Strong or weak supply 2

[ozmosis]

I was reading another thread recently and there was a discussion about starting 7000hp motors DOL. I didn't want to tangent the thread by asking more questions about the actual effect of going DOL with such a large motor (it was on a 13.2kV supply) as I was advised by far more knowledgable members that it was a very stiff supply and would be no problem. Ok, accepted. However, I'm still intrigued as to what can make a "stiff" supply, weak. Is it simply adding more and more loads to a distribution network?

 

[Skogsgurra]

No, that would actually make it even stiffer.

A stiff supply has a very low impedance. In a national power grid, you can consider the grid as a zero ohm voltage source (OK, it may have a few milliohms, but not much more). A 13.2 kV (or any other MV system) usually gets its voltage from a secondary voltage that usually is something like 110 kV or 70 kV, depending on local conditions. The 110 or 70 kV is derived from the national grid (400 or 500 kV). Transformers are used to reduce voltage from national to secondary to tertiary voltage, which usually is what feeds large motors.

The transformers have an impedance that is designed into them so that the short-circuit current isn't too high for the bus-bars and the switch-gear. That impedance plus impedance in long transmission lines is what makes a line "weak". Think of it as a voltage source with an impedance. The more impedance - the weaker the source.

Adding load to the grid, especially motors, makes the grid stiffer. There are two reasons for that. The first is that any impedance parallelled with another creates a lower impedance (think Thevenin). The second is that all motors (induction or synchronous) have a rotating field that generates a counter-EMF that, when there is a heavy load on the grid, helps putting power onto the grid.

That's what it is about, essentially. I am sure many others will fill in what I left out.

Gunnar Englund

http://www.gke.org

 

[Skogsgurra]

I fill in myself. The contribution from motors on the grid is only temporary. So it is only used to calculate short-term effects. Like transient impedance and initial short-circuit currents.


Gunnar Englund

http://www.gke.org

 

[ScottyUK]

Stiff and weak are relative. Thus a supply which appears rock solid to a small load may be weak to a much larger load. The impedance of the transformer supplying the voltage at which the load is connected is usually the dominant factor, neglecting secondary effects such as motor contribution to peak fault level. I use the term 'usually' because there's always some condition which screws up the rule-of-thumb. Such factors include:
[ul]
[li]The presence of, or dependence on, local generation. The behaviour of the generator governor and AVR is highly significant.[/li]
[li]Use of fault limiting reactors which artificially weaken an otherwise very stiff supply, usually for the benefit of the switchgear under fault conditions[/li]
[li]Unusual transformers with especially high or low impedance. There are generally accepted impedance values for common transformers such as distribution transformers. These values are arrived at by compromise between switchgear withstand capability and ability to supply loads without excessive voltage variation under load. Transformers outside of these impedance values are available, but require either unusual (expensive) switchgear or tend to have regulation problems with load change.[/li]
[li]Transformers at the end of long OH lines tend to be the worst for weak supplies. The line impedance is significant and the voltage at the transformer primary can vary substantially with load, causing voltage changes to be reflected through to the secondary side, in addition to the volt-drop due to the transformer impedance itself.[/li]
[/ul]

The following should illustrate the point. The values are rough & ready and not entirely realistic for the real world, but the principle should be apparent.

Take an example of a typical distribution transformer of, say 100kVA rating and 2.5% impedance with a 90kW motor. The motor current is, for the purposes of the example, equal to the transformer FLC and the transformer secondary voltage will drop 2.5% under steady state conditions. Under starting conditions it draws approximately 6x FLC and the voltage drops by 15%.

If we re-run the rough calcs for a 1MVA transformer with 5% impedance, the secondary voltage drops 0.5% under steady state conditions, and during starting it drops 3%.

I've only used transformer impedance in the example, but the effect would be similar using line impedance. I've totally neglected that the impedances are complex and a few other things, but hopefully the simplistic illustration shows what is going on.

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[ozmosis]

Gunnar and Scotty
Thanks for the explanations. That's certainly a lot clearer now.

 

[itsmoked]

Let me add that in these discussions, (such as the 7000hp DOL), 'stiff' generally implies that a start of whatever motor is involved will not cause a disturbance (voltage drop) great enough to cause a problem. 'Problems' could range from non serious but noticeable light dimming to more serious HID lights extinguishing and equipment resetting to very serious where the motor being started or other motor's starting systems drop out during the start, oscillating. What I'm saying is "stiff" can mean different things depending on the conversation's context.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[apowerengr]

There are not a lot of systems which can line start a 7000 HP motor without significant migitation. When I worked on car shredders we had to use a very low (4%) impedance supply transformer and used a liquid rheostat to control the rotor impedance on an wound-rotor induction motor(controlled the locked-rotor current to roughly 2X the full-load current). There are also CBEMA curves and IEEE guidelines describing the tolerance of equipment and people to voltage flicker caused by motor starts and other short-duration transient events.