[tangent]

Quote:

Originally Posted by steinba /img/forum/go_quote.gif I read something Tangent said about it being smart not to exceed 2000uF rail capacitance.

Yes. I doubt you can hear a difference with more capacitance, and if not, then adding more just makes your power supply work harder in startup and shutdown. You might even give it karōshi.

 

[steinba]

Great! I'll leave it as it is, then. Thank you.

(Great word, btw. I'm gonna use it some time.)

 

[steinba]

I couldn't wait for the case, so I soldered the PPAv2, STEPS and crossfeed boards together on the table, hooked up my CD Player and HD-650s and got out a few CD's.

I first played Tarja Turunen's solo album, which has a passage I really like where the soundstage is really full of loud details, so I can't get my speaker setup to resolve it good enough. At least that's what I thought. Initially, it sounded good (not great) on the headphones, but the "reference passage" sounded even more muddy than on my loudspeakers. Bummer. Maybe the PPA needs more headroom. The gain is set at a moderate 8x. I'll try raising it to 11x and try again. Could be that the CD just is a victim of the Loudness War though,

Then, I got out Roger Waters' "Amused To Death" and pressed play...

HOLY SWEET MOTHER OF.... :-O THIS IS ABSOLUTELY, GOBSMACKINGLY AMAZING! (Just had to get that out.)

I carried on with a few classical Telarc recordings, some Dream Theater and some reference CD's from hifi magazines. This all confirmed the impression I got from the Roger Waters CD. I must say I'm well and truly stoked about this.

There was a noticable hum when I touched the bare metal potmeter shaft, but I hope this will be gone when I get the pot mechanically attached to the metal case. (Confirmation, anyone?)

I hope more gain (and headroom) will improve things further.

I already biased the buffers to 30 mA, and the opamps into class A with 1 mA. Can I hope to improve sound quality even further if I raise the opamp bias?

Any tips om improvement options will be greatly appreciated.

 

[tangent]

Quote:

Originally Posted by steinba /img/forum/go_quote.gif Could be that the CD just is a victim of the Loudness War though

That, or they just multi-tracked it to death.

Quote:

HOLY SWEET MOTHER OF.... :-O THIS IS ABSOLUTELY, GOBSMACKINGLY AMAZING! (Just had to get that out.)

I'm glad you like it.

Since you like Roger Waters, you may also have some Pink Floyd, much of which is also reference worthy. They recorded some of the best bass testing material available, particularly if you're not interested in electronica type bass. Delicate Sound of Thunder....yummy.

The only acoustic bass I like better for testing is disc 2 of Talking Heads' Sand in the Vaseline. Life During Wartime from that disc is simply amazing.

Quote:

There was a noticable hum when I touched the bare metal potmeter shaft, but I hope this will be gone when I get the pot mechanically attached to the metal case. (Confirmation, anyone?)

A naked amp board should still be quiet if you added a ground strap to the pot. If you did and it isn't, it could indicate a ground loop. Confirm by testing with a battery-powered portable player. If that cures the hum, the problem is systemic, not something with the PPA in specific.

You might be able to cure it with a ground loop breaker in the PPA. I show one such in this thread.

Quote:

I hope more gain (and headroom) will improve things further.

The only way that would happen is if you have some uncommonly high dynamic range recordings, on high bit depth media. For instance, an SACD where most of the music is at like -40 dB below full scale. Virtually no real-world recordings do this, quite aside from the Loudness War issue.

Clipping isn't one of those squishy matters of audiophile opinion. If you think your amp is clipping, it's east to find out for certain. Set the volume to the correct listening level for the recording in question, then play a full-scale test tone from a test disc on the same player. (Don't have cans plugged in for this.) Measure the volts, multiply by 2.828, and see if that fits within the op-amp's output voltage limits. If it does, it can't be clipping, at least not due to insufficient voltage gain.

For instance, an OPA637 wants approximately +/-3 V between the rails and the output signal. Call it +/-3.5 V to be safe. With a 24 V supply, that leaves 17 Vp-p to play with, or almost exactly 6 Vrms. 6 Vrms into almost any dynamic headphone is destructively loud, to your eardrums for certain and probably to the drivers, too. And at G=8, you can get a 6 V signal with a 0.75 V input signal, which isn't all that many dB off full CD line level.

I'm ignoring things like passive crossfeed filter loss here, of course.

 

[steinba]

Quote:

Originally Posted by tangent /img/forum/go_quote.gif Since you like Roger Waters, you may also have some Pink Floyd, much of which is also reference worthy. They recorded some of the best bass testing material available, particularly if you're not interested in electronica type bass. Delicate Sound of Thunder....yummy.

I'll get to that for sure.

Thanks for the tip.

Quote:

Originally Posted by tangent /img/forum/go_quote.gif The only acoustic bass I like better for testing is disc 2 of Talking Heads' Sand in the Vaseline. Life During Wartime from that disc is simply amazing.

I haven't heard that, but I'll check it out.



Quote:

Originally Posted by tangent /img/forum/go_quote.gif A naked amp board should still be quiet if you added a ground strap to the pot. If you did and it isn't, it could indicate a ground loop. Confirm by testing with a battery-powered portable player. If that cures the hum, the problem is systemic, not something with the PPA in specific.

Ah, but I didn't. That's just a small cable from a screw on the pot to the ground plane, right?


Quote:

Originally Posted by tangent /img/forum/go_quote.gif You might be able to cure it with a ground loop breaker in the PPA. I show one such in this thread.

Thanks. I'll check it out. I'm sure it'll be fine once I add the ground strap, though.


Quote:

Originally Posted by tangent /img/forum/go_quote.gif The only way that would happen is if you have some uncommonly high dynamic range recordings, on high bit depth media. For instance, an SACD where most of the music is at like -40 dB below full scale. Virtually no real-world recordings do this, quite aside from the Loudness War issue. Clipping isn't one of those squishy matters of audiophile opinion. If you think your amp is clipping, it's east to find out for certain. Set the volume to the correct listening level for the recording in question, then play a full-scale test tone from a test disc on the same player. (Don't have cans plugged in for this.) Measure the volts, multiply by 2.828, and see if that fits within the op-amp's output voltage limits. If it does, it can't be clipping, at least not due to insufficient voltage gain. For instance, an OPA637 wants approximately +/-3 V between the rails and the output signal. Call it +/-3.5 V to be safe. With a 24 V supply, that leaves 17 Vp-p to play with, or almost exactly 6 Vrms. 6 Vrms into almost any dynamic headphone is destructively loud, to your eardrums for certain and probably to the drivers, too. And at G=8, you can get a 6 V signal with a 0.75 V input signal, which isn't all that many dB off full CD line level. I'm ignoring things like passive crossfeed filter loss here, of course.

I appreciate your input. Since the muddiness ONLY occurs with that one CD, I'm gonna declare the CD the culprit. With gain=8, I am at maybe 80% volume at loud (but not crazy loud) "normal" listening volume. With the crossfeed, that is. I'm gonna try a gain of 10x and see what happens.

Thanks again!

 

[steinba]

The ground strap worked like a charm. Dead quiet now. Thanks. I guess I should have RTFM first.

I've tried the Tarja Turunen CD again too, and I have determined that everything but electric bass sounds fine from it, but the bass is bloated and distorted on every track. The producer's choice and not a fault in the amp, then.

I still think I'm going to boost the gain a little. I'm gonna drive the amp from an iPod dock as well as my Marantz CD6000 KI Signature that I'm testing with now, and the outputs on the iDock might be less powerful than the CDP as well.

Thanks again!

Edit: So, I've set gain to 10x. Cant say it got very much louder, but it's just a bit more than "loud enough", so I'm happy with it, and gonna leave it like that for now.

Am I right in assuming that changing the gain doesn't affect opamp or buffer bias in any way, so I don't need to recheck those?

 

[jezz]

I wired up my PPAv2 to a Sigma11 and it sounds absolutely great. I followed the directions on the website and everything worked flawlessly the first time. I can truly attest to the quality of the PPA design and documentation, so thank you Tangent for a phenomenal amplifier. With the parts listed on Tangent's site (gain = 11, OPAMPG = OPA627, OPAMPLR = OPA637), it drives my K340's with ease.

An aside about panels: I'm thrilled to report that blanks can be ordered fairly cheaply in packs of 10 from Mouser, so if for those of you as "skilled" with a drill as I am there's plenty of "room to learn." If you think you will ever attempt to create your own end panels in the future, I would highly recommend the purchase of the appropriate chassis punches. Otherwise, like me, your handshake may become fierce after the abuse of a nibbler, and you too may convert to the exclusive use of Front Panel Express.

 

[steinba]

Quote:

Originally Posted by steinba /img/forum/go_quote.gif The ground strap worked like a charm. Dead quiet now. Thanks. I guess I should have RTFM first.

But alas, all is not well.

Everything's fine with the HD-650, but with my Denon AH-D1000, there's a barely noticable hiss (I only hear it when no music is playing) that disappears when I touch the neutrik jack.

Could this be a bad connection to ground on the jack? Will it be OK when the jack's bolted to the metal front plate?
Could it just be that the hiss is oscillating due to too high gain (11x/R4=11K) with the AH-D100's? Should I just drop the gain down to 10 again, or is there another trick to getting the amp to drive the Denons while keeping the high gain?

The "lower R11 to 100-200R-trick; could that solve my problem, or would thar cause otther issues? (When using higher impedance headphones like the HD-650 for example?)

Thanks.

 

[steinba]

Quote:

Originally Posted by steinba /img/forum/go_quote.gif But alas, all is not well. Everything's fine with the HD-650, but with my Denon AH-D1000, there's a barely noticable hiss (I only hear it when no music is playing) that disappears when I touch the neutrik jack. Could this be a bad connection to ground on the jack? Will it be OK when the jack's bolted to the metal front plate? Could it just be that the hiss is oscillating due to too high gain (11x/R4=11K) with the AH-D100's? Should I just drop the gain down to 10 again, or is there another trick to getting the amp to drive the Denons while keeping the high gain? The "lower R11 to 100-200R-trick; could that solve my problem, or would thar cause otther issues? (When using higher impedance headphones like the HD-650 for example?) Thanks.

This drives me crazy.

I'm not entirely sure it's not a grounding problem, since now the hiss is stronger when (or only audible when) i touch the pot. It sure sounds like the oscillation problem, though, since it's only present with the 32Ohm (?) Denons.

The hiss is less intense when I touch the male jack on the headphone cable. (Not the female Neutrik as I said before.)

Could I safely swap R11 with 200 Ohm resistors without risking other issues? Would I have to do the bias adjustments over again?

I tried 10x gain instead of 11x, but with no noticable difference. I'd like to keep the high gain because of the attentuation that the crossfeed causes. I also tried 8x, and the hiss is reduced, but I need more gain than that if I want to keep the crossfeed.

That being said, I built this for the Sennheisers, and they sound SO good driven by it that I could easily ignore the hiss with my Denons, but it would be nice to get rid of it.

Thanks.

 

[tangent]

Quote:

Originally Posted by steinba /img/forum/go_quote.gif The "lower R11 to 100-200R-trick; could that solve my problem, or would thar cause otther issues?

I don't know anyone who's tried it and had it fail. But then, not many have tried it, which is why I haven't yet made the change in the docs.

 

[steinba]

Thanks. I'll be sure to let you know what happens if I try it.

 

[wuerzig]

Quote:

Originally Posted by wuerzig /img/forum/go_quote.gif I think this might be due to the case not being grounded yet. Also, I remember this looked different in analog mode (didnt take a pic sorry).

Just a quick follow-up on the possible oscillation problem. I scoped it again and did an FFT on the scopes memory contents. Its several FM radio stations. I get it even with the probe shorted.

The amp sounds fantastic. I cant wait to build the YJPS.

Cheers, and ein gesundes neues jahr ("a healthy new year") from germany!

 

[whitelabrat]

Almost done with my PPA2. I've used whatever I already had where possible, and a lot of surplus parts. There are even a few SMT chip resistors in there.

 

[whitelabrat]

Up and running. My first solid state amp... that works. First try too!

*pat self on back*

I haven't set the opamp bias for class A yet. Running at 17.5v with OPA602's. Driving Beyerdynamic DT880's.

All is safe so I'm going to break out the OPA637's.

 

[whitelabrat]

Biased the OPA637's a bit over 1ma.

I'm in the happy place.