Solution for compensating circuits with op-amps

[goldsmith]

Hi my friends!
I have this circuit and i want to compensate it . what should i do?


With appreciate
Goldsmith

 

[LvW]

Re: Compensating network

Hi my friends!
I have this circuit and i want to compensate it . what should i do?
With appreciate
Goldsmith

Hi Goldsmith, I am afraid, an answer is not possible without knowing your requirements.
What is the purpose of the circuit - and which parts are essential and may not be removed or altered?
(For example, can the LC circuitry be modified or even removed? )
LvW

 

[goldsmith]

Re: Compensating network

Dear LvW
Hi
Thanks for your reply.
My mean is that now , my circuit is near to oscillation! i want move its poles to the stability region. i think it is possible with adding a compensating network to the amplifier. is it right? if yes , what should i do?
Thanks and appreciate.
Goldsmith

 

[LvW]

Re: Compensating network

Dear Goldsmith,

Question: Is it necessary that the amplifier is inverting?
If a non-inverting unit would be allowed you could desgn it as a phase-lead circuit thereby improving the stability margin.

---------- Post added at 12:32 ---------- Previous post was at 12:03 ----------

As an alternative, you can try to give the inverting amplifier a kind of phase-lead behaviour by adding two capacitors C8 and C9 in parallel to R8 and R9.
Because both resistors are equal C8 should be somewhat larger than C9. The design should be controlled with a corresponding BODE plot.
Both time constants should be approx. in the order of 1/wco (wco=loop gain cross-over frequency).

 

[goldsmith]

Re: Compensating network

Dear LvW
Thanks for your attention.
Thanks for your reply.
I can use a non inverting amplifier . how can i improve it as you said?
And how can i find the value of c8 and c9 specifically? is it possible that you describe its way , please?
With appreciate
Goldsmith

 

[LvW]

Re: Compensating network

Hi Goldsmith,

before designing a phase-lead network for the opamp you should convince yourself that the opamp (resp. its poor phase margin) really is the source of the trouble.
Therefore, at first you should compare the resonant frequency of the LC circuit with the decaying oscillations that you observe.
Perhaps there is the problem.
On the other hand - is it really a problem? If you inject a pulse into a resonant network you shouldn't be surprised that there is a kind of oscillatory behaviour.

---------- Post added at 19:22 ---------- Previous post was at 18:51 ----------

Hi Goldsmith,

in the mean time I am pretty sure that the oscillatory behaviour is caused by the LC network and not by the opamp.
Thus, there is no need for any "compensation" of the opamp as desired from your side.
I do not know your requirements (see my first answer to your posting) but a damping increase of the LC network certainly will reduce the transients (e.g. increase R6)

 

[FvM]

Re: Compensating network

in the mean time I am pretty sure that the oscillatory behaviour is caused by the LC network and not by the opamp.
Thus, there is no need for any "compensation" of the opamp as desired from your side.
I do not know your requirements (see my first answer to your posting) but a damping increase of the LC network certainly will reduce the transients (e.g. increase R6)

You hit the point. The inverting amplifier will be stable, although a small feedback capacitor may be wanted to compensate the amplifier input capacitance and increase the OP feedback phase margin, as suggested in the LM118 datasheet.

A high Q LC circuit can be compensated - at least in theory. practically, you'll have difficulties to hit it's parameters accurately.

The purpose of the LC circuit however isn't clear.

 

[goldsmith]

Re: Compensating network

Dear LvW and FvM
Hi
Thank you for your replies.
I want to learn a compensating in real example and i designed this circuit for this behavior . but you are right. i just want to learn that how can i compensate the circuits that has some op amps. i couldn't find this , in some books that i had read them.
I know about network function and pole and zero but i can't use them in my circuits. Is it possible that you guide me , please?
Thanks in advance
Goldsmith

 

[LvW]

Re: Compensating network

Hello Goldsmith,

at first, the term "compensation" has several different meanings. For example, you can compensate looses with gain or the influence of a damping (positive) resistor with a negative one.
I suppose, you are referring specifically to the opamp open-loop gain and its frequency response that can be modified such that - in case of feedback - the circuit always is stable. This is called "frequency compensation". Without such compensation the opamp gain-vs-frequency curve shows at least two (normally: three) pole frequencies, which may lead to instability after applying feedback. Then, the result of compensating is a so called "single-pole" response with a gain first-order gain roll-off (-20 dB/decade) within the active region (gain>0 dB).
In most cases, this frequency compensation is achieved internally within the opamp unit, however, there are also methods to use external RC networks for such a compensation. If you need some additional and more detailed information I suggest to "google" for keywords like "frequency compensation" or "unity gain stability".
Regards
LvW

 

[FvM]

Re: Compensating network

If we look at the ringing circuit pulse response shown above, we can't see if it's caused by a intentional designed feature, e.g. a LC resonator or more generally, a high Q filter, or if it's an unwanted effect, e.g. caused by insufficient phase margin of a feedback amplifier.

The given cicrcuit is mixing both topics in an inconvenient way. To evaluate frequency characteristics of OP circuits, we should remove additional LC filters.