Previous in Forum: Head Turners   Next in Forum: About Yaskawa Speed control unit
Close
Close
Close
9 comments
Anonymous Poster

Unequal Load Sharing Between Parallel Gens.

10/14/2008 11:16 PM

Hi,

We are facing a unique problem.We have two identical 200 KW generators utilizing isochronous load sharing through Woodward's DSLC(DIGITAL SYNCHRONIZER AND LOAD CONTROL) module and 2301 A speed controller.The problem is that after a combined load of 250 KW,one alternator starts taking more and more load while the other sheds its load,meaning the load sharing no longer remains proportionate(or equal).Also the generator which is carrying a greater portion of the load runs with higher speed and does not unload at all even after 'unload' command is given to its respective DSLC module.Will be grateful if anyone could suggest where the problem is.

Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".
Anonymous Poster
#1

Re: Unequal Load Sharing Between Parallel Gens.

10/15/2008 11:43 AM

When you use isochronous governor mode, one of two generators will be master and other slave. I think you must apply governor droop mode with equal droop factor for equal load sharing.

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

Reply
Commentator
Hobbies - Fishing - New Member but not a new fisherman Hobbies - Target Shooting - New Member

Join Date: Apr 2008
Location: Formerly from NJ USA Planet Earth. Relocated to Space Coast FL
Posts: 79
Good Answers: 9
#3
In reply to #1

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 12:06 AM

Agreed - We generally run one of our 3 TGs in "droop" and the other 2 in "isoch" and load following is not a problem. Ed

__________________
If you understand no explaination is necessary. If you don't understand no explaination is possible.
Reply
2
Power-User
Hobbies - HAM Radio - New Member

Join Date: Oct 2008
Location: Houston, Texas
Posts: 252
Good Answers: 5
#9
In reply to #1

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 6:18 PM

The scheme he is using allows both machines to run isochronously and share load. The problem is in the load balancing.

In an isochronous conventional utility setting, one would run these machines in one of two ways. He is having trouble with a system operating the third way.

1. One machine in isochronous (the line engine) and the other(s) in droop. The line engine takes the load swings to maintain frequency and the other machine(s) is/are base loaded. This is used in utilities with fairly stable loads and allows good frequency control. it also allows the base loaded machine(s) to move very little. The draw back to this scheme is that the operator will have to adjust the base loaded machine(s) to keep the line engine within its control range.

2. Both (all) machines in droop. This works well when the grid is very large or very stable. As the frequency falls, all machines pickup load at their droop settings and when the frequency rises, all machines shed load at their droop settings. This makes the load share across all machines. The frequency however, will vary with the load and must be adjusted by the operator by changing load settings on some or all of the machines in the network.

3. On offshore platforms, island systems or other small isochronous networks which are subject to large load variations, i.e. more load change than one machine can handle by itself without operator intervention, this parallel isochronous system is used to allow frequency control and load ballancing to move all machines on the network through their full range of load while maintaining good frequency control. Woodward, Triconex , Tri-Sen www.TRI-SEN.com and perhaps others can use this third option where it is needed. I know of one system where Tri-Sen (www.tri-sen.com) and Woodward controllers were used successfully in the same network. The beauty of this system is that the frequency is stable over large load swings and the operator can take his readings, watch for alarms and go home on schedule.

Have FUN!

Turbotroll3

__________________
If the software can detect, compensate, avoid, or correct an anomalous condition in the system, it is, by definition, a software problem-regardless of the root cause. In the long run, for most classes of problems, it is cheaper to fix it in the SW
Reply Good Answer (Score 2)
Anonymous Poster
#2

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 12:03 AM

when ever the d.g. start, check for its voltage w/o load & compare with other & then check with full load, & then compare the voltage, i m sure u will find which d.g. is droping voltage & need AVR setting or pump filter cleaning

Reply
Anonymous Poster
#4
In reply to #2

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 1:28 AM

Say what?????????????????????????

Reply
Power-User

Join Date: Oct 2008
Location: Meherrin Virginia
Posts: 319
Good Answers: 6
#5

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 8:06 AM

The isochronous mode of load sharing is sensing load. The mechanical solution to load control was to control speed, The governors were designed to droop a certain percentage of speed from 0 to 100% load (modern US systems droop about 5%), The load is equivalent to 0 to 100% valve movement. This says that if you were running at full speed no load and you immediately loaded the machine to full load the valves would open fully and the speed would decay to 95% rated speed. It follows that if 50% load is added to the machine the valves will open 50% and the speed will decay to 97.5 %. Notice that this doesn't take into account what the machine is rated for. What happens in a real system is the system is so large the speed gets brought back to normal by different governors being bumped back up to speed.

When a unit is placed in isochronous mode that indicates that the speed droop (regulation in the turbine business) will be 0% for 100% load change, which says that if you apply full load to a machine running at full speed no load the valves will move to 100% open and the speed will not change. That works fine with one machine, with two of them no go.

__________________
If you fail to follow through, you will fail.
Reply
Anonymous Poster
#6

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 12:11 PM

Bisically Load sharing depends on Load angle / Power angle. if is high then load sharing is also more and vice versa.

apart from this if speed is praportional to load sharing

if speed is more then load sharing is also more.

in your case also which ever gen set is running with higher speed that will take more load comapare with other genset.

so mainly adjust the speed it will solve u r problem

still having any clarificatin can contact at : venu24502@yahoo.com

u r reply will induce me great honour on u

Regards,

M.vnekata Narayana

Reply Off Topic (Score 5)
Power-User

Join Date: May 2007
Posts: 127
Good Answers: 6
#7

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 12:22 PM

Due to the power curve of the drive motor, running one gen at full capacity gives best efficiency. Your load share device is doing what should.

Mr. Gee

Reply
Power-User
Hobbies - HAM Radio - New Member

Join Date: Oct 2008
Location: Houston, Texas
Posts: 252
Good Answers: 5
#8

Re: Unequal Load Sharing Between Parallel Gens.

10/16/2008 5:40 PM

Check the voltage/current input to the governors on the load (KW) and the parallel line signals. The governors should match the machine's percentage of the load (KW/MW) output to the average system load, AKA, the parallel signal.

If for some reason the parallel line signal does not match the (KW/MW) load on a given machine, the governor for that machine will droop on the difference between these two signals, that is, If I remember/understand how Woodward does it.

The machine with measured Load (KW/MW) higher than the parallel load signal will tend to lag the system and the machine with the load lower than the parallel signal will tend to lead the system.

I suspect that you may have a ground loop between your governor and the load matching module on at least one machine or a ground loop through the two load modules. You should be able to check the Load and Parallel signal DC voltages to common and to ground on each of the systems and find the one with the funny readings. It is also possible that an opamp has failed in one of the load control modules and is giving one governor wonky signals causing the problem.

This can be a sticky problem as the problem can be in either load control module, either governor or in the wiring.

Have FUN!
Turbotroll3

__________________
If the software can detect, compensate, avoid, or correct an anomalous condition in the system, it is, by definition, a software problem-regardless of the root cause. In the long run, for most classes of problems, it is cheaper to fix it in the SW
Reply
Reply to Forum Thread 9 comments
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".
Copy to Clipboard

Users who posted comments:

Anonymous Poster (4); ed fuzzyE kedz (1); Mr Gee (1); otha (1); Turbotroll3 (2)

Previous in Forum: Head Turners   Next in Forum: About Yaskawa Speed control unit

Advertisement