PLCS.net - Interactive Q & A

PLCS.net - Interactive Q & A (http://www.plctalk.net/qanda/index.php)
-   LIVE PLC Questions And Answers (http://www.plctalk.net/qanda/forumdisplay.php?f=2)
-   -   torque control???? VFD??? (http://www.plctalk.net/qanda/showthread.php?t=16081)

Mr_B July 12th, 2005 12:57 PM

torque control???? VFD???
 
How do I do this?

_____O_____O_____
.....a.....b......


Motor "a" feeds material and b pulls it.

Can one program/wire the drives in such a way that b gets a setpoint and a follows but according to motor b's Current drawn.

Hope I'm clear on this. Basicly wot I'm trying is to control a's speed with the current drawn by b so that there is always tension between the two. Not a lot but just a little.

BTW the drives are Yaskawa Varispeed VS mini J7

tx.

rdrast July 12th, 2005 01:48 PM

First, controlling A's speed as a function of B's current is doomed to fail with VFD's. It's possible (mostly) with DC, but not AC.

If you can, just give up that whole idea, configure A as a pure speed control, and B as either:
A) pure vector/sensorless vector torque control.
B) as a speed follower with torque limiting.

Personally, I prefer option B, as that way you won't get any kind of overspeed condition if you lose the load.

I'm not familiar with that drive, but some sort of torque control is probably possible, though it might require a feedback device on the motor.

Good luck

Mr_B July 12th, 2005 01:59 PM

Option B sounds good. Wud u mind to elaborate. I wud think to use the speed as reference but wot exactly do u mean with torque limiting...

msinclair July 12th, 2005 02:05 PM

You can do this if you really want to - it is effectively a "poor man's loadcell". Though it can be done, I strongly advise AGAINST it! I just completed a retrofit where we put in loadcells and eliminated this type of practice which is NOT robust.

I personally would spend the money for a load cell, set one drive up for speed control, set the other up for speed control with tension trim (using the loadcell as your feedback device).

Good luck,

Marc

LJAM July 12th, 2005 06:16 PM

Motors load sharing ?
 
Gents,

are both motors driving the same load?, if that is the case you should consider "droop control". This techniques is employed in many drive applications. Also many standard VFDs have parameters to help configure the drive for the application (talk to your supplier) ... else, there are tons on information in the net about it.

I am not quite sure what you are trying to do but if load sharing is the application, this might be the answer.

Cheers,
LJAM

DickDV July 13th, 2005 08:10 AM

The simplest way to get speed control and tensioning would be to set A up as a speed regulator and B as a torque regulator. A would then determine the web speed between the two rolls and B would determine the tension in the web. The only interconnection between the two would be start/stop control.

As mentioned above, a disadvantage with doing this is that, if the web breaks between the rolls, the B drive would run up to max speed until the system was stopped. If this is a problem, you could add a little complexity by setting up B as a speed regulator following A's speed signal with a torque trim on B's second analog input. This trim signal is NOT the torque signal from A but rather a manual setpoint for B's torque or a signal from a load cell or dancer reading tension directly on the web.

The A drive must be able to hold speed over a large load range so a good sensorless or flux vector drive would be needed.

stasis July 13th, 2005 09:10 AM

Quote:

Originally Posted by Mr_B
Hope I'm clear on this. Basicly wot I'm trying is to control a's speed with the current drawn by b so that there is always tension between the two. Not a lot but just a little.

BTW the drives are Yaskawa Varispeed VS mini J7

tx.

like has been mentioned, you don't want to set this up using current. Bad things will happen when the bearings/gearbox on motor b fail...
What type of material is this? Depending on type, you could just use an ultrasonic device between the motors. Adjust for the material, and program the drives to run at set speeds using the analog input from the sensor as a 'speed pot'. Using the analog control will also allow you to stop the drives if the web breaks.

Mr_B July 13th, 2005 12:26 PM

The Load will never break. The material flow is to the right. At the moment drive a works with a "speed" pot and b also with the same "speed"pot but also through a "ratio"pot.

So the "speed"pot adjusts both the drives'but the "ratio" only drive "a".

So the Operator sets the "ratio"pot to get tension between the 2 motors. This tension is not measured at the moment.

I wanna know if I can do this automatic. This tension. Hope u understand better now.

DickDV July 13th, 2005 03:30 PM

You cannot do both drives as speed regulators even with a "ratio" adjustment because the speed error accumulates constantly increasing or decreasing the tension. (Assuming you don't want web stretch to provide the tension).

That's why the tension must be controlled either by direct tension measurement or by torque on the b drive. Using line speed from the a drive with a torque trim will do the job but no amount of speed trim will get this job done---you'd have to adjust it constantly to keep from building or decreasing tension.

Thomas Sullens July 13th, 2005 04:19 PM

thoughts
 
Mr B do you have a plc close by with 2 spare analog out and one analog in signals. If so you can use these to make A follow B.
It sound like all you need in this case is a little speed difference. If this is the case you can take the input from your pot and scale it for each drive then set your MIN and MAX speeds
in your drives. An example would be if you had a 12 bit input card your INPUT MAX = 0 OUTPUT MAX = 4095 SCALED MIN 0 SCALED MAX = 60
This would be for drive B. Then INPUT MAX = 0 OUTPUT MAX = 4095 SCALED MIN = 0 SCALED MAX = 54 this is for drive A. This should give you a 10% difference in drive speeds. Drive B you would set MIN hz to 0 and max hz to 60. Drive A you would set min hz to 0 and max to 54. This way you could scale and get what ever difference you wanted. There would have to be some strech or slip some where and there must be because if there wasn`t it would drive you crazy doing it manually and on the other hand this maybe the case??
Oh well good luck.

Peter Nachtwey July 13th, 2005 06:00 PM

Use torque/speed limit!
 
Quote:

Originally Posted by DickDV
You cannot do both drives as speed regulators even with a "ratio" adjustment because the speed error accumulates constantly increasing or decreasing the tension. (Assuming you don't want web stretch to provide the tension).

Speed error should not accumulate! One should be able to put encoders on the VFD and make sure they both travel the same number of acounts so at the end of the shift the two motors have traveled exactly the same distance as well as any time inbetween. My customers expect this. This why I must be fanatical about reseting timer and counters. The problem is slippage.

I would make the virtual master for the B VFD go just a bit faster than the virtual master speed for the A VFD. ( Remember what I said about gearing. Don't unless it is the only way. ) This will cause tension. I would then use a torque limit mode that limits the speed and the torque. This way the speed is limited to the virtual master when the load disappears and the torque is limited so it doesn't tear the material or pull it through the A VFD. You can't control both torque and speed at the same time, but you can limit both at the same time. The big problem I see is getting a clean torque signal. Usually there is a model inside the drive that has a filtered estimated value for torque.

No sweat.

Leadfoot July 13th, 2005 06:25 PM

As a J7 has been specified as the VFD in use, an encoder is not an option. Neither is torque control.

You could use encoders on the motors and feed a CONTREX "M-TRIM". The M-TRIM has the ability to make the B lead the A by a given amount and track nicely. Getting the panel mount unit comes with its own HMI.

Contrex used to be called Fenner.

DickDV July 13th, 2005 06:35 PM

Peter, the information in your post is good stuff but it proves my point rather well. You can't do two speed regulators without speed error accumulating with the resulting loss of tension or excess tension. As you point out, you can speed match with encoders but that does not get you tension.

It is very interesting that you mention torque limiting. That can work with two speed regulators because, ultimately, you are setting tension with the torque limit. The undesireable feature of that method is that the tension (torque) is not externally adjustable so the operator has no control over it short of going into the software and changing the limit value. Further, you get the same thing by using a torque trim. This gives you an externally adjustable torque limit.

Also, to be a little picky, you can regulate speed and torque at the same time. Line shaft software does this and must for proper speed and tension control. A paper machine, for example, must regulate all its sections for speed and tension (torque) simultaneously.

Thomas Sullens, your point about programming speed offset into the two analog input signals can be done with a PLC but why? Virtually all modern drives have that same capability built into their analog input configuration parameters. And, again, all you will get is offset speed which will not work for tensioning as I point out above.

DickDV July 13th, 2005 06:36 PM

Peter, the information in your post is good stuff but it proves my point rather well. You can't do two speed regulators without speed error accumulating with the resulting loss of tension or excess tension. As you point out, you can speed match with encoders but that does not get you tension.

It is very interesting that you mention torque limiting. That can work with two speed regulators because, ultimately, you are setting tension with the torque limit. The undesireable feature of that method is that the tension (torque) is not externally adjustable so the operator has no control over it short of going into the software and changing the limit value. Further, you get the same thing by using a torque trim. This gives you an externally adjustable torque limit.

Also, to be a little picky, you can regulate speed and torque at the same time. Line shaft software does this and must for proper speed and tension control. A paper machine, for example, must regulate all its sections for speed and tension (torque) simultaneously.

Thomas Sullens, your point about programming speed offset into the two analog input signals can be done with a PLC but why? Virtually all modern drives have that same capability built into their analog input configuration parameters. And, again, all you will get is offset speed which will not work for tensioning as I point out above.

BITS N BYTES July 13th, 2005 07:49 PM

You will not easily load share with VFD's.
However if all AC motors are powered off the same VFD load sharing is easily and cost effectively implemented.
Take a look at these links:-
http://www.drivesurvey.com/index_lib...?feature_id=19

http://www.drivesurvey.com/dc/dcboar...=735&mode=full

Hope these help.


All times are GMT -5. The time now is 12:26 PM.

.