Discrepancy times

aand74

Member
Join Date
Dec 2005
Location
Deinze
Posts
131
In a project with a safety controller, we have several doors with safety switches and several safety light curtains.
What discrepancy times should I use for these elements (all have two channels)?
 
The discrepancy time is only to deter manual bypass of the input and does not affect the reaction time when the device is activated. We always keep this value under 1 second to make it hard to bypass, but otherwise have no set standard. Usually we measure the actual discrepancy when closing the guard at a lazy pace and add 30% to that time. You don't want operators running around the machine to open and close guards and E-Stops that faulted all the time.

If the guards use magnets and they are mechanically located near the hinge of the door, the natural discrepancy time can be very large, so it is best to have the magnets on the latch/handle side of the guard.
 
There isn't a standard. I can say that AB older GuardLogix this was hard-coded to 500mS. I typically see times of 100mS, 250mS, 300mS or 500mS.

You want it long enough that normal operation of the guard (door or light curtain in this case) causes a false indication. On the other hand you detect a failure of the system (or prevent tampering).

A lot depends on the exact nature of the design. For example two independent switches could take longer to toggle than two channels on the same switch. Location of the actuator can make a difference as well.

Winky's advice is pretty solid. It doesn't have to be exact, give yourself some leeway and round up. You don't want to get called out because someone closed a guard real slow.
 
For Siemens safety I/O, the discrepancy behaviour is selectable so the time you have set may affect the response time.

safety.jpg
 
That's interesting. I wonder what scenario would require a "hold last value" option for a dual redundant safety input.
 
That's interesting. I wonder what scenario would require a "hold last value" option for a dual redundant safety input.

My guess is that it is something in Process safety, instead of Machine safety, but I couldn't say what.

Either that or a setting left over from a really old safety standard and only available for compatibility reasons. Siemens has NEVER done anything like that before.... (I'm still bitter about the S5 timers still being around in the 1500).
 

Similar Topics

Hello everyone, I'm working with an old ACM3710 device and I'm encountering an issue with the data frames I'm sending over an RS232 connection...
Replies
5
Views
653
The sticker on the side of my PLC says it is a Series C and FRN 6. In RSLogix 500 under processor status, the OS FRN shows 0.00 and the OS Series...
Replies
6
Views
3,179
Hello, I am doing a PLC5-Contrologix plant migration. So far with the same exact GSV WallClockTime to DateTime array half of my controllers...
Replies
2
Views
1,560
Hi Friends, I'm working on a Redundancy discrepancy on our VijeoCitect SCADA, The Primary can not take over when the Secondary Server goes down...
Replies
0
Views
1,278
I have a Keyence light curtain, SL-V32HM that is wired to an IB8S in an AENTR. I have the discrepancy time in the IB8S set at 100 ms. The...
Replies
7
Views
2,854
Back
Top Bottom