D

#### Diaa

Hello,

I need to understand the impact of the lead power factor loads on the synchronous generators as i have a problem that sometimes a genset trips on "VAR import"&"gen bus overvoltage".

The plant has 2 x 1.25 MVA gas gensets and 1 x 1.25 MVA standby diesel genset. most of running loads are resistive loads as the plant is not totally commissioned yet: 2 thyristor controlled heaters = 2 x 250KW, resistive load bank = 750KW, lighting, UPS,....etc. the average power factor of the system is 0.96. Also i have a few questions.

1- Why does the gen trip on bus overvoltage?

2- Why does this trip accompanied by a very rapid frequency change which causes the UPS to disconnect the static bypass?

3- Is there a relation between the rapid change in load caused by the heaters and the generator trips?

4- Is there a possibility to reduce the total running machines power factor by adjusting the excitation?

Note: Machines are running in island mode (one in isochronous mode and the other is in droop mode).

#### CSA

Hello,

> 1- Why does the gen trip on bus overvoltage?

Exactly what happens immediately prior to the bus overvoltage trip? Is some load (resistive; capacitive; inductive) being switched on or off when this is occurring?

What happens to the other generator's voltage when one generator trips on bus overvoltage?

> 2- Why does this trip accompanied by a very rapid frequency change which
> causes the UPS to disconnect the static bypass?

Frequency changes (and you didn't say if the frequency increased or decreased--a very important distinction) are the result of an excess of load vs. generation (if the frequency DEcreases), or an excess of generation vs. load (if the frequency INcreases).

> 3- Is there a relation between the rapid change in load caused by the heaters
> and the generator trips?

When load suddenly increases the Isochronous machine must increase it's power output very quickly in order to maintain frequency. HOWEVER, if the Isochronous machine is at or near rated power output (load) when the load increases then it can't increase it's power output very much, if at all. On an islanded system this means the frequency will start to decrease. To some extent the droop machines will try to help support grid frequency by increasing their power output <i>as frequency decreases</i> but frequency control is not their job--it's the Isochronous machines. And, the Isochronous machine can't make more power than its rating, nor less than zero (should the load suddenly decrease and cause the Isochronous machine to reduce its load to zero).

Unless there's some kind of automatic "power management system" or "load sharing" system controlling all the generator prime movers it's the operator's responsibility to make sure the Isochronous machine's load will not go over rated if the load on the system is increased, nor below zero if the load on the system is decreased.

Now, all of a sudden, as expected, 2 MW of load is added to the island. The immediate effect of this is for the island frequency to begin to drop--but the Isoch machine senses the drop in frequency and increases the energy flow-rate into the Isoch machine's prime mover to maintain the frequency of the system at rated. Now the total load on the island is 19 MW, with 7 MW on the Isoch machine, 7 MW on one droop machine, and 5 MW on the other droop machine.

Now let's say the Isoch machine was operating at 1 MW, and the two droop machines were operating at 7 MW each. And, the operator was anticipating a 2 MW drop in load, which would cause the Isoch machine to go into reverse power (-1 MW). He would decrease the load of one or both of the droop machines to increase the load of the Isoch machine, so that when the expected 2 MW drop in system load occurred the Isoch machine would not go into reverse power. So, let's say the operator decreased the load of both droop machines by 2 MW each--so both droop machines would now be operating at 5 MW each, and the load on the Isoch machine would increase by 4 MW to 5 MW. When the expected system load drop occurs, the Isoch machine's load will automatically decrease to 3 MW (and the load on the two droop machines will remain at 5 MW each).

Many operators believe that Isoch machines will automatically maintain system frequency regardless of system load changes--and that's true, <b>as long as the system load changes don't cause the Isoch machine's load to increase above rated power output, or below zero.</b> And, unless there's some control system monitoring system load and controlling the load of all the machines, or at least some of the machines, connected to the system it's the operator's responsibility to make sure expected load changes don't cause the Isoch machine's load to go above rated or below zero--which would cause the system frequency to deviate from rated.

> 4- Is there a possibility to reduce the total running machines power factor

On an Islanded system, the system power factor can't be changed by changing excitation. It is a function of the loads (resistive; capacitive; inductive). Changing the excitation only changes the system voltage--and when there is more than one generator connected to a system with no impedance between it and any other generators on the system (i.e., no transformers between them) then changing the excitation on one machine can cause swings in the voltage of other machines. That's a function of the exciters (the "AVR"s), and can sometimes by tuned or compensated for.

Reactive load can be "shared" like real load (watts; KW; MW). But, usually--and especially if there no impedance (transformers) between generators--it takes tuning and circuitry (reactive load sharing/compensation), and sometimes patience to get it working properly.

Island system operation can be tricky. Sometimes prime mover governors have to be tuned to make them more stable when load changes (watts; KW; MW). And sometimes exciters (AVRs) have to be tuned to make them more stable when reactive power changes (as, say, large inductive or capacitive loads are added or removed from the system).

One other thing I should have asked--and which should always be asked when troubleshooting problems like this: When did this problem start? Has it been ongoing since commissioning? Did it start after some maintenance outage? After some trip from load? After some system component was changed or added? Because these kinds of things can be very important to know and understand when troubleshooting issues like this--they can be the cause of the problem, and if not known can cause a lot of frustration and grief when trying to resolve the problem because steps might be taken which could make the problem worse, or which could cause other unintentional problems. Always try to describe when the problem started, and what is happening when the problem occurs, and what--if anything--might have changed or was done to the system or its major components prior to the start of the problem.

Hope this helps!

D

#### Diaa

Hello,

> Exactly what happens immediately prior to the bus overvoltage trip? Is some load (resistive; capacitive; inductive) being switched on or off when this is occurring? <

what happens exactly before overvoltage trip is a sudden load changes between on and off when the heater starts, as it is controlled by thyristors and it uses rapid on/off method to control the temperature(heater load is 250 KW resistive) which makes the total load on the generators to swing rapidly between 450 KW to 700 KW and the power factor increase from 0.93 to 0.97 when the heater is ON.

> What happens to the other generator's voltage when one generator trips on bus overvoltage? <

the alarm sequence exactly was as following
1- 8:16:42 PM G3 Busbar overvoltage 3
2- 8:16:42 PM G3 Busbar overvoltage warning.
3- 8:24:56 PM G4 Var import.
4- 8:29:31 PM G3 generator undervoltage 3

so i am not sure if G3 trips on overvoltage according to the alarm sequence, also the above mentioned sequence of events happens when the heater load start to swing between on/off, after these fluctuations the switch-gear drop the load by mean of UV/OV protection.

> Frequency changes (and you didn't say if the frequency increased or decreased--a very important distinction) are the result of an excess of load vs. generation (if the frequency DEcreases), or an excess of generation vs. load (if the frequency INcreases). <

the frequency decrease from 50hz to 46.5 hz in 320 ms, then increase again to 50hz in 400 ms

Note: there is a load sharing system controls the load between both running machines. this system shares KW and KVAR, so the active and the reactive power are divided equally between the machines and every machine is loaded by about 30% of its rating.

S

#### saywhat

> 1. Why does the gen trip on bus overvoltage?

If genset is typified beforehand and not properly customized by its own loading time (which includes both modulating resistors and standard increasing commission) overvoltage occurs invariably. Overvoltage is good experiential data for any generator, synchronous or asynchronous alike. What causes the trouble is eliciting further functional expectation from it when it is already functioning beyond its secure typified charge (for the luck of the plant if the solution is acknowledged but for the luck of the generator if it is not).

> 2. Why does this trip accompanied by a very rapid frequency change which
> causes the UPS to disconnect the static bypass?

Light is never static, even in total absence, that is why.

> 3. Is there a relation between the rapid change in load caused by the heaters
> and the generator trips?

Yes, a direct relationship.

> 4. Is there a possibility to reduce the total running machines power factor