IEEE 399 - Motor starting studies thumb rule

[NickParker]

According to IEEE 399:

- For grid-connected systems, if the motor rating exceeds 30% of the supply transformer's base kVA rating.
- For islanded systems, if the motor rating exceeds 10-15% of the generator's kVA rating.

What is the rationale behind this rule of thumb regarding the 30% and 10-15% limits?

 

[LittleInch]

For grid if it all goes wrong, things trip and disconnect but the grid survives as do the rest of the supplies

Island system if it goes wrong the main generator can trip and everything goes dark.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[waross]

- For islanded systems, if the motor rating exceeds 10-15% of the generator's kVA rating.

I was not aware of this rule when I was spec'ing and installing generators.
I successfully used 33% of the generators capacity on a couple of dozen sets.
Cat sizing software would use a slightly lower factor but nowhere near 15%.
If an extreme voltage and frequency dip is acceptable, I have seen motor KVA as high as 40% of generator capacity started on a gen-set.
Any higher factor and it is doubtful that the motor will start. (Been there also.)
My customers were all happy with the voltage dip associated with a 33% factor and were not willing to pay for a larger set to reduce the motor starting induced voltage dip.
IEEE must be using a very very conservative factor for voltage dip and frequency dip with no regard to cost.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[Marke]

Waross

Yes I agree.Have gone to 40% plus on many occasions.
It depends on the generator set up and configurations.
If a three phase averaging AVR is used, plus external excitation (permanent magnet generator for excitation) you can commonly go to full voltage starting with some sag, but no problems. If coupled with turbo boost and current ramp type starting, you can do even more, but some of the "cheap" self excited systems are a struggle at 10%.

Many years ago, I quoted a 120KVA system correctly configured for a 60KW bore hole pump and guaranteed it would work.
The opposition used my sizing to supply a 120 KVA self excited and it crashed, as did the 200KVA and the 300KVA. They ended up with a 600KVA self excited system. - A lot more expensive than correctly sized and configured unit and burnt bulk fuel!!

Such rules need to be qualified as to the conditions because there are many parameters that can have a major influence.

Mark Empson
Advanced Motor Control Ltd

 

[LionelHutz]

Ya, you might want to study the start at that level on a generator.

With the right generator and a soft-starter on an easy to start load you can size the motor at 40-50% of the generator capacity and still start it.

 

[Gr8blu]

IEEE Std 399 is part of the Color Books family. Today's intensely competitive business environment forces plant and building management to be very aware of the total owning cost of power distribution systems. Furthermore, these managers demand assurance of maximum return on capital investments in the power system. ... The planning, design, and operation of industrial and commercial power systems require several studies to assist in the evaluation of initial AND future system performance, system reliability, safety, and the ability to grow with production and/or operating requirements.

In short, the reason the standard sets the level at 15% is because of FUTURE considerations.
Also note - this standard is an inactive (and reserved) standard. There is an ongoing effort to revise/update all the IEEE Color Books; some are further along than others. If you're interested in where things stand now, do a search for the IEEE 3000-series standards.

Converting energy to motion for more than half a century

 

[waross]

Thanks for the support Marke.
The sets that I was installing were residential with A/C loads.
The first tranche, (spec'ed by others.) were single phase, NA, self excited (with a brushless exciter, not capacitor excitation.)
All (about 6.) were 17 KVA and all were too smell.
At each install,the owner and myself would experiment with starting various A/C units to determine which and how many A/C units could be started successfully.
In every case, a 3:1 ratio of set KVA to A/C KVA was acceptable.
Then I would disable automatic starting so that the customer could control which and how many A/C units could be running.
Occasionally there would be one large A/C unit that was a 2.5:1 ratio, and that could be used and the 2.5:1 ratio used only if this was the first A/C too start.
My ultimate limit was the Automatic Transfer Switch contactor.
I was called in to repair a couple of installations done by others.
Too much load automatically starting and the voltage drop would drop out the standby contactor in the ATS.
With no load, the set would accelerate and the voltage would recover.
The contactor would pull back in.
Rinse and repeat.
It took a minute or so to destroy the ATS contactor.
At a 3:1 ratio, I never had a problem or a customer complaint.

Interestingly, I eventually came across the Cat sizing software.
Using the Cat default settings, all of about 20 installs failed as set too small.
After changing the default voltage drop value in the software to a more liberal value,all installs passed.
Had a set been sized to Cat specs, the customer would have started one more A/C and we would be back to the 3:1 factor.

A further note: The 3:1 ratio applies to Standby rated sets.
The KVA of a Prime rated set may be increased by 10% before applying the 3:1 factor,
By definition, a prime rated set plus 10 percent equals the standby rating.

I am willing to bet that the authors of IEEE 399 have never actually commissioned a standby gen-set "In the Wild".




--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[LittleInch]

I would also guess that the rather conservative numbers in the OP allow for a few other loads to be running before starting the motors?

electric pete "out of curiosity, what is the "then" part of this if/then clause?" - I would assume it's get a bigger one....

waross - what volt drop is "more liberal"?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[electricpete]

electric pete "out of curiosity, what is the "then" part of this if/then clause?" - I would assume it's get a bigger one....

Not really. I looked it up and this thumbrule is the cutoff beyond which a "starting study" is required, presumably to further evaluate whether or not the configuration is acceptable. (it's not saying that installations beyond those limits are necessarily unacceptable)

I'm used to seeing some degree of screening calcs for any large motor. I'm not sure exactly where the line is where that turns into a starting study.

 

[waross]

waross - what volt drop is "more liberal"?

It was a long time ago and I forget the exact numbers.
I think hat I was running about 5% below the Cat default percentage.

I would also guess that the rather conservative numbers in the OP allow for a few other loads to be running before starting the motors?

Actually I based my sizing on Available generator capacity.
If there was a 50 KVA load on a 100 KVA set, I used 3 times the 50 KVA unused capacity.
That still didn't get me up to the Cat numbers but it helped.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[Parchie]

That section in IEEE 399 specifies the need to conduct starting motor studies when the condition specified is reached. Besides, that specific section states that one has to do a motor starting study if the motor rating exceeds 30% of the transformer rating -> if there are no generators present. Clearly, IEEE 399 sees that there will be possible problems if large motors are connected to transformers at that size ratio. If only generators are involved, IEEE 399 recommends a motor starting study be done when the rated horsepower of the motor connected to the generator is greater than 10 to 15% of the rated kVA rating of the generator. It is understandable because the terminal voltage of generators are not as stiff as those of transformers when subjected to big block loading.