Standalone Stepper controller

[BrianJensen]

Hi Guys.

I was hoping that I can get a few great ideas of some of the bright heads in here.

For work, I need to build a test device for testing stepper motors. The device needs to be a generic drive for all 2 phase stepper motors.

In basics, I need a few outputs to drive a stepper driver, and a few inputs to read some switches.

The tricky bit is that it need to be stand alone, that is, sort of mobile, and no need to be connected to a pc to make it run.

Can I do something like this on a FPGA ? (I have no experience with FPGA at all)

What would be the most cost efficient way to do a labview device that will live by itself. Can it be done at all, or should I start look at some other alternatives instead.

Any and all ideas are appreciated.

Regards, Brian

[Tom Eilers]

You can buy ISEL 116c (4 Ampere) or ISEL 142c (8 Ampere) stepper motor controllers with these single steppermoter controllers it is possible to upload your test program and start/stop it.

These controllers 4 digital in and 4 digital out. Programming is done through the serial port.

WWW.ISEL.COM/en

Software and Hardware Document is available: please send me an Email

[ned]

QUOTE (BrianJensen @ May 8 2008, 01:49 AM)

For work, I need to build a test device for testing stepper motors. The device needs to be a generic drive for all 2 phase stepper motors.

In basics, I need a few outputs to drive a stepper driver, and a few inputs to read some switches.

The tricky bit is that it need to be stand alone, that is, sort of mobile, and no need to be connected to a pc to make it run.

Can I do something like this on a FPGA ? (I have no experience with FPGA at all)

What would be the most cost efficient way to do a labview device that will live by itself. Can it be done at all, or should I start look at some other alternatives instead.

Have you looked at Copley Motion? They make several stepper motor controls with a CAN interface; you could control it using a CAN card and the FPGA in an NI cRIO. I'm not sure if this would fall in your definition of "cost efficient" since a cRIO isn't inexpensive, and writing code for the CANOpen protocol may be time consuming. Copley does provides libraries and LabVIEW examples but they rely on ActiveX and won't run on a cRIO.

Another option, if you're willing to bypass LabVIEW entirely, would be a unit like the Parker Motion Gemini GT6K. It's an independent controller that can be programmed in a simple text-based language, and it includes a couple of channels of I/O.

[Wire Warrior]

QUOTE (BrianJensen @ May 8 2008, 12:49 AM)

In basics, I need a few outputs to drive a stepper driver, and a few inputs to read some switches.

The tricky bit is that it need to be stand alone, that is, sort of mobile, and no need to be connected to a pc to make it run.

Can I do something like this on a FPGA ? (I have no experience with FPGA at all)

What would be the most cost efficient way to do a labview device that will live by itself. Can it be done at all, or should I start look at some other alternatives instead.

Depending on the complexity of your test and the required data collection an FPGA could do it. On thing you should do is gather your requirements (data rates, motor charactistics, reporting criteria, desired cost range, etc.) and give a phone call to your NI rep. They are usually pretty good on suggesting options.

To be a small portable solution you are probably looking (like the previous posters said) at a compact RIO based system. Once you get it programed and "set free" as I like to think of it, no computer is necessarily required. You could even use the LEDs on a NI9474 to display go-no go type results.

Hope this helps,

Jason

[LAVA 1.0 Content]

Setting the hardware re:current and voltage aside...

I did one of those using a an FPGA that syncronized two steppers such that a poratable welder could cut ovals. If memory serves me (too many projects, so little brain) we just used FTP to transfer the pattern to the FPGA ahead of time.

So yes FPGA's will work.

Ben

[Louis Manfredi]

QUOTE (ned @ May 8 2008, 09:23 AM)

Have you looked at http://www.copleymotion.com' rel='nofollow' target="_blank">Copley Motion? They make several stepper motor controls with a CAN interface; you could control it using a CAN card and the FPGA in an NI cRIO. I'm not sure if this would fall in your definition of "cost efficient" since a cRIO isn't inexpensive, and writing code for the CANOpen protocol may be time consuming. Copley does provides libraries and LabVIEW examples but they rely on ActiveX and won't run on a cRIO.

--I was going to suggest the Copley motion gadgets. Not too expensive. They can be controlled through a serial port, using LabVIEW, using Copley's own setup application, whatever other programming method, perhaps even just a terminal emulator. For simple single-axis or multi-axis non-sychronized work, no need for cRIO or the CAN interface. If your laptop or computer doesn't have a built-in serial port, get a good one from a reliable manufacturer. (eg: NI) In my experience with the Copley gear there's a little learning curve, briefly steep with a good serial card, but likely to stay steep for some time with cheap serial card.

Best, Louis

Edit: purely formatting