ADA4899-1 amp up and running!!!

Sep 5, 2007 at 10:37 PM Thread Starter Post #1 of 21

NelsonVandal

1000+ Head-Fier
Joined
Aug 27, 2006
Posts
1,023
Likes
11
I wanted to try one of the new super spec'd opamps. I never could tame the AD8099's, but these little chips work with no hassle. DC offset -2/7 mV, very low noise, no audible oscillation, no hum. I use them in a Mini3 configuration (same resistor values except there's no output resistors, active ground, TLE2426, ferrite beads on the output). Panasonic FC's, Sanyo OS-Cons, low ESR tantalums and ceramics for decoupling. Absolutely no feedback traces, 0 mm, SMD 805 resistors soldered directly to opamp pins. No feedback caps (not even in the ground channel), maybe I should add some 2 pF like suggested in the data sheet.

Powered by 7 AAA's 1000 mAh for decent battery life, packed in a Hammond C801 with a trickle charger.

Sound is good. I have to burn it in and listen some more. The sound is very close to Mini3 (not burnt in either) i.e. AD8397. Quite forward sounding. Not the same bass slam as AD8397, but that opamp has more slam than anything else.
 
Sep 6, 2007 at 3:41 PM Post #2 of 21
attachment.php

attachment.php

attachment.php

attachment.php


Not so pretty, but it was hard enough to fit everything inside because of the battery pack. I should have used board mount jack sockets, but you live and learn.
 
Sep 6, 2007 at 5:03 PM Post #5 of 21
Whoa indeed. 7xAAA in the C801 case, damn!!! That's one tightly packed case for sure. Custom AAA holder (soldered strips) or using premolded plastic holders?
 
Sep 6, 2007 at 8:50 PM Post #7 of 21
Quote:

Originally Posted by d-cee /img/forum/go_quote.gif
whoa, hang on...

you managed to fit the amp and 7 x AAA in that one enclosure?



Yes, but there's not 1/10 mm left
icon10.gif
. I really had to squeeze it in. I've also squeezed in 6 AAA's, a trickle charger and PPAS board in the same enclosure (not this one but a black C801).

Quote:

Originally Posted by LawnGnome /img/forum/go_quote.gif
That soldering! :O

Congrats on getting it to work.



Sloppy, sure, I was so very eager to put it together very late last night (just after I finished PPAS and Mini3, not sloppy). Pins 7 and 8 are shorted purposely. It's the bias current cancellation circuit. There's "too much" solder on the onboard decoupling caps so they won't break when I use them as a handle
very_evil_smiley.gif
. I'm gonna clean up the mess later.

Quote:

Originally Posted by FallenAngel /img/forum/go_quote.gif
Looks nice, congrats. It's much less messy than the 3-channel class-ab amp I made in the C801.

Do you have a schematic for it or is it from the datasheet?



I just used the same circuit and values as for Mini3 except for output resistors (I didn't want to mess with the feedback loop, that's one of the things that should be good with this pin configuration - loop from pin 1 to 2 just via a 805 resistor). I should probably add some 10 Ohm output resistors outside the loop and remove the ferrite beads. Snub resistor is recommended with capacitive loads. I also added tantalums in proximity to the ceramics as proposed in the data sheet. I'm thinking of adding caps in the feedback loop. I only have 10 pF and 100 pF lying around, think I'll go for 10.
 
Sep 6, 2007 at 9:44 PM Post #8 of 21
I added caps to the feedback loop, 10 pF L/R and 100 pF ground. I changed from the ferrite beads to 10 Ohm resistors on L/R output. It feels safer, and I'm quite sure the sound opened up a great deal. Placebo or instability before?
 
Sep 6, 2007 at 10:33 PM Post #9 of 21
Quote:

Originally Posted by holland /img/forum/go_quote.gif
Whoa indeed. 7xAAA in the C801 case, damn!!! That's one tightly packed case for sure. Custom AAA holder (soldered strips) or using premolded plastic holders?


No room for any holders. Just about room for the batteries. I soldered directly on the battery terminals and used thin but strong plastic film as isolation. I'm not planning on changing batteries since they are rechargeable. It was very fiddly I can tell, but a single 9 V would be drained in no time. Quiescent current is about 45 mA.
 
Sep 11, 2007 at 9:12 PM Post #10 of 21
Impressions on this after some burning in time:

This is the least coloring opamp I've ever tried. It doesn't seem to add any character of it's own. It's not dark or bright. Not warm or cold. Not fast or slow. Not a hint of sibilance, and the best way to describe the treble is to say the amp doesn't sound at all. The treble is there but I just don't think of it at all. The mid is very clean and articulate as well and maybe a bit pronounced. Voices come through clearer than anything I've heard. The bass is also very neutral, but I think the amp loses some control over it when used with low impedance phones.

The soundstage isn't as wide and spectacular as with Mini3 or PPAS. It's a bit "in your face" and forward sounding. This is my major concern. It sounds wider and lighter when I bypass the ground channel opamp. If I could have this refinement and sound signature (and maybe a warmer mid) with the soundstage of AD8397 it would be absolutely perfect.
 
Sep 13, 2007 at 9:02 PM Post #11 of 21
The problem with the forward and narrow soundstage is fixed. It was a bent cable from the trickle charger that caused some current flow to ground making it sound more mono. It's such a tight fit that these things easily happen. Now it sounds just as wide as PPAS and Mini3. The details, especially in the mid, like voices, are still amazing even though the centre isn't emphasized now. It's a bit brighter than Mini3, but the Mini3 is a bit on the dark side I think.

So now it's perfect? No, there's something missing, "weight" or "body" or something "organic". Maybe it's tubes I'm missing
rolleyes.gif
. Still, this is the amp Ilisten to most of the time.
 
Sep 14, 2007 at 8:42 AM Post #12 of 21
NELSON?? WITH OPAMPS?? WHAT'S WRONG WITH THIS WORLD?

Hey who are you!??! Where are the discreet open loop buffers?
tongue.gif


Anyway, awesome work. I'm very curious about the layout and the wires hiding beneath. Do take some pictures kay?

I see you're using some Lumberg PSU Socket Jacks, they are small and sturdy but i cant find a proper jack to use it with? Do you mind taking pictures of the charging adapter?
 
Sep 14, 2007 at 2:20 PM Post #13 of 21
Good to hear from you, TzeYang, what have you been up to?

I'll post some pictures tonight, right now I'm saving lives
biggrin.gif
.

I've messed up my discrete buffer amp and made it into some kind of 2 ch LISAIII and planning to make it 3 ch. I've posted a question on Headwize about JISBOS, hoping they could be used at +/- 4.5 V and make a portable amp out of them. They are closed loop with low output impedance as oppesed to "our" open loop buffer. I've realized they could be made really small, like 5x10 "perfboard holes" each (excluding caps). I was hoping to get rid of the crosstalk issue and be able to not have to use any stinking opamps at all. If this low supply is possible, this could be a real killer amp for IEM's and low Z phones, and high Z phones at moderate levels.

The sound of the amp above is very clear and uncolored, almost the same quality as the open loop buffers, just missing a bit of warmth. It's addictive to have this clarity, other opamps sound muddy in comparison. It's too fun to mess with opamps, can't really let go of them, and sometimes I need more gain than unity.

Now I want a better source (I have an old H/K all discrete CDP and iRiver IHP120 BlackGate HiQ modded, both with optical out). I think an iBasso D1 would be the natural step toward a better sound. Probably have to mod it with discrete buffers or LMH6321 and add a ground channel. I'm not ready for it yet.
 
Sep 14, 2007 at 4:01 PM Post #14 of 21
Been busy at college. Not quite active in this hobby these days ever since i entered college. busy busy busy. BUT, i'm currently sourcing parts for some fully discreet amps i sorta designed, and i'm working on a LISA III-ish portable amp which can be switched to close loop mode or open loop mode easily. (the prototype is done for the latter but it's not really optimised yet.)

Anyway, the Jisbos doesnt really do well at such low voltages right? Perhaps you should make a minified Jisbos with transistors that works on lower voltages and current. i'd say go with AAA/AA batteries with the portable open loop buffers you have in mind instead of 9V batteries.

Oh, as for your source, do consider the monica II /
biggrin.gif
I can give you the schematics if you want.

Somehow even though you might not understand chinese but i feel the need to show you this:

http://www.andaudio.com/phpbb2/viewtopic.php?t=26525

Loads of diamond buffer variants. Parts matching is necessary if you wish to use them open looped.
 
Sep 14, 2007 at 5:42 PM Post #15 of 21
Quote:

Originally Posted by TzeYang /img/forum/go_quote.gif
Anyway, the Jisbos doesnt really do well at such low voltages right? Perhaps you should make a minified Jisbos with transistors that works on lower voltages and current. i'd say go with AAA/AA batteries with the portable open loop buffers you have in mind instead of 9V batteries.


Well, I don't know, that's why I asked. BC550/560 and BD137/138 sound excellent at 9 V, but I don't know about the JFET's. And for batteries, I use AAA's in this amp and in PPAS, and my plans is to fit the JISBOS in a C1201 with 8 AAA's.
Quote:

Oh, as for your source, do consider the monica II /
biggrin.gif
I can give you the schematics if you want.


So you like Monica, can it be made portable? Does it have an optical in. Sometime I have to go balanced, I think. Is Monica balanced?

Thanks for the link. I haven't read it yet, but will do tonight (using translate.google.com).
 

Users who are viewing this thread

Back
Top