Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › PPA builders' thread
New Posts  All Forums:Forum Nav:

PPA builders' thread - Page 2

post #16 of 646
could someone explain me how does the 'amplified and buffered' ground work? I know that symetricaly driven headphones sounds better, but I don't understand the PPA ground circuit..
post #17 of 646
I have an Elpac wm071 wall wart. It's a +15, -15 volt supply. Could I use this to power the PPA? I'd just connect the + and - leads from the wart and leave its ground unconnected.
post #18 of 646
But you mean BUF634 is OK at a meta42 pcb with same opamp and the same multiloop feedback? I don't the think the difference is that dramatic.
post #19 of 646
Quote:
Originally posted by Glassman
could someone explain me how does the 'amplified and buffered' ground work? I know that symetricaly driven headphones sounds better, but I don't understand the PPA ground circuit..
This is only a active voltage divider. If you have done this with two resistors it would have been rather thirsty. This is an electronic way to create a low impedance node with too much current consumption.

I personally don't understand the benefit of this if you have no trouble to get +- voltage.
post #20 of 646
So the groundchannel is just a fancy TLE2426 that can be omitted and bridged if you have a dualvoltage supply?
post #21 of 646
It can certainly be omitted if you have a split power supply.
post #22 of 646
I meant something else, not that TLE2426 rail splitter, I know how this works but there are 3 channels each with AD8610 and HA-5002s, one for left channel, one for right and the third is for ground and I don't understand this.. is that something like active current sink or what??
post #23 of 646
The ground channel serves as a power buffer for the rail splitter. It may seems to be a little bit odd but this is the designers priviledge to whatever he/she wants.
post #24 of 646
peranders> think of this as the headphone Equiv to a BTL (Balanced to load Amp with the exception of the Negitive part not Inverted as in a Normal BTL design and the Negitive portion operating at unity gain.
post #25 of 646
again i believe all specs have been posted on the above websites ...

If it Sounds Good, It is Good - Duke Ellington aka JMT sig.

And it does sound good...
post #26 of 646
Quote:
Originally posted by xtreme4099
again i believe all specs have been posted on the website ...
Can't find anything really. Am i blind? Where do you find any figures?

EDIT: I found it!
post #27 of 646
in other words its something new and improved that hasnt been done before except by our fellow diyers .. thx everyone...
post #28 of 646
Quote:
The real problem may be that there isn't any sound card out there that can provide the performance necessary to elucidate the differences between a PPA and a META42, so RMAA dutifully reports that there is none - not something you'd really want to share with the masses, despite the fact that the lack of any difference might be entirely due to the sound card.
Not true at all. You can very easily see differences in noise performance, dynamic range performance and especially channel crosstalk. I am not talking about PPA specifically as I haven't built it yet but I did measure different amps, sources and variations of the same design and you can always pick up differences. Distortion - unless to a headphone - is usually low enough that your point may have some merit but otherwise... Maybe what you need is a better soundcard. I bought a Revo just for the sole purpose of measurement as it has the second best ADC chip from AKM. And indeed it has 104dB + noise performance. Audigy is much worse and indeed if sound card is worse than what you're measuring then the results are meaningless.

Also in my experience, test results are very repeateable and don't really vary from run to run. You can of course debate merits and demerits of the methods they use. For example the real value of channel crosstalk is masked by the noise floor, i.e. if it's better than noise floor then the value of noise floor will be shown. You could probably use auto-convolution over a long period of time to get the real value but of what use it would be is questionable. You can also debate if certain flaws are audible or harmful to the sound or even if the whole thing matters subjectively. But RMAA is very useful to pick up potential trouble with a particular amp.
post #29 of 646
Controversy aside, I wanted to ask about something. How does the PPA do w/ EL2001's?
post #30 of 646
Thread Starter 
back to the topic

tangent's updated his step-by-step guide!
http://tangentsoft.net/audio/ppa/amp/steps.html
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › PPA builders' thread