LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

PID setting recommendations (and LabView tips)

Hi guys.

I'm measuring resistance of a sample by varying temperature from 300K to 2.6K in intervals of 0.5K. Temperature is adjusted using a Sumitomo CKW-21 Helium Compressor Unit and a Lakeshore 335 Temperature control. Because the compressor doesn't provide a constant flow, but rather in short bursts, it appears to be essentially impossible to stabilize temperature below 10K and it jumps all over the place.

I was wondering if you guys have any suggestions as to the optimal PID setting for the Lakeshore for different temperature ranges.

I'm thinking about trying to write a code in LabView that will automatically adjust it during the run. I have absolutely no idea how to do that right now, so any hints in that area would be greatly appreciated too!

0 Kudos
Message 1 of 5
(2,008 Views)

PID control algorithm is meant for linear systems. From your description it looks like the Temperature control system itself is non linear below 10K. You will have to understand the system behavior thoroughly before deciding on the control strategy selection. 

 

NO one would be able to suggest the PID parameters. It is solely depends on your system. Even you will have to decide on P or PI or PID your system requires.

 

I think you need some one expert in Control System Design.

0 Kudos
Message 2 of 5
(1,985 Views)

Yeah, I understand it's not possible to stabilize the temperature at each interval perfectly given the setup. I was just hoping to optimize it. For some reason Lakeshore PID autotune fails at stage 3, which has reasons for failing listed as "system response is too slow, or the heater is too underpowered for the system to autotune", but the system response is actually really fast and I don't think the heater is underpowered, so I'm not sure why it's complaining 😞

0 Kudos
Message 3 of 5
(1,979 Views)
If you check the wikipedia article there's a acouple of way to achieve the pid values. Half the P of self oscillation,, 40% of the oscillation time as D and so on, as a astarting point.
/Y
G# - Award winning reference based OOP for LV, for free! ADDQ VIPM Now on GitHub
"Only dead fish swim downstream" - "My life for Kudos!" - "Dumb people repeat old mistakes - smart ones create new ones."
Certified-LabVIEW-Developer
0 Kudos
Message 4 of 5
(1,974 Views)

I guess the Lakeshore uses the Relay-Based Auto Tuning algorithm. It will generate pulse (Periodic square wave, period should be more than the time constant of the overall system) output from the controller and check for the feedback response. Check if you have any setting with respect to period of the pulse generated. 

 

Can you share the overall block diagram of the system for better understanding?

0 Kudos
Message 5 of 5
(1,970 Views)