Reverse fed xfm

[Cold Fusion]

Years ago I was told it had something to do with how the feeding Y transformer is reacting with the load Y transformer, if the feeding transformer is out of balance with voltage then the load transformer tries to balance the feeding transformer, basically both windings are in parallel, so it kind of made sense?

I'm still working on this one.

I'm thinking it has to do with the same reasoning the utilities ocassionally use a Y-Y transformer and the primary Ho is connected to the utiity MGN. I have never understood what that gets them.

cf

 

[steve66]

The physics is foggy. I'm not understanding all I know - yet.

cf

I'm with you CF, I haven't ever seen an exact explination either, so I think its as foggy to me as you. It would be nice to see a diagram with some current flows that explain this.

Or maybe we can figure it out. It might help to look at this backwards. Ignore the neutral connection for now - assume it is unconnected. And assume there is a slight unbalance on the input voltages (maybe one phase is only 200 volts, and the other 2 are 208 volts.)

With a balanced input, the sum of the voltages around the Delta equal zero. No load - no current flow.

Now with our unbalanced input, the voltage around the delta should be some non-zero voltage. So we would expect some circulating current is needed to cause a voltage drop to make the sum of the voltages around the delta zero.

So, to me, there should always be circulating currents in a wye to delta xformer with an unbalanced input.

So I think the real question is how unhooking the neutral on the input prevents the circulating currents?

Somehow, I think the answer has to do with the fact that with an unconnected neutral, all the line currents have to add up to zero.

Does that get us halfway to an answer??

Steve

 

[steve66]

I think I've got it. If you draw the primary "Y" with an open neutral, and draw currents into each of the windings, you will see the sum of the currents have to be equal to zero.

Since the voltages across the secondaries are some constant times the currents in the primary, we can write an equation for the voltage around the delta. (Draw the delta open at one point, and write an equation for the voltage across that open. We want the voltage across that to be zero so there won't be any circulating currents when we close the delta.)

I think we can show that the voltage across that open is zero as long as the primary neutral is unconnected.

Once we connect the neutral at the primary, I think we can show that the voltage won't be zero if there is some current through the neutral connection.

Steve

 

[Cold Fusion]

I think I've got it. If you draw the primary "Y" with an open neutral, and draw currents into each of the windings, you will see the sum of the currents have to be equal to zero.

Since the voltages across the secondaries are some constant times the currents in the primary, we can write an equation for the voltage around the delta. (Draw the delta open at one point, and write an equation for the voltage across that open. We want the voltage across that to be zero so there won't be any circulating currents when we close the delta.)

I think we can show that the voltage across that open is zero as long as the primary neutral is unconnected.

Once we connect the neutral at the primary, I think we can show that the voltage won't be zero if there is some current through the neutral connection.

Steve

Nice model - I'm liking it. I'll sketch that out tonight.

cf

 

[brian john]

I have seen several with XO grounded, 3 had burnt conductors on the XO/neutral/grounded termination. All 3 also had an ungrounded secondary.

 

[hurk27]

I have seen several with XO grounded, 3 had burnt conductors on the XO/neutral/grounded termination. All 3 also had an ungrounded secondary.

The secondary of the load transformer has nothing to do with the X/0 connection on the primary side, it is the un-balanced voltage from the supply transformer that cause the problem in the X/0 connection, I'm thinking the supply Y transformer being in parallel with the load Y transformer is causing current in the X/0 because of this imbalance voltage, I need to draw out two transformers Y-Y in parallel and maybe I can see it, but why is it that we don't have this problem when separate tanks that don't have a common core, (like the utility uses) are used ????

 

[hurk27]

Ok after drawing a little diagram of the two Y-Y transformers, I think I can see what now causes it:

when we wire two Y transformers and have the X/0 connected, the supply transformer (utility) See's the other transformer as a impedance load, if the voltages are not equal then each phase will have a different amount of current on it, this off set of current is now being shunted through the X/0 back to the supply transformer as the extra current causing even more imbalance, I think the only thing that keeps it from running away is the resistance of the conductors between the two transformers, as each star point is trying to balance the voltage but since they are in parallel it cause a loop and more and more current is built up in the X/0 conductor, kind of like having two phases bucking against each other just out of phase.

 

[__dan]

Load neutral is connected to an egc

Load neutral is connected to an egc

The X0 of the Y primary, if it's unconnected to ground, can be expected to have a small floating voltage on it due to unequal loading between phases on the load side. If the primary side Y X0 is now connected to ground, the imbalance of current flow can be expected to flow on it, as a grounded neutral. So, the source is supplying a neutral, not an EGC, and must be marked / wired appropriately. The primary Y X0 is not separately derived and cannot be jumpered from X0 to the ECG and the case, this will result in neutral current flowing in parallel over the ECG and metal conduit paths.

Since this is existing, the first thing I would do is see if the connection can meet code as a neutral (insulated everywhere). If it connects to the case or the conduit, the connection would have to be opened. If it looks like a neutral I would amp clamp it under load and see what it's doing.

As already noted the primary side earthing does nothing for earthing the secondary side. This needs correction also.

 

[brian john]

The secondary of the load transformer has nothing to do with the X/0 connection on the primary side, it is the un-balanced voltage from the supply transformer that cause the problem in the X/0 connection, I'm thinking the supply Y transformer being in parallel with the load Y transformer is causing current in the X/0 because of this imbalance voltage, I need to draw out two transformers Y-Y in parallel and maybe I can see it, but why is it that we don't have this problem when separate tanks that don't have a common core, (like the utility uses) are used ????

I mentioned the ungrounded secondary as a side note to the installation, an ungrounded secondary in a commercial office building is bad idea in my opinion.

 

[steve66]

Nice model - I'm liking it. I'll sketch that out tonight.

cf

Hopefully it helps make it a little clearer. I think I must of seen this before (but mostly forgotten about it) because it seemed to fall in place pretty quick. So I guess I lied earlier when I said I have never seen this explained

After thinking about it a little more, I did realize that my analysis is basically assuming 3 single phase transformer banked together - not a single three phase transformer on a common core. Not sure if that makes any real difference or not.

Steve

 

[dicklaxt]

How would the imbalance in a primary WYE winding effect the delta secondary and more than a delta/delta would,an imbalane in a primary winding doesn't know that it is a WYE or Delta setup ,does it????

dick