PID Control Question

[mdstitt]

Hello,

I'm looking for suggestions for controlling a furnace using PID control for the heaters. This particular furnace will require a ramp to temperature over a specified variable amount of time using the LabVIEW PID control toolkit. The problem I am facing is the feedback for PID control. A pyrometer is to be used as the only process feedback for control. Unfortunately, the pyrometer does not register any temperature until 1000°C (the furnace will go to 2500°C). Does anyone have a suggestion as to how to control the furnace temperature until the pyrometer kicks in? My only thought is to specify an amount of power to begin the initial heat up, and then switch to the pyrometer once it begins to register. This approach may work fine, but it seems kind of "Clunky" from a programming point of view. It is proving difficult to determine when that switch may occur and to constantly monitor it for the appropriate time to switch while still maintaining a "recipe" approach to running the furnace.

Maybe some of you LabVIEW champions have a suggestion or two that may help my situation. Any suggestions are appreciated.

Thanks,

Mark

[JK1]

Hai,

Whats the hardware that you are using for your control application.

Coz few of the connector blocks (eg: scb 68) has a CJC built within it.  So you can connect  a thermocouple to it, measure the temperature until the 1000 deg mark.

Add some info the the post reg hardware and application that will give more clarity.

with best regards,

JK

[johnsold]

Mark,

Clearly, you cannot use PID without feedback. Until the pyrometer becomes active you will need to use some open loop approach.

How does the pyrometer output behave as the temperature approaches and exceeds 1000°? If it jumps from zero (or some other out of range value) to a meaningful reading, that could easily be detected with an In Range and Coerce function. If the output of In Range and Coerce is True, then use PID, else open loop.

Lynn

[mdstitt]

The hardware used is NI Fieldpoint. The pyrometer's output is maxed at 20ma until a valid reading is detected, then output is 0-20 ma. A thermocouple is not an option because the furnace temperature exceeds the limit of any thermocouple, and the furnace is an inert gas chamber where having an open port for a thermocouple would not be feasible. Pulling the TC out at 1000° would allow atmosphere into the chamber even with the best setup.

I think I will have to ramp up my output power slowly while monitoring the pyrometer. Once a valid reading is detected, I can switch to PID closed loop control. I just thought someone might have a slick way of doing this that I hadn't thought of.

[AnalogKid2DigitalMan]

Mark:

There are themocouples out there that can easily take the heat:

http://www.omega.com/pptst/XTA-W5R26.html

I use an XMO unit to measure furnace temperatures up to 1800C easily in an inert atmosphere. They are a C curve response, if you can add a port to the furnace and seal it with the probe installed, this may be an option for you. Just don't shut down the gas purge! I retracted one one at 1800C into ambient by mistake and the sparks flew!

Nanmac also makes exotic high temperature T/C's:

http://www.nanmac.com/