Throw me a life line, PID problem.

[lcclower]

I am running a PID controller in front of a laboratory hotplate, I got the hotplate on the very cheap because the thermostat's out.
(Now the thermostat is completely out).
I drilled a hole in the plate the same diameter as the thermocouple, got a snug fit.
I set the PID at 450F to preheat, molds, pit a moldon the plate and started checking the temp with a laser thermometer.
up to 250F the reading on the PID and the laser thermo match each other, but at 450F indicated on the PID I am getting 350F on the laser thermo.
Anyone got any ideas?

[Mike W1]

Just a thought, but maybe when the TC heats up there's some shrinkage with the hole in the plate and you're losing a bit of the contact. That possibly would throw things off. I used a different type TC. Mounted a small piece of aluminum tightly to the aluminum plate on the top of my hot plate. It's tapped for the TC and screws in tight and used the thermal heat grease. Also don't have any faith in the Laser thermo's. When mine crapped out there was never gonna be a replacement.
http://i.imgur.com/n6w8MbP.jpg

[Charliemac]

It most likely you're controlling the temperature closer to the heater than the surface of the hot plate. Also sometimes the surface finish of the test piece will affect the reading, especially if its shiny. A surface thermocouple is a better was to go.

[jsizemore]

Where is the actual heating element? Is it bonded to the bottom or interspersed throughout the plate? Could be the thermocouple is real close or butted up to the element.

[17nut]

IR thermometers dont like shiny surfaces.
Cant it 15-20degrees and you'll probably get a completely different reading.
In your case i would trust the PID.

[mold maker]

The IR thermometer reading is only approx in most instances. It will determine between hot and cold where you can't reach, but reflectivity, the color of a surface, and even oxide on the surface will alter the reading.