Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › PPA v2 construction discussion
New Posts  All Forums:Forum Nav:

PPA v2 construction discussion - Page 71

post #1051 of 1082

Tangent supporting this amp after all these years is effin awesome...I love him.

post #1052 of 1082
Originally Posted by nini_knoxville View Post

I'm using the OPA627. I guess with g you mean the op-amp biasing voltage?


It's shorthand for "gain".


Originally Posted by vixr View Post

Tangent supporting this amp after all these years is effin awesome...I love him.


I, um, love you, too?



post #1053 of 1082

Did anyone try the Dyno-bias Buffer in PPAv2? I have read something good about it and want to try it. In the schematic, 1n4935 is used to provide a voltage drop for working condition so can i change it to something like 1n4148?

post #1054 of 1082
I thought I had resolved my 'ground' issues at my new location  (my Aliendac failed chip gets REALLY hot, but no audio output). Ha no ground problem at all!
I got feed up with no PPA at work, so I brought it back in and plugged it into the built in sound on my MB.
I didn't have my STEPS with me so I used a 16V laptop power supply (sucky, but better than nothing, SOO MUCH BETTER).
The next day I brought the 24V Steps in and hooked it up. I got horrible white noise and muffled-weird-phased music.
When I touched the (grounded) case of the STEPS the white noise changed! I made sure the steps and ppa were not touching chassis or anything.
I'm happily back on the laptop power, but any ideas what is wrong? Could my work have a poor ground? The power DOES run through a cubicle tray system.
I checked the STEPS, I get continuity between chassis and AC ground pin, and no continuity between +- DC or either phase of AC.


** Update **

Update, after an hour of bliss I walked away and came back to some hiss (not like before though, the audio was not muffled or out-of-phase sounding). The hiss does not increase/decrease in volume in relation to input volume (pc volume) or analog volume on the amp. For kicks I pulled the gnd opamp and hiss went away. I swapped chips (no change), then I got crazy and shorted pins 3 to 5 of the gnd opamp together (+ in to output), NO HISS! I've since pulled the ground opamp out and shorted from pin 3 to pin 5 (R1G to R10G), the socket holds the wire well.


No parts were too hot to touch and opamps were cold as ice. Maybe I have some instability? Can that sound like constant hiss? Or maybe it is just an artifact of running the PPA without an (presumply) isolated power supply.


AD8610 in all 3 locations. Should I buy/try a different model GND opamp, or add some resistance/capacitance between pins 3/5?

Edited by tsaavik - 1/29/14 at 11:25am
post #1055 of 1082
 Maybe I have some instability? Can that sound like constant hiss?




What have you got in C6?

post #1056 of 1082

First off, YOU ROCK for replying to me (AGAIN) with an issue, so sorry!

Please trust that I've had many years (8 now) of enjoyment between the problems, and its why I don't give up on this great little amp :D


C6G is:


    standard dipped silvered mica capacitor

    CUST REF #: C6G

    Digikey 338-1061  $1.14 x2 $2.28 (I bought 2, so I should still have a spare at home)


    datasheet: http://www.cde.com/catalogs/STD-DIPPED.pdf


So I could quickly replace/add the 2nd cap if you think it might help (or order something else and wait).



P.S. I tried the steps again with no gnd opamp and pins 3-5 shorted. I got no audio on left channel and the same weird muffled/staticy/badness on right channel I had initially.

Edited by tsaavik - 1/29/14 at 3:49pm
post #1057 of 1082

I doubt the STEPS has any part in your problem. I believe the only reason the symptom changes is that the voltage changes. If you were to run this amp from a bench supply, I believe you'd find that it behaves up to a certain point, then stops working well.


Any chance of trying that? You'd also learn whether the amp is oscillating, since current draw would spike at the same time.


What are the results from trying the standard troubleshooting steps? Measurements at PPA-specific nodes could also help.


There's a lot of knowledge in this thread, too.

post #1058 of 1082

I made a PPA v1 a long time ago (2003 or 2004?), and I absolutely loved it with the Sennheiser HD650.  I sold all my audio & headphone equipment in 2004/2005 and got out of the hobby.  I am getting back into headphones now (have Beyer DT-880 600ohm on the way, and already catching the Head-Fi bug again) and was thinking about building another PPA, but I see the boards are no longer for sale.


ExpressPCB is prohibitively expensive for a small board run.  It looks like Futurlec will accept ExpressPCB files.  I have a question, were the PPA v2 boards made with 1 oz copper?


Also, I haven't gone looking for parts availability yet.  Are any of the PPA v2 parts out of production or otherwise hard to source?

post #1059 of 1082


Are any of the PPA v2 parts out of production or otherwise hard to source?

The PN4392s are no longer available at Mouser or Digi-key but a couple of DIY sites offer them.

post #1060 of 1082
 ExpressPCB is prohibitively expensive for a small board run.  It looks like Futurlec will accept ExpressPCB files.  I have a question, were the PPA v2 boards made with 1 oz copper?

Just a modest remark... before going into the trouble of having boards made in very small numbers... wouldn't it be a better choice building something readily available, like an M3?

I know it's the PPA2 thread, but ... :D

post #1061 of 1082



Although I have not listened to the PPA-V2, I still have to agree with stixx.  I built my M3 a little over a year ago and could not be happier with it.  Easy to build, allows op amp rolling, parts are easy to find and it has an excellent support base.

post #1062 of 1082
Originally Posted by jboehle View Post

were the PPA v2 boards made with 1 oz copper?


ExpressPCB uses 1-¬ľ copper.


The difference shouldn't matter much.  It's not like the PPA v2 was so close to the resistance margin that we need that extra quarter ounce.

post #1063 of 1082

A question for anyone who may know.

I built a PPA V2 several years ago with the STEPs power supply. I fell out of the hobby for a few years and lost all my documentation I had saved and Tangents site doesn't seem to be accessible.

I'm trying to  tune my PPA V2 as after a few years being moved around in boxes I think some of the Trim pots may need to be readjusted.

First issue is, I tried tuning the buffer bias points. I was looking for somewhere around 30ma bias. So I was looking to set around 65mv on R24/34 on all three channels. However I found my meter would read 63mv, and then slowly travel up to 65mv on before settling on the right rail. Also, as I adjusted one rail, and went back to check the others, I found they fell out of balance and needed to be readjusted. Is this normal operation for the PPA V2 or indicative that something may need to be repaired/changed?

Also, my 2nd issues I came across is biasing the op amps into Class A. On R10 I balanced all three channels to 1.7v, though I only did this for balance's sake, as I have forgotten how to properly bias the PPA into Class A. What sort of readings should I be looking to get across R10 for a properly biased setting?

The op-amps used are 8610A if that makes any differences.

post #1064 of 1082

As for the web site itself, it's back now. (Some file permissions accidentally got reset.)


As for your PPA v2 buffer biasing symptom, that sounds like a tempco issue. I'm not sure I ever learned how a simple JFET CCS is supposed to behave as a function of temperature, but I'm pretty sure it isn't inherently tempco-stabilized. One of the whole points of this sort of biasing, though, is to force a constant current through a part specifically in order to keep temperature stable, assuming the rest of the environment has reached equilibrium. In order to get to that equilibrium point, though, you have to case it up and stop wafting strange breezes across the board. :)


As for the op-amp biasing, it's simple Ohm's Law. V=IR, so given the fixed R and the desired I, you know what V should be. Or, given that V=1.7 and R=1000, we know that I must be I=V/R, or 1.7 mA.

post #1065 of 1082

Excellent! It seems everything is working as it should then. Good to see you still supporting your projects. Even after its been 7 years since I built it.

Great work all around Tangent.


New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › PPA v2 construction discussion