EmbDev.net

Forum: Analog Circuits op amp driving capacitive load


Author: Thomas (Guest)
Posted on:

Rate this post
0 useful
not useful
Hi guys,
I read that when an op amp is driving a capacitive load, its transfer 
function is disturbed, a new pole is added into it which reduces the 
phase margin. This leads to instability at the output voltage of the op 
amp.
Now I'm wondering if this problem occurs only when the input voltage of 
the op amp is AC, or also for DC?
I have an op amp (Voltage follower), which has a cap of 10uF in fornt of 
it.
Do you think this cap will influence its stability even if the input 
voltage is DC, and I need to use a compensation method?
Thx

Author: Cubus (Guest)
Posted on:

Rate this post
0 useful
not useful
If you have a stability problem, you will see oscillations at the 
output. The type of input voltage doesn't matter. Or your settling 
behaviour is bad when applying a step at the input. In the datasheet 
there should be a value for maximum capacitive load. The op amp can 
handle a greater load than written in the datasheet when putting a 
resitor in series to the capacitance. But then the maximum output 
current and slew rate are reduced.
You will find opamps whitch can drive unlimited capacitive load, but 
then there could be worse values for offset voltage, 1/f-noise etc. You 
should compare.

Author: Ina (Guest)
Posted on:

Rate this post
0 useful
not useful
>Do you think this cap will influence its stability even if the input
>voltage is DC, and I need to use a compensation method?

Yes, of course. There's always enough noise within the circuit to make 
the OPamp finally run into wild oscillations. And do not forget the 
voltage step when turning on the supply voltage.

Remember, there are some OPamps which can directly drive such high loads 
though.

Author: Thomas (Guest)
Posted on:

Rate this post
0 useful
not useful
Thank you for your answer.
I read a lot of technical papers in the net, explaining what happens 
when an op amp is driving a cap load. But I'm not sure if this problem 
occcurs only when the input is AC, or regardless of the input... So when 
we talk about bause diagramm for phase and amplitude, and phase margin 
etc... which frequency are we talking about, isn'it the freq of the 
inout signal to the op amp?
Thx

Author: Thomas (Guest)
Posted on:
Attached files:

Rate this post
0 useful
not useful
I'm using TLV4110 op amp.
I'm using it as voltage follower.
In the data sheet I found the following graph. Could you help me 
interpret it?
As far as I know, the bigger the phase margin, the more stable is the 
output of the op amp.
On the other side, according to this graph, the stability depends on the 
output resistive load too...

Author: Cubus (Guest)
Posted on:

Rate this post
0 useful
not useful
In your circuit is always noise and noise is broadband (as Ina said). So 
you will never have only a dc input voltage. And  even your supply isn't 
clean and constant. It includes noise and ripples from ac input or even 
peaks from switching power supply.

Author: Ina (Guest)
Posted on:

Rate this post
0 useful
not useful
>In the data sheet I found the following graph. Could you help me
>interpret it?

It just states, that you shall insert a series resistor of at least 20R, 
if you intend to drive large capacitive loads.

Tell us in more detail what you plan to do...

Author: Cubus (Guest)
Posted on:

Rate this post
0 useful
not useful
You can read from the diagramm:
Load resistance has a small influence on phase margin too. But the load 
capacitance is dominant. For higher values you have to put a resistor in 
series (RNull). Have you had a look at the application information at 
page 12? There's even given a value. You can try.

Author: Thomas (Guest)
Posted on:
Attached files:

Rate this post
0 useful
not useful
Guys,
thank you very much for all your answers.
What I'm intending to do is, to measure the current consumption of a 
uprocessor. So I have to take care abbout the voltage regulation of it 
too.
So now I assume that my uP (Load) is my resistor R2.
The current consumed by R2 s the same one flowing through R1 (shunt).
Later I will amplify the voltage drop across R1 to know how much current 
is flowing.
So The caps C1 and C2 are decoupling caps, needed by the uP.

In my testing board, I used a battery at Vg1. So its voltage is pretty 
much stable.
At the output of the op amp, the voltage is also pretty much stable, BUT 
the voltage drop across R1 is not stable!!!
Unfortunately I can't use a series resistor of 20 Ohm, as stated in the 
data sheet, because imagine when my uP is consuming 100mA, I will have a 
voltage drop of 2V across that resistor, and that is too much!

Author: Thomas (Guest)
Posted on:
Attached files:

Rate this post
0 useful
not useful
In 
http://www.analog.com/library/analogdialogue/archi... 
some compensation methods are proposed.
Let's concentrate about Snubber Network which seems to be a very well 
known method, It is said that the values of Cs and Rs are determined. 
based on the freq of the signal.
I attached the sweep of the voltage across the shunt resistor in the 
feedback loop. The signal is periodic, and has a period of 133kHz.

Author: Ina (Guest)
Posted on:

Rate this post
0 useful
not useful
>What I'm intending to do is, to measure the current consumption of a
>uprocessor. So I have to take care abbout the voltage regulation of it
>too.

The approach you follow is highly unsuited. There are way simpler 
circuits to accomplish that. Measure the voltage drop across the shunt 
resistor by using a "high side current monitor", for instance. Google 
for that.

Reply

Entering an e-mail address is optional. If you want to receive reply notifications by e-mail, please log in.

Rules — please read before posting

  • Post long source code as attachment, not in the text
  • Posting advertisements is forbidden.

Formatting options

  • [c]C code[/c]
  • [avrasm]AVR assembler code[/avrasm]
  • [code]code in other languages, ASCII drawings[/code]
  • [math]formula (LaTeX syntax)[/math]




Bild automatisch verkleinern, falls nötig
Note: the original post is older than 6 months. Please don't ask any new questions in this thread, but start a new one.