Servo Mode Question

Hello all. I want to apologize up front for the lengthy post.

I am currently working on a job for a large sporting event. We are developing a curtain that we want to control with a servo motor, for the repeatability factor. In the past, we have relied on a 'poor man's servo', using dc drives with absolute encoder feedback. For our current system, I don't use the motion commands in Logix 5K (or motion cards), but rather our own code to figure out accels, decels ,speeds, and target positions. Due to the nature of the (entertainment) industry, the error factor can be .125" (or greater, within reason) without causing operational problems. Also, if the HMI control system sh*** the bed, we can change to analog control, and if that fails, batteries will allow the show to go on.

Knowing that we like to have multiple options for backup in case of catastrophic failure, my question is this:

What is the difference (in laymen's terms, please - I have read this post http://www.plctalk.net/qanda/showthread.php?t=3267&highlight=velocity+torque , but don't really grasp the explanation) between torque mode and velocity mode in a servo app?

My local AB rep said that we could have an external control setup (fwd/rev switches and a speed pot) that we could select for control if the MO2AE card (and control rack) fails, BUT............. my amplifier has to be setup in velocity mode. What happens to the torque output?

Forgive my ignorance, but I have worked with DC control systems almost exclusively for the last 15 years, and have worked with servos a few times, but only in torque mode. (and they have never failed) We must have redundency in control for this gag

Thanks in advance to anyone who can help. Again, sorry for the lengthy post.

B

Mmmm...

Redundancy for motion control with CLX. Gonna be interesting topic

Do you need SIL2 or SIL3 or any else certification?

I am still a little confused as to your setup.........But I'll give you this bit of info.

When your drive is in VELOCITY mode, the drive (rather than the motion controller) closes the velocity loop. Along with that, the drive closes the current loop and performs commutation.

The current loop and commutation is always done in the drive regardless of what mode you are in. The motion controller generates the position profile and closes the position loop. The output of the position loop is a velocity command, which is fed to the drive.

In terms of control, you are requiring the motion controller AS WELL AS the drive to have some form of intelligence.

When your drive is in TORQUE mode, the velocity loop as well as the position loop are closed within the motion controller. This means that the drive is a little more "dumb". I prefer to say that the drive is "allowed" to be more dumb.

The one versus the other argument could go on for ever..........The application and other requirements should help make the decision for you.

There is alot more convenience with a drive in torque mode. Because both the V and P loops are closed in the motion controller, this means that all the servo gains can be changed inside a single device. Since servo gains may vary from one application to another, locating all the servo gains in a single device simplifies this feature.

When it comes to high-performance systems, velocity mode is usually the way to go. This has to do with servo loop sampling rates. When it comes to multi-axis systems, requiring the motion controller to close these loops at high speed is sometimes alot to ask (since the motion controller must spread its computational power across all the drives). The use of high speed digital drives will make up for this. Especially when it comes to cost.

I'll leave it at that for now....... Keep in mind that the application is what is controlling the decision of one versus the other.

What exacly is your setup going to be? Are you using an M02AE card or not?

Yes, I am going to use the MO2AE. I have a mock-up sitting on my bench right now with an Ultra 200 drive. Eventually, I'll tap into the J1 connector, and wire in the external controls. I realize that this seems 'over-the-top', but there has to be a way to control the motor in the microscopic chance that something goes haywire in the CLX rack. As I stated before, the few times that I have used servos in the past have never failed, and they weren't permanent installs. The were torn down and set up every night in a different venue/city. Ah, the rigors of the touring industry.

You need to have a little more faith in your control hardware.....

If by some 'microscopic' chance the CLX rack goes belly up (yet to see it in a controlled environment), then I would think that your redundancy would have to be outside the whole servo control system. You might as well take the whole thing out of the picture and introduce some more primitive way of moving the curtain.

There is alot more that can go wrong that just the CLX rack. You have the rack itself which consists of a power supply, processor, backplane and motion control card. You also have the drive (Ultra 200) as well as the servo motor and associated encoder and power wiring. And if I had to say which one would go belly up first, I would bet some serious money that it WILL NOT be the CLX rack.

Good luck with your app......

Oh yeah. You mentioned that your drive has to be in Velocity mode. The Ultra 200 can be configured for V or T mode.

I have faith. The higher powers-that-be don't always, though. Thanks for the explanation. It did help me out. We did toss the idea around of a PowerFlex drive with AC motor connected to a MO2AE (hassle?!?!). Then if it's necessary to override the system, we could jump the motor to line via a motor contactor/overload setup.