Star/delta Wiring Diagrams

[JohnD]

Hi guys

 

Sorry if this has been mentioned before, but the drawings of the star/delta starters on the website http://www.lmphotonics.com/star_delta.htm are inconsistent. The wiring and control circuit diagram correctly shows the connections U1-V2 (L1), V1-W2 (L2) and W1-U2 (L3) when in delta connection, but the last of the three drawings of the three stages of open-transition from star to delta incorrectly shows the connections U1-W2, V1-U2 and W1-V2.

 

In my experience, only the first diagram (U1-V2 (L1), V1-W2 (L2) and W1-U2 (L3)) gives acceptable transition currents, and the latter connection gives currents 50-100% higher at transition, although it does appear to work OK and keeps the same rotation. I have seen connection diagrams in pump manufacturers' literature stating the wrong method of connection, and on some respected control panel manufacturers' drawings statements to the effect that if the motor rotation is wrong, then swap U1 for V1 and U2 for V2 at the starter output terminals, which is effectively the same as the second example and hence gives the much higher transition currents.

 

According to my information the simplest and most reliable way to correct a wrong motor rotation is to swap two phases upstream of the contactor assembly (e.g. at the panel isolator load terminals providing there are no phase-rotation sensitive devices downstream), but leaving the motor connections and contactor wiring as they should be. This effectively reverses the phase rotation of the whole starter (and hence the motor) without messing with motor connections. Doing so should always be labelled up within the panel for the next guy who has to change the pump - the new one may have a different rotation to the previous one.

 

Regards

 

John Daniels

[marke]

Hello John

 

There are many discussions and claims about the merits of connecting the delta to the following phase rather than the preceding phase etc, and this is certainly pushed by some manufacturers.

In a perfect environment where the motor is at synchronous speed before the transition, the connection does make a difference because the voltage generated by the open circuit rotor is at almost an equal frequency to the supply and just lags it in phase by a small amount.

The reality however, is that the frequency generated by the open circuit motor is well less than the supply due to the motor being well below synchronous speed, and so the phase angle at the point of reconnection in delta is purely random and there is no appreciable advantage in either connection.

 

If you have a star delta starter "wired correctly" for minimum transients (assuming that you are achieving synchronous speed before the transition) then reversing the incoming phases to reverse the motor actually configures the connections incorrectly. The motor will rotate in the opposite direction, but the windings are then connecting onto the preceding phase rather than the following phase. If you are concerned about the connection, then the correct way to reverse the motor si to actually swap two windings. That way, the ends always go back to the following, to the leading phase.

 

Best regards,

[JohnD]

Thanks for the reply Marke.

 

I guess it's a case of which works best in any given situation, given variables such as motor speed at transition, time between disconnection of star and reconnection of delta and the rate of deceleration. I have had several instances of the panel being changed (new contactors) but the motor and the star timer kept the same, and flicker complaints occuring from that point on. The only real difference I can imagine is the transition time and resulting deceleration of the motor before and after. It's probably down to the pull-in speed of the contactors and the finite time the interlocking mechanism takes to operate.

 

One further question - a foul water pumping station I am working on at the moment has two 30kW 400V 62A 50Hz pumps, each typically runs at around 20kW electrical load on star/delta starters. Using an Amprobe Skylab, I have recorded star inrush currents of up to 135A and transition currents up to 309A. This gives a star/delta transition of up to 5 times rated running current which sounds a little high - 3½ times being the norm. The supply authority claims to have measured up to 600A using their Linecorder instrument which apparently captures down to a couple of microseconds (well faster than the eye can tell).

 

Do my measurements sound about right? I suspect the difference is down to the sampling rate of the instrument - the faster it is the higher transient value it can capture rather than being weighted to something more akin to the naked eye. Flicker is after all only what the eye can see (and be irritated by).

 

The reason I ask is that I am struggling to determine whether the connections are optimised for this installation, without (as yet) trying anything different for fear of making an already bad flicker complaint worse. The panel manufacurer in this instance has chosen to reverse all the motor windings (i.e swapped U1 & U2, V1 & V2, W1 & W2) so U1, V1 and W1 are connected together in star with U2 - L1, V2 - L2 and W2 - L3. The delta connections are thus U2 & V1 - L1, V2 & W1 - L2, W2 & U1 - L3. This is opposite to the accepted way of all the 2's together as the star point.

 

I presume that as the wiring is consistent (i.e. no single phase winding reversed) and the flux is rotating in the same direction, the motor will turn the same way as if the windings were not reversed, and this alone should not cause high transition currents.

 

Any comments?

 

Pining for the fjord? What kind of talk is that Tosh? If I hadn't nailed it to its perch......

[marke]

Hello JohnD

 

Yes, it is totally dependent on the phase of the generated voltage at the point of changeover relative to the supply voltage and this is dependent on the transition time and the slip that the motor is operating at.

One of the main issues, is that very few motors actually reach synchronous speed in star due to the torque curve in star and the torque curve of the driven load. It is not uncommon for pumps and fans to transition to delta at half speed, and at this speed, there is both an open transition switching transient, plus a step to close to locked rotor current.

 

This gives a star/delta transition of up to 5 times rated running current which sounds a little high - 3½ times being the norm.

I suspect that this is actually quite low from my experience. I would suggest that you are measuring the steady state currents rather than the transient current.

It sounds as though the motor is only up to part speed, (less than 80% speed) at the point of the transition and you are measuring the delta current at that speed. I would expect the "transient" to possibly be 2 - 3 times higher than that, but only lasting for a couple of cycles. This will not always be read correctly with standard instrumentation.

 

Checking the wiring is not really a major issue, what is important is where the delta conactor closes to. If you take the winding which starts on phase 1 at the main contactor, then the other end of that winding should go back to the preceding phase. Assuming the phase sequence is measured as L1 - L2 - L3, then the ned of the winding on L1 should go to L3 etc. This however assumes that the motor is operating at virtually no slip and there will be no change in phase when the delta contactor closes. - Not all that common!!

 

Best regards,