Surge Protector on Every I/O

An MOV is a surge protector....but 1 time only.
Far cheaper choice and may be what spec intended.

NetNathan said:

An MOV is a surge protector....but 1 time only...

Click to expand...

Frequently, but it shouldn't be. That's one of the reasons I said transorbs are better.

From Wiki (Link):
"The main parameter affecting varistor life expectancy is its energy (Joule) rating. Increasing the energy rating raises the number of (defined maximum size) transient pulses that it can accommodate exponentially as well as the cumulative sum of energy from clamping lesser pulses. As these pulses occur, the "clamping voltage" it provides during each event decreases, and a varistor is typically deemed to be functionally degraded when its "clamping voltage" has changed by 10%. Manufacturer's life-expectancy charts relate current, severity and number of transients to make failure predictions based on the total energy dissipated over the life of the part. Note that in consumer electronics, particularly surge protectors, the MOV varistor size employed is small enough that eventually failure is expected.[7] Other applications, such as power transmission, use VDRs of different construction in multiple configurations engineered for survivability.[8]"

NetNathan said:

An MOV is a surge protector....but 1 time only.

Click to expand...

A $3 power strip with ten cent protector parts maybe sells for $30 or $85. When that ineffective protector fails on a first surge, then wild speculation says protectors are only a one shot device.

A proven solution, that might cost $1 per protected appliance, earths numerous direct lightning strikes. And remain functional for decades. Both use MOVs. But some are provided by companies known for integrity. So MOVs do not fail to promote profits. Companies such as Belkin, Tripplite, APC, Panamax, or Monster do not even claim to protect from typically destructive surges.

Or one can read spec numbers. A hundreds joules surge, too tiny to damage any appliance, may also damage that near zero joule protector. Then a naive consumer uses wild speculation to assume, "My protector sacrificed itself to save my ....".

A surge too tiny to damage any appliance also confronts a "surge protector....but 1 time only". Undersizing joules gets the naive to recommend it and buy more.

Unfortunately, most who make protector recommendations do not read specification numbers. Never ask a relevant question. Where do hundreds of thousands of joules harmlessly dissipate? A protector is only as effective as its earth ground. Ineffective "1 time only" protectors do not even have that always required and dedicated earth ground connection.

Effective protectors earth direct lightning strikes; remain undamaged for decades. Provided by companies known for integrity. Then robust protection inside equipment is not overwhelmed by a rare and potentially destructive surge.

Transorbs feature significantly less less energy levels. Cost more. But provide other advantages.

These types of Surge Protectors are considered for the benefit according to me, as they help keep signals strong. Whether you’re controlling a manufacturing process with a computer, distributed control system (DCS), programmable logic controller (PLC), or field device —surge protection barrier reduce or eliminate the risk to your equipment from lightning strikes and other harmful transient voltage or surge currents.

jjohnson21 said:

These types of Surge Protectors are considered for the benefit according to me, as they help keep signals strong.

Click to expand...

How do they do that? Read its specifications. A let-through voltage of 330 volts means it does nothing - remains inert - until 120 volts well exceeds 330 volts. How often does your AC main voltage spike to well above 330 volts? Where is this signal strengthening?

These protectors are also harmful if connected to signal wires (ie TV cable, telephone, etc) due to excessive capacitance. So again, where is signal strengthening? It is not found in numeric specs. It is not found in how these devices are constructed. What exactly does it do to strengthen signals?

An honest recommendation includes specification or other numbers that define a recommendation.

Surge protection at I/O points (which is what the OP inquired about), is not the same as surge protection for a service entrance.

I/O point surge protection is about protecting the panel and PLC against surges coming from the field wiring as well as surges originating inside the control panel. Surges from things like unsuppressed contactors, relays, solenoids, line induced transients from high amperage wiring, the operator running through the conduit with the forklift, etc. These are the sorts of things that can take out your entire control panel and cost thousands to tens of thousands of dollars to repair. This is the reason for the spec asking for I/O protection. It is purely an insurance policy against field induced surges costing the customer a ton of money. This protection can be as simple as a diode, MOV, a transorb, or more elaborate protection such as the previously mentioned Phoenix Contact modules.

Service entrance surge protection does nothing to protect the panel against these types of issues. In these cases service entrance protection is nearly worthless as it will only serve to prevent the transient from rolling into the rest of the service area. The customer will still be out money to replace components in the control panel. Not to say that service entrance surge protection is useless, it has it's place and is a good investment. It will not however protect from all instances of surge, because not all surges originate from outside of the service entrance.

In my opinion, I/O protection as well as service entrance protection is nearly always worth it in the long run. Even if you build the panel right in the beginning to insure against issues, someone, someday will modify it and when that happens, you never know what they will do that will mess things up.

Just my two cents.

The OP's application is for water/wastewater treatment. In a previous life, I worked for an integrator that did a lot of this type of work. In about every case, this same spec was boilerplated into the requirements. Finally figured out that this industry uses a lot of underground cabling, in a lot of wet locations, and for some reason seems to insist on 120VAC signals for every limit switch. We quit trying to fight it, especially when the spec called out Phoenix or equal. Everyone was using them; so did we.

icky812 said:

Service entrance surge protection does nothing to protect the panel against these types of issues.

Click to expand...

Those tiny noise transients cause by switching must be inside, already solved, and part of the design. And using devices designed for frequent transients (ie avalanche diodes, transorbs, transil). Anything on its AC cord does nothing to protect from that noise.

View some numbers. A 120 volt protector has a let-through voltage of 330 volts. That means it does nothing until a voltage spike well exceeds 330 volts. Furthermore, read charts for life expectancy. If protection is needed from frequent (ie hourly) spikes from switches, et al, then that protector is degraded in a week or month. Those are designed for spikes that occur maybe once every seven years.

Induced surges are just as irrelevant. The concern is for surges that actually do damage such as linemen errors, spikes created by tree rodents, and direct lightning strikes.

Lightning is a connection from a cloud (three miles up) to earthborne charges maybe four miles distant. Shortest path is never 5 miles across the sky. Shortest path is down to earth and then four miles through earth. If that station is in that current path, then either damage results. Or that current is connected harmlessly to earth - effective protection.

A direct lightning strike many blocks away is a direct strike incoming to everything. Nothing inside even claims to protect from destructive transients. Only service entrance protection that is low impedance (ie less than 10 feet, hardwire not inside metallic conduit, no sharp bends, etc) will protect from destructive transients created by stray cars, wind, or direct lightning strikes.

Only then are electronics on (for example) 4-20 ma current loops protected. Since those transients (not noise transients created inside equipment) must be eliminated inside equipment.

No protector does protection. Critical to whole system protection is single point earth ground. Protectors are only connecting devices to what actually absorbs potentially destructive energy - earth ground. Any protector that would 'absorb' a surge is only doing what is already done better inside all electronics.

Near zero protectors are rated at hundreds or thousand joules. A surge that tiny is often converted by electronics into rock stable, low DC voltages to safely power semiconductors. Destructive transients have must higher energy. Those can overwhelm what is already robust protection inside all electronics. Those are never avert by adjacent magic boxes with tiny joule numbers.

Best protection at each controller is already inside each controller. Concern is for transients so large as to blow through near zero joules protectors. And maybe even larger as to overwhelm best protection inside controllers. That means a solution that also answers this question. Where do hundreds of thousands of joules harmlessly dissipate?

mbartoli said:

The OP's application is for water/wastewater treatment. In a previous life, I worked for an integrator that did a lot of this type of work. In about every case, this same spec was boilerplated into the requirements. Finally figured out that this industry uses a lot of underground cabling, in a lot of wet locations, and for some reason seems to insist on 120VAC signals for every limit switch. We quit trying to fight it, especially when the spec called out Phoenix or equal. Everyone was using them; so did we.

Click to expand...

Have been in the same boat as "mbartoli" and agree with everything he said.

If it is in the specs to provide this protctection then I'd bet it's also in the specs exactly what is required. The last thing the OP should do is just put in some of the jury-rigged solutions mentioned here without first submitting his plan and getting it approved.

As the old saying goes: There's always enough time and money to fix it right the second time.

Westom, I wish I lived in your fantasy world where every system was designed perfectly. For major installations where you have someone like Black and Veach or GE or Fluor Daniel designing, I bet what you are saying is true.

In reality most systems designed by smaller integration firms don't get designed as you suggest. I wish they did, my job would be a lot easier.

icky812 said:

Westom, I wish I lived in your fantasy world where every system was designed perfectly.

Click to expand...

Fortunately we live in a free market. Equipment that fails due to inferior designs does not get purchased again.

Properly designed equipment even comes with numbers that define what it can withstand without damage. Sometimes they make a mistake. In one case, we found someone epoxy painted inside of a switch panel. 20,000 volt protection routinely found in switches was compromised. Since we live in a free market and since that product was designed by engineers (not business school graduates), then they immediately fixed that defect and welcomed the feedback. So that external transients would not cause any further problems to us or any other customer.

We did not cure it by adding transient protectors to a control panel. We identified and then solved what was (in this case) directly traceable to a human mistake.

Transient suppressors have another advantage. They can be powerful diagnostic tools to help locate a transient's source. Then that transient is averted at its source.

If equipment needs protection at its inputs, then something else is wrong. Suppressors are being used to cure a symptom - not solve a problem.


Another (separate) point. MOVs as one shot protection is best called a scam. Any protector that requires frequent replacement is ineffective. In the case of MOVs, it is grossly undersized (fails catastrophically) and used in violation of the manufacturer's Absolute Maximum Parameters.