PID for Thermocycler Oven/Freezer


Thread Starter

Jeff Hulett

I am working on a custom control system for an Oven/Freezer thermal cycler. Is a PID system appropriate for a process that utilizes both electric heat and liquid Nitrogen cooling?

We plan to implement this with National Instruments LabVIEW PID software. Is this appropriate,and if so, how do we handle controlling both the heater setting and the liquid nitrogen cooling for good control over +150C to -70C?
The PID could output the amount of heating or cooling effect needed, e.g., varying from +100% for cooling to -100% for heating. The heating and cooling actuators might use the PID value directly, or the value could be used in cascade
to control each through another loop. For instance, the electric heat control might operate a duty-cycled timer or SCR controller for 0 to 100% heating effect, while the liquid nitrogen might be controlled using a valve opening from 0 to 100% for cooling effect. A case for cascade might be if the cooling medium can be controlled on flow rate; then the primary PID output provides the flow rate setpoint, and a flow control loop worries about maintaining flow at that setpoint.

Ken Irving <[email protected]>
I can't comment on the LabView PID from experience, but I have used standard PID many times for Heat/Chill systems with electric heat and LN2 chill. If the process is well-lagged, and the response is integral+deadtime, you will
probably need P+D (no integral) tuning (though this is standard jacketed- reactor practice). Depending on the relative sizing of the heating and LN chilling, there may be a significant difference between heating and chilling gains - manual step-tests and measurement of heat/chill rates will show this - so the tuning may differ for heating and chilling. This should *only*
require changes in PID gain, as the other dynamics will stay the same. If you change the Proportional gain 'on-the-fly' take care that the PID algorithm doesn't exhibit glitches at the changeover - many (commercial) PID algorithms weren't written with dynamic gain in mind. If necessary, use a multiplier between the PID and the Manual station. You will also find that the effective chill gain depends on the operating temperature, due to the deltaT between LN2 (-196'C ish) and your oven, but over the range you
mention, this will be a secondary effect.

Hywel Thomas


Johan Bengtsson

The comment above is good, but I just want to point out that changing the gain on the fly or adjust the split point to a proper value (see my other post) is two ways of doing the same thing.

/Johan Bengtsson

P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]