Motor Control Recommendations

[TCjr]

I’m looking for some opinions and product suggestion on a controller/motor system that will utilizes a 6” valve to produce near mach speeds. I would like the system to be highly responsive & accurate in order to update and position the valve very quickly in order to maintain a nearly flat output. Any suggestions or recommendations would be greatly appreciated.

tks,

[Jochen]

I can think of several options concerning the controller but before I can give you any advise it would be great if you could provide any specific values that quantize "very quickly". Ideally you could provide information about the required control loop rate but it would also help if you could give me some numbers about the dynamics of the system (e. g. rate of change of the flow rate).

What exactly is your process variable (PV)? Is it the flow rate? What disturbances have an impact on the PV (e. g. pressure fluctuations)?

Another question: Do you feel comfortable with LabVIEW programming? Do you have any experience with Real-Time applications or FPGAs?

Best regards,

Jochen Klier
National Instruments Germany

[TCjr]

Thanks Jochen for your response….I can give you some specifics. Here’s the concept. I have several very large holding tanks that hold a limited amount of compressed air. The idea is to release this air, using a 6” valve, through a nozzle to obtain various speeds up to mach. The speed is solely determined by the change in pressure, or in other words, the difference between atmospheric pressure and the pressure between the output nozzle and the 6” valve. Ideally, I would like to ramp up in about 500ms and then level off to a preselected velocity (Pressure), which means that the valve opening must be constantly changing and updating to maintain the correct pressure differential for several seconds. I would like to maintain control of the selected velocity (pressure) by updating the valve position in 1.0 to 10.0 ms increments, if possible. I’m looking for the best possible solution. I’m also curious about what type/size motor would provide the best response characteristics. Maintaining a tight control over the valve position would give the best velocity output profile.

I’m comfortable using LabVIEW….I’ve programmed several systems using it, but I’ve not used any of your motion control type software. I’ve not used FPGAs but that dosesnt mean I not willing to learn about them…..if that is what it takes I’d do it. 

tks a lot, Terry

[Jochen]

Terry,

thank you for this information. This sounds like a very interesting application.

Let's have a closer look at your speed requirements. Ramping up in 500 ms requires control loop rates that run much faster than that. You are talking about update rates of about 1.0 ms to 10 ms which sounds quite reasonable.

From this perspective there are two options:

1. A high-performance motion control device like the PCI-7352
2. A system based on LabVIEW Real-Time (PXI or standard PC that supports LV RT) and a DAQ board

Option 1.) is easier than option 2.) but you would need to be able to provide a single analog feedback that is proportional to the velocity. If you can use a pressure sensor that outputs the pressure difference between the output nozzle and the 6” valve as a single analog signal, option 1.) would be feasible for you. This signal could be used as analog feedback for the onboard PID controller.

Option 2.) requires more programming as you need to implement the controller by your own but on the other hand there are more degrees of freedom. For the case that you need to combine multiple sensor signals to generate the feedback signal or if you want to implement your own control algorithm (e. g. model based approach) LV RT gives you the power to that.

If you are looking for an industrial solution with a small form factor a cRIO system might be an option for you. As this is a FPGA based device LV FPGA would be required to program the application which brings some more complexity. On the other hand the FPGA can run control loop rates with up to 200 kHz which provides a lot of headroom for your application. By the way: You can expect a control loop rate of up to 16 kHz with option 1.) and about 20 kHz with option 2.) which should both meet your requirements.

I hope that helps,

Jochen Klier
National Instruments Germany