RSTune

[AGENTTINFOIL]

Hey Guys,

just looking for some opinions on RSTune... Has anyone used this software? and what did you think? Is it worth buying and how easy is it to use?

thanks Travis

 

[Andybr]

Travis.

I do not use RSTune but I have used Expertune (RSTune is a badged version of this) for years. For anybody who needs to tune process plant loops regularly it is worth its weight in gold. Do not just think of it as an autotune package however. The analysis tools are where the real power lays. I am not sure whether it would be as useful for fast acting loops such as are found in motion control applications. The real cost savings come when you are trying to tune loops with long cycle times where you might need to wait several hours to see the effect of a change.
Andybr

 

[Ron Beaufort]

you might be interested in the basic overview I covered in this post ...

http://www.plctalk.net/qanda/showthread.php?postid=84553#post84553

also note that the last time I checked (several years ago) I think there were extra charges for the "professional" or "optimizer" version which had extra bells and whistles ... personally I've never tried that one ...

 

[Peter Nachtwey]

I am not sure whether it would be as useful for fast acting loops such as are found in motion control applications.

The theory is the same. It may work for slow motion control systems that can be controlled by a PLC PID. The real limitation is the PLC's ability to collect the PV as function of the CO at regular intervals.

The real cost savings come when you are trying to tune loops with long cycle times where you might need to wait several hours to see the effect of a change.
Andybr

Yes. There is enough information in the advanced control thread to roll your own auto tuning program in fact a Scilab version of Autotuning exists in the Advanced Control thread. The key is to save away the data from a RSTrend an convert it to a file with three columns. Time, control output and PV.

The advantage the 'packaged' version have is that they have more bells and whistles.

The key to making any auto tuning system work is the excitation of the system. If the excitation isn't done right the auto tuning results will not be accurate.

 

[akreel]

I've been using RSLoop Optimizer.

I haven't taken advantage of many of the features. I've only used it to create a historical trend of the process, and to use that data to get recommended tuning parameters.

I can't say that I'm getting optimal performance out of my PID loops, but the response is good enough for me to walk away. I think that the tool has saved me much heartache in the long run.