[ninjikiran]

I see it like this, money is not free....

 

[Sganzerla]

 Is there any consensus already of what capacitors one could change for a better sound?
 Something like smoother highs, more articulate bass and mids, instrument separation, etc...?

 

[zibra]

Good question
I would also ask what opamps do you recommend for matrix, something like AD797 class or better? Any discrete opamps would be good, no matter of size? I rather prefer first option but would like to know possibilites for Matrix without mods

 

[purrin]

 
Quote:

 Is there any consensus already of what capacitors one could change for a better sound?
 Something like smoother highs, more articulate bass and mids, instrument separation, etc...?

 
Pansonic FMs? Maybe too bright though. Panny FCs? Elna Silmics for smooth? Might need to mix and match for voltage gain and current gain stages. Can't really do much with the existing output transistors - not that they are bad - but there will always be some graininess in the treble.
 
Only thing I could think of that I'm sure would work are oil-filled caps (e.g. ASC x386s, Obligatos), but those things are so huge that even one of them wouldn't even fit in the case.
 
A quick and dirty solution would be to bypass the existing caps with a small film cap, e.g. WIMAs. I could only fit 0.01uFs on the bottom of the board, but the mods made a small improvement in all of the aspects that you have mentioned. I did this to the four caps feeding the output transistors. I left the caps feeding the opamp alone - they don't need it as they are already bypassed with 0.022uf film caps.

 

[purrin]

Quote:

2 - My one question for M-Stage owners - "IF"? I wanted a slightly warmer presentation with my k701s... I have the class A OPA627 upgrade in - I have some Opamps on hand...incl LT1364, LM4562, and OPA2107. What would you suggest for my k701s - if I favored a bit of a warmer feel? 

Definitely not OPA2107 = controlled bass and bright. I love using this opamp with an output stage that is slowish and warm.
 
LM4562? (more it gets pushed into class A, the warmer it sounds). OPA627 pushed into class A sounds brighter. This only applies when using resistor mod on the Matrix. I actually think the LM4562 and OPA627 have a sound balance that converges with light class A biasing.
 
I disagree with folks who say LM4562 sounds bright. I've had at least a dozen of of those chips (purchased directly from Digikey) that I've modified and sent off to other head-fiers. LM4562 even without class A mod sounds warmer than stock OPA2134 (and I've got a dozens of those sitting around too for other projects). YMMV.
 
JRC4556? It's warm all right!

 

[BmWr75]

Quote:

Quote:

2 - My one question for M-Stage owners - "IF"? I wanted a slightly warmer presentation with my k701s... I have the class A OPA627 upgrade in - I have some Opamps on hand...incl LT1364, LM4562, and OPA2107. What would you suggest for my k701s - if I favored a bit of a warmer feel? 

Definitely not OPA2107 = controlled bass and bright. I love using this opamp with an output stage that is slowish and warm.
 
LM4562? (more it gets pushed into class A, the warmer it sounds). OPA627 pushed into class A sounds brighter. This only applies when using resistor mod on the Matrix. I actually think the LM4562 and OPA627 have a sound balance that converges with light class A biasing.
 
I disagree with folks who say LM4562 sounds bright. I've had at least a dozen of of those chips (purchased directly from Digikey) that I've modified and sent off to other head-fiers. LM4562 even without class A mod sounds warmer than stock OPA2134 (and I've got a dozens of those sitting around too for other projects). YMMV.
 
JRC4556? It's warm all right!

Purrin,
 
When I search Digikey for LM4562 and OPA627, multiple products are returned for each.  Which part #s are correct for the mstage?
 
Thanks,
Scott

 

[purrin]

LM4562: Digi-Key Part Number: LM4562NA-ND
 
OPA627:
Digi-Key Part Number: OPA627BP-ND (DIP8) -or-
Digi-Key Part Number: OPA627AU-ND (SMT)
        
OPA627s will need a appropriate 2x single to dual adapter.  DIP8 will plug into the adapter.  SMT will require soldering (see alternative below).  Here is a single to dual adapter for DIP8:
 
http://cimarrontechnology.com/single-to-dualop-ampadapter-dipversionpn021001.aspx
(order with two DIP sockets installed and x2 DIP8 OPA627s) This will be the most flexible plug-n-play solution.
 
Alternatively for OPA627 try: http://cimarrontechnology.com/single-to-dualop-ampadapterpn020302.aspx
(order with extended pins, then you can easily solder resistors for class A bias directly on the extended pins over the chips) This would probably be the best sounding solution as it keeps everything in the same package without additional class A piggyback adapters, etc. The distance to the power decoupling caps is critical to opamps. You start putting some distance and extra resistance to the opamps from the socket on the board... not ideal.
 
I've heard stories about fake chips before, but now I'm more convinced than ever to order only from trusted sources. If it's too good to be true, it is, e.g. "These are special OPA627s made in a special plant for domestic consumption in whateverland so that's why they are $4 dollars each, etc."  Ya right. I've heard Bernie Madoff can get you 50% returns on your investment consistently year after year - and I'm sure he's got a special explanation on how he does it too.

 

[BmWr75]

Purrin,
 
Thanks for the details on part numbers.  I ordered the OPA627 from China, am surprised to see the US sourced ones cost about 4X-5X more.  Think I'll try the US sourced ones to see if I can hear a difference.
 
Scott

 

[avexdevil]

 
Quote:

I'm just posting for 2 reasons:
 
1 - those who follow this thread - which is slanting to a "now it's time to sell the M-Stage for the next big thing" OR the "the FOTM hype was off and this thing just isn' what I thought it would be" posts...I bought the M-Stage very early on - I'm not a hype freak. This amp is flat out married to my k701s - not even a question. I would sell the k701s if I didn't own the M-Stage. Just saying - they are beautiful together - I've a/b'd against the LD MKIV which I am a HUGE fan of and the M-Stage is the huge winner for the k701s
 
2 - My one question for M-Stage owners - "IF"? I wanted a slightly warmer presentation with my k701s... I have the class A OPA627 upgrade in - I have some Opamps on hand...incl LT1364, LM4562, and OPA2107. What would you suggest for my k701s - if I favored a bit of a warmer feel? 

 
1. WORD. qft. i had the k701s before and i got the m-stage to compliment it. people who go the other way, amps before headphones will find themselves wanting to try the next best combination and that usually becomes a vicious cycle of neverending upgrades (not to mention incredibly costly). Best you pick a pair of headphones you like, stick with 'em, then get a nice amp to maximise its potential.
 
2. maybe it's because i own a m50 and several warm presentation iems, I too have the class A opa627 modded in, but the k701 will never sound warm enough for my standards. In fact, I would go as far to say that I only use my k701s for acoustic guitar and pure vocal tracks (I don't listen to much classical unfortunately, just orchestral compositions from movie soundtracks), the k701's intepretation for any other type of music becomes too thin and lifeless for my taste. They are great for movies and relaxation music; only reason why i'm still holding on to them in favour of getting some hd650s.
 
I'd also be interested to know what OPamps can give the k701 extra warmth/thickness

 

 

[purrin]

I've got my modified M-Stage for sale if anyone is interested. I figured I'd let people on this thread know because I get questions on and requests for mods from time to time.
 
http://www.head-fi.org/forum/thread/529933/fs-matrix-m-stage-modified
 
Nothing wrong with it. IMO it's probably the best entry-level amp of this type of design (opamp + diamond buffer) you can buy. I'm just moving up to another league.

 

[DaEMoNteNTAcLe]

anyone knows what caps i can replace the pair near the opamps to make this amp sound darker? mine is elna.
or should i change the 2 big power caps instead? also elna.
 
im kinda done playing opamps, and dont want to change them, or topology of the circuit.
i just need it to be darker sounding overall. im using lovely cube tho, but same mod should affect should apply.

 

[K3cT]

Thanks to my dear friend Mike, I got a M-Stage for a loan and I immediately took it for a ride with the Audez'e LCD2. Plenty of power, no problem for the LCD2 as it is able to drive it quite effortlessly. The M-Stage is no β22 for sure but I still like it. 
 
Well, after a while I noticed how the sound was a bit too distant with the class A-biased OPA627 so I swapped it with a personal favorite of mine, the cheap-but-amazing-sounding OPA1642. Well, the first thing that I notice is the sound is no longer distant but now a little bit forward, just the way I like it. However, this doesn't do high as well as the OPA627 as for example, cymbals now lack shimmer and air. Overall the sound has become a little bit darker and more forward. 

 

[purrin]

Quote:

anyone knows what caps i can replace the pair near the opamps to make this amp sound darker? mine is elna.
or should i change the 2 big power caps instead? also elna.
 
im kinda done playing opamps, and dont want to change them, or topology of the circuit.
i just need it to be darker sounding overall. im using lovely cube tho, but same mod should affect should apply.

You can try Silmics, although I don't know if they will actually be more warm than the stock Elnas. I suspect changes to the four medium size caps ~440uF? feeding the output transistors will do more than changes to the opamp caps. You can try increasing the value of the caps for more warmth, but that could slow things down a bit.

 

[dw1narso]

Quote:

Thanks to my dear friend Mike, I got a M-Stage for a loan and I immediately took it for a ride with the Audez'e LCD2. Plenty of power, no problem for the LCD2 as it is able to drive it quite effortlessly. The M-Stage is no β22 for sure but I still like it. 
 
Well, after a while I noticed how the sound was a bit too distant with the class A-biased OPA627 so I swapped it with a personal favorite of mine, the cheap-but-amazing-sounding OPA1642. Well, the first thing that I notice is the sound is no longer distant but now a little bit forward, just the way I like it. However, this doesn't do high as well as the OPA627 as for example, cymbals now lack shimmer and air. Overall the sound has become a little bit darker and more forward. 

 
Brian,
 
have you play with the gain setting, both with the OPA627 and OPA1642?
 
Before you, I also borrowed Mike's M-Stage... gain setting definitely affect the sound... for Mike's class A biased OPA2134, I like the highest gain the best...

 

[lescanard]

Quote:

Quote:

2 - My one question for M-Stage owners - "IF"? I wanted a slightly warmer presentation with my k701s... I have the class A OPA627 upgrade in - I have some Opamps on hand...incl LT1364, LM4562, and OPA2107. What would you suggest for my k701s - if I favored a bit of a warmer feel? 

Definitely not OPA2107 = controlled bass and bright. I love using this opamp with an output stage that is slowish and warm.
 
LM4562? (more it gets pushed into class A, the warmer it sounds). OPA627 pushed into class A sounds brighter. This only applies when using resistor mod on the Matrix. I actually think the LM4562 and OPA627 have a sound balance that converges with light class A biasing.
 
I disagree with folks who say LM4562 sounds bright. I've had at least a dozen of of those chips (purchased directly from Digikey) that I've modified and sent off to other head-fiers. LM4562 even without class A mod sounds warmer than stock OPA2134 (and I've got a dozens of those sitting around too for other projects). YMMV.
 
JRC4556? It's warm all right!

Thanks for the input purrin - I'm going to give the LM4562 class A a whirl I think - I love the 701s - I'm seeking a bit of a better low end (although I'm not a bass head - I enjoy well presented lows) like I hear in my K271s - we'll see.