DC Motor


Thread Starter


we are operating GE frame 9 gas turbine. During 88QE DC motor availability checks during startup, the current rises below set point for the high current set point alarm which is normal. But when the speed begins to build to normal, the current drops to 12amps and gradually builds to 28amps AND REMAINS THERE which is below 30amps limit for low limit alarm set point. This prevent startup of the turbine.

I have checked the starting resistor the readings are ok. I have checked the transducer it is ok. Please can it be brushes, commutator, field, or armature winding problems? The motor type is LSK1804 cvl,75KW,125VDC.

Please help.
If your motor FLA is 600A and you're running at 28A, the motor is not loading for some reason. That's what the control system is objecting to. I question your motor data, though. 125VDC seems like a very low supply voltage for a 75kw motor. I would have expected 300VDC or 600VDC.
Steve Myres is probably correct--the motor is not producing pressure or flow. Have you checked the coupling between the motor output shaft and the pump shaft?

What is the L.O. pressure during and after the start-up of the pump?

Most Emergency L.O. Pump motors on GE-design heavy duty gas turbines are 125 VDC motors, because the battery for the emergency pump motors and the turbine control system is usually 125 VDC.

But with ANY electric motor, the current drawn by the motor is directly related to the power the motor is providing to the device it is powering.

Are you sure you are not monitoring the field current, instead of the armature current?

How did you confirm the current transducer is working correctly?

Is there L.O. in the tank when you are running the 88QE motor?

And, again, is the coupling installed, and is it firmly connected to the shafts?

If this is an usual application where your motor has a constant field current, then the motor armature current will be directly proportional to the torque produced. My guess is that your turbine is not loading your motor enough, and your low limit current protection is acting correctly.

Now, I don't know what you are using a starting resistor for. If the resistor is utilized for field weakening, then it probably remains bypassed and this causes the low current issue.
The typical GE-design heavy duty gas turbine Emergency L.O. Pump Motor does have a constant field current applied to it, and uses a two-step starter which has resistors in the armature circuit to step up the armature current over a few seconds as the motor is started. There is also a rheostat that is used to adjust pump speed/pressure when full voltage (125 VDC) is applied to the armature. The application of constant field current keeps the motor ready for duty, helps to drive off condensation and reduces the possibility of overspeed of the DC motor due to loss of field current. Stepping up the armature current reduced the initial load on the 125 VDC battery when the Emer. L.O. Pump hss to be started.

The potential problem is that GE has given responsibility for the design of the Frame 9 heavy duty gas turbines and auxiliaries to the division in Belfort, France, and the Belfort Bunch have been steadily changing nearly every aspect of the design of the auxiliaries from proven and time-tested designs which were used for decades. Previously, there was no "transducer" in the DC motor starter to monitor current in any circuit. And there was no logic in the turbine control system to monitor low current; there was only a simple, single thermal overload relay to alarm on excessive armature current. The intent was to keep the <b>Emergency</b> L.O. Pump control scheme--every aspect of it--as simple as possible to make it as reliable as possible. There was an undervoltage relay to prevent a START if there was no 125 VDC power for the Emergency L.O. Pump motor, and an overload relay to alarm if the armature current was excessive. And an alarm to indicate if the Emergency L.O. Pump motor was running (which prevented synchronization). Simple and uncomplicated. But, GE Belfort cannot leave things simple or uncomplicated.

GE Belfort has a proven history of over-complication of simple and proven philosophies and desgns. Left unchecked, they have repeatedly shipped needlessly complicated systems and logic to their Customers, with horrendously increased numbers of alarms and misleading alarm text messages--with little or no documentation and poor factory support for resolving the issues caused by the over-complication. While the intent is good, the testing and implementatio is not, and once in the field it's very difficult to get the factory engineers to correct or acknowledge design flaws and to simplify for the benefit of the Customer and GE.

Adding to the problem is the use of poorly trained people to install and commission these over-complicated systems and schemes, leading to all sorts of perceived problems, many of which result in equipment failures and schedule delays. Couple this with poor factory support and this results in posts like this to World Wide Web foruns looking for help with self-inflicted problems caused by lack of experience and training and lack of procedures (which all divisions of GE power generation engineering are guilty of). So, many Customers are left with systems and "logic" that just don't work. And they can't get support from the factory.

Without the ability to see the DC motor starter schematics and the application code in the Mark VIe we will be unable to provide any other information. We have no way of knowing how the Emergency L.O. Pump was being tested, and if the test was realistic or not. It's entirely likely the DC motor was purposely uncoupled from the pump for some kind of "check" and the alarm text messages and "logic" are causing confusion. And, commissioning personnel are not trained or experienced in how electric motors work, and are having a difficult time getting GE Belfort factory support because GE Belfort engineers don't believe the system can be over-complicated or designed incorrectly.

We'll likely never know how this is resolved. Unfortunately.

Finally, at 746 Watts per horsepower, a 75,000 Watt (75 kW) motor is a 100 horsepower motor, which is a little big for an Emergency L.O. Pump motor for a GE-design Frame 9E heavy duty gas turbine. 7.5 kW (10 horsepower) is more likely. As Steve Myres said, 600 FLA (Full Load Amps) is much more than most 125 VDC batteries could provide; 60 Amps is much more reasonable.
CSA and the big guys thanks for your response.

The rheostat is a a two step armature current control during start up since the armature resistance is very small. The resistance are control by timers and I have checked the timers controlling the resistance they are ok.

For the current transducer I used a DC clamp meter and a new transducer to measure the armature current. the result were the same.
The L.O level in the sump is Ok.

This is why I needed more technical analysis concerning field current, armature current, commutator if any of this relationship can cause the armature current to be low. Though I have not checked the coupling.

In case any of the contributors have access to the markv code the transducer signal name is 96QE. the low current level is l51QE.


The current drawn by ANY electric motor is directly proportional to the work being done by that motor. In the case of the DC Emergency L.O. Pump Motor, driving the Emergency L.O. Pump, that would be the pressure/flow produced by the pump when the motor is running.

It's pretty doubtful there is something amiss with the Mark sequencing; digital control system setpoints don't drift or change unless someone changes them (knowingly or unknowingly).

Also, it would be a good idea to have a look at the L.O. System P&ID and see if there are any check valves or foot valves that might not be working properly in the L.O. system. GE, decades ago, used to teach something called the "Binary Troubleshooting Method." Basically it's premise was: When faced with a problem, you "divide" it into two basic parts (say, in this case, controls and mechanical parts). You make your best estimate of where the problem may lie and then you start further dividing that part into successively smaller parts. If you find the problem, that's great! If you don't find the problem in the part you started with and you're satisfied you have eliminated the possible causes in the part you started with, you move to the other part, and do the same thing: divide it into successively smaller parts until you arrive at the problem. It has worked many times for me when I'm baffled as to what the problem might be. In this case, you have checked some of the controls-related issues and feel you have eliminated some of them. You can pretty much eliminate the Mark* portion--again, because digital control system setpoints don't drift. So, start focusing on some of the pump and mechanical system issues, as well as the motor issues, and start eliminating them--until you find the problem.

Be sure to write down the tests and checks you perform <b>AND</b> the results of the tests and checks! Sometimes just the act of writing things down will cause you to say to yourself, "That wasn't really a valid test/check. I need to revise my thinking on that and re-test or re-check."

Do you have any archived data about what the armature current value was prior to the problem starting?

You were specifically asked:

>What is the L.O. pressure during and after the start-up of
>the pump?

There is zero substitute for going out to the Accessory Compartment and watching the motor and pump, and observing the pump discharge pressure gauge. None.

Knowing what pressure the motor/pump is developing when it is running would be extremely helpful when trying to understand the current being drawn by the armature of the motor. If the pressure is lower than it should be, that would correspond to a low armature current draw. However, if a check valve was not sealing properly somewhere in the L.O. system and allowing Emer. L.O. Pump flow to "leak" somewhere it shouldn't that could be cause for problems, also. Or, if there is a foot valve in the suction of the Emer. L.O. Pump that wasn't allowing enough suction that could also be a cause of low pressure/flow/armature current.

What happens when the Aux. L.O. Pump is started and runs? What is the pressure of the Aux. L.O. Pump discharge, and what is the pressure of the L.O. bearing header?

Try running the Emer. L.O. Pump when the Aux. L.O. Pump is NOT running (and when the unit is at zero speed, and Cooldown is OFF). And monitor the Emer. L.O. Pump discharge pressure, the Aux. L.O. Pump discharge pressure, AND the Main L.O. Pump discharge pressure. If either or both of the Aux.- and Main L.O. Pump discharge pressures increase when the Emer. L.O. Pump is running, that could be an indication of problems with check valves or something like that.

If the Emer. L.O. Pump commutator brushes are good and don't need to be replaced, and the Emer. L.O. Pump motor starts and runs up to proper speed (which is to be determined by the L.O. pressure <i>at the collector end of the generator</i> (the furthest point away from the L.O. pumps)) AND you know the coupling between the Emer. L.O. Pump Motor and the Emer. L.O. Pump shaft is good, then that could be a problem with the Emer. L.O. Pump itself (the impeller has come loose from the shaft, or the foot valve (if present) is not allowing sufficient suction flow), or something similar.

Also, try rotating the Emer. L.O. Pump shaft by hand. If it doesn't spin relatively easily it could be a pump bearing problem, though that would usually cause the motor armature current to be higher than normal. If it spins too easily, again, there could be a problem with the pump impeller coming loose from the pump shaft.

It's not uncommon for the set screws of couplings to loosen over time, and it could be as simple as the set screw of either the motor or the pump half of the coupling is not securely screwed to the shaft and the motor is not transmitting any torque to the pump impeller.

Centrifugal pump discharge pressure and "resistance" to flow are somewhat related; so, if there's no resistance to the discharge flow of the pump--meaning that the discharge flow is somehow "leaking" to somewhere it should not be, and doing so "freely"--that could be a problem. Again, a loose L.O. piping fitting, or a non-working check valve, could be the problem.

Most of the L.O. Pumps I have seen on GE-design heavy duty gas turbines have a "loop" of piping next to the pump motor coupling area which has a plugged pipe tap on the top of the loop. (The piping loop is sometimes a 3- or 4-inch 180 degree loop on the top of the pump assembly.) You can remove the pipe plug and install a gauge (temporarily or permanently) in the tap to observe and check against the gauge on the Accessory Gauge cabinet.

Yes; worn DC motor commutator brushes could cause a problem. Usually the motors have access covers on opposite sides of the main body of the motor (which is a vertical motor) which can be used to check the condition of the brushes. Do this CAREFULLY: With the brush covers removed, apply power to the DC Emer. L.O. Pump Motor Starter (which applies power to the field circuit of the motor). Then, with a face shield on two people (one on each side of the motor looking at the commutator), have a third person TEST run the DC Emer. L.O. Pump from the motor starter (there's usually a TEST button on the door of the starter). Look for sparks and arcs, especially on the commutator under the brushes--there SHOULD NOT be any! If the unit was originally provided with a Mark V, then it's probably time to replace the DC motor commutator brushes anyway. And, it can't hurt. Once the power to the DC motor is locked-out and tagged-out and verified to be off, it's usually a pretty simple matter to remove the brushes for inspection--of the brushes and the commutator beneath the brushes. Spin the motor shaft slowly to check all of the commutator segments and insulation segments between the wedges. They should be shiny and smooth.

What is the RPM of the pump motor when it should be running at full speed (the actual, measured RPM--not the nameplate RPM)? (When the timers on the stepped starter sequence have timed out?)

Have you checked the current being drawn by the field circuit? If so, what is it? What is it when the motor is not running, and what is it when the motor is running at proper speed?

Have you measured the field circuit resistance? If so, what is it?

A LOT of people are probably blaming the Mark V because it is preventing the turbine from starting because of the issue with low current during test running the Emer. L.O. Pump during a START. But, this is just exactly what the purpose of the test run of the Emer. L.O. Pump is for--to bring problems with the Emer. L.O. Pump, and it's motor, or related issues (check valves; foot valves; couplings; wiring issues; etc.)--to the attention of the operators and maintenance staff BEFORE the Emer. L.O. Pump NEEDS TO START AND RUN TO PROTECT THE TURBINE AND GENERATOR BEARINGS! Yes; the problem is being annunciated by the Mark V--but, as can be seen from the descriptions above (and all of the responses to this post) there can be MANY things which can cause problems with Emer. L.O. Pump operation which can be reflected in the current being drawn by the DC motor driving the pump.

So, if you're confident that the current input to the Mark V is correct, I would say forget about the problem being in the Mark V. And go find out what is causing the problem outside of the Mark V. Again, digital setpoints (as opposed to analog potentiometers used in the olden days!) don't drift. And, digital sequencing doesn't change. The likelihood of a problem with the Mark V is somewhere between slim and none--unless there are Diagnostic Alarms you haven't told us about.... So, while it's the Mark V that's preventing a START, and you are confident the input to the sequencing in the Mark V is correct, it's most likely NOT something in the Mark V. It could be something in the DC motor starter, but, realistically, that's not too likely, either.

The current drawn by any electric motor (AC or DC) is a function of (is directly related to) the work being done by the device the motor is driving. If the current is higher than normal or higher than it should be, then something is wrong with the motor or the bearings or the device the motor is driving, or the device is somehow being overloaded. If the current is lower than normal or lower than it should be, then something is wrong with coupling between the motor and the device it is driving, or the device isn't performing the work it was designed to perform. It's most likely something mechanically wrong with the pump, or the coupling between the pump and the motor, or some valves in the system are not working correctly--but it's not likely the problem is caused by the Mark V.

Hope this helps! Thanks for writing back, and please let us know how you fare in resolving the problem. If you write back for further assistance, PLEASE provide the answers to the questions asked (even if you, or someone at your site, don't think the questions are relevant). And, when telling is what you've done--PLEASE tell us HOW you've done it and what, specifically, the results of what you did were. Just telling is you checked something and it's okay doesn't really tell us if it was checked correctly and what the results were.