I am using Lakeshore 330 temperature controller to study Raman spectra of oxide materials in the temperature range 300 - 10 K. I have to record the Raman spectra at the desired temperature using Lakeshore 330 controller. From 300 to 200 K, the controller works very well in the auto mode. Below 200 K, the stability of the controller begin to deteriorate and below 100 K it is very difficult to control the temperature. This becomes very serious when it is required to record the spectra at close temperature intervals. I would like to buy the necessary Labview driver set along with Labview software for the temperature controller.
This doesn't sound like a LabVIEW problem to me, as the software should only be adjusting the set-point and heater range. Does your system operate well under manual operation?
What are you using for a temperature sensor? How far is the temperature sensor from the heating element? Is the Lakeshore 330 loaded with a calibrated temperature curve or generic? What system are you trying to control? type of heating element? Are you adjusting the power output range for the heater?
If you are trying to use the HIGH output for heater power to achieve fine control at low temperatures, you're going to have a bad time.
At present I am using the controller without any software. I change the temperature manually in the "Autotune" mode. I am seriously facing the problem below 100 K where the controller does not control temperature within specified limit. I very badly require someone's help with a Labview control software along with Labview driver to carry out the measurements.
The controller should never be run in Autotune mode; this is for calibration only. This mode re-calculates PID settings based on overshoot/settling time, etc. (is it running in PID mode? if the controller is running under P only, you will get massive overshoot)
I understand that you think software drivers will solve the problem, and while they may help a bit, you have to consider all aspects of your system. If you can not get good temperature control in manual mode, you will likely NEVER get it via software control (the software drivers will simply take over the role of you pushing buttons).
General advice since I don't know your system specs:
1) use a diode temperature sensor (not thermocouple); calibrated values loaded into Lakeshore preferred.
2) distance between heater element and temperature sensor should be a small as possible (less than 1 inch)
3) mechanical interfaces between heating element and temperature sensor should have proper amount of thermal grease (or approved brand of epoxy if permanent) to minimize thermal lag
4) for temperatures < 100K, use the Medium (or LOW) heater output range.
5) the lowest temperature you can reach will be dictated by your physical system (vacuum achieved, external energy sources, etc)
Drivers can be found here: http://sine.ni.com/apps/utf8/niid_web_display.model_page?p_model_id=2168
Hello Mr. Proland,
Thanks for your suggestion. I tried with manual mode and I could adjust the P and I values ( D is always constant at 100) to get the stability within +/- 0.5 K . However adjusting P and I values are bit time consuming. During 6 hours of operation, I could record the Raman spectra only for 4 temperatures. I do not know how to go by P and I values for different temperatures that will allow me to record the spectra at least for 10 different temperatures per day.
The sensor is silicon diode which is very close to the sample and the calibrated vales are loaded in the controller.
I use a turbo molecular pump for vacuum and vacuum level achieved is 10-6 torr.
I feel that through software control, better temperature stability may be achieved. This will also allow me to record the spectra for large temperature range per day.
I request your suggestions.
That derivative value sounds insanely high, it is typically the smallest of the three. Here's a guide for determining PID settings http://newton.ex.ac.uk/teaching/CDHW/Feedback/Setup-PID.html
If your goal is to have your software application set the PID values for given temperature ranges, you'll need to determine what values you want to use.
Also, temperature control should be based of a sensor that is close to the heating element to minimize thermal lag, overshoot, and settling time. These can be greatly effected by distance to the heater and the thermal resistance of each interface between the heating element and control sensor. Use a second sensor near the sample to record the actual test temperature.
In regards to the software implementation, I don't have anything on hand (I used lakeshore temp controllers at a previous job; our closed cycle He cryostat had +/- 0.8K stability using only a single set of PID values from 10 K to 330 K). The drivers that I directed you to should allow you to build what you want.
One note from my previous experience, it is typically faster to cool to lowest temp, then step to higher temperatures for your measurement (heating to a stable point was faster than cooling); so you'll get more measurements per day.