Intrinsically Safe Relay/Solenoid Coil Circuit 1

[TurboXS]

Looking for an "intrinsically safe" relay driver circuit. In particular it must be fail safe which suggest series circuits rather than parallel circuit.

thanks

 

[BrianG]

What does "intrinsically safe" mean to you? It is NOT the same as "fail safe".

Whole industries like oil, gas, mining etc., have specific definitions of what intrinsically safe means to them, for example: that any fault will not cause sparks which can ignite flammable vapour or dust that may be present, or that it will not cause toxic fumes/smoke under fault conditions, or start a fire (e.g. overloaded resistor burnout), generate dangerous voltages, etc., etc. So, there are a whole host of these sort of conditions grouped under the heading of intrinsically safe. You need to define your requirement to us.

Fail safe for a circuit or component means exctly that: the failure defaults the system to a known, benign state (not always easy to achieve!)

Hope that helps!

 

[itsmoked]

BrianG is right on! We need VERY specific info on your application to help you correctly.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[TurboXS]

Your definition of intrinsically safe is what I understand it to be, however there is also an additional element to that in that the circuit configuration should be such that certain faults won't lead to an non-intrinsically safe circuit configuration, hence fail safe.

So the main question is;

A relay coil stores energy which must be controlled particularly when turning off the relay. The usual technique (neglecting speed of operation) is to use a diode across the coil. As this circuit configuration is in parallel if the connection to the diode fails the circuit now has no control over the energy stored, thus the circuit has not failed in a safe manner.

So what other techniques can we employ to make this safer?

 

[Ashereng]

TurboXS,

I will try to further BrianG's posting:

In my oil and gas industry:

"Intrinsically Safe" is a protection concept employed in potentially explosive atmospheres. Intrinsic safety relies on the electrical apparatus being designed so that it is unable to release sufficient energy, by either thermal or electrical means, to cause an ignition of a flammable gas. In addition, "intrinsically safe'" products are also incapable of storing large amounts of energy which might spark an explosion when discharged.


"Fail Safe" is another way of saying "Fail Position".
For a relay, fail safe has no inherent meaning. It fails either open circuit or fail close circuit. A human assigns fail open to be the safe mode or fail close to be the safe mode. The "safe mode" is in relation to the circuit, and the effect of the circuit.



"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[TurboXS]

I understand what you guys are saying about the differences between intrinsically safe and fail safe.

If it helps then ignore my remarks regarding "fail safe" aspect.

 

[itsmoked]

You are talking about a relay's coil inductance. With respect to intrinsic saftey you cannot have a single point failure that can cause a spark. If your relay coil "opens" in the middle I don't see how you are not going to get a spark no matter where you put a diode. Some devices can just never be made intrinsically safe. Are you sure you can use a relay?? Are you sure you need a mechanical relay?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ScottyUK]

Hi Keith,

A number of I.S. valves use coils to drive the spool within the servo. They aren't so different from a relay coil. They do require a barrier which limits the energy let-through under fault conditions, and they are specified as to the load characteristics that they may be used with. Energy storage capability in the load is one obvious parameter which must be considered.

Hi TurboXS,

First question for any hazardous area design: why can't the equipment be located in the safe zone?

Have you considered using an ExD enclosure? ExD allows you to build pretty much anything you want, but the enclosures become physically massive and expensive once they start to increase in size. The construction is akin to that of a pressure vessel and is designed to contain the products of an explosion long enough to release them slowly and at a low temperature by venting through carefully designed 'flame paths'. I.S. is only mandatory for Zone 0 environments, and there are very few reasons for putting a switching element in such a location.

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[Ashereng]

ScottyUK, I am interpretting ExD to be explosion proof?

If yes, explosion proof of course is also a valid option. Of course, the enclosure must be closed to maintain rating.
If you need to service, the precautions are more involved (eg. different type of permit, gas detectors, etc).

I agree. If you can put the switch out of harms way, the better.

However, from reading the OP, I am interpretting the question to be more about failure mode, rather than ex-proof/IS.



"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[ScottyUK]

AsherEng,

Yes, ExD = flameproof in the UK, broadly akin to explosion proof in the US.

I agree, the OP needs to detail his requirements a little more closely in the form of a spec for us to know what he is trying to do.



----------------------------------

I don't suffer from insanity. I enjoy it...

 

[TurboXS]

Intrinsically safe design can take on a number of different forms such as placing the design in explosion proof cases, removing the hazards out of the area, etc.

However the main goal is to control and prevent sufficient energy in a circuit both in operation and in a fault condition such that the energy is below that needed to ignite any flammable material or gases. Typical energy levels limits may be 50uJ or less. Intrinsically safe design cannot be achieved by enclosure design alone.

In this particular case the relay cannot be relocated outside the hazardous area and the relay is needed in the circuit. The relay itself is encapsulated to exclude flammable material or gases from either the contacts or the coil. The energy stored in the relay must be controlled by the use of external components. This is where the problem lies, the use of a diode configuration to control back EMF is acceptable to one standards authority but was not to another. It was deemed not to comply due to its parallel configuration. That is, if the diode failed or the connection was broken between the diode and the coil then the coil energy would not be controlled and it would be unsafe. The former standard authority recognises multiple diode configurations and "infalliable" PCB tracks widths.

Thus the problem is what circuit configuration can make this relay coil circuit safe. The normal way is to limit the current into the coil which reduces the energy stored (0.5*L*i^2) but then will the relay operate? Just to add to the problem, the intrinsically safe design considers any two faults as being possibility.

Slightly off the subject but you may think that fibre optics should be intrinsically safe.

Quote:
New standard offers answers
The ISA SP12.21, Fiber Optics, subcommittee is working to determine how much power it takes to create explosive energy levels with fiber optics, and what might be done to prevent it—setting limits for using optical fibers in hazardous areas (April InTech Standards department).

One of the reasons the committee is writing the report is to educate installers of optical fiber that this is a potential hazard. "In flammable or classified areas, granted there's a very restrictive set of conditions that has to occur to get it to ignite," said Rich Harner of Dow Chemical. "But if there's any possibility, we don't want to cause an ignition in a production area or actually in a telephone or communication situation that uses fiber."

 

[itsmoked]

If the AHJ sez you can't have a separate diode away down some wire that could break and spark because the coil is no longer diode protected then why not pot the diode into the relay thereby utilizing infallible traces:

http://news.thomasnet.com/fullstory/19813

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[TurboXS]

It is located within the potting.

 

[ScottyUK]

Have a look at

http://www.extronics.com/listing.asp

There are quite a few relays listed under "Remote I/O, Isolators, Barriers and Converters" on the products list. One style uses opto-isolation and a FET output for use wholly within the hazardous zone or as interfacing between the hazardous and safe zones. The other style only provides interfacing between the safe and hazardous areas and uses conventional relay design with what amounts to a zener barrier integrated within the relay design.

Note: I have not yet had cause to use these products, I am merely aware of them.

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[TurboXS]

I think careful selection and testing of the relay and a close examination of the internal construction of the relay is called for.

In any case I have a careful look at the links suggested and quiz the authority on the issue.

 

[kontiki99]

I only scanned through the posts.

I believe Fail Safe for a relay implies protection for the eventual failure of coil or coil energizing circuit or contactor failure (inclusive ors). Ex, design so engine keeps running despite loss of electrical power.

Of course failure of the spring to maintain the contacts in a particular position could be statistically significant in some designs too.

Relay sticking open or closed should probably be a non catastrophic event.

It doesn’t hurt to do a good Failure Effects Mode Analysis.

Protection against sparks and explosion is something different.

There is some good basic relay info here ->

http://www.leachintl2.com/english/desref.html

 

[kontiki99]

Its worth noting that relay open/close response times can degrade over time too.

For a given design, that can be significant too.

I've been involved in projects where we exercised complex relay circuits inside a black box on an automatic tester. If a relay was found to operate too slow a certain number of times we replaced it.

That was a reliability improvement project for that box. If poor relay timing could have caused a catastrophic timing or sequencing failure, the effect would have to be considered in fail safe design.

I think FEMA is the current best design practice.