Dear Sir,

It is a good practice to use a PLL to maintain synchronicity in grid modelling. If your system is limited in operation between 49.9 to 50.1Hz, the system will remain stable by itself.

Hi Sachin,

You can try - at least on simulators (where there is no practical risk of blowing up things :-)...

But on a more serious note, you are running a risk - it may or may not work depending on the connected apparatus, and the exact frequency variation.

(Difficult to describe in greater detail on an internet posting !)

PLL gives the information about grid angle  based on that you can program your current to inject desired power in grid through VSI.

It is mandatory to know the grid angle position in order to control the active reactive power flow.

In case you generate angle based on grid frequency you don't know the phase difference between your injected current and grid voltage that will cause uncontrollable power flow.

in grid connected operation mode, always you have to use PLL synchronization system. PLL estimate the grid voltage angle which is essentially used to generate the distributed generator voltage. for every frequency oscillation we need PLL. A deep study has to be establish to robustness to the control loop. 

Stratum 1 clock synced PLL's do the job. 

At one time 40 yrs ago in Manitoba, power stabilty could be compared with OCXO or VLF with 1e-10 stability 

Hi all,

To make my question clear. Lets say we have a PV-Inverter where in the aim is to inject the current into grid at unity power factor. Assuming grid is not distorted and at fixed frequency.

If grid voltage is v(t)=Vsin(wt)

if we estimate V and then divide v(t) by V,  the template of grid voltage is obtained.

now if we inject the current into grid with the template ie sin(wt) obtained using hysteresis controller, we will have unity power factor current injection.

So where would the need for PLL be?.

Hi Sachin, your assumption of grid is not distorted and at fixed frequency is eliminating all the factors which are to be considered and which have been discussed earlier. Then your way of estimating will work.

If the grid frequency of the system and voltage varies..how will you proceed? 

How will the estimation work with varying conditions.

Essentially a PF tracking generator is an Injection Locked Loop (ILL), just like the old TV horizontal sweep voltage generators would also adjust vertical sweep to lock in within a certain tuning range.  So essentially it behaves like a type 1 PLL mixer with resonant generator..

QUestions may arise for power interruption and synchronization for various durations on output suppression and re-sync power output and surge withstanding as well as short circuit fault detection/recovery..

Also what is your generator response to an online automatic tap change transient voltage change from a Distribution transformer? 

Mr Somesh,

I am not trying to estimate grid frequency/angle which is not required once we have template. Secondly unit template is not a variant of PLL ,(PLL is a closed loops system while unit template estimation is open loop).

Please clarify the point you want to make regarding the question of necessity of PLL in case of undistorted and frequency invariant grid.