Problems With Toshiba/mototronics Mvc3 Soft Starter

[ahawkins]

Hi there, this is my first time posting here, but not my first time to visit this board. I have gained a lot of insight into drive and soft start theory and operation by reading the postings left here over the past few years.

 

Now I have a problem of my own, one with a Toshiba/Motortronics built JKSSS+/MVC3 soft starter. The starter drives a 1750 HP 2300V 3600 RPM motor, which drives a pump used to cut coke out of process drums at the refinery using high pressure water. The reason for the use of the soft starter is to limit the affects of water hammering on start up of the system, as water is moved hundreds of feet from ground level up to the top of coker cutting derrick. To date, we have two soft starters installed on two 1750 HP motors. The system setup has proven to work on both pumping systems, but not without problems.

 

In the case of one of the pumping systems, we have experineced erratic trips on the soft start system, with the controller reporting a short circuit trip running the startup operation of the soft starter. Sometimes the soft starter will successfully accelerate and startup the motor for 3 to 4 successful attempts, and the it will fail to startup the next time. When soft starter problems occur, we place the starter in bypass and start the motor DOL. We are able to still do this within the system, as we have a three-way motor operated bypass valve installed on the line, which will slowly fill up the line and bypass a portion of the water flow back to the holding tank. Once the line is filled with water, a limit switch in the line is made, which tells the valve to go to full flow. This is used as opposed to using soft start on the motor. The valve currently is use today does not suffer from leakage issues, thus the valve does not affect soft start usage during startup when the valve is effectively disabled from the system. Thus we can switch between the two system setups, but it requires disabling the three-way valve before operating soft start, as the more reliable mode of operation today is still to use the three-way bypass valve.

 

I have performed maintenace checks with our plant technicians, following the troubleshooting procedures outlined in the maintenance manual, as well have had a Toshiba service engineer come on site to take a look at the equipment. Static checks have been performed on the SCR stacks (2300V, 600A sized), as well as having made fiber optic cable connections verified as properly connected and output signals from the controller boards verfied as being within tolerance. None of the SCR stacks were shorted, with resistance values falling within tolerances. We had experienced failed startup soft starts and successful soft starts with a device setting of 50% initial voltage with 500% FLA current limit, and a 20 second acceleration time on the motor. Note, I only requested Toshiba field service after I was able to simulate reoccuring trips during system startup before asking for assistance. During his visit, he found nothing wrong with the physical equipment or with the SCR components. We were able to successfully run a soft start, with water slowly filling the line with no water hammering occurring, and the motor was able to successfully accelerate out of the 20 second soft start without tripping.

 

Also, the motor and feeder cable systems have also been checked. The motor is just over one year old, being in service since May 2007. The motor was specifically built to handle loads at reduced voltages and for a longer acceleration time. At the time the motor was installed, the feeder cable was meggered and a PdMA MCE test was performed. The cable passed the meggering, and the PdMA test indicated no resistive imbalance on the cable or insulation problems. And when the starter is put into manual bypass and the motor starts DOL, we have not experienced a short circuit trip from the starter (blowing one of the main line power fuses).

 

The only thing on the system setup that is out of place is voltage imbalance measured per phase by the digital controller. There is a 2 to 3 percent imbalance between minimum and maximum values, but nothing high enough to cause a current imbalance trip. Could this cause a reported short circuit trip on the soft starter? Or could it be a load issue with the pump system? Each time the pump starts, there is a varying amount of water left in the piping, so I assume that the pump is pushing against varying pressure every time the pump starts. Could this affect soft starter performance? Could it be a failed controller board as well? We have replaced the programmable control panel before, but not the main controller boards before. As it has been discussed previously with Motortronics MVC3 products (which is the Toshiba device that we are pretty much using), the device firmware changes often. Could this be a problem that has been fixed in a later firmware update, and I just do not know about it?

 

As the performance of the soft starter has remained erratic, we have kept using mechanical means to slowly fill up the piping and avoid water hammering. The ultimate goal is to still get away from the use of this three-way valve, as this is a costly annual maintenance expense to pull and shop the valve, and operate without the three-way valve as we have done with the other cutting water pump system.

 

If any of you could provide some insight or informaiton from past experiences that may be able to help me out, I would greatly appreciate your advice. Thanks.

 

Andrew

[marke]

Hello Andrew

 

Just a few questions that may help to isolate the problem.

1. Is this problem occuring with one starter only, or are you seeing similar characteristics on both?
2. Is the trip occuring partway through the start, or the moment that a start is called for?
3. How is the line contactor being controlled?
4. If the two starters are located together, is it possible to swap the output connections so that starter 1 supplies motor 2 etc? may help to eliminate starter or motor.

Best regards,

Mark.

[jraef]

You also didn't mention how old the soft starter is. At one time, Motortronics had some bad components on what are called the Temp/CT boards. The Temp/CT boards collects the status of the temperature sensors and the current from the CTs for each SCR stack, converting it to a fiber-optic signal to be sent to the mP. It alternated between sending current values and temperature values in order to cut down on the component costs, because it isn't imp[ortant to send the temperature very often.

 

The problem they had was that the the on-board power supply would fail intermittently and in failing, the values sent to the mP go high. This was/is an indirect form of "fail safe", meaning that instead of NOT reading current that was there, it reads values of 65536 on the registers so that the unit trips off line on ridiculous values, such as by saying that the heat sink was 1000 degrees C, or the current was something like 60 times the normal peak values. If the board power supply winks out while transmitting the CT value (which is most of the time), you were likely to get the ridiculous value during a CT value transmission, which the starter would sometimes interpret as an instantaneous Short Circuit.

 

If, however, you were getting something such as upstream fuses blowing, then this would not be the case. The Temp/CT board issue had nothing to do with real currents, it was essentially a "false positive" situation.

[ahawkins]

Mark,

 

To answer your questions below.

 

Is this problem occuring with one starter only, or are you seeing similar characteristics on both?

To date, this problem is only occuring with one starter, and not the other.

 

Is the trip occuring partway through the start, or the moment that a start is called for?

 

The trip is occuring during the soft start, and not at the moment that the start command is given and soft start commences. When the start signal is given from the field, you will hear the line contactor close in and the SCRs begin to fire shortly thereafter. The trip occurs at varying times during the soft start ramp period. This has made me feel like it is some kind of controller board issue.

 

How is the line contactor being controlled?

 

The line contactor is controlled via remote start from the field to the main controller board. There is only start signal that can operate the soft starter.

 

If the two starters are located together, is it possible to swap the output connections so that starter 1 supplies motor 2 etc? may help to eliminate starter or motor.

 

 

 

I would like to think that this would be a feasible troubleshooting option, but the setup time to perform this test would require both motors/pumps to be shutdown for a prolonged period of time. The operating units cannot have both machines out of service at the same time. I am also afraid to tamper with the working soft start setup on the working setup, as causing a working setup to fail would not help the current situation.

 

Andrew

[ahawkins]

jreaf,

 

Sorry for leaving that information out in my first post.

 

You also didn't mention how old the soft starter is. At one time, Motortronics had some bad components on what are called the Temp/CT boards. The Temp/CT boards collects the status of the temperature sensors and the current from the CTs for each SCR stack, converting it to a fiber-optic signal to be sent to the mP. It alternated between sending current values and temperature values in order to cut down on the component costs, because it isn't important to send the temperature very often.

 

The two soft starters were built in the 2002/2003 time frame, with the starters commissioned in mid-2003. Several months later, the SCRs in both units were replaced with the largest stack size that can be installed at 2300V (went from 400A rated stacks to 600A rated stacks), due to blown stack issues on commissioing of the two pump motor starters (blown stacks occurred on the motor starter that is currently working without problems today, and the stacks in the other motor starter were subsequently upgraded as well). So the stacks in each starter are about the same age and built it 2003.

 

The problem they had was that the the on-board power supply would fail intermittently and in failing, the values sent to the mP go high. This was/is an indirect form of "fail safe", meaning that instead of NOT reading current that was there, it reads values of 65536 on the registers so that the unit trips off line on ridiculous values, such as by saying that the heat sink was 1000 degrees C, or the current was something like 60 times the normal peak values. If the board power supply winks out while transmitting the CT value (which is most of the time), you were likely to get the ridiculous value during a CT value transmission, which the starter would sometimes interpret as an instantaneous Short Circuit.

 

If, however, you were getting something such as upstream fuses blowing, then this would not be the case. The Temp/CT board issue had nothing to do with real currents, it was essentially a "false positive" situation.

 

This is the register value that I have seen on several of the trips. Like I said, we have not been blowing upstream fuses, and all electrical equipment checks showed nothing wrong with the motor or feeder cables. If you were to lose a CT value coming from one of the Temp/CT boards, could that also cause an induced ground fault trip as well? What has happened on several of the reported short circuit trips from the soft starter controller is if you check the three phase and ground circuit current values at the time of the trip, a high ground current is listed, and low or no phase current values are listed, as well as those phase currents do not properly sum to the listed ground fault value, yet the device is still tripping out on a reported short circuit trip, not a ground fault trip. After checking all SCR stacks for shorts, I am convinced that we are not seeing a true ground fault, so I believe that it is a controller board problem of some sort.

 

With all that said, what is the repair path to persue? Has there been a redesigned Temp/CT board that is more fault tolerant, do the ring transformers need to be replaced/calibrated/adjusted, or do the supply transformers need to be checked or replaced? Are these transformers highly sensitive to phase voltage imbalance? The phase voltages are slightly imbalanced, but the phase magnitudes are slightly above 2400V at all times, so there should not be a voltage drop problem of any sort.

 

 

Andrew

[jraef]

jreaf,

 

Sorry for leaving that information out in my first post.

 

 

 

The two soft starters were built in the 2002/2003 time frame, with the starters commissioned in mid-2003. Several months later, the SCRs in both units were replaced with the largest stack size that can be installed at 2300V (went from 400A rated stacks to 600A rated stacks), due to blown stack issues on commissioing of the two pump motor starters (blown stacks occurred on the motor starter that is currently working without problems today, and the stacks in the other motor starter were subsequently upgraded as well). So the stacks in each starter are about the same age and built it 2003.

 

 

 

This is the register value that I have seen on several of the trips. Like I said, we have not been blowing upstream fuses, and all electrical equipment checks showed nothing wrong with the motor or feeder cables. If you were to lose a CT value coming from one of the Temp/CT boards, could that also cause an induced ground fault trip as well?

Yes, very possible. The MVC uses the Residual Current method of Ground Fault sensing, so if one phase reads extremely high or extremely low, it could interpret that as a GF. it really ends up being kind of a race as to which trip function it goes out on first.

 

...

With all that said, what is the repair path to persue? Has there been a redesigned Temp/CT board that is more fault tolerant, do the ring transformers need to be replaced/calibrated/adjusted, or do the supply transformers need to be checked or replaced? Are these transformers highly sensitive to phase voltage imbalance? The phase voltages are slightly imbalanced, but the phase magnitudes are slightly above 2400V at all times, so there should not be a voltage drop problem of any sort.

 

 

Andrew

 

Yes, they were continually redesigning the Temp/CT boards, so most likely they have a fix available for your unit. You would need the serial number so they can track the firmware version of the microprocessor it goes with since there may be issues with compatibility, or at least get the firmware number from the display function; it's in there. being that they were bought so long ago, you may not be able to get them to provide them for free, but they are not terribly expensive. I would get spares too if I were you.