in search of a cheap and easy to build cmoy

[enJr]

i am thinking about building a cmoy. i wandered around the internet and read bunch of stuff.
 
seems that there was a guy, nwavguy, who did an awesome home amplifier called objective 2 (which is currently sold on jds labs web page, i am sure most of you aware of), which is a nice objective by the way. he also did some measurements on a standart cmoy with a flawed design.
 
i know there are lots of better desings, but they are too complicated to build (it would be bothersome to bug) or they are no portable (objective 2 for example)
 
i am thinking about morphing those two. i mean cmoybb and standart cmoy with jrc4556ad.
 
it is like this with 82nF//22k bass boost: image
 
any ideas how it is gonna turn out? any insights would be great.

 

[tomb]

  i am thinking about building a cmoy. i wandered around the internet and read bunch of stuff.
 
seems that there was a guy, nwavguy, who did an awesome home amplifier called objective 2 (which is currently sold on jds labs web page, i am sure most of you aware of), which is a nice objective by the way. he also did some measurements on a standart cmoy with a flawed design.
 
i know there are lots of better desings, but they are too complicated to build (it would be bothersome to bug) or they are no portable (objective 2 for example)
 
i am thinking about morphing those two. i mean cmoybb and standart cmoy with jrc4556ad.
 
it is like this with 82nF//22k bass boost: image
 
any ideas how it is gonna turn out? any insights would be great.

His measurements on the Mini3 were seriously flawed and then got worse from there.  He was trying to discredit the CMoy after smearing the Mini3, but like all very simple circuits, it tested quite well - at least for the things he tested, those which he considered important.  There's a lot more to it than that.
 
Regardless, a CMoy is great fun and performs like a champ, especially with high-impedance phones, which is the basic reason Chu Moy invented it.  In the DIY-world, though, the CMoy is sort of a rite-of-passage.  All of us who've been here awhile recommend Tangent's excellent version and tutorial as a first build:
http://tangentsoft.net/audio/cmoy-tutorial/
 
It will pretty much give you priceless experience in almost every area of DIY: soldering, wiring, picking parts for making up a BOM, even case work (hint: a paper hole punch can do a number on mint-tins).  Except for the opamp, Tangent even wrote the tutorial so that you could pretty much purchase everything at the local Radio Shack.
 
P.S. I would stick with the opamps that Tangent recommends.  An OPA2134 or 2132 will give spectacular results, doesn't cost an arm and a leg, and you can always change it out later.
 
P.S.S. Note that you if you go through a legitimate DIY resource such as Tangent's, the entire principles of gain are discussed and options for using different resistors to obtain different gains are clearly spelled out.  There is no trumped-up "gotcha" with an isolated CMoy that happened to be built with no-gain.

 

[enJr]

thank you, i appreciate it. i think i am going to build the same circuit i have described with OPA2134. and like you said, all it takes is to change op-amp with another, jrc4556ad for example.

 

[00940]

On topic: the njm4556 is a great chip if you have low impedance, low sensitivity cans (Grado for example). But it's a bjt opamp, so you need to take care of balancing the input bias current to avoid offset. Makes it quite a lot more difficult to design. See this page by Tangent for basics. A simple alternative for those headphones is to build a cha47 (schematic can be found here ).
 
For other headphones, a standard cmoy with opa134 isn't too bad at all, jfet input so no worry about bias current, not too sensitive to layout, decent current output. If you have 300ohms cans, seriously consider using two batteries. If you are using low impedance headphones, one might be enough but then you should reduce the resistors in the divider (r1+/r1-) to something like 1K5 each and increase the caps to 1000uF.
 
 
 
Quote:

  His measurements on the Mini3 were seriously flawed and then got worse from there.  He was trying to discredit the CMoy after smearing the Mini3, but like all very simple circuits, it tested quite well - at least for the things he tested, those which he considered important.  There's a lot more to it than that.

 
 An ax to grind ?
 
How were his measurements seriously flawed ? Reading the "good old" thread on amb's forum, the only "flaw" I can identify is trying to get the quoted max power into 33r (which the mini3 components obviously cannot deliver).
 
And what didn't he test that it is so important ?

 

[tomb]

  On topic: the njm4556 is a great chip if you have low impedance, low sensitivity cans (Grado for example). But it's a bjt opamp, so you need to take care of balancing the input bias current to avoid offset. Makes it quite a lot more difficult to design. See this page by Tangent for basics. A simple alternative for those headphones is to build a cha47 (schematic can be found here ).
 
For other headphones, a standard cmoy with opa134 isn't too bad at all, jfet input so no worry about bias current, not too sensitive to layout, decent current output. If you have 300ohms cans, seriously consider using two batteries. If you are using low impedance headphones, one might be enough but then you should reduce the resistors in the divider (r1+/r1-) to something like 1K5 each and increase the caps to 1000uF.
 
 
 
Quote:

  His measurements on the Mini3 were seriously flawed and then got worse from there.  He was trying to discredit the CMoy after smearing the Mini3, but like all very simple circuits, it tested quite well - at least for the things he tested, those which he considered important.  There's a lot more to it than that.

 
 An ax to grind ?
 
How were his measurements seriously flawed ? Reading the "good old" thread on amb's forum, the only "flaw" I can identify is trying to get the quoted max power into 33r (which the mini3 components obviously cannot deliver).
 
And what didn't he test that it is so important ?

You got a PM.  I'm not going to get baited on this stuff and it's not productive.  I expressed an opinon - and it's not in a vacuum.

 

[mcandmar]

I've built a few from parts stabbed into a breadboard, to simple point to point wiring on a proto board crammed into an Altoids tin, to Grado RA-1 clones in cigar box's and they can sound pretty decent, but no where near the O2.
 
As to schematic / component values, sounds like your already on top of that. Experimentation is all part of the fun so build it on a breadboard and work from there..
 
 

 

[vixr]

   
 An ax to grind ?
 
How were his measurements seriously flawed ? Reading the "good old" thread on amb's forum, the only "flaw" I can identify is trying to get the quoted max power into 33r (which the mini3 components obviously cannot deliver).
 
And what didn't he test that it is so important ?

if you think tomb has ANY ax to grind, then you are hopeless... and that's Mr. B to you...

 

[enJr]

  For other headphones, a standard cmoy with opa134 isn't too bad at all, jfet input so no worry about bias current, not too sensitive to layout, decent current output. If you have 300ohms cans, seriously consider using two batteries. If you are using low impedance headphones, one might be enough but then you should reduce the resistors in the divider (r1+/r1-) to something like 1K5 each and increase the caps to 1000uF.

i am gonna use it with superlux hd668b, it is around 50 ohms, i suppose it wouldn't hurt to build it with 2 batteries.
 
i am wondering if there are any other opamps which can perform better, like ad826 or ad8066. anybody have any idea? by the way, thank you all, for all your insightful comments.

 

[tomb]

 

  For other headphones, a standard cmoy with opa134 isn't too bad at all, jfet input so no worry about bias current, not too sensitive to layout, decent current output. If you have 300ohms cans, seriously consider using two batteries. If you are using low impedance headphones, one might be enough but then you should reduce the resistors in the divider (r1+/r1-) to something like 1K5 each and increase the caps to 1000uF.

i am gonna use it with superlux hd668b, it is around 50 ohms, i suppose it wouldn't hurt to build it with 2 batteries.
 
i am wondering if there are any other opamps which can perform better, like ad826 or ad8066. anybody have any idea? by the way, thank you all, for all your insightful comments.

00940 gives good advice, but I'd hold the horses on thinking that 50 ohms is that much higher in impedance - it's not.  You might check the efficiency of the Superlux. If it's not very high, then use the low impedance advice that he gives.  Otherwise, the single 9V should be plenty.
 
The trouble with most Analog opamps is that they're mostly SOIC-8, which means SMD.  You can't use them in an opamp socket unless you solder them into a Brown Dog adpater (Cimarron Technology), first.  It's more than that, though - it usually means the opamp is not stable with the trace/lead lengths in standard opamp sockets and through-hole construction.  The AD8066, for instance comes in only SOIC-8 or smaller.  I think you can get the AD826 in DIP-8, but again - maybe it's just me, but I've not had much luck using Analog opamps in most circuits that are designed for opamp rolling.  I can recall plugging an AD8066 with a Brown Dog adapter into a SOHA-1 and the result was smoke.

 

[enJr]

  00940 gives good advice, but I'd hold the horses on thinking that 50 ohms is that much higher in impedance - it's not.  You might check the efficiency of the Superlux. If it's not very high, then use the low impedance advice that he gives.  Otherwise, the single 9V should be plenty.
 
The trouble with most Analog opamps is that they're mostly SOIC-8, which means SMD.  You can't use them in an opamp socket unless you solder them into a Brown Dog adpater (Cimarron Technology), first.  It's more than that, though - it usually means the opamp is not stable with the trace/lead lengths in standard opamp sockets and through-hole construction.  The AD8066, for instance comes in only SOIC-8 or smaller.  I think you can get the AD826 in DIP-8, but again - maybe it's just me, but I've not had much luck using Analog opamps in most circuits that are designed for opamp rolling.  I can recall plugging an AD8066 with a Brown Dog adapter into a SOHA-1 and the result was smoke.

i never thought about socket difference. i was thinking that they should be the same.
 
about ad826, do they sound like the dual version of ad825? there was a thread on the forum, somebody reviewed alot of opamps suggesting ad8066 and ad825.
 
and thank you for your insightful reply. i owe you one