paralleling op amps

Do you have a question? Post it now! No Registration Necessary

Threaded View

I've seen a few web sites where better performance
into low R loads is achieved by paralleling op amps
together. Has anyone tried this?, I've tried some
TI NE5532 op amps: they run hot but seem to work.


Re: paralleling op amps

"Mark Harriss"   =  lying, PITA, criminal, artistic fool from Cairns
Quoted text here. Click to load it

** Like this one:

Post the damn links   -   you lazy,  damnable  ASS    !!!!!!

Got a clue what an op-amp " voltage follower "  is ??????

Yours wired that way ?

Quoted text here. Click to load it

**  Hot with a load or at idle???

 Idle dissipation will be about 300mW ( ie 10mA x 30 volts ).

Under full load  ( 7 volts rms into 220 ohms)  add  another 300mW at least.

600 mW in an 8 pin DIL is hot.

...........  Phil

Re: paralleling op amps

Use a more suitable opamp.
There are many high-drive opamps on the market. Paralelling them will not work,

Try the TI website

-Andrew M

Quoted text here. Click to load it

Re: paralleling op amps

Quoted text here. Click to load it

You can parallel opamp voltage followers to reduce the total noise
floor, but I've never seen it done for higher output drive or any other
purpose. Just use a suitably rated opamp for driving the load required,
there are plenty of ones designed for driving 50ohm loads and are
stable into capacitive loads etc. Or if your drive needs are greater
you would typically use an external tranisistor driver.

Dave :)

Re: paralleling op amps

Quoted text here. Click to load it

It's mainly to reduce distortion into a given load, rather than
increased drive capability. If you scroll down to the bottom of
this page you can see what I'm after. The NE5532 gives good low
distortion and I would like to minimise it as much as possible.


Re: paralleling op amps

Re: paralleling op amps

Quoted text here. Click to load it

You shouldn't go just paralleling all of the pins, that would not be good.
You could make two complete amplifiers, or voltage followers, or whatever
it is, and then connect each of these to the load through a small resistor.
You need to think what will happen if the DC offset of the two op-amps is
not equal, in this case they will tend to fight each other, and the output
resistors will help to prevent the currents due to this 'fighting' from
being too high.  Each op-amp must have its own feedback, taken from its own
output pin, and then a resistor from the output pin to the load, then you
should be OK.  If you try to share the feedback connection, it will be
impossible for all of the op-amps to be happy with their feedback
(inverting input) voltages at the same time since they have slightly
different offset errors, and so they will fight.  The feedback needs to be
taken from the output pin of each op-amp, not from the load, because the
different op-amps will never all be happy with the load voltage, but it is
possible for them all to be happy with their own individual output pin
voltages, which may differ from each other by some millivolts. The size of
the series resistors will need to be larger if the output referred DC
offset of the two op-amps is more different, so use precision resistors.  I
would go for the largest series resistors that are acceptable in your

Unless you're driving a really low load impedance, where you are sure that
the output current IS causing distortion, it all sounds like more trouble
than it's worth.  The only exception I can think of is if you have unity
gain buffers driving cable through series resistors, where it wouldn't be
much of an extra pain to do it.    If you are driving cables then it is
better to have some series resistance anyway so that it won't be close to


Site Timeline