[Mad Max]

Those are the ES bipolar electrolytics.  I found them muted-sounding when I rolled caps in my ol' Creative soundcard long ago.  Whether on outputs or power supplies, I don't like them one bit.

 

[SpudHarris]

Trying these as all my other caps are 16v only. These are 470uf Nichicon 25v...

Got 35v Simlic ll on order.

 

[Audio Addict]

Trying these as all my other caps are 16v only. These are 470uf Nichicon 25v...

Got 35v Simlic ll on order.

Nice amp for rolling.

I am using 4 AD797 in my RSA Apache. I have no soldering skills, any recommendations to take it to another level?

 

[SpudHarris]

Nice amp for rolling.

I am using 4 AD797 in my RSA Apache. I have no soldering skills, any recommendations to take it to another level?

Yes it's a great amp period. I am using it with a Graham Slee PSU1....

I personally love the AD797 but my favourites are:

AD744 OBCA
NE5532 (Signetics) OBCA
LME49990
Muses 01

The V5 Burson is an astounding unit but limited by its size.

 

[Audio Addict]

How did adding the capacitors improve the AD797? I would to try it but no DIY skills here.

 

[Mad Max]

Generally, everything about the sound improves to some extent.

 

[trooper05]

I still cannot believe how much the sound has changed with the 10uF Silmic bypass, that dark black sound cant get enough of it.

 I did a bit of research looking at datasheets of various OPAMP's, it seems implementation varies depending on which OPAMP used:
 
The AD797 within the datasheet states:
"A 1.0 μF to 4.7 μF tantalum in parallel with 0.1 μF
ceramic bypass capacitors are sufficient in most applications.
When driving heavy loads, a larger demand is placed on the
supply bypassing. In this case, selective use of larger values of tantalum capacitors and damping of their lead inductance with
small-value (1.1 Ω to 4.7 Ω) carbon resistors can achieve an
improvement" Pg 13
 
The LME49990 states within its datasheet:
"SUPPLY BYPASSING
To achieve a low noise and high-speed audio performance, power supply bypassing is extremely important.
Applying multiple bypass capacitors is highly recommended. From experiment results, a 10μF tantalum, 2.2μF
ceramic, and a 0.47μF ceramic work well. All bypass capacitors leads should be very short. The ground leads of
capacitors should also be separated to reduce the inductance to ground. To obtain the best result, a large
ground plane layout technique is recommended and it was applied in the LME49990 evaluation board." Pg10
 
It seems as each OPAMP has specific bypass requirements in order to function within its specified optimum values. When the circuit does not have the specific requirements detrimental performance can be expected such as oscillation with associated overheating along with a potential veiled sound which is impossible to notice without a oscilloscope. The information within the datasheets seem a little vague on what is the best bypass arrangement is. 
 
@SpudHarris
 Post us your impressions on the Simlic ll capacitors  I look forward to hearing your impressions! 

 

[zilch0md]

Spoiler: A summary of the recent cap discussions

I have found that putting a 10uf Silmic II on the power pins makes a more profound improvement than a 0.1 film. The board will already be bypassed with various capacitors to keep the opamp stable but the power might be filtered with "average" capacitors in the first place. Putting a quality capacitor on the chip in essence cradles the opamp in its sonic signature vs the signature of the other capapcitors. If you want really deep and full bass put a 10 uf Silmic II on the chip. It's better than any other capacitor I have tried.

 
 

http://m.ebay.de/itm/12-pcs-10uF-25V-ELNA-RFS-SILMIC-II-AUDIO-Capacitor-85-C-br-NEU-Kondensator-/361425093253?nav=SEARCH

That should do it for the opamps. What value do you need for the board?

 
 

   
Aren't Polymer caps bad for audio?
 
@trodas did some explaining here: 
http://www.techpowerup.com/forums/threads/x-fi-sound-bugs-i-think-i-have-a-solution.23582/page-3#post-887544
 
"Polymer caps is great for high-frequency voltage filtering (just check their frequency response curves, for christ sake!) while using them for anything getting close to audio is UNFORGIVABLE mistake that completely erase any filtering in low-frequency, hence the strong bass line kill the quality effectively.
You can't measure that on the RMAA, to show this problem you need the DAC/opamps output a loud bass line and then the polymers crumble and kill the voltage filtering = the card output get much less that satisfactory.

Just because todays there are some kind of "hype" about polymers does not mean that they are usefull for everything. They have a very strict usability around 200 - 500kHz, and that it is.

The key to understand this problem is thing called ripple rating coeficient. It basicaly says what multiplier has to be applied on the rated ripple for different frequencies. Nothing show this better that this table:

Samxon X-con URL polymers: 120Hz = 0.05; 1kHz = 0.30; 10kHz = 0.70; 100kHz = 1.00
Samxon elyctrolyte GA caps: 120Hz = 0.50; 1kHz = 0.80; 10kHz = 0.90; 100kHz = 1.00

The specs are very much the same as with all other polymers - even spoken - the better polymers, the worser they run at low audio frequency."

 
 

  "Whatever you end up using, be sure to add a WIMA cap across the power pins (pins 4 & 8 on duals, pins 4 & 7 on mono/singles)"
 
Will this provide any benefit on a Asus STX? 
 
I currently run LME49990 super analytical OPAMP, I am after some OPA627's (Apprently sound more relaxed, laid back) but they are way too expensive and too many fakes around. Can anyone recommend something similar? Would OPA827 be a good choice?
 
Any advice would be appreciated! 

 
 

To trooper
Try a wima so you know the sound, but then try a 10 uf Silmic on your 49990. I use the 49990 in my xonar bypassed with a Silmic and it might be what you are looking for. You keep the detail of the 49990 but add warmth and space.

The way capacitors are used close to the opamps is for stability. If you look at the product sheet you will generally see 10uf or 4.7uf and 0.1uf going from + to ground and from - to ground. This stabilizes the rails by removing any jitters and sending them to ground. However, the operation of the opamp is powered by the + and - and not by ground. While a 0.1 across the + and - is oftentimes used to calm an unstable opamp down (compensation would be much more effective) it is way to small to improve the sound. You will notice it if the capacitor you choose is sibilant. You will hear it in the treble, but 0.1 is not enough for the entire audible spectrum. When adding a capacitor to + and - you are in essence adding a power reserve to the opamp directly. In this position you will "hear" the capacitor. So it needs to be big enough to supply the whole dynamic range (think bass drum kicks which move the speakers in your headphones more than any treble sounds).

The immediate reaction to this might be "you don't want to hear a capacitor!" Well, there is no way around it. In your STX, your power comes from the computers PSU. Lots of "sufficient" capacitors there for powering your computer but all of them awful for audio. From there it goes to the STX which will have some filtering using more capacitors to clean up the noisy power of the computers PSU. From there it goes to your opamps that are bypassed to ground for stability using ceramic (yuk) and film and you also have some electrolytics there (nichicon fine gold?) for power reserve. All of these capacitors have an effect on your sound. This is why you need to choose the very best sounding capacitor to be the very last one in the chain.
For me it is ELNA Silmic on my xonar. it goes to my headphone amp which is a mixture of Silmic and SEPF and then finally on the opamps in my headphone amp I have ELNA Cerafine because they have a beautiful clarity that complements the full and rich Silmics.

So the fact is, you will be listening to capacitors so you have to choose which ones sound best to you.

 
 

  Thanks for the info @pelopidas How do you bypass on them small LME49990, you put in between pins 4 & 7 on both OPAMPS? Or is there a better way to do it?

 
 

I am assuming you have your 49990's on a dual adapter. In this case just go from + to - so pin 4 to 8. That will take care of both. If they are single soic then 4 to 7.
But you have a STX don't you? So you have sockets for dual opamps. Try it like the pic I attached. That's what I use in my xonar

 
 

  I look forward to your impressions. Leave the emi shield of for a while so you can play around with different capacitors.
You can read about my experience a few pages ago.
www.head-fi.org/t/432749/the-opamp-thread/4905#post_11796572
 
As a side note, I started this by being dissatisfied with the bass on the AD797. 10uf Silmic totally changed that opamp, put some meat on the bone so to say.
As a side note, ( i guess the other side) is once you do this to the AD797 and LME49990 other opamps such as the 49720, 49710, 2107, 5534, 827 and even the LT1028 prove to be really inferior. My no.1 is 49990 with silmic and next is 797 with silmic. The 49990 is more spacious, noticeably better noise rejection and has an ease about it while being super detailed.
 
Can anyone recommend anything that tops those? Are the muses really worth all that extra money?

 
 

  I've tried all popular opamps in my Xonar in the last 4 years. I kept coming back to AD797 in I/V and 49720HA in buffer.
Thats a pretty decent sound but it wasn't right. The desire to keep messing with it, straining to listen deeper into the recording leads to trying all kinds of opamp variations.
"its too sibillant.. the bass is flabby... where is the bass?..the soundstage is small or muddy..."
Well, after playing around for 4 years, the solution is stupidly simple.
The LME 49990, 49720, 49710, AD797 and all other awesome opamps need better bypassing.
After reading these different fora for years, you will come across adding compensating capacitors in the 50pf range from the output to the neg feedback.
Did that. Didn't make me happy. Something else was wrong. Try adding 0.1uf from + to -... well that didn't do it either.
The trick for me is to put 10uf on the chip from + to -. I use Silmic II because they are generally my favorite. Before, the 49990 in buffer would get between real warm and actually hot. Obviously oscillating.  Now it is barely above ambient. The AD797 sounded good in I/V but I kept wishing it had more Bass, more fullness in sound.

What changed after adding the caps?
1. Most noticeable is that the music has gained two octaves of bass. It goes so low, I had no Idea how much i was missing out on. Its low, low and full and rich and controlled and has volume that moves air.
2. The separation of instruments, their definition and texture is much more visceral. Everything is locked in its place and there is clear separation. The locations of each sound are rock solid.
3. The background has become black. I thought I used to know what that meant when people would say that music comes out of a black background and floats in space. It removes a haze that you couldn't put your finger on.

So, anyone wondering about the AD797 and LME49990 absolutely need to try this cheap and simple mod. 10uf Elna Silmic II across the power pins of the opamp. Its the cheapest, fastest mod out there and has spectacular results.
 


I cannot overstate how phenomenal this simple mod is. I will be trying different caps and bypasses soon to see what else can be achieved.

 
 

  the Xonar HDAV 1.3
Probably the least of all Xonars right out of the box. After bypassing the output caps with wire, it opened up completely into something quite lovely. But, obviously, it wasn't good enough because I have been modding and swapping opamps from the beginning.
Only two mods sofar have had an appreciable effect.
1. bypassing output capacitors with wire.
2. adding a 10uf Silmic II to the opamp + -
 
If you have a pair of AD797B for your I/V, you must try adding 10uf Silmic to it. Perhaps other caps will also be as awesome. Ill let you know as I am now testing capacities and manufacturers to see what difference it makes.

 
 

  Burn in for capacitors? Absolutely.
Burn in for Opamps... Well, I know that some think you need to. I personally don't know that I have ever heard a difference. I have listened for it but a new lme49720ha sounds the same as one I have used for years.
If you can hear the difference and want to burn it in, then open foobar or some similar player and put a track on repeat. Chose something that has deep bass and lots of treble. Let that play over night.
On my xonar, when im burning in capacitors on the opamps like i am right now, i turn off my headphone amp and turn the volume to 100. Let it do its thing for a while, come back, don't forget to turn the volume down and test with a familiar song.

 
 

  I bypassed the LME49990 with 10uF 25V ELNA today. I am actually speechless on how much the sound has improved, there is more separation, detail and control in the lower frequency's. The background is black, the sound has become dark. Everything you stated in your earlier posts (www.head-fi.org/t/432749/the-opamp-thread/4905#post_11796572) was accurate. The sound is more relaxed exactly what I wanted.  I honestly did not expect so much improvement from this cheap mod

  
 
I even de soldered  and re soldered them to double check it was not my imagination
 
Next thing to try is a wima cap, I think I will stick with this for a few hundred hours first let them caps burn in, not even past the four hour mark yet 

.
 
The LME49990 also runs way cooler, anyone who is having trouble with hot OPAMPS should do this mod!
 
Thank you @pelopidas for the information!  

 
 

  @pepopidas
 
I just read your experience from a few pages ago, it seems you have experimented with various OPAMPS and capacitors 

. I have 10uF 25V ELNA on the way, I will post impressions once I have installed them (cant wait). 
 
I have only experience with 49720, 49710 and 49990, and stock OPAMP on STX. If you do find anything better than the LME49990 let me know! The only OPAMPS I hear that are better are discrete and are super expensive. 

 
 

  Here is the pinout for both a dual or single OPAMP:
                           DUAL                                                  SINGLE

              
 
The longer leg is normally the positive, so the shorter leg will go to pin 4. Unless of course you have a bipolar capacitor then it wont matter. 

 
 

  I still cannot believe how much the sound has changed with the 10uF Silmic bypass, that dark black sound cant get enough of it.

 I did a bit of research looking at datasheets of various OPAMP's, it seems implementation varies depending on which OPAMP used:
 
The AD797 within the datasheet states:
"A 1.0 μF to 4.7 μF tantalum in parallel with 0.1 μF
ceramic bypass capacitors are sufficient in most applications.
When driving heavy loads, a larger demand is placed on the
supply bypassing. In this case, selective use of larger values of tantalum capacitors and damping of their lead inductance with
small-value (1.1 Ω to 4.7 Ω) carbon resistors can achieve an
improvement" Pg 13
 
The LME49990 states within its datasheet:
"SUPPLY BYPASSING
To achieve a low noise and high-speed audio performance, power supply bypassing is extremely important.
Applying multiple bypass capacitors is highly recommended. From experiment results, a 10μF tantalum, 2.2μF
ceramic, and a 0.47μF ceramic work well. All bypass capacitors leads should be very short. The ground leads of
capacitors should also be separated to reduce the inductance to ground. To obtain the best result, a large
ground plane layout technique is recommended and it was applied in the LME49990 evaluation board." Pg10
 
It seems as each OPAMP has specific bypass requirements in order to function within its specified optimum values. When the circuit does not have the specific requirements detrimental performance can be expected such as oscillation with associated overheating along with a potential veiled sound which is impossible to notice without a oscilloscope. The information within the datasheets seem a little vague on what is the best bypass arrangement is. 
 
@SpudHarris
 Post us your impressions on the Simlic ll capacitors  I look forward to hearing your impressions! 

 

 
You guys are going to make me learn to use my soldering iron.  It's late, here, but I'll probably be back with some questions, later.  

 

[pelopidas]

I ordered some small 10uf caps in the following flavours.
Nichicon Muse BP - the bipolars. I am really quite curious to see what a bipolar sounds like.
Panasonic FC - the "go to" when building an amp for a friend of a friend who does not know better.
Nichicon Muse KZ - really quite decent caps from previous experiences.
Once these arrive, I am going to retest all caps and opamps again
These were my previous experiments but this needs to be better organized.

http://www.head-fi.org/t/432749/the-opamp-thread/4905#post_11796572
http://www.head-fi.org/t/432749/the-opamp-thread/4920#post_11804552
 
since so many of us have very different equipment, it would also be interesting if you guys would try some of these and let us know how it responds to your equipment.
Since I primarily listen through my HD650 it is very well possible that I am tuning the sound to be excellent for this scenario. What do you guys listen to and what are your impressions of any mods you make?

 

[Stadtpirat]

Hi,
Thank u for the great MOD!
i want use the DUAL 49990 in the O2 and ZXR.
What i need for DUAL 49990 SILMIC Mod and what are the numbers of the pins for solder?
 
I have the DUAL 49990 on DIP8.
 
In germany i can buy:
WIMA MKS4 63V 10uF 10% tol.
 
How Many V are good?
What is better WIMA or SILMIC?

 

[SpudHarris]

I am using the Lycan from Burson. Man, this thing is the OpAmp rollers dream... It's a test bed. Reasonablely priced but transparent as they come, powerful and easily changeable OpAmps.

Listening to an old favourite tonight from my Graham Slee Solo days, the AD823. Using the GS PSU1 with the Lycan which has to help. When my Silmic II's get here I will be trying them on AD797 / ADA4627 (current favourite with the Lycan) OPA111BM and LME49990.

 

[pelopidas]

Hey Spud
How does the Lycan compare to the Fi-quest you were using last time you were talking Amps? Have you tried the output bypassed 5532 on there? I looked it up on Bursons website and it looks like they are using silmics on the board already. Are the 4 big caps silmics?

 

[SpudHarris]

Hi pelopidas,
I haven't compared them like for like but can tell you that the Lycan is a more powerful amp, and from memory it is more transparent than the Fi-Q. That said, the Fi-Q is more flexible and of course portable. I have had the AD823 in place for a couple of days and could quite happily keep it where it is... I have never heard vocals so intimate and clean.

The output bypassed Signetic 5532 is amazing and I will take the AD823 out tonight to test.

The Lycan is an amazing amp, even more so when you consider the cost.

 

[zilch0md]

I ordered some small 10uf caps in the following flavours.
Nichicon Muse BP - the bipolars. I am really quite curious to see what a bipolar sounds like.
Panasonic FC - the "go to" when building an amp for a friend of a friend who does not know better.
Nichicon Muse KZ - really quite decent caps from previous experiences.
Once these arrive, I am going to retest all caps and opamps again
These were my previous experiments but this needs to be better organized.

http://www.head-fi.org/t/432749/the-opamp-thread/4905#post_11796572
http://www.head-fi.org/t/432749/the-opamp-thread/4920#post_11804552

since so many of us have very different equipment, it would also be interesting if you guys would try some of these and let us know how it responds to your equipment.
Since I primarily listen through my HD650 it is very well possible that I am tuning the sound to be excellent for this scenario. What do you guys listen to and what are your impressions of any mods you make?

I would love to participate in the experimentation here, but given that my soldering skills are deplorable, I will just keep following this thread and see which caps you guys end up using on the LME49990. I realize you might never settle on a "final" solution, as I thought you were "there" with the Silica II bypassed with teflon:



Your description of that mod was mouthwatering, but it seems you're after something better still, so I will remain patiently dependent on your diligent research.

I really appreciate your passion and energy!

Mike

PS> A quick shout to Nigel - How's it going?

 

[SpudHarris]

Hey Mike, I'm good pal thanks...

I'm still awaiting my Simlic II's :rolleyes: I'll definitely report back ones I have tried them on the 49990's. If you want me to solder one up for you with caps just shout me, I'm quite nifty with my soldering iron I have a few dual modules lying around.

The nice thing about the Lycan is that rolling is soooooo simple and even those modules in your post above will fit...