Needing help with supporting "PID VI"

[Shuai]

Hi, All:

       I am building an STM (Scanning Tunneling Microscopy) using labview . I need to use PID to control the current feed back loop. I am still using the old version from NI company named "simple PID Lv7.1". It doesn't work well. Does any one know if there is new version of PID with the same feature but doing better? Thanks a lot.

[natasftw]

If you hit ctrl+space and type "PID" do you see anything?

[GerdW]

Hi Shuai,

It doesn't work well.

Ok, in which way it "does not work well"? Did you try to improve it?

Does any one know if there is new version of PID with the same feature but doing better?

Yes, there is!

Since LV8 (?) NI offered the "PID & Fuzzy control" toolkit. This toolkit is still available and since (maybe) two years (LV2014?) it is for free with your LabVIEW installation.

So your question boils down to this question: Which LabVIEW version do you use right now?

[Shuai]

Hi, GerdW:

       Thanks for your reply. 

       "It doesn't work well", because the voltage feedback loop keeps overshooting, and the current can not get stable easily to the setpoint value. I tried to improve it for two days, but still not working well. 

        My labview is 2010 I believe, I will double check tomorrow. For the PID& Fuzzy control toolkit, does it come with the software CD? Or is it a sub vi that I can download and use? Thanks a lot. 

[GerdW]

Hi Shuai,

the PID toolkit is available on your installation DVDs, but in LV2010 you need to pay a license fee to be able to use it (after some limited trial period).

About the license part you should phone your local NI sales guy!

"It doesn't work well", because the voltage feedback loop keeps overshooting, and the current can not get stable easily to the setpoint value.

Sounds like unoptimized PID gains.

Maybe combined with a non-linear system behaviour: PID controls are made for systems with (mostly) linear behaviour…

[Blokk]

The PID& Fuzzy control toolkit is an add-on. I guess if you have LV2010 then you need to buy it...? Depends on your actual LabVIEW licence. You can contact a local NI representative to clarify licencing of NI products ( http://www.ni.com/company/worldwide-offices/ )

http://sine.ni.com/nips/cds/view/p/lang/en/nid/209054

But I feel if you have problem with the old PID functions, you will likely get problems with the new ones as well. Did you tune your PID control loop? Do you know how to do it? The above toolkit also contains autotuning algorithms, but you can tune a PID also "manually". One example is the Ziegler-Nichols tuning rule, but there are many other techniques (open/closed loop, etc...):

http://www.chem.mtu.edu/~tbco/cm416/zn.html

edit: GerdW was faster 🙂

[Shuai]

Hi, Gerdw and Blokk:

      Thanks for your reply.

      I did optimize the old PID, and not work well yet. I may go back and reoptimize it later. 

      I will try the new toolkit as well at a later time, and will let you know how it works.