The Opamp thread

[SpudHarris]

^ Thanks leeperry 'not helping'.......

Seriously though, will DC Offset kill over a period of time? Is it a heart attack or a long term illness??

 

[leeperry]

all I can say is that there's loads of caps on my board, and hopefully they'll chime out any excess DC offset.

and I love the LT1364 SS, wider than the 49720..yet very natural, I'm hearing a lot of stuff in the mids that used to be hidden/resonating

 

[12Bass]

AC-coupled circuits (where capacitors are used between stages) shouldn't have DC offset issues. It is often difficult to know the design unless you have a schematic. The best way to tell for sure, however, is to measure the output with a multimeter. Under 10mV is best.

 

[12Bass]

Unlike a normal AC music signal, DC offset sends a constant current through the headphones. If it is sufficiently high, it can burn out the fine wire windings. A smaller DC offset can lead to other issues, as this asymmetry displaces the voice coil from its resting position, a condition which may lead to bottoming out or clicks on transients.

 

[SpudHarris]

Oh man...... How do I know if I'm harming my JH13's?? I own a multi-meter where do I need to measure?? Any help would be great, I don't have $1000 to waste.

 

[leeperry]

ok, what about that 30" FLAC sample? MEGAUPLOAD - 04 - Roy Budd - Plaything.flac

are the first notes of the keyboard on the right channel distorted and full of ugly aliasing?

if not, then it's time for me to roll op-amps

PS: I'm also hearing clipping on the right channel of this song...no wonder, it's been through some major loudness war treatment

 

[Slash47]

Quote:

Originally Posted by SpudHarris /img/forum/go_quote.gif Oh man...... How do I know if I'm harming my JH13's?? I own a multi-meter where do I need to measure?? Any help would be great, I don't have $1000 to waste.

Just put it on DC, mV's.

Measure from each channel to ground. It should be below 20mV ideally. It's as simple as that.

 

[leeperry]

ok I've just done it, I manage to measure 1 mV if I'm lucky...manual says there's a 1% error margin on this $10 voltmeter

 

[murrays]

Quote:

Originally Posted by ecclesand /img/forum/go_quote.gif I read that and it's still over my head. I'm an MBA with enough knowledge of electronics and circuitry to be dangerous. In other words, I know squat!

Ok, the basic concept is this:
Op amps by design have a small bias current flowing at the input.
The current flows in the input circuit resistance resulting in a voltage.
(Ohms law gives: voltage = resistance x current).
This input voltage is further multiplied by the gain of the stage to appear at the output.
Many circuits have capacitors to block the DC, so it is not a problem.
However, removing capacitors deemed "unnecessary" is a mod that some folks do. So just beware of these possible consequences. It is easy to check for DC offset after making changes.

Quote:

Originally Posted by ecclesand /img/forum/go_quote.gif I have been looking at various TI datasheets for the Burr Brown opamps and sometimes this input bias current is measured as nA and sometimes it is pA. What is the difference and why are 2 opamps (by the same company) measured differently?

milli, micro, nano and pico are all metric prefixes (as kilo & mega).
All just different scales for measuring the same thing, current in Amperes. 2 uA = 2000000 pA.

Quote:

Originally Posted by ecclesand /img/forum/go_quote.gif Also, what is the best (or in my case, the easiest) way to measure DC offset? Can I simply touch the probes of a multimeter to the output RCA jacks on my DAC (which is where I'm rolling the opamps)? I certainly don't want some opamp to blow away my Millett MiniMax.

A good way is to make up a simple test adapter to use with your multimeter. Solder some wires or terminals onto a spare headphone plug to make it easier to attach the mutimeter probes. I have one with a pair of load resistors soldered on to it.
A quick and dirty test for DC offset is to set the output to zero (i.e. stop playing any audio) and plug and unplug the headphones several times. If you hear clicking in the headphones, this will be the DC current switching off and on as the contact is made & broken. Of course, don't use your expensive phones for this!

Quote:

Originally Posted by ecclesand /img/forum/go_quote.gif Also, how can you tell if an opamp is oscillating? I've read it described as a whooshing sound. I haven't heard this yet so maybe I haven't experienced it?

An oscilloscope is best, but not always available.
Yes, I have heard whooshing (hiss). It may vary as you move the headphone cable around, or near to sources of interference. Also the sound may change as you move your hand near the metalwork of the amp (case, heatsinks, knobs, etc).
Another indication is excessive heat or power drain - sometimes there is more power being used at RF than at audio frequencies. This may even be destructive at the extremes. I have heard that you may generate enough RF to be picked up by a radio tuner (maybe around 1 MHz).
I have had amps oscillating where there is no obvious audio artifact and been quite surprised to see the bad signal later on the 'scope. The load on the power supply will invoke additional distortions, but they may be subtle ("colourations") and you may need a distortion analyser to find these (even less likely to be available than an oscilloscope).

 

[leeperry]

so, any new contenders for the LT1363/64? I know mine won't get off the HD2 anytime soon. at some point I'll try the 1361/1469 I guess, eventually

 

[ecclesand]

Quote:

Originally Posted by murrays /img/forum/go_quote.gif Ok, the basic concept is this: Op amps by design have a small bias current flowing at the input. The current flows in the input circuit resistance resulting in a voltage. (Ohms law gives: voltage = resistance x current). This input voltage is further multiplied by the gain of the stage to appear at the output. Many circuits have capacitors to block the DC, so it is not a problem. However, removing capacitors deemed "unnecessary" is a mod that some folks do. So just beware of these possible consequences. It is easy to check for DC offset after making changes. milli, micro, nano and pico are all metric prefixes (as kilo & mega). All just different scales for measuring the same thing, current in Amperes. 2 uA = 2000000 pA. A good way is to make up a simple test adapter to use with your multimeter. Solder some wires or terminals onto a spare headphone plug to make it easier to attach the mutimeter probes. I have one with a pair of load resistors soldered on to it. A quick and dirty test for DC offset is to set the output to zero (i.e. stop playing any audio) and plug and unplug the headphones several times. If you hear clicking in the headphones, this will be the DC current switching off and on as the contact is made & broken. Of course, don't use your expensive phones for this! An oscilloscope is best, but not always available. Yes, I have heard whooshing (hiss). It may vary as you move the headphone cable around, or near to sources of interference. Also the sound may change as you move your hand near the metalwork of the amp (case, heatsinks, knobs, etc). Another indication is excessive heat or power drain - sometimes there is more power being used at RF than at audio frequencies. This may even be destructive at the extremes. I have heard that you may generate enough RF to be picked up by a radio tuner (maybe around 1 MHz). I have had amps oscillating where there is no obvious audio artifact and been quite surprised to see the bad signal later on the 'scope. The load on the power supply will invoke additional distortions, but they may be subtle ("colourations") and you may need a distortion analyser to find these (even less likely to be available than an oscilloscope).

Wow...great information. Thank you very much!!

 

[SpudHarris]

Quote:

Originally Posted by leeperry /img/forum/go_quote.gif so, any new contenders for the LT1363/64? I know mine won't get off the HD2 anytime soon. at some point I'll try the 1361/1469 I guess, eventually

Well I have soldered up the OPA132UA's that came in the mail today and all I can say is 'Oh my God!' these things are bloody fantastic. Andrea kept bangin' on about them so I thought I'd give them a try as the aren't uber expensive.

You guys must try them, they are just amazing. Impactful, fantastic airy detail and such depth. Of course applications etc will make a difference but my application is different to Andreas but I agree with everything he said of them.

My LT1363's came today also but I'm ****ting myself at the thought of killing my JH13's and not 100% sure how to check for DC Offset which judging by earlier posts there most probably is with LT1363's

 

[leeperry]

oh, yes OPA132UA! I have them too...I like the 2132P but it was maybe a bit too laid back to my taste, this one would be an improved version...but so far I'm still very unimpressed(in a good way) by the LT1364, as I said fantastic U87 voice recordings just sound so natural that it's scary.

talk about "disappearing headphones"

well, plug a voltmeter on a RCA cable and try to measure any DC voltage...if your board has lotsa caps, it's a good guess you won't read any anyway. a proper design won't let any DC offset get to your phones IMHO.

 

[SpudHarris]

I don't think I'll be taking out the OPA132's for a while, Ive ordered some more this evening to play around with some new Class A modules (resistors the size of a grain of salt on top of brown dogs).

I assume from an earlier post that I can just put a jack plug like a 3.5 Neutrik in the headphone out and measure from there.

 

[ROBSCIX]

I tried the dual channel 2132 but found them cold and boring... I am pretty sure I have some OPA132's in my parts bin. Will have to give them a try and hear how they sound.
Single channel version usually surpass thei dual channel siblings.