Generator Drop Out with large block load 1

[CanuckEngineer]

Hello everyone,

Great website, great knowledgeable people here resolving key issues found on sites.

I work for an Independant Commissioning Agency which is responsible for the verification of all mechanical/electrical systems within buildings.

I work up here in Southern Ontario, Canada.

I am working on a new construction hospital project.

The project has:

qty:2 CAT gensets 3516 Engine, 825 Alternator Frame, 2000kW, 2500kVA rating at 0.8pf. its is a three phase system with a line voltage of 4160V. Excitation is PM.

There is an operating standards Hospital code in Canada where Emergency Life System power has to be restored via the generators within 10 seconds of a loss of utility power.

The two generators are meant to parallel and synch together to share the load however to meet the above code, the generators race to the emerg. bus and the first one there will pick up the Emergency Life Safety Load.

An issue has come up, we have an advanced load shedding system designed and implemented in this project, however all of the power control for tripping the load shedding breeakers are not on UPS and/or DC battery power.

Therefore the Generators have to come online and assume all of this load before it can shed via its controls.

There is an estimated design of up to 3MW of power on the Emergency Bus. Which means that since these loads cannot be dropped until the Generators come online the first Genrator that comes onto the sytem will see a 3MW block load hit.

With your various experiences, what is your expectation of one generator taking a 150% block load hit. Each generator is rated a 2MW. Will the field voltage in the voltage regulator break down thereby losing the generator and essentially all source of emergency power.

Much thanks for your help.

 

[rasevskii]

We have an expert on the forum: "catserveng" whom I would expect to respond professionally soon to CAT items.

But you cannot expect any 2MW gen set to take a 3MW load at all let alone a sudden block load. The unit would (has to) trip on all sort of protection functions, or it will stall with damage...electrical and mechanical...

Just a fast answer to your question.

rasevskii

 

[X49]

That would most definitely not work. The alternator might be able to supply that load for a short time (though voltage dip would be high and the protection could trip), but the engine would have no chance of taking 150% load.

The obvious solution seems to be to add a UPS to your load shedding controls.

 

[catserveng]

A 2MW 3516B engine can't even do a 100% block load consistantly, and depends greatly on your entire systems tolerence to voltage and frequency deviations. There is no overload capacity on modern standby rated engines, so 2MW is all you'll ever get.

Not having any details of your paralleled system hard to guess what's actually taking you down, but likely a race between under-frequency or under-voltage.

Depending on your allowable voltage and frequency tolerance, you may be able to block load somewhere between 1500 and 1800 ekW, but issues such as V/Hz setting in AVR, 27 and 81U settings in protective relays, governor and AVR dynamic settings, and the types of loads you're trying to pickup all have an impact on transient response.

Your system design needs to be changed to either have both units paralleled before accepting any load, or load shed below a single engines rating, bring that load on then allow second unit to parallel. Care should be taken to prevent active parallel attempts while on-line unit is accepting transients, problems such as reverse power and VAR trips can happen if phase and voltage match synchronizing is used, controls are trying to hit moving targets. And unfortunately sometimes to make these systems work (paralleling large engine gensets in less than 10 seconds), phase and voltage windows get opened up beyond what we would see in prime power applications. You can pass a test, but problems will occur later in system life.

Hope that helps,

Mike L.

 

[CanuckEngineer]

Thanks everyone for your input.

Hi catserveng nothing is tripping as of yet, because this 3MW load is the future expected load, i.e. the load that will be hooked next year after building hand over to the owner.

Right now the emergency bus load is well beneath 2MW which is why the testing has been going well, however I am trying to see in the future.

When I raised the issue with the construction, the Engineer of Record, Consultant Engineer dismissed my concerns stating the CAT's have a 300%-10 sec short circuit current rating which they state will be sufficient.

This rating does not apply to this situation correct?

 

[rbulsara]

150% step load is out of question.

Refer to ISO 8528 standards. They have performance categories like G2 and G3, specifying voltage and frequency deviations and recovery rate for various step loads for various categories.

For 100% step load a diesel-gen set need to meet category G3, I beleive, which most units in the USA do not. G2 compliance is more common.

Even after all that it takes a good field technician to tune the system right and a good engineer to test it. I have done a few 100% step load tests and not many have passed.

Rafiq Bulsara

http://www.srengineersct.com

 

[dpc]

The short circuit rating has nothing to do with block loading capability, unless your load will be a zero impedance short circuit.



David Castor

http://www.cvoes.com

 

[catserveng]

I guess I say after doing this for a number of years, hearing a statement like that from the engineer who designed the system would certainly cause me concern. Hopefully whoever provided the switchgear was experienced.

Yes, the unit has 300% field forcing for 10 seconds for short circuits, David already answered that nicely. Rafiq gave you the reference which is the general guideline for generator performance and response. Problem is that these days with all the different types of loads in a facility, things can be hard to predict.

So a few things you may want to get a handle on before this goes too far,

What is best estimate for emergency load?

If it is beyond capability of a single unit, does you operation sequence assure reliable ability to pick up that load?

What is your voltage abd frequency deviation tolerence?

How will you define acceptable system performance?

Hope that helps, Mike L.

 

[rbulsara]

That 300% amp for 10 second is mainly for alternator's thermal capacity, mainly to accommodate motor starting, where the the actual kW (power factor) is low and the current is high. kW still have to remain within rating of the prime mover.

It not meant for 300% of kW rating for 10 seconds.

Also during a short circuit voltage goes down to almost zero and does not recover!

Rafiq Bulsara

http://www.srengineersct.com

 

[ScottyUK]

A short circuit consumes virtually no power, so the engine doesn't see much load. If the engine was working hard prior to the short circuit then it will actually speed up if a short is applied to the alternator, counter-intuitive though that may sound. Your consultant appears to be confused about how synchronous machines behave if he believes block load capability can be surmised from short circuit withstand rating.


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

If we learn from our mistakes I'm getting a great education!

 

[waross]

Two identical plants will be up and ready within a second or so of each other.
Step 1> Inhibit transfer until both sets are on line.
Step 2> If the second set fails to come up, use power from the first set to initiate load shedding. Then transfer the remaining load.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[TheBlacksmith]

I'll show my ignorance here, probably generated by all my years in the Navy and working on Navy designs. Most shipboard systems shed load when the power goes away; the controllers open when the "drop out" voltage on the coils is reached and only the vital loads restart upon restoration of power. Is not a simialr scheme possible here?

 

[CanuckEngineer]

Thanks to everyone for their input.

I have since went back and said that a re-design is necessary.

TheBlackSmith, you make a great point, this would apply when your undervoltage trip coils are installed on the breakers, they were not on this project.

However ironically this was the suggestion that I had mentioned to them for the quickest solution.

Only on the breakers needed to bring initial load at about 1900ekW or so.

I am not sure how much of the UV coils run about? They are using Schneider Masterpact breakers (200A-1000A) frame size.