Part Wind Motor Starters

[Walden]

I design and build electrical control panels for refrigeration systems. Very often the refrigeration compressor has a part wind start using two star connected, electrically separate windings. Each winding carries about 50% of the total running current of the compressor. One winding is energised and 1 second later, the second winding is energised. This system of starting a motor is, I believe, North American in origin and seems to have been used in Europe since the end of WW2. It appears to be used only for refrigeration compressors.

The vast majority of these motors and their starters work satisfactorily for years but there is a substantial minority of failures with the contactor tips welding together and the compressor winding being destroyed. Each winding has its own contactor and thermal overload relay and the motor as a whole is protected with HRC fuses or circuit breaker.

What appears to happen is that with a low current fault in one winding, the thermal overload operates and the contactor opens and draws an arc across the contactor tips. The arc melts the silver on the tips of the contactor and the tips become welded together. The faulty winding is now permanently connected to at least one phase of the mains supply by the welded contactor tip or tips and through the thermal overload relay: the winding is quickly destroyed and shorts out to earth. Only now do the protective fuses or MCBs “see” a large enough current to operate. But by the time they do, the contactor, the thermal overload relay and the compressor motor have all been destroyed.

Does anyone know of any research into these part wind starters?

Is the “good” winding generating a voltage in the faulty winding which enhances and maintains the arc I wonder?

Martin Pirrie.

 

[laguna]

Hello, Martin Pirrie.
We use YCAJ chillers on our CW comm. application, with a part wind starting for the compressors, operating as you said. All I can say is that instead of thermal overload relay, York put the motor's PTCs at work and also, the controller is monitoring the absorbed current with an ordinary current transformer, looking for an exceeding of a user predefined FLA rate. They also insert an AMPOHM supressor (0-47/220) on the auxiliary contacts of a part wind compressor's contactors. Maybe this will help you.
Anyway, who "see" the low current fault? Do you have a CB for this feature? Why don't you try to adjust the tripping time and level for your protections?
Bye.

 

[Walden]

Hi Iaguna!

Thanks for your reply. The suppressors that York fit, sounds interesting I shall investigate more. I have never seen them here in UK.

The problem, I believe, is protecting the compressor as a whole with three fuses (or MCBs). The fuses have to be sized to carry the full load current plus a bit but each winding is designed to carry only about 50% of the full load current. There are thermal overloads in each winding circuit but often the motor winding is burnt–out before the main fuses blow. On some sites I know, it appears that the compressors have a death wish! Every month, literally, another compressor bites the dust!

I have suggested fitting a set of fuses for each winding, but was told by the compressor manufacturer that it would invalidate their guarantee.

In the UK, part wind starting is seen on refrigeration compressors only. Ask a UK electrical engineer not in the refrigeration industry what a part wind start is and they will have no idea! I see that in the US it is used quite a lot; Hubbel have some fire pumps using pws: not so on this side of the Atlantic.

 

[busbar]

Part-winding starting is viewed as a less-expensive reduced-voltage starting method—sometimes suggested as being a {questionably equivalent} substitute for Y-∆ starting. In the US, for a given motor size, nondescript “reduced-voltage starting” is mandated by electric utilities in some locations, and the hermetic-compressor folks seem to have adopted P-W as the default quick-and-dirty solution.

Another weakness is the use of so-called “definite purpose” contactors of much lower duty—having smaller contact surfaces and clearances compared to NEMA-rated devices ordinarily used for motor-control tasks. In many cases, the packaged-chiller manufacturer’s attitude is, if there is a suspected contactor problem, discard it and install a new replacement, in place of the more traditional industrial custom of obtaining repair parts {like contact sets and coils.}

 

[Farkel]

You might want to try puting MOV's across the contactors contacts, to obsorb the arc, so as to rid one self of the flash over of silver welding.

 

[laguna]

Ya, try using the MOV's for each contactor. Calibrate'em carefully.

 

[Walden]

OK all, thanks for your suggestions and thoughts, I'll see if I can come up with a closer form of protection which will satisfy the compressor manufacturers and allow me to sleep easier at night!

A recent report by a compressor manufacturer put the failure of their compressors because of " burnt or welded contactors " at 22% of all returns. By far the highest reason. An electrical engineer will say that it is putting the cart before the horse! Rather like blaming your neighbour's fence for damaging your car when you drive into it! It is the winding fault which causes the contactor tips to weld not the other way round! Often all I am shown is the blackened contactor, thermal overload relay, blown fuse links and a compressor on its way back to be rewound! The ususal comment is, "There you are, your contactor did this!"

It was ever thus!

Thanks again,

Martin Pirrie.