GaryS
Member
I hate to be different but I stand by my statement that torque control for the slave is the only way to do it. Do not consider a flux vector drive for any slave drive in that configuration.
Here is a link to Yaskawa elearning video explaining it. I would encourage everybody to heck out there videos. I know they are from Yaskawa and most of you don’t use them but they are free and the principles apply to all drives.
https://www.yaskawa.com/support-tra...let_selectprodsubgrp=&_yastabstemplate_WAR_ya
I have been doing drives for over 30 years, doing factory service for a numberof those years. And have seen many try to do a common load system with all the drives in speed control. Just looking at the lines run it does look like they achieve that. But if you look deeper and watch what the drives and motors are doing you will see they are actually fighting each other. One will pull high current and drop the output frequency back a little while the other will deliver less current while the frequency go up a little. And them the condition reverses. If you try to set up both drives as flux vector drive the y will both try to be speed masters on the system with no motor slip so one or both drives will over current and shut down.
As we all know there can be only one master in any system at a time.
Think about the
Chain conveyor and how it’s made I this case it is over 1300 feet long the average ling of this type of chain is about 6 inc long with a pin at each end. Each pin has fitting tolerance of maybe 1/32 of an inc each but you have over 2600 pins the slop on each will add up and increase over time ( they never get better) We all know you can’t push a chain it can only be pulled so each drive will have to pull its share of the chain. But the load on the chain is never going to be loaded equally over the length of the chain. And that load will be constantly changing as it runs. You couple that with the accumulated ware of the pins and other parts of the system it becomes increasingly impossible to have both drive run at the same exact speed and actually share the load. You have to decide either to match speed or share the load you can’t do both .
I has a system a few years ago the chain conveyor was a loop of about 2000 feet with 3 identical drive systems same VFD same gearbox everything was identical on each system. The original designer set it up as a speed follower. In the PLC they had over 10 rungs of very complex math code to try and get them to share the load. When I observer the PLC code it was commanding a constantly changing speed command to each VFD of between 5 and 10 HZ none were getting the same speed command. The calculations were using the VFD current to calculate the necessary VFD speed command not working very well. To the human eye it looked like the line was running very well. So they were happy with it. All 3 of the VFD’s were just V/F control allowing the motor speed to slip back from the VFD output speed depending on load. Using flux vector VFDs on those would be a disaster.
I actually installed and set up the first Flux Vector drive brought in to the US. That was about 1993 I think
They were for a pair of 30 ton bridge cranes, for the bridge drives the bridge was 80 wide, motor and gearbox on each end of the bridge. A Flux Vector VFD for each motor, new matched gearboxes on each and new precision machined wheels on each drive as well. We had 2 cranes that had to run together with a common load shared on the hooks of both cranes so it was 4 drives in total all set up identical in every way the crane bay was about 800 feet long. But when we ran the cranes they would not run more than about 50 ft before the bridge would rack so far that it would try to jump off the rails. We ended up having the VFD design engineers come in from Japan to do a set and discovered the only way to make it work was to detune one of the VFDs to allow it to have less precise speed control over one motor on the bridge ( More slip in the motor) It ran great for over 5 years they normally replace their VFDs about every 5 years or so these were replace with a newer models.
A thought to keep in mind, no 2 VFDs no matter who’s they are will ever produce exactly the same output frequency you may not be able to see the difference but over a long time it will show up.
As for the comment that you need to replace the existing VFD because it can’t do torque control to add a second one is just not true. I can add the second VFD in Torque mode and leave the original in place. The original would be master speed control. One has to run as master speed control. Also no encoder is necessary on this conveyor system at all. You would be better off using a limit switch or photo eye for positioning you eliminate the problem with slack in the chain due to load and wear.
I have see other examples of this to many to go into here.
Here is a link to Yaskawa elearning video explaining it. I would encourage everybody to heck out there videos. I know they are from Yaskawa and most of you don’t use them but they are free and the principles apply to all drives.
https://www.yaskawa.com/support-tra...let_selectprodsubgrp=&_yastabstemplate_WAR_ya
I have been doing drives for over 30 years, doing factory service for a numberof those years. And have seen many try to do a common load system with all the drives in speed control. Just looking at the lines run it does look like they achieve that. But if you look deeper and watch what the drives and motors are doing you will see they are actually fighting each other. One will pull high current and drop the output frequency back a little while the other will deliver less current while the frequency go up a little. And them the condition reverses. If you try to set up both drives as flux vector drive the y will both try to be speed masters on the system with no motor slip so one or both drives will over current and shut down.
As we all know there can be only one master in any system at a time.
Think about the
Chain conveyor and how it’s made I this case it is over 1300 feet long the average ling of this type of chain is about 6 inc long with a pin at each end. Each pin has fitting tolerance of maybe 1/32 of an inc each but you have over 2600 pins the slop on each will add up and increase over time ( they never get better) We all know you can’t push a chain it can only be pulled so each drive will have to pull its share of the chain. But the load on the chain is never going to be loaded equally over the length of the chain. And that load will be constantly changing as it runs. You couple that with the accumulated ware of the pins and other parts of the system it becomes increasingly impossible to have both drive run at the same exact speed and actually share the load. You have to decide either to match speed or share the load you can’t do both .
I has a system a few years ago the chain conveyor was a loop of about 2000 feet with 3 identical drive systems same VFD same gearbox everything was identical on each system. The original designer set it up as a speed follower. In the PLC they had over 10 rungs of very complex math code to try and get them to share the load. When I observer the PLC code it was commanding a constantly changing speed command to each VFD of between 5 and 10 HZ none were getting the same speed command. The calculations were using the VFD current to calculate the necessary VFD speed command not working very well. To the human eye it looked like the line was running very well. So they were happy with it. All 3 of the VFD’s were just V/F control allowing the motor speed to slip back from the VFD output speed depending on load. Using flux vector VFDs on those would be a disaster.
I actually installed and set up the first Flux Vector drive brought in to the US. That was about 1993 I think
They were for a pair of 30 ton bridge cranes, for the bridge drives the bridge was 80 wide, motor and gearbox on each end of the bridge. A Flux Vector VFD for each motor, new matched gearboxes on each and new precision machined wheels on each drive as well. We had 2 cranes that had to run together with a common load shared on the hooks of both cranes so it was 4 drives in total all set up identical in every way the crane bay was about 800 feet long. But when we ran the cranes they would not run more than about 50 ft before the bridge would rack so far that it would try to jump off the rails. We ended up having the VFD design engineers come in from Japan to do a set and discovered the only way to make it work was to detune one of the VFDs to allow it to have less precise speed control over one motor on the bridge ( More slip in the motor) It ran great for over 5 years they normally replace their VFDs about every 5 years or so these were replace with a newer models.
A thought to keep in mind, no 2 VFDs no matter who’s they are will ever produce exactly the same output frequency you may not be able to see the difference but over a long time it will show up.
As for the comment that you need to replace the existing VFD because it can’t do torque control to add a second one is just not true. I can add the second VFD in Torque mode and leave the original in place. The original would be master speed control. One has to run as master speed control. Also no encoder is necessary on this conveyor system at all. You would be better off using a limit switch or photo eye for positioning you eliminate the problem with slack in the chain due to load and wear.
I have see other examples of this to many to go into here.