Verify that LED indication light is working

[rQx]

Hi,
We are thinking of implementing a light to our safety system that indicates that a machine is in a ceratian more dangerous state then normal.

Talking to a safety specialist he said (but didn't know how to) that maybe I could have an indication if the light was broken. For example a relay is activated when the indication light is broken.

I can't figure out how to achieve this. Also I don't know HOW the indication light behave if it's broken. Does it still have the same internal resistance? Is it lower, higher or zero. It seems that he took this out of thin air without knowing there is a solution for it.

/Tim

[L D[AR2,P#0.0]]

Make it so the led is on when it is not more dangerous, if the LED fails then it indicates a more dangerous state than normal. Make it an illuminated pushbutton so you press it to test the LED.

[rQx]

Originally Posted by L D[AR2,P#0.0]

Make it so the led is on when it is not more dangerous, if the LED fails then it indicates a more dangerous state than normal. Make it an illuminated pushbutton so you press it to test the LED.

I would say that the "norm" is to have my flashing light indicate a more dangerous state. A non-lit light is not a good way to inform of a more dangerous state in my opinion since it does not catch the eye or make you aware that something is as it shouldn't be.

I could have two light, but then I would depend on that the faulty light is fixed before the other fails

/Tim

[tarik1978]

You can try a lamp failure detection like this one:
https://www.tele-online.com/en/case-...shy-detection/

[ndzied1]

Perhaps an acceptable approach would be to have redundant LED lights and include a testing feature. Then have instructions that an operator or maintenance person must verify the operation of the lights periodically.

I have also seen a lot of safety circuits specify that they be supplied by SIL rated power supplies and include a specific overcurrent rated device.

[cardosocea]

The traffic light people already solved this by monitoring the current going to the lamp. That's one way of doing it, though if it's an LED it will be tricky I think.

[rQx]

Thanks!
I now have two approaches.
I found a relay from Schneider Electric RM35JA31MW that was an under/overcurrent relay, seems to fit my need. Also the redundant lights are an option
/Tim

[Puddle]

I've heard of applications before where LED indicators were specified to have transistor feedback to show they were energised. I wonder if a faulty LED would be detected on the dark test from the failsafe IO though?

[widelto]

I recall having seen many years ago a special output board that had a resistance and that resistance was used to measure if the light (bulb o led) at the end was ok or not. That was on Allen Bradley.

[TWS]

Preliminary Publication 1756-UM058D-EN-P - October 2004, Page 4-23 describes no load detection on diagnostic output modules.

1756-OA8D
1756-OB16D.

[James Mcquade]

we use AB safety gate locks and use the LOTO method.
the gate lock cannot be closed when a lock is on and the machine safety logic will not allow the machine to run without the gate lock being reset.
Can you do something like that?
james

[jholm90]

Please take a moment and look into muting lamps as they describe the concern noted. 1734-IB8S point-io input modules can have a "test output" (normally used for wiring in the estops) programmed as a muting output with current monitoring

https://literature.rockwellautomatio...t057_-en-p.pdf

[rQx]

The lamp and lower safety level is achieved with a key switch and a contactor relay (rated PLc).

In this safety level the machine can run with the lids open but only at 15% speed, this is what I want to indicate. As soon as the key switch is turned the light will flash indicating that IF the lids are open the machine can still turn at 15% speed. So I can't use any locks.

I appreciate the output modules examples but we use Siemens and the safety functions doesn't go through a PLC.