or Connect
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Need an explanation of potential causes of DC Offset.
New Posts  All Forums:Forum Nav:

Need an explanation of potential causes of DC Offset.

post #1 of 12
Thread Starter 

I built a CMoy because I am noobish. One channel has 36mv of DC offset and the other has 8mv. True the Opamp in it is about 5000 levels below sub par (have a nice BB coming in the mail soon) but I'd like to know if there is anything other than the amp that could be at fault here (the amp is a TL082C from RadioShack). 

 

Any help is much appreciated and I would like to have /causes/ explained so I can do some debugging on it (Have some nice cans that I dont want to damage and on tangent's site it said that more than 20mv is a bad thing)

post #2 of 12

 

It's easy...

 

The op amp has two inputs, one inverting

and one non-inverting.

 

You need to balance the impedance so that

each input see the same impedance looking

out from the input.

 

Random variation in component tolerances

and using high value feedback networks can

lead to results such as yours. I like to keep

the parallel impedance of my gain and feedback

resistor to a value around 600. This tends to

keep my offsets pretty low.

post #3 of 12
Thread Starter 
Quote:
Originally Posted by Avro_Arrow View Post
 

 

It's easy...

 

 

The op amp has two inputs, one inverting

and one non-inverting.

 

You need to balance the impedance so that

each input see the same impedance looking

out from the input.

 

Random variation in component tolerances

and using high value feedback networks can

lead to results such as yours. I like to keep

the parallel impedance of my gain and feedback

resistor to a value around 600. This tends to

keep my offsets pretty low.

thank you for this explanation. So my current setup is 4.7k coming out of the output into the feedback loop (leaving out bass boost circuit atm as I'll get that calculated later) and the resistor to ground is a 1k. The input has a 100k resistor to ground. does this mean I need to lower the value of the 100k?

post #4 of 12

 

In reality there are some practical limitations.

 

If you are using a DC blocking capacitor on the input,

it is pretty tough getting a low impedance here.

If you raise the feedback/gain resistors to try and

match the input impedance, you will have excess

noise.

 

This is one of those situations where the difference

between bi-polar input op amps and JFet input op amp.

 

Op amps have a bias current that flows either into or

out of (depending on the op amps design) the inputs.

As the current flows through the impedance seen by

the inputs, it develops a voltage in accordance with ohms law.

The greater the bias current, the greater the voltage.

 

Bi-polar input op amps have, in general, much higher

bias currents than JFet op amps.

 

When you switch from the TL082C to the OPA3134/OPA3132

the offsets will be much less.

post #5 of 12
Thread Starter 
Quote:
Originally Posted by Avro_Arrow View Post
 

 

In reality there are some practical limitations.

 

 

If you are using a DC blocking capacitor on the input,

it is pretty tough getting a low impedance here.

If you raise the feedback/gain resistors to try and

match the input impedance, you will have excess

noise.

 

This is one of those situations where the difference

between bi-polar input op amps and JFet input op amp.

 

Op amps have a bias current that flows either into or

out of (depending on the op amps design) the inputs.

As the current flows through the impedance seen by

the inputs, it develops a voltage in accordance with ohms law.

The greater the bias current, the greater the voltage.

 

Bi-polar input op amps have, in general, much higher

bias currents than JFet op amps.

 

When you switch from the TL082C to the OPA3134/OPA3132

the offsets will be much less.

 

Ok. Until then, is it safe to use headphones on this?

post #6 of 12
Thread Starter 
Quote:
Originally Posted by Avro_Arrow View Post
 

 

In reality there are some practical limitations.

 

 

If you are using a DC blocking capacitor on the input,

it is pretty tough getting a low impedance here.

If you raise the feedback/gain resistors to try and

match the input impedance, you will have excess

noise.

 

This is one of those situations where the difference

between bi-polar input op amps and JFet input op amp.

 

Op amps have a bias current that flows either into or

out of (depending on the op amps design) the inputs.

As the current flows through the impedance seen by

the inputs, it develops a voltage in accordance with ohms law.

The greater the bias current, the greater the voltage.

 

Bi-polar input op amps have, in general, much higher

bias currents than JFet op amps.

 

When you switch from the TL082C to the OPA3134/OPA3132

the offsets will be much less.

  

 

also, I googled the OPA3134 but didnt come up with a place to get it.....I had originally planned on the opa2132/34....is this still ok? I'm new to opamps and have no idea what the "best" are

post #7 of 12

 

I must have been tired (thats my excuse and I'm sticking to it...), but other sharp eyes should have caught it...

 

What you are looking for is OPA2134, not OPA3134, no wonder you couldn't find it.

The NJM4556 is also pretty decent in a CMoy.

 

When designing an amp, we target less than 10mV offset.

Less than 50mV offset should be safe even in the long term

for pretty much anything but the most delicate headphones.

post #8 of 12
Thread Starter 
Quote:
Originally Posted by Avro_Arrow View Post
 

 

I must have been tired (thats my excuse and I'm sticking to it...), but other sharp eyes should have caught it...

 

haha ok.

 

Quote:
 What you are looking for is....

Wow...I believe i was looking at the P not the PA so it was $7 vs the 3ish....and that other one...thats only 78c! I've been wanting to sell these locally (I'm in high school and some friends have expressed interest in my CMoys [especially since I'm printing my enclosures because I have access to a 3d printer]) but it wasn't looking like it was going to be possible with a $7 opamp and most everything else sourced to RadioShack....i've been resourcing a lot of parts but that opamp will help A LOT with the price....is theNJM as good as the OPA? or are they about the same? 

post #9 of 12

 

The NJM is a bi-polar input op amp so you will have to be more careful

with resistor selection to avoid offset but it has good output current

at lower impedance than the OPA.

 

The OPA is a JFet input op amp and is more forgiving as to component

selection.

 

You should be using 1% metal film resistors if you can.

 

Try them both and see which sounds better with your design, the music

you and your friends listen to and with the headphones they use.

You might find one always sounds better than the other and you might

find some prefer one and some prefer the other...give them a choice.

post #10 of 12

 

If your up for some fun, once you get a design working well, try soldering

one op amp right on top of the other...stacking them...

post #11 of 12
Thread Starter 
Quote:
Originally Posted by Avro_Arrow View Post
 

 

The NJM is a bi-polar input op amp so you will have to be more careful

 

with resistor selection to avoid offset but it has good output current

at lower impedance than the OPA.

 

The OPA is a JFet input op amp and is more forgiving as to component

selection.

 

You should be using 1% metal film resistors if you can.

 

Try them both and see which sounds better with your design, the music

you and your friends listen to and with the headphones they use.

You might find one always sounds better than the other and you might

find some prefer one and some prefer the other...give them a choice.

I only use 1% metal film as per tangent's recommendation. My bad experience with bi-polar input makes me lean towards the OPA but I understand that lower output impedance gives better dampening factor as well.....Since they are so cheap I'll get both

post #12 of 12
Thread Starter 
Quote:
Originally Posted by Avro_Arrow View Post
 

 

If your up for some fun, once you get a design working well, try soldering

 

one op amp right on top of the other...stacking them...

what will this do exactly? will it harm the amps at all?

 

nevermind. I googled it. I'll try it with my two sucky opamps when I get to the shop today. :)


Edited by backspace119 - 2/4/14 at 8:36am
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Need an explanation of potential causes of DC Offset.