Emergency switch

M

Marc Sinclair

> > I suppose that your NC contact is electromechanical, if this is the
> > situation you could replace it for a solid state NC contact and forget any
> > type of vibration.

You can forget safety and standards compliance too.

> > However the idea to put two contact blocks in parallel is
> > convincing.

If the switch can fail NC one time in a million, putting two in parallel makes the chance of failure one in half a million, surely a degradation in safety

> > Another simple means we follow is to wire two such switch contacts in
series.
> > The operator has to press both these switches simultaneously in the
event of
> > Emergency.

I'm horrified, Safety is the issue here, peoples safety. This setup would not be allowed in Europe. the regulations specify that there should be no ambiguity in emergency stops. You should work harder to find a solution that does not degrade safety.

<moral> If this is to be an automated world it is our job to make it safe. If people are injured or killed by machinery and robots then trust in
automation will diminish. </ moral>

marc sinclair
 
R

R A Peterson

I'm sure the original poster meant "wire in parallel" as wiring them in series would not provide the solution he suggested (requiring depressing both PBs to estop the machine). I think the rest of us agree this is a really bad
idea, and one that ought not be given any more thought to.
 
B

Bruce Axtell

I agree with Marc and others who stated the contacts should not be in parallel, but my take on this was that the two parallel contacts were on the same operator. The issue was vibration, and the likelihood of two contacts opening at the same time due to vibration would be reduced if in parallel, thus maintaining the integrity of the circuit, yet both would open when the operator was pressed. Never would I advocate two separate stop buttons in parallel in this circumstance.

Just another interpretation of the original post, FWIW (for what its worth).

Bruce Axtell
Engineered Control Systems
 
J
<snip>
-> > > The operator has to press both these
-> > > switches simultaneously in the event of
-> > > Emergency.
->
-> I'm horrified, Safety is the issue here,
-> peoples safety. This setup would not be
-> allowed in Europe.
<snip>

This would not be allowed here in the states, either, were OSHA to read this thread.

Bottom line: is the priority safety, or convenience? Most people don't like to be asked this directly, but to some, the priority is truly convenience, and not the safety of the operator. If this is the case, take the switch out. If you are more concerned with safety, mount the button somewhere it is not going to be subjected to that type if vibration.

The other note (snipped) about reducing from 1 in a million to 1 in 500,000 is actually conservative. You actaully have 3 potential failures : Block 1 fails closed, block 2 fails closed, or the contact bar gets dislodged and
shorts from the first set to the second set. This puts your safety down to 1 in 333,333.

This is a battle which I fight, sadly, almost daily. Emergency stop is just that, an EMERGENCY! ( I have had some managers suggest that the E-Stop just tell the PLC to stop the machine, like a machine stop!) Nothing should be
done to comprmise the safety of the operator on the machine. My experience shows that they do that enough themselves.
 
J

Johan Bengtsson

Well, isn't that quite standard behavior?
The E-stop breaks the circuit when pressed (NC=Normally Closed contacts) one button pressed will stop the machine. This will not solve the original posters problem, putting two contact blocks on the same pushbutton in parallel will perhaps (at a cost of safety).

A better E-stop pushbutton may have enough better reliability to accept the cost in safety of connecting two contact blocks in parallel. (someone will probably argue against this...)


/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
R
Something I ran into that has not been mentioned in the many replies to this question is: What is the voltage in use in the E-Stop circuit and the wire length of the total circuit?

I had a project where we were using 24VDC E-Stop circuits through (6) PBs and a total wire length of almost 100 meters. The voltage drop through the wiring at the current required for the E-Stop relay was high enough that it made the relay unreliable. Sometimes we could not get it to pickup, sometimes it held for a few minutes/hours and then might open, sometimes it was fine for days. It sounds similar to what Tanweer Ahmed is seeing in his application.

The two things we learned are:

1. Check the voltage at the relay if you can get it energized. Be sure it is close to the rating for the coil. If it is less than 90% of the nominal coil voltage, you may have a problem.

2. Be sure to use low voltage rated contacts on switches as appropriate. Higher voltage rated devices often don't work well at the more and more
common lower control voltages. Also note that most low voltage contacts are rated for resistive loads only, be careful what you hang on the end of that circuit.

Russ Kinner
AVCA Corporation, Maumee, OH, USA
 
T

Tanweer Ahmed

Thanks to all those responding to this thread raised by me.

As a short term solution I have placed two contact blocks (NC) in parallel at all the push buttons. Note that the operator will not need to press two buttons but pressing a single button will open both the contacts. The probability to open both the contacts at the same time due to vibration is reduced by doing so. Thus improving the operation without degrading the safety aspect.

As a long term solution we are redesigning the circuit in such a way that instead of connecting all the E-Stop push buttons in series we will wire
individual push buttons to energize one relay each in latching mode i.e. if a NC push button is pressed for a short time or the contact opens by itself for any other reason the relay will deenergize and will not energize back unless and untill the reset button is pressed. The reset button is not located in the operator console but in the engineers control room. Contacts (NO) of all these relays will be connected in series to get the output signal for E-Stop command. This E-Stop command goes directly to the machine to freeze all the wheels immediately (no PLC or other electronics in between). By doing so the operator will not face any difference in operation
but the plant technicians, in the event of any mallfunction will immediately know which part of the circuit failed and can rectify the fault. The
parallel contact blocks will remain in place.

I will wellcome your comments on this design. Please note that its a legal requirement here to keep the E-Stop circuit independent of the normal
machine control system i.e. PLC, PC or DCS etc. controlling the normal machine operation.

Best regards,

Tanweer Ahmed
Phone: +9221-636 5519
Fax: +9221-568 2972
E-mail: [email protected]
 
E
Tanweer, I R-E-A-L-L-Y don't like the idea of two NC contacts in parallel...way too risky! What if ONE contact falls off the switch? It stays
CLOSED! If you are having nuisance problems with one or more switches, replace them or try a new brand. Perhaps it's not even the switch that's
causing the problem. Have you checked wire terminations for this circuit at terminal strips, etc.?

How long has this system been in service before you started having problems with the E-stops? If they've been trouble-free for some time, then something has changed/broken/worn-out/etc.. Do a little more investigating. Your quick-fix "solution" will only cause grief (or worse, injury!) in the future.

Russ Kinner's idea about voltage drop is a good one and should be explored. You may have been "borderline" for a long time without knowing it. But at least it was SAFE!

Since you've already spent the money for those "additional" contact blocks, THROW AWAY the old ones and REPLACE them with these new ones,
instead of doubling-up each switch.

I also wouldn't recommend your circuit "redesign". It will only add more complexity to the circuit, and more components that can fail. If you want to monitor which E-Stop was pressed, use that second set of contact blocks for your latching relays, but in a separate circuit for monitoring purposes only.

Continue reading that "legal requirement" you mentioned.... I'll bet you will also find that it's a legal requirement NOT to have two NC contacts in parallel for Emergency Stop circuits ;o)

Regards,

- Eric Nelson
[email protected]
Controls/Software
Packaging Associates Automation Inc. [email protected]
Rockaway, NJ, USA
 
M
Well that is one solution. I'm still curious, what is the make and model of the switch and contacts that are giving you the problems?

Mark
 
B

Bob Peterson

>As a long term solution we are redesigning the circuit in such a way that instead of connecting all the E-Stop push buttons in series we will wire individual push buttons to energize one relay each in latching mode i.e. if a NC push button is pressed for a short time or the contact opens by itself for any other reason the relay will deenergize and will not energize back unless and untill the reset button is pressed.
> <clip>

this sounds like you are asking for trouble. you are making the emergency stop circuit more complex and adding delay into it (by the relays) and adding potential failure points (the relays).

My suggestion is to monitor each point in the chain with a fast PLC input. That way you can catch when it trips and figure out which one of the PBs has tripped and fix whatever is wrong with it.

One time I had a similar problem. Turned out one of the estop pbs had a contact block on it that was damaged. sometimes it would open due to
vibration. we only found it because a millwright was hammering on the structure nearby to this PB and it repeatedly tripped the estop circuit but
the latching PB was not tripped. we were lucky the millwright decided to hammer on the structure the PB box was attached to or we might never have
found it.

Bob Peterson
 
J

Johan Bengtsson

Interesting idea, since the safety is discussed for every soulution i would say that this is more dangerous than connecting the contacts in parallel: If the input card in the PLC is broken in a way that allow a current to pass between two inputs the pushbutton will not stop the machine.
Perhaps an unlikely error but I think it is more likely than a contact block in an E-stop pushbutton not to open when the pushbutton is pressed.

Your post raises an interesting fault finding tool: a hammer, more useful than a lot of people may think. So far I have mostly seen it at work on valves however...


/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
Perhaps I read the post wrong regarding the PLC I/O . . . but I interpreted it as a method of monitoring the contact pair only. We do this on most every system we design. One set of contacts in series for the E-Stop Loop and the second set of contacts individually wired into the PLC to let the operator know which set of contacts were responsible for the shut down. If the post below was correct in their interpretation . . . I am in agreement with their response.
 
Doing it the way you interpreted it is completely ok, but it does not solve the problem in the way the original poster wanted. Since vibrations was (supposed to be) the orignal problem and the
contacts opened due to this and that was what was meant to be monitored using a second pair of contacts don't solve the problem. It does solve the problem as to see what E-stop pushbutton was
pressed but not which one opened due to vibrations if only one of the pushbuttons contacts have opened.

/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
T

Tanweer Ahmed

I'm compiling here the suggestions and comments so that we should reach a CONCLUSION:

DELAY TIMER:
Try a safety timing relay. Honeywell makes such a beast - Anthony Kerstens P.Eng.
Just take the input directly into a flag-bit and use timer to insert time delay - Hiannie.
-----I don't want to put any delay timer with E-stop circuit because
emergency is EMERGENCY.

TWO PUSH BUTTONS:
wire two such switch contacts in series. The operator has to press both these switches imultaneously - Santhiraj
the regulations specify that there should be no ambiguity in emergency stops - Marc Sinclair
This setup would not be allowed in Europe - Bob Peterson
This would not be allowed here in the states, - Jansen, Joe
Never would I advocate two separate stop buttons in parallel - Bruce Axtell


TWO CONTACT BLOCKS:
We use two Emergency switches in the operator desk with 2 sets of contatcts - Anandac
Likelihood of two contacts opening at the same time due to vibration would be reduced if in parallel - Bruce Axtell
Putting two contact blocks on the same pushbutton in paralell will perhaps solve the problem (at a cost of safety) - Johan Bengtsson
What if ONE contact falls off the switch? - Eric Nelson
Surely a degradation in safety - Marc Sinclair


VOLTAGE DROP IN WIRING:
The voltage drop through the wiring at the current required for the E-Stop relay was high enough that it made the relay unreliable - Russ Kinner


SOLID STATE CONTACTS:
replace it for a solid state NC contact and forget any type of vibration. - Jos Castell
You can forget safety and standards compliance too. - Marc Sinclair

Regards,

Tanweer Ahmed
Phone: +9221-636 5519
Fax: +9221-568 2972
E-mail: [email protected]
 
Since this thread has been identified as an "ENGR" problem, and you are sure that vibration is the cause, then substitute alternate-action HS
devices, i.e., Push (alt pull) to Trip and Pull (alt push) to Reset.

Additional concerns: make sure that contacts are suitable for the applied current and voltage, especially if low value. Also insure that contact surfaces have a wiping action when they close. Lastly, insure that they will function reliably for a small number of operations. The latter will avoid the "relay operation" fiasco that occured in the nuclear industry. Relay "life" was certified for millions of operations, not for few operations.

If all else fails, then attack the cause, not the symptom.

Regards,
Phil Corso, PE
Trip-A-Larm Corp
 
M
I'd still like to know what make/model switch you were using. We've never had any problems with switches sold as ESTOP switches that had late-break NC contacts.

Mark
 
J

Johan Bengtsson

My conclusion:
1. Check if the voltage drop is the reason in this particular case, if it is check if this can be easily fixed (thicker wires or rerouting the cable, it may be unnecesarily long) All contacts
should remain in series. I suppose you have relay(s) connected to the E-stop buttons, there might exist relays that are more suitable (less
necessary current, less necesary voltage) without loss in safety.
2. If that isn't the problem try to isolate the puchbuttons from vibrations and/or replace them with "better" ones (less sensitive to vibration
but not less secure). If the problem remains and you find an E-stop pushbutton with a so much higher safety on each contact block that you
not will loose overall safety by connecting them in parallel (safety compared to a "standard" E-pushbutton without this type of connection)
Then I see no real risk in doing this type of connection. The requirements of the individual contact blocks are higher in this case.

I know it sounds risky and that we are talking about human security perhaps even life/death, but when it comes to E-pushbuttons and their
reliability it comes to that anyway, regardless of how you do it. It may be hard to motivate such a thing but too frequent false trips will
probably result in someone disabling the E-stop entirely some day, and where is the security then? I have seen people disbling safety equipment to rise the production rate a little bit!


/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
R

Ravish kamath

try for 'PILTZ' make solid state emergency protection for feed back to plc with 3-emergency switches. At the field end use permanent magnet type switches,which can with stand vibrations. ravish
 
Top