reducing gain and report back.