Load modelling: Using Resistor or current sink as a load?

[vieha007Electronic]

I am confusing what is the difference in modeling a load (unknown load). Some simulations model load as a pure resistor while the others use a current sink. I would like to know what is the difference between them?

P/S: For example, Moura (in his famous LDO book) recommended that in order to check the stability of the LDO: use both of resistor and current sink as the load for the simulation as 2 extremes.

I am wondering why?

Appreciate your comments.

Sincerely,

 

[crutschow]

A simulated current source has an infinite resistance so it causes a different phase-shift in the feedback loop as compared to a low resistance load, thus both are used for simulation purposes to check the feedback loop stability at the extremes of output load resistance.

 

[vieha007Electronic]

A simulated current source has an infinite resistance so it causes a different phase-shift in the feedback loop as compared to a low resistance load, thus both are used for simulation purposes to check the feedback loop stability at the extremes of output load resistance.

Thank for your quick reply. But I am still wondering why.

For example for checking stability of my LDO (targeted Iout is: 1mA to 1A, Vout_LDO = 2V), this means the Rload is equivalent to Rload(min) = 2Ohms to Rload(max) = 2kOhms as 2 extremes. But when checking with 2 current sinks (1mA and 1A, respectively) as 2 extremes, the results from R and current sink are totally different.

Could you give me more inside, please?

Thank you!

 

[vieha007Electronic]

I am still wondering why, for example, if we use R and current sink as 2 extremes, why don't we just use R with min and max values for checking stability?

Also, should I understand that with iload min, and max, we have to use both of R and curent sink for each case respectively?
That means:
Case 1: iLoad min: Rmax and current sink = iLoad_min
Case 2: iLoad max: Rmin and current sink = iLoad_max

Thank you

 

[KlausST]

Hi,

It seems you think in DC only.

For sure a constant current sink of 100mA connected to 5V - in terms of DC - is the same as connecting a 50 Ohms load.

But "stability" means AC stability.
And here is the difference: A current sink/source has infinite AC impedance. Although it is drawing DC current.

Imagine you have 5V DC and a ringing of +/-100mV (ringing cuased bay anyy error influence). So the output voltage is somewhere between 4.9V and 5.1V.
With a resistive load of 50Ohms you get an output current range of 98mA ... 102mA (it changes by 4mA, and these 4mA are damping the ringing)
But with a current sink the current is constant... no current here to damp the ringing.

To avoid such situations some circuits use a "zobel" circuit. Basically an RC series combination. Here just imagine a 100R / 1uF combination. It draws no DC current (no power loss) but it generates AC current to dampen oscillations.

Klaus

 

[vieha007Electronic]

Dear Mr. KlausST,

Great! I understand the problem now. Really appreciate!

Thank you very much!