OPA627 - Class A Biasing?
post-759206
Thread Starter
Post #1 of 28

Duncan

Headphoneus Supremus
Moderator
Joined
Jun 26, 2001
Messages
13,166
Reaction score
1,318
Joined
Jun 26, 2001
Posts
13,166
Likes
1,318
Hiya... not too often that I post here


Quick question to all you DIY gurus... looking at converting the 627s that I have put into my CD player into class a biasing, but am not sure of the value of the resistors to get...

I have a service manual / schematics for this player if there isn't a straight forward answer, and is dependent on the rest of the circuit...

Thanks all!
 
     Share This Post       
post-759212
Post #2 of 28

Glassman

Headphoneus Supremus
Joined
Mar 22, 2003
Messages
1,830
Reaction score
11
Joined
Mar 22, 2003
Posts
1,830
Likes
11
you need to get around 2mA per opamp, the equation is quite simple:

your negative supply voltage divided by 2 = resistor value in kiloohms..

put that resistor from the opamp's output to that negative supply..
 
     Share This Post       
post-759222
Post #3 of 28

Duncan

Headphoneus Supremus
Moderator
Joined
Jun 26, 2001
Messages
13,166
Reaction score
1,318
Joined
Jun 26, 2001
Posts
13,166
Likes
1,318
15v... so thats a 7.5k ohm resistor per channel...

Sweet, thanks Glassman


(btw... any particular grade or type that I should get, or just general purpose?)
 
     Share This Post       
post-759334
Post #4 of 28

PinkFloyd

Headphoneus Supremus
Joined
Jan 13, 2009
Messages
9,511
Reaction score
21
Joined
Jan 13, 2009
Posts
9,511
Likes
21
So Duncan would connect a 7k5 resistor between pins 6-2 ?

 
     Share This Post       
post-759339
Post #5 of 28

00940

Headphoneus Supremus
Joined
Nov 6, 2002
Messages
4,493
Reaction score
44
Joined
Nov 6, 2002
Posts
4,493
Likes
44
from 4 to 6
 
     Share This Post       
post-759348
Post #6 of 28

PinkFloyd

Headphoneus Supremus
Joined
Jan 13, 2009
Messages
9,511
Reaction score
21
Joined
Jan 13, 2009
Posts
9,511
Likes
21
That's what I initially thought but Duncan pointed out "or would it be two... if its a feedback 'device' then it'd be two rather than 4"

So....... 6-2 or 6-4 ? I still think 6-4 but it's always best to get a third and 4th opinion :)
 
     Share This Post       
post-759361
Post #7 of 28

KTpG

100+ Head-Fier
Joined
Jan 17, 2003
Messages
436
Reaction score
10
Joined
Jan 17, 2003
Posts
436
Likes
10
6-4.

Class A biasing is for the output... and I don't think it would do well on the inverting input (pin 2).

BTW- 2mA per chip may not be exactly what you want. You can use other biases. You might try different values to see what works best- perhaps as low as 2.21k ohms and as high as 10k or so.
 
     Share This Post       
post-759363
Post #8 of 28

00940

Headphoneus Supremus
Joined
Nov 6, 2002
Messages
4,493
Reaction score
44
Joined
Nov 6, 2002
Posts
4,493
Likes
44
Quote:

Originally posted by Glassman
put that resistor from the opamp's output to that negative supply..


or also from the HA-1 project :


the biasing resistor is the 1.5k, 0.6W one

or even, from Tangent's article on opamp biasing :

 
     Share This Post       
post-759373
Post #9 of 28

PinkFloyd

Headphoneus Supremus
Joined
Jan 13, 2009
Messages
9,511
Reaction score
21
Joined
Jan 13, 2009
Posts
9,511
Likes
21
Quote:

Originally posted by Duncan
Hiya... not too often that I post here


Quick question to all you DIY gurus... looking at converting the 627s that I have put into my CD player into class a biasing, but am not sure of the value of the resistors to get...

I have a service manual / schematics for this player if there isn't a straight forward answer, and is dependent on the rest of the circuit...

Thanks all!


So the question is fourfold from where I stand.

1: Will converting the 627's into class A biasing bring about any sonic improvements and, if so, why?

2: Is it worth spending money on milspec resistors or will standard 1% metal film be ok ( I know the answer to that one..... use standard 1% varieties in this application)

3: connect between pins 6-2 or 6-4 ?

4: Is Glassman's reply 100% on the button or are there any other views on this "you need to get around 2mA per opamp, the equation is quite simple:

your negative supply voltage divided by 2 = resistor value in kiloohms..

put that resistor from the opamp's output to that negative supply.."

Comments appreciated and anything else that should be taken into consideration welcomed.

Pinkie.
 
     Share This Post       
post-759387
Post #10 of 28

guzzler

Headphoneus Supremus
Joined
Jan 17, 2003
Messages
1,851
Reaction score
12
Joined
Jan 17, 2003
Posts
1,851
Likes
12
1. It should provide a small improvement in sound, as the opamp is always loaded, and so there is no crossover distortion

2. 1% resistors will be fine there

3. 6 and 4. 6 and 2 would make it act as a feedback device, I'm not sure what affect it would have, but it wouldn't be nice!

4. 2mA is a decent starting value, some opamps will respond better at lower a lower bias, but that is something experimentation determines.

A more advanced approach would be to use a JFET cascode (see PIMETA/PPA schematic) or a CRD (see MINT schematic)

g
 
     Share This Post       
post-759398
Post #12 of 28

PinkFloyd

Headphoneus Supremus
Joined
Jan 13, 2009
Messages
9,511
Reaction score
21
Joined
Jan 13, 2009
Posts
9,511
Likes
21
Quote:

Originally posted by guzzler
1. It should provide a small improvement in sound, as the opamp is always loaded, and so there is no crossover distortion

2. 1% resistors will be fine there

3. 6 and 4. 6 and 2 would make it act as a feedback device, I'm not sure what affect it would have, but it wouldn't be nice!

4. 2mA is a decent starting value, some opamps will respond better at lower a lower bias, but that is something experimentation determines.

A more advanced approach would be to use a JFET cascode (see PIMETA/PPA schematic) or a CRD (see MINT schematic)

g


Cheers Guzzler.

Pinkie.
 
     Share This Post       
post-759426
Post #13 of 28

PinkFloyd

Headphoneus Supremus
Joined
Jan 13, 2009
Messages
9,511
Reaction score
21
Joined
Jan 13, 2009
Posts
9,511
Likes
21



This is all very interesting
Can anybody work out what Graham slee has done by placing a 33K resistor between pins 7-8 and 1-8 on the AD823 op amp?

This was one of the circuit mods for the 2004 Solo :wink:

Pinkie.
 
     Share This Post       
post-759447
Post #14 of 28

PinkFloyd

Headphoneus Supremus
Joined
Jan 13, 2009
Messages
9,511
Reaction score
21
Joined
Jan 13, 2009
Posts
9,511
Likes
21
That was probably an unfair question as the mod also involved replacing a BC337 with a BC184C, a 4k7 with a 33K, a 10uT with a 220nF, aa 100pF with a 220pF, a 47K with an 18K, a 1k with a 4K7 and a 220uF with a 1000uF........... oh and a 4K7 was replaced with a 66K.


here's the consumers view:


I've got the full schematic of the 2004 Solo but I would be really interested to hear your opinions on the 33k op amp bridge ..... I know the answer as I've got the blueprints in front of me but it's an interesting exercise


The answer will follow
 
     Share This Post       
post-759517
Post #15 of 28

Glassman

Headphoneus Supremus
Joined
Mar 22, 2003
Messages
1,830
Reaction score
11
Joined
Mar 22, 2003
Posts
1,830
Likes
11
positive bias, nothing more.. most opamps sounds better when biased to negative, ie. they are sourcing current, but a few others sound better when they sink current.. this guy apparently believes AD823 is one of them
 
     Share This Post       

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top