Motors synchronising using VFD

Status
Not open for further replies.
M

Muhammad Umer ansar

Member
Location
karachi pakisatan
Occupation
junior electrical engineer
I Have two motors, both connected with two different Vfds. First Motor is connected to VFD-E series delta vfd and other to MS-300 delta vfd. I need to sink first motor with the second. I remained successful in getting a variable volt using AFM of the first vfd, and connected AFM and ACM(ground) to AVI and ACM of other VFD. But speed of 2nd motor is not varying with the first one. even after inserting parameters.
Help needed!
 
I believe in order to synchronize two motors vfd should support wired link between them. call vfd manufacturer or check the manual.

Sent from my ONEPLUS A6013 using Tapatalk
 
Muhammad Umer ansar said:
.... I remained successful in getting a variable volt using AFM of the first vfd, and connected AFM and ACM(ground) to AVI and ACM of other VFD. But speed of 2nd motor is not varying with the first one. even after inserting parameters.
Help needed!
Click to expand...
I have no clue what those TLAs mean (Three Letter Acronyms)...

If by "synchronizing" you mean ratio control between them, then that is called "electronic gearing" and only the highest level of VFDs are capable of that with any degree of precision, almost always requiring encoder feedback to Flux Vector Control (FVC) or Field Oriented Control (FOC).

If what you mean is that you have two motors that are MECHANICALLY linked (i.e. by belts or shafts) and you want the motors and VFDs to share the load between them, then you set them up as a Master and Slave, with the master following your speed command (velocity control), and the slave using torque following using a torque value signal from the Master drive. I've done that using simple analog signals and been within 2% load sharing. The trick is though that your Master drive must have the ability to have its analog output signal tied to Output Torque, and the Slave drive must be capable of operating in Torque Control mode. Not all VFDs are capable of either of those.
 
I think what he is trying to do is to use an analog input of one drive to set the speed and have an analog output from that drive tied to an analog input of the second drive to set that speed. I don't think you can ever get that to work very reliably if there's actually tied to the same shaft. I have done this with drive systems where I wanted one drive to be just a little bit faster than the other but they weren't actually sharing the same shaft. I think jraef has the right idea.
 
You have to make the second drive a follower. The normal scheme is set up the first drive. Then the analog output is programmed as a current output. Second drive follows current output of the first drive. Make sure second drive accelerate/decel times are pretty short. First drive is/should be slow times.

Do not try to synchronize on speed. If you do and the drives are off at all they will fight each other. Also just sending a speed reference to both and running open loop (V/Hz) usually results in 80%’load on one motor depending on their torque/slip curves. Current feedback avoids this. Flux current is relatively small and you want equal torques so if they output the same current torque is shared

More fancy is to use virtual line shafting...use two dual output encoders. Follower follows the master drive encoder. Or use a virtual axis in a servo system and sync all drives to that. But this requires full vector control drives with encoders.
 
paulengr said:
You have to make the second drive a follower. The normal scheme is set up the first drive. Then the analog output is programmed as a current output. Second drive follows current output of the first drive. Make sure second drive accelerate/decel times are pretty short. First drive is/should be slow times.

Do not try to synchronize on speed. If you do and the drives are off at all they will fight each other. Also just sending a speed reference to both and running open loop (V/Hz) usually results in 80%’load on one motor depending on their torque/slip curves. Current feedback avoids this. Flux current is relatively small and you want equal torques so if they output the same current torque is shared
Click to expand...
I don't think I have ever tried this. Seems like it ought to work.
paulengr said:
More fancy is to use virtual line shafting...use two dual output encoders. Follower follows the master drive encoder. Or use a virtual axis in a servo system and sync all drives to that. But this requires full vector control drives with encoders.
Click to expand...
I have often wondered why more designers do not use electronic gearing. It tends to get rid of a lot of bulky and expensive mechanical components in favor of an extra motor and drive.
 
petersonra said:
I don't think I have ever tried this. Seems like it ought to work.

I have often wondered why more designers do not use electronic gearing. It tends to get rid of a lot of bulky and expensive mechanical components in favor of an extra motor and drive.
Click to expand...

Because encoders are fairly expensive, historically vector control was only on servo drives and even then we used analog tricks like running the armatures in series on the same drive, and because troubleshooting closed loop control especially with multiple motors is not easy. It also doesn’t help that few VFDs have this built in. Historically you’d use a separate control system, a servo controller, and the drives (normally DC) are called amplifiers...they are totally dumbed down to simple current regulators.

As an example of troubleshooting problems if your gears are worn you can get mechanical oscillation inducing electrical oscillations. You will fight everything in software until you realize it’s not an electrical/software problem. Running a virtual axis instead of master/follower largely helps prevent this but ultimately the trick is recognizing when it’s a mechanical and when it’s an electrical issue. Plus the best way to troubleshoot encoders is with an oscilloscope. This puts it out of the reach of many technicians.

But when it works, you get extremely tight gearless control that just seems to work like magic.
 
I have used electronic gearing a number of times. It is almost magic. Never had any mechanical issues with it but by the time things start wearing out I have been long gone.
 
Muhammad Umer ansar said:
I Have two motors, both connected with two different Vfds. First Motor is connected to VFD-E series delta vfd and other to MS-300 delta vfd. I need to sink first motor with the second. I remained successful in getting a variable volt using AFM of the first vfd, and connected AFM and ACM(ground) to AVI and ACM of other VFD. But speed of 2nd motor is not varying with the first one. even after inserting parameters.
Help needed!
Click to expand...


Hello Muhammad,

That’s an interesting question.
In your query-- you are mostly [concerned] on the frequency although there can be more that you haven’t shown your curiosity on.
The general concept of the Master/Slave configuration is dependent upon the setting of the Master VFD.

The torque reference is communicated through its “control loop” that is sent to the Slave VFD.

This reference is then referred to as the “TORQUE FOLLOWER”.
This basically is the important aspect of the VFD performance for both Master and Slave-- and expect a smooth functionality of the two VFDs working in concert
I’ve seen this setup with both identical drive motors and VFD drives.

I’m not familiar with the two models motor you have presented. So, I’m assuming they could have a miniscule difference in terms of torque capability.
If you set the Master for example the “proper” torque to handle the load-- the Slave VFD may not have the same response (or result) even though the parameter that was set on the Master--it may not have the same effect on the Slave.

This setting may show that is visible and promptly displayed on the operator’s console.
This disagreement on the Master/Slave usage is more evident on a super long conveyor for example whose functions have to be in perfect unison. Conveyors used in mining operation.

This is not to say that your problem will require a “Pyrrhic Victory” to overcome –but there are better and smart designs of controllers that will overcome this issue.
A web search will certainly help and make comparisons--depending on the usage that you have.

BTW:

If our Brit friend-- who has extensive experience on conveyors would be kind enough to chime in-- perhaps you will find a definitive answer.
This is not to say that contributors to provide answer to your inquiry—would not lead to the answer you expect.
There is a website that focuses on this kind of problem. I don’t recall the address however.

Good Luck and may good things come your way.

As an aside:
Achi Taqdeer, Husn Ittafaq. (good Luck)

You speak URDU?
 
Status
Not open for further replies.
Top