Problem on Cage Motor with Vector Drive 3

[sunrays]

I recently changed a 55kW 750RPM FLC: 101 Amps, Slip Ring Motor with Rotor Resistance Controls with a 55kW 750 RPM, 101 Amps Squirrel Cage Induction Motor and a Closedloop Vector Control Drive. At full speed and maximum torque requirement the current drawn by the motor was to be 200 Amps i.e. 200% of its FLC. The drive is also of 55kW with a continuous current of 125 Amps & 150% for 60 Sec. What is the reason for the abnormal increase in motor current as compared to the Slip-Ring motor. Could some one help on this?

 

[DickDV]

First, sunrays, I don't see any specific mention of abnormal currents in your post. Did you leave them out?

Second, it would appear that your drive is too small. 150% of 125amps is only 188amps unless the drive has a higher 5sec rating that covers your requirement.

Please, a little more info.

 

[Barry1961]

What is the "full speed and maximum torque requirement"?

Barry1961

 

[electricpete]

Calculate FLA

I = P/sqrt(3)/V/PF/EFF

If I choose V=460V, PF=0.8, Eff=0.85, I get 101A

I=55000/SQRT(3)/460/0.8/0.85=101A

What's the voltage?
What's the question?



=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

What is the maximum torque requirement?

It is possible and likely that the squirrel cage will not give you as high of a torque.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Also at what speed does this peak torque requirement occur?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[aolalde]

Sunrays

Post the load type, voltage and current to the driver, voltage and current to the motor and operating frequencies and motor shaft speeds when the abnormal current was measured.

 

[jraef]

As we await your answers, I hope that you realized when you selected a replacement system that Wound Rotor motors tend to produce more peak torque than Squirrel Cage motors. While this is not an abolute, your posted experience would tend to make me think you may need a larger SC motor and VFD than what you pulled out.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[sunrays]

Let me explain the application. This is for a wagon tippler with 2 nos. of 55kW Slip-Ring Motors driving a common shaft. The speed changes for the clamping and declamping of the wagon to the system is done by resistance cutting. The total operation of the system is to rotate the wagon from 0 - 175 Deg. and back. The electric controls is to be replaced with a 2x55kW, 415V SC Motors with Vector Drives is closedloop. The peak torque required is from 90 - 175 deg. when the mecanism moves against gravity. During this period the SC motor was found to stall even though the currents recorded in the drive was around 200Amps. The overload setting in the drive was increased from 150% to 200% with no effect.

 

[ozmosis]

Apart from the fact that the system could be under-sized, there is a possibility that as you have two independant drives both in their own closed-loop control, but driving the same shaft, they could be 'fighting' against each other and the increased load could be caused by the two drives rather than the actual load. Tyically, closed loop vector drives are designed for high performance, instant reactions etc so if there is any sort of delay between starting the two drives, the error created would be corrected by the other drive and hence the possibility of working against each other.
This is assuming I have understood your system correctly. I take it you do have two drives running their own individual motor? You mention motors (plural) and then you mention the "SC motor was found to stall" in the singular. Have you monitored both motors at the same time to determine if the loads are balanced?

 

[sunrays]

Thanks sed2developer.

You are correct in that there are two drives (both in closed loop) running individual motors, each motor connected to a gearbox and both the gearboxes coupled to a common shaft. One of the drives is the master and the other works on the current controlled slave. The currents and speed profiles of each of the drives was monitored independently. There was no delay in starting of one of the drives and the load sharing by both the motors and drives was equal. The mechanical response of the system was also good. Any sluggish response in the eclectics will be reflected in the connected mechanicals which was not the case.

I wont give in to the argument that the system could be under-sized. The performance of a slip-ring motor can be matched by a cage motor & drive combination.

Has any one attempted to change a slip-ring motor with a cage motor and a drive? If so I would like to know the motor sizing and the application.

 

[ozmosis]

sunrays
there's a section on the following website from Imphotonics

http://www.lmphotonics.com/slipring.htm

:
"Starting a high inertia load with a standard cage motor would require between 400% and 550% start current for up to 60 seconds. Starting the same machine with a wound rotor motor (slip ring motor) would require around 200% current for around 20 seconds"
increasing the overload to 200% on a vector drive will do little. This overload will be time related and probably rated only for a few seconds, probably not enough to overcome the inertia of your system +gravity.
Has your gear-train ratios remained the same or has this changed?

 

[DickDV]

OK, sunrays, I am going to take a flyer on this one and make a few assumptions. Correct me if I assume wrong.

First, it sounds like the new system is a pair of VFD's feeding a pair of motors that are mechanically tied together thru a common shaft. Whoever configured this system has wisely set one drive/motor as a speed regulator with the other set arranged as a torque follower. Given this arrangement, it is essential that the two motors are sharing load equally especially when at the heavy load point on the swing arc of the machine. It sounds like you have checked this but you must be sure about it.

Second, since we don't have precise torque-speed curves for the new motors we have to assume that peak torque on the motors is around 200% nameplate rating and that the current at that torque is also about 200% of nameplate full load amps. If this is right, the the VFD's must be able to supply that 200% current for as long as the overload exists. Since you said you have increased the current limit on the drives to 200% (of 125amps, I presume), it sounds like you have already anticipated this issue. However, VFD's, just like motors, have a time limit for operation at these overloads and that time limit may not be enough to get your machine thru the high-load range. A good indicator of the weakest link would be whether the drives fault when you say the system "stalls". If the drives do not fault, I assume that the motors continue to draw heavy currents, the drives drop down to near zero Hz, and the motor shafts stops turning. I wouldn't think that this condition could go on for very long before either the drives fault or the motors go up in smoke!

sunrays, if you are with me up to this point, please report back to us the output voltage, output current, and frequency of each drive when stall occurs. Also, if the drives fault when stalled, please tell us what type of drive fault occurs---ie, overcurrent, overtemperature, etc.

I'll do my best to help you thru this when you report this data for us.

Good luck.

 

[CJCPE]

With the slip ring motor, you can operate up to the breakdown torque of the motor with something like 200% of rated current. With the vector drive, you may not be able to get as close to the breakdown torque of the motor and will probably need more current to get the same torque as you approach the breakdown torque.

The drive should automatically adjust the motor voltage to the optimum voltage for the frequency and load. However, the optimum voltage may be more than the drive can put out when operating at maximum frequency. You may be able to get more torque if the frequency is set below the maximum.

If the drive is rated for 150% couurent for 60 seconds, it may not be able to supply much more than that regardless of the apparent range of adjustment. The adjustment that determines the short-time current limit is usually called the "current limit" adjustment or perhaps the "torque limit" adjustment. "Overload setting" implies an adjustment that is time related. The drive may have both types of adjustment. The limiting adjustment is the one that must be increased.

 

[jraef]

Some important subtle points brought up by DickDV may be contributing to the problem as well. Dick correctly stated that the second drive must be a TORQUE FOLLOWER, but Sunray actually stated that it may be a CURRENT FOLLOWER. The second drive MUST be in torque follower mode, and not all vector drives are capable of that. Current follower only will not be accurate enough provide true load sharing, I know from experience. If you are attempting to use an analog output proportional to current on the master to run the slave in current limit against that signal, it will not work unless the motors are seriously over-sized!

Also just because it is a closed loop vector drive doses not mean that it can provide torque follower capability, and even if they say it can does not mean it is good at it. That feature also requires a high precision interface between the drives. The ones that are best at it use fiber optics for noise immunity, because any noise on the feedback signal delays the slave drive response and interferes with load sharing, causing one motor to pull the other.

If you are stuck with the drive you already have and all it has are analog I/O for comms, make sure you are able to have the slave run in torque follower mode (torque control), and that the output from the master is proportional to torque, not speed or current. Then make sure you shield the heck out of that interface cable.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[sunrays]

Thanks DickDV & jraef.

First of jraef's points. The two drive communicate on a dedeicated bus. No analog I/O interfacing is used. So no time delays or noise interfearance is experienced. The slave is receving the current reference from the master.

Now to DickDV is point of view. The design of the master/slave option was mine to ensure that no skewing of the mechncials is experienced. The plots of the currents of both the drives shows no difference. The slave slave current virtully follows the master reference. This confirms that the load sharing is perfect between the two.

On the issue of stalling the drive hits its current limit and the motor speed drops to zero. If left in this condition I get a I²t trip. The best part is the motor does not experience a temp. raise even at this moment of time. The output voltage, current & frequency is as follows: 400V, 200 Amps, 50 Hz. on each of the drives. The entire operation from 0 - 175⁰ and back to 0 is over in 50 secs.

The comparitive currents on the slip-ring motor based system is around 100-110 Amps max.

Has slip compensation got any thing to do with this? Will it help in any way?

 

[DickDV]

sunrays, thanks for this important information. I have the following conclusions:

First, if, under stall conditions, the drive output frequency stays at 50Hz, the drive is operating as a scalar open loop V/hz drive and not as a Flux Vector Drive. This is very important and is worth double-checking. If the drive is properly programmed, when the current limit is met, the drive output frequency should drop down with motor shaft speed to around 2Hz when the motor fully stalls. The voltage will fall also with the amps staying right at the current limit level.

Second, I wouldn't be too worried about the current comparisons with the wound-rotor motor. The lower currents were occurring at 400V while, on the inverter, the currents may well be occurring at lower voltages. Its kw or kva that counts, not amps.

Third, does the drive you are using offer a torque regulator option of operation. jraef mentioned how desireable it is to operate in that mode and I fully agree. You do seem to have a balance of sorts now, but if the first item above is not right, then the balance you have is of no value. I have not had any trouble using analog input/output coupling between torque follower drives as long as the drive response is faster than the load demand. With a large machine like you are describing, fairly slow load demand will avoid problems.

Please take a look at those voltage and frequency values at stall again. I strongly suspect that you are not operating in Flux Vector Mode.

 

[aolalde]

This application is not merely a speed change control, the load torque is variable
( TL = k*sin a) and the torque is negative when the wagon is going horizontal.
When the operation begins, the wound rotor works as an induction generator holding the load (which torque increases as the angle to the vertical increases). The motor works as a generator and the rotor resistance should match the speed with the maximum or breakdown torque. When the wagon is lifted the starting torque requirement is at a maximum and it reduces as the wagon approaches the vertical position.

Note that SCIMotors have around 200% BDT as compared to 275-300% BDT for wound rotor motors. By the other hand the actual VFD driver set up does not seem to match the described wound rotor motor performance.

 

[CJCPE]

Regarding the slip compensation question, slip compensation is a slight increase in operating frequency as load increases to avoid the slight drop in speed due to motor slip. It provides tighter speed regulation and should not be an issue in this application. In a vector drive, the tuning adjustment for speed regulation may not be called slip compensation.

As jraef and DickDV stated, it is better to operate the slave drive in the torque regulator mode of operation. However, the current regulating mode of the drive you are using may be essentially the same function. Equal current does not guarantee equal torque. You should also compare frequency to make sure that slip is equal. (You know speeds are equal because of the mechanical connection.) Applied voltage or V/Hz effects torque per amp and should also be equal, but difficult to measure accurately.

I would also suggest that the selected SC motor and drive combination may not be capable of providing the same performance as the slip ring motor when the torque requirement approaches the motor's breakdown torque.

 

[sunrays]

Thanks once again DickDV for the valuable inputs.

First the slave drive is in torque mode the reference being the current from the master and hence I used the phrase "current controlled slave"

The drives selected has an option to switch between V/f, Openloop Vector & Closedloop Flux Vector by software means. I have obviously selected the Closedloop Vector option. No doubts about that. So the question of openloop V/f mode of operation is out of question.

The comparison of the currents was at full speed for the cage motor which is equal to full voltage.

To recheck the values I would have to wait a while. This is running plant and I had to revert back to the old system when I could not get the currents under control. All the trails had been on an empty wagon. With an additional load of another 70 tons in the wagon I wonder what would be the results.

Another point to be noted is that as aolalde has pointed out the load GD² varies as the wagon is rotated from 0 - 175⁰. The reverse operation is completely an overhauling one with the mechanicals driving the motors. I had considered this in my design and provided for good dynamic braking which is working extremely well. The breaking currents during this operation is well with in control and hovers around 80 - 110 Amps. which the drives handle perfectly. I had done a comparison of the currents during the reverse operation for both SC motor & Slip-ring motor and found the values are at the same level.

That leads to question if the drives can handle the breaking torque with out hitting the current limits then is it not correct to expect the same combination of drives to handle the motoring torque as well?