I'm wondering. I was researching into the Johnson Controls unit, and in the PDF document from the Johnson Controls site, there is this warning:
Originally Posted by http://cgproducts.johnsoncontrols.com/MET_PDF/125188.PDF
The document does state that the unit will shut off in the case of a sensor failure, and will also shut off in case of a program failure.
However, this unit is relay controlled, and relays can stick in the closed position.
So with the Johnson Controls unit, at least, the only way to ensure safety is to use two units - one as the main unit, and a second one set a couple degrees higher so that it will shut down the primary unit if it fails.
I couldn't figure out what the Ranco and HerpStat units use for voltage control. Do either of those use solid state switching as opposed to a mechanical relay switch?
Here's what I found for Ranco:
http://www.icca.invensys.com/uniline/d/d71.pdf
And for HerpStat:
http://www.spyderrobotics.com/produc...ual_online.pdf
Interestingly, HerpStat makes no claim that the unit will shut off in the event of a program failure.
So, any thoughts on redundancy of temperature controls? Considering a BP can live over 20 years, I think one could assume that the risk of a temperature controller failing in that time frame is fairly high. So does anybody think the cost of an extra unit to maintain redundancy is justified, or is it okay if a snake is exposed to a 125 degree strip of flexwat for several hours (the time it may take for you to get home from work and notice the temperatures are way high)