hai frends please help me

what do u mean by stability of a system? Is means keeping the gain of the system constant or increase the gain with time? If your answer is former then please explain me why the open loop control system is said to be stable compared to closed loop control system.

Anjalikrishna1985 said:

what do u mean by stability of a system? Is means keeping the gain of the system constant or increase the gain with time? If your answer is former then please explain me why the open loop control system is said to be stable compared to closed loop control system.

Click to expand...

Stablity can mean many things. Usually its keeping a Uniform *something* over a period of time, temperature, or other range of things.

Example; A Amp that is said to have High stablity at high temps means that at high Temperatures the amp would Normally start cutting the signal, However because it has high stablity at High temps, You can let the Amp get hot and still get the same performance as if it were cool.

but my second question is under doubt ,why we are calling open loop control system a stable one than a closed loop control system

Normally when we talk about the stability of an amplifier we're talking about how prone it is to oscillation.

https://en.wikipedia.org/wiki/Frequency_compensation

sir
ur just deviating frm my doubt which one is more stable open loop control system or closed loop control system?if ur answer is open loop system then why it is so?

It depends on the circuit and on its wiring layout whether the circuit is stable with or without negative feedback.

An LM393 comparator has a high input impedance and a voltage gain of 200,000. If the output gets near its non-inverting input then it oscillates due to the capacitive-coupling through the air. The comparator doesn't have frequency compensation so it also oscillates if it has negative feedback.

An LM358 opamp is nearly the same as the comparator because it also has a high input impedance and a gain of 200,000. If the output gets near its non-inverting input then it also oscillates due to the capacitive-coupling through the air. It has frequency compensation so it is stable with negative feedback.

He isn't talking about that audioguru.

Open loop control is without feedback, for example using a lamp dimmer circuit to control the temperature of a soldering iron.

Closed loop control has negitive feedback, for example using an electronic controller with a thermocouple to control the temperature of a soldering iron.

A properly designed closed loop system is generally more stable since negitive feedback reduces the effect of any changes.

An open loop temperature controller will allow the soldering iron tip to cool down when you clean it with a sponge whilst a closed system will try to stop this from happening my supplying more power to the element.

a closed loop system is always stable IF the feedback provided is negative.

if under any condition the feedback becomes positive, then as long as the loop gain is less than unity, the system will remain stable.
when loop gain becomes equal to 1,the system starts to oscillate.this is the beginning of unstability.
if loop gain becomes greater than unity, the oscillations occur with increasing magnitude,until the system breaks down or some other mechanism brings down the loop gain to 1 or less than 1.

The second question may be best answered using mathematics. Unfortunately, having been out of university for too long, I am no longer competent in this, so one of the others will have to chime in to help. But as I recall, an open loop system is one where the elements have been cascaded without any feedback, whereas a closed loop system includes feedback.

To best understand the stability issue, you can resort to mathematics to see that a feedback system has the potential to oscillate whereas an open loop system does not. It is this potential that causes people to claim that a closed loop system may be less stable, and in this case they consider "stability" to mean lack of oscillation. (stability may also refer to how steady a circuit performs versus temperature, and it may also refer to the propensity of a microcontroller to become insane as a result of poor programming). The advantages of feedback in a control system, however, are so great that feedback is universally used in such systems. The potential to oscillate must be dealt with by the system designer. Oscillation can be avoided by carefully insuring that the feedback cannot become positive at any frequencies at which the system has gain. Common practices include adding damping, by restricting the bandwidth, by keeping gain to the minimum necessary, by avoiding unintentional feedback paths, and so on. Practical applications of how to stabilize a feedback system usually depend on the kind of system and are typically not for the beginner.

Stability is not keeping the gain constant or anything like that. It's how the output of the system might go to infinity or oscillate to a certain input. Although not exactly the same thing, you can sort of compare it to how exciting a vibration system at its resonant frequency will cause the amplitude of the vibrations to go towards infinity (like wind blowing on a bridge, or pushing a child on a swing, you have to push in synchronization with the swinging to increase amplitude).

Open-loop systems are ALWAYS stable because they have no feedback from output to input, and so the input can't really "build up on itself" like a resonant frequency.

CLosed-loop system CAN BE UNSTABLE but DON'T HAVE TO BE UNSTABLE because they have feedback and this can happen. Properly designed closed-loop systems are stable, at least within a range of interest.

A lot of this is really simplified so don't take all this as exact. You need to take a control course or Google up some more stuff to fully understand. These are just to kind of give a mental picture of what is going on. If you really want a full answer you have to go into the math in a control systems course. (This also means that if this is for homework you will need to do some actual research).

dknguyen,
u told "Open-loop systems are ALWAYS stable" then explain me what is the use of negative feed back in amplifier with transistor in CE configuration ? i heard that this closed loop one is more stable than open loop one

It depends on what you mean by stable.

If you're talking about the system non oscillating then open loop systems are always stable.

If you're talking about the system not being affected by changes in gain then closed loop systems are more stable.

Also positive feedback doesn't always mean oscillation, it's often used to prevent ocillation, for example if used with a compatator it adds hysteresis which will stop it operating in its linear region.

Here's an example: A low voltage disconnect for an inverter.

Without positive feedback the inverter might start to turn of at 9.999V then when te voltage creeps back up to 10 (because the battery voltage is higher without a load) it turns back on again causing the voltage to drop; the cycle repeats causing oscillation. Adding a feedback resistor on the comparator might cause it to turn off at 10V but require 12V for it to turn back on again which will require the battery to be recharged.

http://www.ecircuitcenter.com/Circuits/op_comp/op_comp.htm

Anjalikrishna1985 said:

dknguyen,
what is the use of negative feed back in amplifier with transistor in CE configuration ? i heard that this closed loop one is more stable than open loop one

Click to expand...

in ce configuration, -ve feedback can be provided by adding a resistor from emitter to ground. this makes the amplifier more stable to temperature variation.for example,due to rise in temperature, the emitter current may increase.this causes a larger drop across the feedback resistance.this in effect reduces the base-emitter voltage, bringing down the emitter current.
but the only problem is that if this -ve feedback exists for the ac signal as well, gain is reduced significantly(while it improves input and output impedances).hence the normal practice is to bypass this resistance using a suitable capacitor so that emitter is virtually at ground with respect to the ac signal,while -ve feedback exists for the quiescent emitter current(dc).

my tiny brain,
Close lope system is more stable than open loop,
why
because in close loop system for example we want to maintain the temperature of a water in tank. in close loop our sensing element countinously measure the process variable(Temperature) and give feedbak to the processor, processor compare this measured value with our desired set point(our desired temperature in this example), then it operates the final control element (may be heater or cooler) according to the difference of both(measured temperature and our desired temperature)values. System remains stable because it countinously detect the changes in teamperature and take action according to the error(difference between measured temperature and our desired temperature)

But in open loop,
there is no feed back from the sensor to the processor and it is difficult to mantain the temperature of water.

my poor english.

Ayne said:

my tiny brain,
Close lope system is more stable than open loop,
why
because in close loop system for example we want to maintain the temperature of a water in tank. in close loop our sensing element countinously measure the process variable(Temperature) and give feedbak to the processor, processor compare this measured value with our desired set point(our desired temperature in this example), then it operates the final control element (may be heater or cooler) according to the difference of both(measured temperature and our desired temperature)values. System remains stable because it countinously detect the changes in teamperature and take action according to the error(difference between measured temperature and our desired temperature)

But in open loop,
there is no feed back from the sensor to the processor and it is difficult to mantain the temperature of water.

my poor english.

Click to expand...

Different kind of stable. You are talking about the the "stability" of steady state output (or the steady state error), whereas I'm talking about the loop stability where the system goes haywire when it's unstable. IN control system, the stabilty I am talking about comes first since, obviously...if the system starts oscillating to infinite it's pointless how small the steady state error is.

Your example of the hot water with the feedback is only true if the closed loop system is properly designed. If you designed the closed loop system incorrectly, the temperature of the water could keep on increasing towards infinity or oscillate a lot- just behaves very strangely. Although it's hard to do this since temperature chances so slowly. IN a system where the changes are fast it's a lot easier to run intot his problem. In an open loop system, no matter how you design it, it wouldn't increase the temperature of the water to infinity (unless you specificaly designed it to do that) and the temperature of the water would not oscillate between enormous numbers.

If you don't speak English and you take a control system course, there may be another english equivelant word you use for the stabiltiy I'm talking about. it's probably some translation issues (or you didn't take a control system course yet).

hai guys
i had one doubt .what do u mean by "oscillating output"?is this means output resembles to a sinusoidal wave all the time or it means the output shouldnt have a stable state? again by mathematics we will find a system stable or not by analysisng system by various techniques like ROC of z transform , region occupied by system in S plane in laplace forms ..... etc.my question is by these techniques stability on which parameters we are finding.(i mean parameters= temperature ,gain........)

Yeah, what you are saying is what I am trying to get at (very simply anyways). It's like if you build an amplifier with feedback. If you build it wrong (which I am told happens all the time), then the output looks like an oscillator. The output oscillates (or oscillates to infinite) when it is not supposed to. An oscillator is supposed to oscillate though when you build it correctly (but not to infinite).

The analysis methods you're talking about (I don't remember all of this since I forgot it as soon as I learned it), but temperature isn't one of them since it's unique to what you are doing. I think the parameters are gain, bandwidth, , dampening, gain margin, phase margin, settling time, and steady state error. Then there's step response and a few other things I don't remember. But stability is dependent on multiple parameters working together though. It's most liklely that all (or all but one) of these parameters affect stability. You can't just say "if this single parameter is within this range, then the system is stable".

sir
then which are parameters should be stable for a system to be stable.

read the above post . "....think the parameters are gain, bandwidth, , dampening, gain margin, phase margin, settling time, and steady state error...."
and "....You can't just say if this single parameter is within this range, then the system is stable..."

then which are parameters should be stable for a system to be stable.

Click to expand...

it depends on the system under examination , for eg : for an oscillator the main params are the gain(Beta) and the +ve feedback factor (A)

Anjalikrishna1985 said:

sir
then which are parameters should be stable for a system to be stable.

Click to expand...

A fast car isn't fast because the wheels are fast, or the engine is fast, or the transmission is fast, and not any one of these being right will make the car fast. They all have to work together indirectly to make the system stable.

In the same way a system is not stable because the dampening is stable, or the gain is stable, or the bandwidth is stable, and not any one of these being right will make the system stable. They all have to work together indirectly to make the system stable.