Transient Voltage Variations for Generators - ABS requirement 1

[NickParker]

The generator vendor cannot meet the ABS requirement for momentary transient voltage variations for the step load of 60% of the generator's rated current. However, these voltage variation requirements can be met for a step load of 33%, which aligns with the diesel engine's maximum step load capacity (sudden load steps of 0-33%-67%-100%, in three steps).



Given this, I believe the deviation from the generator vendor is acceptable.
Any opinions?

 

[waross]

60% AT 0.4PF is only 24% kW loading;
So, the prime mover has to support 24%. to avoid a frequency dip and to avoid UFRO action.
This is a governor and prime ,over issue.
The AVR has to support a voltage drop caused by the 60% loading.

A prime power set has a 10% size/capacity advantage over a standby set.
This is reflected in the standard.

 

[waross]

The generator vendor cannot meet the ABS requirement for momentary transient voltage variations for the step load of 60% of the generator's rated current. However, these voltage variation requirements can be met for a step load of 33%, which aligns with the diesel engine's maximum step load capacity (sudden load steps of 0-33%

As noted. this will be a 24% step kW load on the engine.

 

[Gr8blu]

If this is really on a marine application - does the generator supply power for propulsion? If so, what is the relative power between the subject generator output and a LARGE load (like a propulsion motor)? The test is not only load ADDITION, it is load LOSS. If, for some reason, a propulsion motor dropped off line while under significant power, it could easily meet or exceed the 24% (of the generator output) power criteria.

 

[waross]

I doubt very much if the stated standards apply to electric propulsion systems.
Propulsion systems are more related to MG sets with variable generator speeds thrown in for good measure.

 

[NickParker]

If this is really on a marine application - does the generator supply power for propulsion? If so, what is the relative power between the subject generator output and a LARGE load (like a propulsion motor)? The test is not only load ADDITION, it is load LOSS. If, for some reason, a propulsion motor dropped off line while under significant power, it could easily meet or exceed the 24% (of the generator output) power criteria.

The propulsion is by Main Engine; ofcourse the auxiliaries of main engine requires power supply from these generators (Auxiliary)

 

[waross]

I read that as a test condition;
With block loading of 60% KVA or 24% kW the voltage dip must meet the standard.
I don't read anything that would prohibit block loading or dropping greater KVA or kW loads.
In such cases the voltage dip would be greater.
Said another way, the standard is specifying specific test conditions.
If the generator passes the test under those conditions, that does not prevent greater block loading than the test values.

 

[RRaghunath]

May be, the vendor needs to supply a bigger Generator (higher kVA), keeping the Prime mover same in order to meet the Voltage regulation requirement while starting the largest motor.
I have seen such a thing done while sizing a Black start Generator in a Gas Turbine based Power plant.

 

[NickParker]

I read that as a test condition;
With block loading of 60% KVA or 24% kW the voltage dip must meet the standard.
I don't read anything that would prohibit block loading or dropping greater KVA or kW loads.
In such cases the voltage dip would be greater.
Said another way, the standard is specifying specific test conditions.
If the generator passes the test under those conditions, that does not prevent greater block loading than the test values.

Actually the spec says, "transient voltage variation upon sudden loss of rated load, or the addition of 50% of rated load, shall not exceed ± 20% of rated generator voltage. The generator output voltage shall recover to/ within ± 5% of rated voltage, with no more than one overshoot / undershoot, in less than 1 second. For engines driving generators, the transient frequency variations shall meet the ABS."

Therefore, I interpret that this 50% loading applies only to the alternator. Since the specification does not mention the power factor during sudden load addition or loss, it seems to be less stringent than the ABS requirement.

I will request the vendor to comply with the ABS requirement.

 

[waross]

or the addition of 50% of rated load, shall not exceed

That begs some questions.
1> Is motor starting considered to be the rated current or the starting current.
2> Is the load the load in KVA, (motor starting 6 x rated KVA), or the load in kW, (motor starting 3 x rated kW)

Therefore, I interpret that this 50% loading applies only to the alternator.

Maybe not a valid interpretation:

For engines driving generators, the transient frequency variations shall meet the ABS."

Anecdote Alert;
Years ago, a country faced a power crisis with 6hr on and 6hr off power rationing.
A supplier ordered a large number of standby sets.
Whoever sized the sets failed to take into account the large amount of A/C loads.
By trial and error, at over 6 sites, I developed a good understanding of just how much a standby set could handle.
My findings, that were borne out by success sizing a lot of standby sets in the years following:
1. A standby set will only start a motor rated at 50% of the set capacity if the motor is the only load.
2. A standby set will start a motor rated at 33% of the set capacity with acceptable voltage drop for other loads.

Notes:
20% voltage drop or not? 20% drop corresponds to 80% voltage.
A common drop out point for magnetically held devices is 80%.
At 50% sizing, most magnetically held devices would drop out, implying more than 20% voltage drop.
At 33% sizing, all magnetically held devices would hold in, implying less than 20% voltage drop.

Notes on Notes:
For a prime rated set, add 10% to the set capacity before using these guidelines.
Example, a 50 KVA rated prime power set will be a re-rated 55 KVA standby set.
Consider it to be a 55 KVA set for the purpose of these sizing suggestions.

And another Note:
If you use the free generator sizing software from Caterpillar the software will recommend a larger set that suggested by these guidelines.
The software uses a default "allowable voltage drop" setting.
You may change the "allowable voltage drop" setting in the software.

After sizing the set to ABS standards, you may wish to double check it against these suggestions.
Or use the Cat sizing software, (with the default "allowable voltage drop" changed to ABS values).

 

[waross]

A note on generator voltage drop due to block loading:
Example; A motor that draws 50 KVA, is started on a 150 KVA standby set..
The motor will draw about 150 kW on starting. That will be seen as a 100% load on the engine. There will be a brief frequency drop as the governor and the engine respond to the load.
Note: When the set is running no load, it will be running 3% fast or at 61.8 Hz.
If the block load pulls the frequency below 3 Hz or 57 Hz, Under Frequency Roll Off will become active and drop the voltage setting.
So, 100% block loading on the engine.
KVA loading:
This is the loading seen by the generator end.
300 KVA represents a 200% block load on the generator. The voltage drop will be determined by the ability of the AVR to compensate for the internal voltage drop.
How will this be affected by UFRO?
UFRO may move the voltage set-point down a few percent, but the KVA block loading will have dropped the voltage so far below that lowered setpoint that UFRO action will not be a factor.
Hope this helps.

 

[TugboatEng]

Some of these rules are very antiquated I think you would have a hard time finding an AVR that couldn't meet the requirements. Digital voltage regulators are even more responsive. More difficult to meet is the 300% overload requirement that may or may not apply. Permanent magnet or separate excitation can really boost performance under overload conditions.

 

[NickParker]

That begs some questions.
1> Is motor starting considered to be the rated current or the starting current.
2> Is the load the load in KVA, (motor starting 6 x rated KVA), or the load in kW, (motor starting 3 x rated kW)


Maybe not a valid interpretation:

Anecdote Alert;
Years ago, a country faced a power crisis with 6hr on and 6hr off power rationing.
A supplier ordered a large number of standby sets.
Whoever sized the sets failed to take into account the large amount of A/C loads.
By trial and error, at over 6 sites, I developed a good understanding of just how much a standby set could handle.
My findings, that were borne out by success sizing a lot of standby sets in the years following:
1. A standby set will only start a motor rated at 50% of the set capacity if the motor is the only load.
2. A standby set will start a motor rated at 33% of the set capacity with acceptable voltage drop for other loads.

Notes:
20% voltage drop or not? 20% drop corresponds to 80% voltage.
A common drop out point for magnetically held devices is 80%.
At 50% sizing, most magnetically held devices would drop out, implying more than 20% voltage drop.
At 33% sizing, all magnetically held devices would hold in, implying less than 20% voltage drop.

Notes on Notes:
For a prime rated set, add 10% to the set capacity before using these guidelines.
Example, a 50 KVA rated prime power set will be a re-rated 55 KVA standby set.
Consider it to be a 55 KVA set for the purpose of these sizing suggestions.

And another Note:
If you use the free generator sizing software from Caterpillar the software will recommend a larger set that suggested by these guidelines.
The software uses a default "allowable voltage drop" setting.
You may change the "allowable voltage drop" setting in the software.

After sizing the set to ABS standards, you may wish to double check it against these suggestions.
Or use the Cat sizing software, (with the default "allowable voltage drop" changed to ABS values).

If I use VFDs for all the motors and implement a form of pre-magnetization for transformers to minimize inrush current, this should address transient voltage issues caused by the sudden addition of large loads. However, would a sudden rejection of these loads still pose a problem?

 

[waross]

Starting a motor will be a kW load of +300% and a KVA load of +600%
Stopping the same motor will be a load of 100% dropping to zero%, or minus 100%.

Energizing a transformer will be a KVA load of from 280% to 2500% in the worst case.
That is 280% for a fully offset peak plus the added transient effect of saturation.
De-energizing the transformer will be similar to stopping a motor: Minus 100%.
And if the block loading and unloading is at 50% of the gen-set rating, that becomes load dumps of minus 50% of the set rating.

 

[edison123]

Starting a motor will be a kW load of +300%

huh? why?

 

[waross]

The starting current of a motor is accepted to be about 600%.
But that current is at a very low power factor.
The real current is probably less than 300% , so 300% is erring on the safe side.

Anecdote in support:
You may find, in old text books a proposal to reduce motor starting current by adding a lot of capacitors to supply the reactive starting current.
The theory held that motor staring current could be much reduced by this method do to the low PF of starting current.
However, the capacitors were ideally switched out in steps as thee motor accelerated.
I am not aware of the scheme ever being implemented due to the cost and complexity of the arrangement.
The proposed scheme does illustrate the relatively low real current component of the total starting current.
Reactive current takes no power.
The kW loading is based only on the real current component of the starting current.

 

[edison123]

I do the testing of LV motors with inrush capacitors compensation and without inrush capacitor compensation all the time in our shop and never the KW demand has crossed 5 to 6 % of motor KW because it's all KVA with pf less than 0.1.

No diesel engine can survive 300% overload however short the duration may be. Your assumptions are way off.

 

[waross]

No diesel engine can survive 300% overload however short the duration may be. Your assumptions are way off.

That is why I size generators at 3 times the motor running current, not 6 times the running current.
When I was faced with a number of residential sets that were all undersized I went through a trial and error procedure many times.
Each home had several smaller A/C units. I would work with the home owner to determine which and how many A/C units could be started on his new set.
The 3:1 ratio was always confirmed.
The governor would go full rack almost instantly, BUT,
Turbo'd sets would slow down and blow black smoke until the turbo spooled up.
Using a 3:1 ratio for a 300% load equals a 100% block load on the diesel set.
I admit that a successful start was a combination of speed drop and voltage drop, but the action of the diesel sets was consistent with close to 100% block loading.

I do the testing of LV motors with inrush capacitors compensation and without inrush capacitor compensation all the time in our shop and never the KW demand has crossed 5 to 6 % of motor KW because it's all KVA with pf less than 0.1.

Something not right here.
A running, loaded motor will have kW demand equal to motor kW plus losses.
An oveloaded motor may have a peak kW demand of 250% of motor kW plus losses.
A motor under test with enough capacitors to substantially reduce starting current will be an a low leading power factor if the capacitors are not switched out.
I can accept an unloaded motor with a very large capacitor bank running at 10% PF.
My sets would not blink at 5% to 6% block loading. The action of the sets was closer to 100% block loading.