Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Synchronization of an isolated generator with load to grid 4
- Thread starter mbous
- Start date
Waross,
I dont anticipate your needing my help anytime soon! I'm certain I haven't got anything to enlighten you with, but was trying instead to fill in the blanks in the discussion topic.
I think we're coming at this from different angles, with different working defitnitions of real power control. I was a plant operator in the Navy and we had basically the same systems you're describing. Governors and Voltage regulators with a droop aspect to their control.
In your example the only way to vary kw load-sharing between the two gens is for the operator to adjust their governor speed settings. That means that the operators are performing the function of "real power controllers" by the definition I was using.
I am actually pretty surprised to hear you say that gens were left with nothing but droop control and operated in parallel for years without operator adjustment. I know that we would see imablances on the order of 10% develop between identical paralleled machines within about a day or so. I always chalked this up to slight differences in the percentage droop characteristics of the governor.
We're agreed that two similar machines can roughly share a varying load for a while if their droop and speed setting are the same. Let's remember, for the purposes of the discussion started in the OP, that they do this by varying system frequency. If the machines had a 5% droop, then no load system freq would be 5% higher than full load system frequency. Since the OP asked about paralleling to a grid, I'm assuming a sufficiently large grid that the frequency can't be changed by the machine in question. That means that the un-checked action of a droop circuit will either drive the machine to over-load or reverse power very quickly if left in parallel. The grid will give or take MWs many times the machine rating without changing frequency, so the droop governor can't share with it.
If mbous has well trained operators and direct reference control to a drooping governor, he can parallel with the utility, transfer load quickly, and then breaker parallel. This is the procedure used in the Navy. If he has a real and reactive power control (load sharing panel), he can remain in parallel indefinitely.
I dont anticipate your needing my help anytime soon! I'm certain I haven't got anything to enlighten you with, but was trying instead to fill in the blanks in the discussion topic.
I think we're coming at this from different angles, with different working defitnitions of real power control. I was a plant operator in the Navy and we had basically the same systems you're describing. Governors and Voltage regulators with a droop aspect to their control.
In your example the only way to vary kw load-sharing between the two gens is for the operator to adjust their governor speed settings. That means that the operators are performing the function of "real power controllers" by the definition I was using.
I am actually pretty surprised to hear you say that gens were left with nothing but droop control and operated in parallel for years without operator adjustment. I know that we would see imablances on the order of 10% develop between identical paralleled machines within about a day or so. I always chalked this up to slight differences in the percentage droop characteristics of the governor.
We're agreed that two similar machines can roughly share a varying load for a while if their droop and speed setting are the same. Let's remember, for the purposes of the discussion started in the OP, that they do this by varying system frequency. If the machines had a 5% droop, then no load system freq would be 5% higher than full load system frequency. Since the OP asked about paralleling to a grid, I'm assuming a sufficiently large grid that the frequency can't be changed by the machine in question. That means that the un-checked action of a droop circuit will either drive the machine to over-load or reverse power very quickly if left in parallel. The grid will give or take MWs many times the machine rating without changing frequency, so the droop governor can't share with it.
If mbous has well trained operators and direct reference control to a drooping governor, he can parallel with the utility, transfer load quickly, and then breaker parallel. This is the procedure used in the Navy. If he has a real and reactive power control (load sharing panel), he can remain in parallel indefinitely.
Hi JBinCA;
I pretty much agree with you. One misunderstanding. The machines were not left untouched for years. They were operated with droop control for years, but the machines were started, paralleled, and then the load picked up by governor setting adjustment.
In regards to running parallel with the utility on droop control, for example consider a machine with with 2 Hz. droop (3.3%)(60Hz. system).(Normal on many smaller diesels is 1.8Hz or 3%)
At full load the speed will drop 2 Hz. from the governor setting.
With the governor set to 60 Hz. the machine will not take any load.
With the governor set to 61 Hz. the droop will open the throttle 50% and the set will take 50% load.
With the governor set to 62 Hz. the machine will take 100% load. These are stable operating parameters.
However, a quadrature circuit is a minimum requirement to prevent excess reactive current if the system voltage varies.
The problem with paralleling with the grid is not frequency or real load. it is reactive currents.
If you parallel with a grid with stable frequency, droop control will maintain a stable KW output until their governor wears out.
However, the absorbtion or export of vars is determined by the excitation or voltage setting of the set relative to the grid voltage. If the excitation is too high relative to the system voltage you will export VARs. If the excitation is too low relative to the system voltage you will import VARs.
Either way you may overheat the machine with excess current.
A quadrature circuit compensates for reactive currents by biasing the voltage sensed by the voltage regulator. It is just a CT and a resistor and often the resistor is built into the Automatic Voltage Regulator.
If we meet, I'll buy the coffee and we will discuss generation systems late into the night.
respectfully
I pretty much agree with you. One misunderstanding. The machines were not left untouched for years. They were operated with droop control for years, but the machines were started, paralleled, and then the load picked up by governor setting adjustment.
In regards to running parallel with the utility on droop control, for example consider a machine with with 2 Hz. droop (3.3%)(60Hz. system).(Normal on many smaller diesels is 1.8Hz or 3%)
At full load the speed will drop 2 Hz. from the governor setting.
With the governor set to 60 Hz. the machine will not take any load.
With the governor set to 61 Hz. the droop will open the throttle 50% and the set will take 50% load.
With the governor set to 62 Hz. the machine will take 100% load. These are stable operating parameters.
However, a quadrature circuit is a minimum requirement to prevent excess reactive current if the system voltage varies.
The problem with paralleling with the grid is not frequency or real load. it is reactive currents.
If you parallel with a grid with stable frequency, droop control will maintain a stable KW output until their governor wears out.
However, the absorbtion or export of vars is determined by the excitation or voltage setting of the set relative to the grid voltage. If the excitation is too high relative to the system voltage you will export VARs. If the excitation is too low relative to the system voltage you will import VARs.
Either way you may overheat the machine with excess current.
A quadrature circuit compensates for reactive currents by biasing the voltage sensed by the voltage regulator. It is just a CT and a resistor and often the resistor is built into the Automatic Voltage Regulator.
If we meet, I'll buy the coffee and we will discuss generation systems late into the night.
respectfully
Waross,
Great post as always. I think you cleared up a misconception that I didn't realize I had. Of course it must be reactive currents that could rapidly go unstable without a quadrature circuit or other reactive power controller, because it's voltage that normally swings at an industrial plant's point of common coupling - not frequency.
What I remembered from my operating days was that gen currents would go unstable on a machine with only droop control of both governor and voltage regulator if left in parallel with a utility grid. For some reason, I attributed that to real power, but you're right that it's reactive power that's the culprit.
You're on for the coffee should we ever meet, but I'll buy because I'll be the one who gets the most from the meeting.
Best Regards,
JB
Great post as always. I think you cleared up a misconception that I didn't realize I had. Of course it must be reactive currents that could rapidly go unstable without a quadrature circuit or other reactive power controller, because it's voltage that normally swings at an industrial plant's point of common coupling - not frequency.
What I remembered from my operating days was that gen currents would go unstable on a machine with only droop control of both governor and voltage regulator if left in parallel with a utility grid. For some reason, I attributed that to real power, but you're right that it's reactive power that's the culprit.
You're on for the coffee should we ever meet, but I'll buy because I'll be the one who gets the most from the meeting.
Best Regards,
JB
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question