Op Amp Start up Time

bob.jo...@gmail.com

unread,

Dec 30, 2011, 1:24:44 PM12/30/11

to

How long does an op amp take to "startup"? I have a very
straightforward non-inverting amplifier to amplify an audio signal. It
works fine but when enabling power through a switch the audio takes
about 0.5 seconds to be heard. I have monitored the power to the op
amp as it uses a rail splitter and all the power comes on within 20ms.
The op amp is a very low power version(uses < 0.3mA) with 10uF IO
coupling.

I have monitored the output of the op amp and it is off when the power
is off, when the power comes on it it takes about 0.5 seconds before
it jumps to input voltage(with gain of course).

I'm using a TLE2426 and if I add too large of a capacitor for noise
reduction then the split rail takes about 1.5 seconds to come on. I
removed the capacitor and it comes on in about 20ms. The op amp itself
though still takes a while. I imagine this is an issue with low power
op amps(takes longer to charge the internal capacitance's?)?

Is there a parameter on the datasheet that specifies how long the op
amp takes to "start up" or a way to calculate it from the parameters?

Or is there some other thing possibly going on? I do see that the
input also takes a bit of time to ramp up but not as long as the
output of the op amp. This maybe due to the scope too as I am using AC
coupling. What I do know is that the audio output takes about 0.5
seconds for me to hear it when I power the circuit.

Joerg

unread,

Dec 30, 2011, 1:37:15 PM12/30/11

to

Bob.Jo...@gmail.com wrote:
> How long does an op amp take to "startup"? I have a very
> straightforward non-inverting amplifier to amplify an audio signal. It
> works fine but when enabling power through a switch the audio takes
> about 0.5 seconds to be heard. I have monitored the power to the op
> amp as it uses a rail splitter and all the power comes on within 20ms.
> The op amp is a very low power version(uses < 0.3mA) with 10uF IO
> coupling.
>

Ah, 10uF. Did you check the input? If you have a lot of gain it might
take a long time to get the input to where the opamp is in the linear
range because then every millivolt counts.

> I have monitored the output of the op amp and it is off when the power
> is off, when the power comes on it it takes about 0.5 seconds before
> it jumps to input voltage(with gain of course).
>
> I'm using a TLE2426 and if I add too large of a capacitor for noise
> reduction then the split rail takes about 1.5 seconds to come on. I
> removed the capacitor and it comes on in about 20ms. The op amp itself
> though still takes a while. I imagine this is an issue with low power
> op amps(takes longer to charge the internal capacitance's?)?
>
> Is there a parameter on the datasheet that specifies how long the op
> amp takes to "start up" or a way to calculate it from the parameters?
>
> Or is there some other thing possibly going on? I do see that the
> input also takes a bit of time to ramp up but not as long as the
> output of the op amp. This maybe due to the scope too as I am using AC
> coupling. What I do know is that the audio output takes about 0.5
> seconds for me to hear it when I power the circuit.
>

If you have a lot of gain the input needs to be within millivolts or
tens of millivolts of target value. That can take a long time if the
resistor values are high.

Simple test: Double the value of that 10uF cap. If that doubles the
turn-on time you know what causes it :-)

--
Regards, Joerg

http://www.analogconsultants.com/

Jamie

unread,

Dec 30, 2011, 2:04:37 PM12/30/11

to

Saturated inputs are always a pain on start up. Also damaging in some
cases.

Jamie

JOF

unread,

Dec 30, 2011, 4:32:28 PM12/30/11

to

It's probably the time required for the 10 uF coupling cap on the
input to charge to the split supply level. You might be able to
connect a diode from the split supply directly to the input to shorten
that time without affecting operation very much, since it will be
almost an open circuit when the voltage across it has been reduced to
close to zero by the operation of the circuit.

--
John

Joerg

unread,

Dec 30, 2011, 4:33:55 PM12/30/11

to

If a saturated input (without over-stressing parasistic substrate diodes
paths) causes an opamp to die upon power up or become damaged, then the
IC designer should be dunked into a moat until there is a vow to do
better next time :-)

Jamie

unread,

Dec 30, 2011, 5:53:08 PM12/30/11

to

I think it's cheaper to replace the chip and patch the problem! :)

Jamie

Joerg

unread,

Dec 30, 2011, 6:02:19 PM12/30/11

to

Or let a Rottweiler and a Shepherd loose on the team :-)

Did the pain ease up? We've got three big dogs but so far never had a
real bad fight happen. Had a Rottie mix until Thanksgiving 2010, when
his hind muscles finally gave out at age 13-1/2 and we had to put him to
sleep. I still miss him.

Tim Wescott

unread,

Dec 30, 2011, 6:22:57 PM12/30/11

to

I'm with Joerg on this: your issue is more likely with the circuit
surrounding the op-amp than with the amplifier itself.

Post a schematic somewhere? And maybe tell us _why_ you need it to work
faster, and _how short_ a time is acceptable?

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com

Phil Allison

unread,

Dec 30, 2011, 6:41:37 PM12/30/11

to

"Bob.Jo...@gmail.com"

>
> Or is there some other thing possibly going on? I do see that the
> input also takes a bit of time to ramp up but not as long as the
> output of the op amp. This maybe due to the scope too as I am using AC
> coupling. What I do know is that the audio output takes about 0.5
> seconds for me to hear it when I power the circuit.
>

** Reminds me of the slow start up exhibited by most condenser microphones -
the kind that use a 60 volt polarising supply.
Circuit resistance is about 2Gohms and total capacitance to be charged is
about 200pF plus it takes several time constants to settle.

So about 2 seconds in all to get full output level.

..... Phil

Jamie

unread,

Dec 30, 2011, 8:52:08 PM12/30/11

to

We have some one at work that had to do that a couple of days ago,
they had a rescue rottie that was up there in age to start with, she had
that dog for at least 5 years. The other day she tells me that he is bad
and could not get up from the floor, peed on the floor then got scared
and try to clean it up because he knew he did wrong.

He was 15 years old. She loved him like one of her own kids. actually
that dog got treated better than her husband and he'll confess to that :)

As for my dog, yes, my arm is still a little sore, it is all black
and blue. it'll need to go through a healing process but I am sure I'll
make it :) I'll just need to watch her teeth a little closer next time
she goes for that full size basket ball.

Jamie