DC and AC Sharing A Wire

[gar]

110213-0812 EST

So far there is no clear cut definition for the statement of the problem.

I think it is clear this is not equivalent to a burglar alarm. In a burglar alarm system you are interested in the change of state from an intrusion. This sets a latch that must be reset by some means other than a return to the non-intrusion state.

The problem here I believe is personnel protection related to a machine (process). The type of system defeat you want to protect against is someone opening a panel or door and then with the item (door or panel) open having the person over-ride the sensor to make the monitoring system think the doors are closed so the machine can be run with the doors open.

A resistor at the end of the monitoring loop can not solve this type of over-ride. You have to in some way prevent over-ride of the door sensor.

But the starting point is an accurate definition of the logic of this monitoring system. Once you have a definition of the goal, then you can work on the system design.

.

 

[G._S._Ohm]

So far there is no clear cut definition for the statement of the problem.

. . .the starting point is an accurate definition of the logic of this monitoring system. Once you have a definition of the goal, then you can work on the system design.

.

Yes.
A simplified schematic of the existing system and a truth table or state machine representation for the new design would help.

 

[dbuckley]

I think it is clear this is not equivalent to a burglar alarm. In a burglar alarm system you are interested in the change of state from an intrusion. This sets a latch that must be reset by some means other than a return to the non-intrusion state..

No, no, and I say again No.

Although you can use an alarm panel just the way you say, and almost everyone does, by using the programmable outputs on alarm panels, they can do many non-traditional-alarmy things. And a pair of outpouts, non-latched, one following the tamper state, and one following the contact state is something that (some) panels can do.

Just because it say "Burglar alarm" on the front, it doesn't mean you have to use it that way!

 

[gar]

110213-1712

dbuckley:

big john has magnetic reed switches on his doors.

The switch opens when the door opens. The operator, job setter, or electrician with the door open pulls out his little magnet and places it next to the reed switch. If the system was latched on trip, then the person resets the latch, door is still open, then we have a potential safety problem.

The monitoring system has no way of knowing that this door is open.

This is a real world situation and occurs much of the time to do service work. May not be a good idea.

.

 

[dbuckley]

I certainly agree the sensors have "limitations", and could be easily bypassed with nothing more than a magnet.

But there's no latching that I'm aware of; take the magnet away, and the circuit is open again.

Of course, if someone tapes a magnet up there.....

 

[hurk27]

After reading what this is for, all I can say is you can not automate bad management, management must set rules that workers must follow, and have repercussions when they violate those rules, we have discussed this at length on here, no different then not making workers adhere to LOTO requirements, keeping power rooms locked, and or keeping control cabinets shut and or locked. the best way for something like this is to have a indication of a panel being open like a strobe light that will let others know when a panel is open, and a rule that if a strobe light is flashing they must find out why?

No matter what system you can design, there will always be someone who will figure out how to over ride it, management must come first.

But to offer a suggestion, I once put a motion detector in a control cabinet to alarm if it was opened when said controls was in operation, I used a key switch for the only bypass for maintenance personal when trouble shooting is being done.