Calculating an Adjustment angle from the Residual Flux

[magnX]

Hello Gurus,

I am looking at an algorithm that reduces the inrush in a 3 phase power transformer by using the controller switching method of a static transfer switch between two sources. To compensate for the remnant flux Br, the algo computes an adjustment in degrees that eventually gets added to the firing angle to figure out the optimal firing moment.

Anyways, the part that I can’t understand is how you can go from Br to degrees?
In my understanding, if you know the B-H curve and you know the flux at the transfer time you can interpolate to get H (magnetic field) or I (current) but how can you get degrees?

I appreciate any help.

 

[Skogsgurra]

You need to think of the volt-seconds absorbed by the core.

If there's no Br, your best choice is to switch on on top of the waveform, or 90 degrees from zero crossing. A well designed transformer will then reach design flux density just before the sine hits zero in negative direction. The core than has a positive flux and the negative voltage starts working the flux towards zero. Zero flux is reached at 270 degrees and after that point, the flux grows in negative direction until 360 degrees.

Now, if you have a remnant flux, you have to adjust the switch-on point accordingly. If the flux is positive, you have to switch on later. Simply because you don't have that much flux left to utilize. If the flux is negative, you shall switch on earlier so that the flux is at zero when voltage is zero.

Mathematically, you see to it that the volt-seconds integral balances the available flux and you do that by calculating the integral of the sine voltage (which is a cosine plus an arbitrary constant) and adjust the switch-on so the resulting flux remains lower than the saturation flux.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[magnX]

Thank you very much for the answer.

I understand all that and I totally agree. As mentioned in my first post the algo that I am looking at does something that I still don’t get:

It builds a lookup table to come up with degrees vs flux, meaning it takes into account the flux(V-sec) at time the transfer is done and using the lookup table you get some degrees (positive or negative) in adjustments to fire either earlier or later as you mentioned.

The formula they use is adjustment (in degrees) = A* (flux)^2 +B * flux +/- offset depending on the polarity of the flux. The above equation looks to me like the B-H curve if plotted.

My questions is H or I linearly dependant on the angle? That way if you get H you can scale to get angle? How can you get an angle using the above?


Thx,