Wye and delta transformers

Status
Not open for further replies.
GoldDigger said:
Yes, it would, and given a difference in inductance of a factor of ten or more when the core is fully saturated, I suspect that it is possible.
And, of course, at full load the higher idling current would be adding (in quadrature) to the load current.
Also, the circulating currents if you connect the input wye point can easily be several times the full load current.

Both effects are bad.
If you apply, for example, 208 volts to one 120v wye winding the delta secondary will look a lot like a short circuit and so the resistive component of the idling current will be far greater than even the increased magnetizing current.
The delta winding will, in effect be trying to have L to L voltages of 480, 480, and 600. That triangle does not add up!
Click to expand...

Isnt the high leg of a high leg delta that way because of the neutral being center tapped on the opposite two phases? If there is no neutral on the primary side, does a 240V high leg delta really differ from a straight 240V delta or even a theoretical 240/138Y service? L-L are all 240V.
 
JFletcher said:
Isnt the high leg of a high leg delta that way because of the neutral being center tapped on the opposite two phases? If there is no neutral on the primary side, does a 240V high leg delta really differ from a straight 240V delta or even a theoretical 240/138Y service? L-L are all 240V.
Click to expand...

Correct. All this discussion now is about what would happen if you were to bring four wires to the low side of the transformer and land the neutral on the XO. Why would someone want to do that? IDK. I'm not worried about it.
 
GoldDigger said:
The other way to look at it is that the power dissipation in the transformer is a function only of the currents in primary and secondary windings.
The voltage that produces those currents is nice to know, but is irrelevant to calculating the power dissipation.
The heating will be the sum of the heat from both windings, but even with the secondary open it will not take more than a 40% excess of current in the primary to drive the transformer to its full heat limit.
Click to expand...

But it's not really. The RESISTIVE heating in the windings themselves is a function only of the currents in the windings. But a lot of the heat produced by a transformer is from magnetic losses and eddy currents in the core, not resistive losses in the windings. Core heating is not closely related to current - it's much more closely related to voltage (since at a fixed frequency, it's peak-to-peak voltage that determines magnetic losses and saturation).
 
mpoulton said:
But it's not really. The RESISTIVE heating in the windings themselves is a function only of the currents in the windings. But a lot of the heat produced by a transformer is from magnetic losses and eddy currents in the core, not resistive losses in the windings. Core heating is not closely related to current - it's much more closely related to voltage (since at a fixed frequency, it's peak-to-peak voltage that determines magnetic losses and saturation).
Click to expand...
On the contrary, the inductor and core have no way of knowing directly what the applied voltage is. That applied voltage will determine both the magnitude and waveform of the current and the current is what produces the magnetic field and in turns causes core saturation.
I agree 100% that just knowing the peak, average, or RMS value of the current is not enough. And it is true that knowing that the applied voltage is a sinusoidal waveform of a particular magnitude does tell you what the current will be, and that core losses (from hysteresis rather than saturation AFAIK) adds to the I2R heating. Hopefully in a good transformer design the eddy currents will be minimized, and will again depend mostly on the current waveform that results.
I suspect that if we continue this line of discussion it will end up being like whether it is voltage or current that electrocutes you. :)

mobile
 
Last edited:
GoldDigger said:
On the contrary, the inductor and core have no way of knowing what the applied voltage is. That applied voltage will determine both the magnitude and waveform of the current and the current is what produces the magnetic field and in turns causes core saturation.
I agree 100% that just knowing the peak, average, or RMS value of the current is not enough. And it is true that knowing that the applied voltage is a sinusoidal waveform of a particular magnitude does tell you what the current will be, and that core losses (from hysteresis rather than saturation AFAIK) adds to the I2R heating. Hopefully in a good transformer design the eddy currents will be minimized, and will again depend mostly on the current waveform that results.
I suspect that if we continue this line of discussion it will end up being like whether it is voltage or current that electrocutes you. :)
Click to expand...


If we're talking only about an unloaded transformer, then I would agree that current is just as good a metric as voltage since both are directly linked mathematically (at a fixed frequency).


...And we all know the only right answer is that current, not voltage, is what kills you. :thumbsup:
 
Status
Not open for further replies.
Top