Emergency Stop Circuit

[sauceboss]

I'm working on a project that has e-stop pull cords around the full length of a conveyor system which is about 750 feet long and it has 16 e-stop pull cord stations (8 per side). Due to the length, I'm looking at using 120vac for the e-stops and to avoid too much voltage drop by going thru all the e-stop contacts, I'm looking at running each e-stop signal back to the control panel and using it to activate the coil of a cube style safety relay. I'm then wanting to loop a 24v signal thru a contact of each relay and use it to activate a monitored safety relay which then activates the coil of a safety contactor to run my motors. So essentially I have 16 relays in my panel that mirror the state of each of the 16 field e-stops.


Here's a diagram of my idea (the actual design has 16 e-stops, 16 relays, and a PLC-controlled contactor after the safety contactor to activate once a reset button is pressed): https://i.imgur.com/2Qo5JQT.png


I've got a few questions though because I feel iffy about this setup.

1. Do I even need to bother with the relays or would voltage drop not be an issue here?
2. Is using the e-stop contact to active a safety relay and using the relay's contact as my actual safety contact allowed? I'm thinking what if the relay contact welds?
3. Related to (1), the relays are DPDT so if I used them as dual channel, would this help against the welding issue by having the monitored safety relay use dual channel inputs? Essentially, if the relay does weld, do both contacts stay closed when the coil is de-energized or just the welded contact?
4. How are e-stops located far away from the control panel normally done?

For reference, the parts I'm looking at all Allen-Bradley and are the Lifeline 4 e-stop pull cord, a 100S-C safety contactor, 700-HPS safety relay, and a 440R GuardMaster monitored relay.

 

[L D[AR2P#0.0]]

Here's an example from an Allen Bradley document that you should start with:


https://literature.rockwellautomation.com/idc/groups/literature/documents/at/safety-at059_-en-p.pdf

 

[jholm90]

If this is dual channel, then your incoming logic would require two relays per device. After completing a quick search there are no force guided relays with a 120 AC coil so you would have to use two of the 100S-C contactors (one per channel) as your incoming two relay solution instead of the little 700-hps pcb sized ones. If this is all one zone then I would tie all of the estop devices in series (in the panel on a common terminal strip) to turn on the contactor pair. Use an auxiliary contact back from each switch as a plc input or status indicator light on the panel for monitoring which switch was pulled. (Six wires plus ground for each device)

As for the wiring between the two input contactors above one each should go to the input of the safety relay wired like an estop. Both normally closed contacts of your output contactors need to be wired in series with the reset button. Because you have two of the incoming contactors if one gets welded both inputs will not complete with the same timing and you will be able to troubleshoot this. Welded contacts on force guided relays and contactors will prevent all contacts from changing state as all poles are firmly secured together (force guided).

 

[sauceboss]

L D[AR2,P#0.0]
That's one way I may do it in a system that was smaller. But with this system, I do not have a way to get 24v through all of the e-stops because they are so far away. In your image, where the NC contacts of each e-stop are, I'm wanting to instead put the contact of a 120v safety relay and use 120v with each e-stop to activate the corresponding relay.


jholm90:
According to the datasheet for the 7000-HPS series PCB relays "These small relays feature mechanically linked contacts required for safety circuits as per EN 50205 (Type B )". "mechanically linked" and "force guided" seem to mean the same thing according to Allen-Bradley.

If I tied the e-stops in series at the panel, then there would be over 7,000 feet of wiring that the signal has to travel through as well as numerous contact points. I'd be concerned that the voltage drop would be too high to actuate the contactor.

 

[sauceboss]

Would this be the better way to do it? https://i.imgur.com/4zT5n1v.png


Each e-stop actuates two PCB relays, one for Channel l of the monitored relay and one for Channel 2. It seems this would prevent a fault if one of the relays welds as the monitored relay could detect it. I also added a second safety contactor as before it could weld and the 480v would stay one. With two, the monitored relay could detect a weld and not restart the second one. I also added the PLC-based restart circuit where the aux of the monitored relay tells the PLC when the safety circuit is good, and then someone has to push the reset button for the PLC to energize its contactor.

 

[mass89]

I don't know exactly what 440 Relay you are using but the MSR125 can take 40ohms on its input circuits (and 500ohm for the MSR7).

Did you calculate 7000ft as a loop or as distance from panel to each estop?
7000ft of 1mm2 copper (don't know what this is in AWG) is around 36 ohm.
If you used the MSR7 you'd be fine, if you used the MSR125 you'd be cutting it fine with by the time you had contact resistance increased.
This case I would have 2 of the safety relays, have all of one side conveyor on one relay, and the other side on another relay. Wiring each E-stop contact to a relay then to the safety relay seems excessive.

Whatever you do, I wouldn't daisy chain the e-stops directly to each other, this is a nightmare to fault find if a contact/cable/connection was to fail in one of the switches.

 

[sauceboss]

I don't know exactly what 440 Relay you are using but the MSR125 can take 40ohms on its input circuits (and 500ohm for the MSR7).

Did you calculate 7000ft as a loop or as distance from panel to each estop?
7000ft of 1mm2 copper (don't know what this is in AWG) is around 36 ohm.
If you used the MSR7 you'd be fine, if you used the MSR125 you'd be cutting it fine with by the time you had contact resistance increased.
This case I would have 2 of the safety relays, have all of one side conveyor on one relay, and the other side on another relay. Wiring each E-stop contact to a relay then to the safety relay seems excessive.

Whatever you do, I wouldn't daisy chain the e-stops directly to each other, this is a nightmare to fault find if a contact/cable/connection was to fail in one of the switches.

The 7,000 feet total is loop distance if all the e-stops were wired to the panel and in series. 75 feet to first e-stop, 75 back to panel, 150 to second e-stop, 150 back to panel, 225 to third e-stop, 225 back to panel, etc.


1mm2 copper is around #18 AWG which has 6.385 ohms/kft or around 45 ohms for the 7000 feet. I was looking at using the 440R-S12R2 which is 900 ohm max input resistance although I don't know what that actually means. Does that mean the 24v signal could run through 140k feet of wire (900/6.385) and the relay would still detect it? Typically what I do is use a voltage drop calculator with a load current of 0.5A and it says I'll have 2 volts back at the panel.

 

[mass89]

The 7,000 feet total is loop distance if all the e-stops were wired to the panel and in series. 75 feet to first e-stop, 75 back to panel, 150 to second e-stop, 150 back to panel, 225 to third e-stop, 225 back to panel, etc.


1mm2 copper is around #18 AWG which has 6.385 ohms/kft or around 45 ohms for the 7000 feet. I was looking at using the 440R-S12R2 which is S ohm max input resistance although I don't know what that actually means. Does that mean the 24v signal could run through 140k feet of wire (900/6.385) and the relay would still detect it? Typically what I do is use a voltage drop calculator with a load current of 0.5A and it says I'll have 2 volts back at the panel.

Yeah, input resistance is the maximum resistance on each input channel, so from S11-S12 you can have 900 ohm and S21-S22 you can have 900 ohm.
I'm sure the manufacturer will understate these figures a little but I would use some of your own margin of safety as well (don't forget contacts resistance), there are some other small factors at play.

In terms of volt drop, one of the channels is normally 24vdc and the other like a floated voltage, in the case of safety relays I'd just rely on the manufactures spec of input resistance. At best this device will take 150mA to power the entire device (at power terminals), input circuits will only have a fraction of that running through them, volt drop will be nothing

 

[JasonTheSparky]

If voltage drop is a concern, then increase the size of the conductors used to link the various e-stops. I wouldn't be using an 18 awg conductor for field wiring anyway - pulling that small of conductors that far is just asking for issues...

 

[rjswagel]

Using force guide contacts on the input side and applying the concepts of the output side is wrong. The whole idea of a force guide contact is to provide feedback if a contact welds shut.
The only way way to maintain channel integrity is to have one relay for each channel of each estop ie... 32 relays for 16 estops. Because each relay will only use a single contact having force guided contacts is a moot point. You could use standard contacts then apply IEC60204 to determine the PFHd.

 

[RobertB]

Sauceboss,

I have a similiar set up here at my facility. Already engineered and running, with no permission to change program. Panic pulls were a continuous loop, so it was always a 'hunt' to find the culprit zone that was left open/triggered. We decided to put indicator lights above each zone location that could be seen from the main Control Console... or by a by a short walk.
Since it was a 3 wire system, we tied in continuous green indicator lights at each zone. If faulted, the green light would go out. System has helped a lot. However...you have 16 zones though...and even using LED indicator lights you still have extra mA to figure into your V.Drop formula.
If indicator lights are connected in parallel with your NC contacts, indicator lights would be OFF during NORMAL/non-faulted state.... nearly eradicating extra mA load. Downfall... you could only show 1 faulted zone at a time.
Hope this suggestion helps?

 

[Rson]

Are you locked into Allen Bradley?

Something like this would be perfect for a Jokab system that checks pulses IMO. I’m not aware if Allen Bradley has a similar operation.

 

[sauceboss]

I'm locked into using Allen-Bradley but their GuardMaster series of monitored relays I'm looking at does do dual-channel pulse testing. I just wasn't sure it could go as far as I needed it to but it seems it will.



Normally we'd have 24v Ethernet I/O blocks in the field and each e-stop would have it's aux tied to a port as well as a red beacon next to it so it's easy to know what cord was pulled as well as it would be shown on an HMI. This customer didn't pay for that nor even just lights. I will see if they would be down for pulling 7 wires to each e-stop instead of 5 (2 in, 2 out, ground, 2 extra) and then interconnect them in the panel so the ins and outs are daisy-chained and you can meter out any issues there. The two extra wires will be there in case they decide in the future to either have a beacon with 24v and common or run the aux back. I may also have them wire each pair of e-stops back instead of each e-stop since there's an e-stop on one of the conveyor and then 3' across it there's one on the other side. Cut down on the amount of wires and such.


Thanks for all the help everyone. I learned a couple new things.

 

[rkukl]

Have you considered AB GuardLink?

YouTube Video


https://ab.rockwellautomation.com/allenbradley/safety/guardlink.page

 

[sauceboss]

Have you considered AB GuardLink?

YouTube Video


https://ab.rockwellautomation.com/allenbradley/safety/guardlink.page

Too expensive for us at the moment but it's something that we keep on eye on.