Why Do I Need Electrical Isolation?

Why Do I Need Electrical Isolation? An Acromag Whitepaper

Examining the important aspects of electrical isolation; what it does, why we need it, and how to test for it.

Acromag is in the business of signal conditioning. We manufacture circuits that amplify, isolate, filter, and convert one signal form to another. Most of these circuits also provide electrical isolation. However, added isolation has a cost. Sometimes customers question their need for isolation or fail to recognize the need for adding isolation in their application. This paper covers the basic aspects of electrical isolation.

Briefly, electric current refers to the conceptual flow of atomic particles or electrons through wires and electrical devices. Conductive materials like metal and water allow electric current to easily pass through them. The force that drives electric current to flow through a conductive medium is potential difference or voltage. The opposing force that curbs or limits this current flow is resistance. Comprised of 60% water, the human body is an excellent conductor; except that electric current from a source allowed to pass through the body can induce injury via electric shock. Materials that are weak conductors of electricity have high resistance to current flow; these materials are often used to add isolation or insulate circuits. In general, greater force (voltage) along a conductive path will drive higher current flow; if poorly controlled in the absence of resistance (insulation/isolation), it may result in circuit damage, personal injury, or even death.  

Table of Contents

  1. What is Isolation?
  2. Common Methods of Signal Isolation
    1. Transformer or Inductive Coupling
    2. Galvanic Isolation vs Galvanic Isolator
    3. Optical Isolator, Optical Coupler, or Fiber Optic Link
    4. Capacitor
    5. Magnetoresistance
  3. Why Do I Need Isolation?
    1. Block High or Hazardous Voltage
    2. Protection from Electric Shock
    3. Reject High Common-Mode Voltages
    4. Protect Circuits from Transient Noise
    5. Break Potential Ground Loops
  4. Safely Testing Isolation
    1. DC Voltage Insulation Test
    2. High-Potential (HIPOT) Testing
    3. Common-Mode Noise Rejection
  5. Conclusion
  6. More Resources

Click here to download Why Do I Need Electrical Isolation?


What is Isolation?

With respect to electric circuits and electronic instruments; “isolation” means the deliberate introduction of a non-conductive separation to inhibit current flow. Galvanic Isolation is the process of blocking current flow to prevent a direct conduction path between circuits is called . This term sometimes causes confusion because “galvanic” refers to metal and the electrochemical process in which one metal corrodes to another when both metals are in electrical contact and in the presence of an electrolyte. But galvanic isolation refers to the absence of metal or a conduction path.

Galvanic isolation is accomplished by physically adding distance, clearance, or insulating material around a circuit to block unwanted current flow. But how do we preserve a circuit signal and allow it to be transmitted across an isolation barrier? We can additionally isolate the signal by transmitting it magnetically using transformers or magnetoresistance. We could transmit it optically using optical couplers, optical isolators, or fiber-optic media. Or we could capacitively couple the signal across an isolation barrier using capacitive isolators.

Signal isolation is usually accomplished by a combination of actions; physical separation and insulating material, combined with a method of isolated signal transmission (magnetic, optical, or capacitive). The important thing is that regardless of our isolation method, isolation prevents the electrical conduction of unwanted current between circuits; while still allowing our wanted signal to cross an isolation barrier without providing a conductive metal path.


Acromag offers the industry’s best selection of process signal isolators. Click here to view them.

Common Methods of Signal Isolation

So, the real trick in isolation is not how to add insulation or separation to a circuit. The trick is adding electrical isolation to block unwanted signals, while still allowing the wanted signal to transmit through the circuit; and without providing a direct (galvanic) path for signal conduction. Below are some common ways to isolate a signal between two points without providing a direct conduction path between them.

Transformer or Inductive Coupling

The most common example of a galvanic isolator would be the transformer. The primary and secondary windings of a transformer are insulated from one another. They don’t connect to each other electrically, so there’s no metal to metal contact. Instead, they use magnetic field flux, generated by coils of wire overlapping a ferromagnetic material; signals are inductively coupled to/from the ferro-magnetic material using a varying magnetic field.

Transformers buffer or change voltages by stepping them up or down. They’re also used for isolating signals for safety, as well as isolating a circuit from AC line voltage. A transformer allows its secondary windings to be offset from a ground reference on the primary side. Thus, breaking potential ground loops between the primary and secondary circuits. Because it involves the mutual inductance of magnetic fields from coils, it can be more susceptible to magnetic interference. Further, unless properly shielded, it can also be a source of magnetic interference to adjacent circuitry (inductive and radiated emissions). Transformers are traditionally bulkier than optical or capacitive isolators. However, there’s newer technology that uses chip-scale transformers, encased in integrated circuit style packages, to magnetically isolate signals. (For one example of this technology, see Analog Devices isoPower® and iCoupler® technology.)


Galvanic Isolation vs Galvanic Isolator

Galvanic isola-tion” should not be confused with galvanic isola-tors.” A galvanic isolator is used to block low voltage DC currents from coming on board boats, via shore power ground wires. These DC currents can accelerate galvanic corrosion on underwater metals of boats and cause extensive damage; metal in hulls, zinc anodes, prop, drive-shaft, etc. Galvanic isolators are used because boats plugged into shore power at marinas each act like giant batteries; contributing DC voltage to the power signals via the ground wires. This produces corrosive electric currents through all the metals that contact the water. The metal and water form a giant battery, causing the metals to corrode in galvanic fashion; the way terminals and plates of a battery corrode as current passes through them. Zinc anode is a sacrificial metal added to a boat’s conductive metal surface; concentrating the resultant corrosion to itself.

How Galvanic Isolators are Used

Galvanic isolators are inserted in-line with the green safety ground as they enter the boat, between the shore-power inlet and the boat’s electrical panel. It allows AC fault current to pass through it while blocking DC current. Thus, AC faults are transmitted back to the power source, where they can safely trip a breaker or open a fuse. Simultaneously, destructive galvanic DC battery currents are blocked/minimized to reduce galvanic corrosion. This enables the zinc anodes of your boat to help protect its underwater metals and not those of other vessels that surround it; as they act to control the corrosion of the metal attached to your own boat. Most galvanic isolators are designed to be fail-safe; meaning that if they fail, they do not also open the path to ground for fault current.

Your first instinct might suggest, “Why not just remove the ground-wire?” However, this would be dangerous. The ground wire must be present to carry fault current back to the dock power source or transformer. Otherwise, if you accidentally contacted the shore power AC line by some type of wiring fault, you could become the medium to carry fault current back to the transformer; this could be fatal.


Learn more: Why You Need USB Isolation for Industrial I/O

Back to table of contents

Optical devices transmit information through their medium or across their barriers using varying levels of light intensity; with no direct electrical conduction path. A light source (transmitter, typically an LED) sends light waves to a photo-sensitive device (receiver, typically a photo-transistor). The combination is often held in place with insulating plastic, like that of an integrated circuit IC. Alternatively, transmit and receive functions are separated using a transmitter linked to a remote receiver via fiber optic cable. One major benefit of optical isolation is its inherent immunity to EMI (Electro-Magnetic Interference or electrical and magnetic noise).

Some comparative disadvantages to optical isolation are its:

  • Generally higher power dissipation
  • Susceptibility to temperature effects
  • Traditionally slower speed (specifically optical couplers, not fiber optic links)
  • Finite life of its transmitter (LEDs degrade over time)
Acromag 612T DC Voltage/Current Input Dual-Channel DC-Powered Transmitter Drawing with Optical Isolation

Capacitor

Remember that capacitors generally allow AC current to flow, but block DC current. Thus, they efficiently couple AC signals between circuits, at different DC voltages, via a varying electric field. There are many capacitive isolation devices available, and it is a common technology of digital isolators. Many modern devices will even use isolation-rated capacitors to connect between grounds on each side of an isolation barrier. This provides a conduction path for transient signals; perhaps to earth ground (also helpful in quelling radiated emissions). Capacitive isolation is faster than optical isolation.

Unfortunately, capacitors are more prone to failure when stressed by voltages above their voltage rating. And for some capacitors, this failure mode can result in a short circuit condition; abruptly ending its isolation-ability, as well as possibly rendering its circuit unsafe or hazardous. Safety rated Y-type capacitors are used in line to ground applications and are designed to fail open; while X-types are used in line-to-line filtering applications and may fail short. Also bothersome when used to isolate digital signals; often the first bit transmitted after power-up using capacitive digital isolators is used to setup the data stream, and must be ignored (only the trailing bits contain useful data).

Magnetoresistance

Magnetocouplers use Giant Magneto Resistance (GMR) to couple from AC to DC. An explanation of GMR isolation is beyond the scope of this paper. Briefly: GMR refers to an isolation scheme that relies on the property of a material to change the value of its electrical resistance, when an external magnetic field is applied to it. It’s important to remember that GMR operates like a transformer; it uses the variable magnetic field of an AC coil. However, it does this to linearly alter the DC resistance of a physically isolated sensing element.


Click here to download Why Do I Need Electrical Isolation?


Back to table of contents

high-voltage-signs - Why Do I Need Electrical Isolation? Acromag Whitepaper

Why Do I Need Electrical Isolation?

We have two principal reasons for introducing isolation into an electric circuit:

  1. To block the transfer of high or hazardous voltages
  2. To break ground loops

Block High or Hazardous Voltage

We use isolation to prevent the transfer of high or hazardous voltages between circuits. We typically block these voltages using isolation for safety reasons and protection from electric shock; but also to block high common mode voltage present in our signals, which can prevent its measurement and damage equipment. Isolation can also block transient voltages for the same reasons. High voltage may drive injury via electric shock and the unintended flow of electric current through the body. Additionally, it may also drive damage to an electrical circuit because of unintended electric current flowing between conductive circuits.

Protection from Electric Shock

One reason we isolate a circuit is to help prevent electrical shock. That is, by introducing isolation between conductive bodies, we minimize or eliminate the potential for unintended current flow. With no shared common reference or conductive path between two conductors or circuits you cannot complete a circuit for current to flow. This is because of potential differences between them sufficient to produce electric shock; the sudden and rapid flow of electricity between potentials when crossed with a conductor. Shock currents in the body can be felt at about 0.5mA; they can drive an erratic heartbeat and potentially be fatal above 10mA; and they can stop a human heart at 2A. Isolation blocks voltage potentials that could drive dangerous current levels through a body if contacted/crossed.


Learn more: How to Select the Right Isolator

Reject High Common-Mode Voltages

Isolation blocks the dangerous transmission of high voltages between circuits which can drive electric shock to personnel or equipment. Another key use of isolation is to enable the measurement of a signal with a high common-mode voltage that prevents valid measurement and could damage equipment. The reality is that most instruments will have a common-mode input range inside of ±10V; unless specifically designed to reject high common-mode voltage. Thus, signals with an offset from measurement ground greater than 10V cannot be converted properly and could damage the instrument. Isolation rejects the unwanted high common-mode voltage present in some signals, allowing the real signal of interest to be discerned.

Remember that electromagnetic noise is ever-present in most environments because of nearby machinery and electric motors, relays, fluorescent lighting, etc. As a result, common-mode noise can be capacitive-coupled, inductively coupled, or radiated into the measurement system. And it will typically take the form of a DC offset, combined with a continuously variable 50-60Hz component (and even higher frequency harmonics of 50-60Hz) that can mix with and obscure your measurement. Isolation blocks the transmission of this error through our system (see Ground Loops below). But some applications will naturally contain a greater offset voltage than this.

Example

If an input is restricted to voltage potentials in the ±10V range; how would you measure one cell of a large array of solar cells connected in series? Or measure the individual cell of a large hybrid battery? Since these signals are offset from circuit common by larger amounts (they have high common-mode voltage potentials), certainly greater than ±10V; this makes their measurement difficult and potentially dangerous to your equipment. Note: the common-mode portion of an input signal is normally computed as the sum of the voltage potential of the positive lead, with respect to measurement return or common; and the voltage potential at the negative lead, with respect to measurement return or common, divided by two (Vcm = [Vin+ + Vin-]/2). Signal isolation blocks the high common-mode portion of input signals like these; which otherwise make our measurement difficult and can damage our equipment.

Protect Circuits from Transient Noise

Isolation also breaks the conduction path for high transient voltage noise between circuits; i.e. voltage spikes generated from switching inductive loads, lightning, or operation in the presence of nearby electric motors and machinery. Isolation is the first best defense in preventing transient noise from propagating through a measurement system.

Back to table of contents

Break Potential Ground Loops

In electric circuits or conductive paths, a potential difference or voltage is the force that drives current flow through it. This is great for transmitting signals over long distances. However, as we link conductive circuits together, there will be differences in reference potentials between circuits that can drive unwanted current flow between the circuits that adds noise and signal interference increasing error.

Electrical power to most instruments includes a circuit common that may be referenced to earth ground. Although we idealize common and earth ground as a reference point of 0V or no potential; not all circuit commons are 0V and at equivalent potentials. These circuits linking together to transfer signals and connect earth grounds/commons create potentially unwanted current flow between them. This is because of their different potentials. In fact, for devices connecting to the same building earth ground, it’s not uncommon to encounter differences in earth ground potential from 100-300mV; even over relatively small transmission distances. Given this potential difference between separate earth ground points and conductive paths between them, current will flow in the resulting circuit. As an illustration, Figure 1 (below) shows an earth grounded input source connected to an earth ground referenced input; as well as the resultant overlapping ground loop circuit in the return signal path.

I/O Connection with Ground Loop via Two Separate Earth Grounds
I/O Connection with Ground Loop via Two Separate Earth Grounds - Why Do I Need Electrical Isolation? Acromag Whitepaper
Figure 1: I/O Connection with Ground Loop via Two Separate Earth Grounds

Usually, current flowing along the intended design path (circuit conductors) is predictable, and controlled by the resistance/impedance of the circuit. However, if a ground loop is formed due to varying earth ground potentials between circuits, unwanted current will mix in the signal path and return along an unintended path between the earth grounds. This will drive measurement error, especially if our wanted signal is comparatively small. Chances are very high and probable that more than one earth ground point could be made In circuits covering some distance. This is because all conductors have some resistance; current flowing through this resistance will always produce a voltage difference along that conductor. The important thing to remember about earth ground loops is that they result in unwanted noise and interference. This drives measurement error, and severe earth ground loops can create risk of electric shock.


Note: Ground loops can be created accidentally, and discerning extra connections to earth ground may not be obvious.
Example:

Devices connecting to the USB ports of personal computers connect to earth ground through the computer. Computers have the earth ground of their AC power plugs connected in common to their chassis and to the USB signal and shield ground. Likewise, double-shielded Ethernet cable may also connect to earth ground at the PC network interface. Thus, simply making a USB or Ethernet connection between your PC and your instrument could create an unexpected ground loop circuit. Or, if you use a grounded oscilloscope to probe your circuit (most scopes also connect to earth ground at their AC power connection); earth ground could inadvertently be connected at more than one point.

For this reason, some Acromag instruments, and many USB-connected devices, are at risk when connecting to the USB port of a PC without a USB signal isolator. The connected device may either be driven by an earth grounded input signal already, or have earth ground applied at another point. With two connections to earth ground in the circuit, and perhaps some distance between earth ground connections, there’s potential for ground loop current to flow; interfering with/preventing circuit operation. Therefore, we manufacture and encourage the use of USB isolators. You might think using battery-powered laptops would quell the need for USB isolators when connecting to USB devices. That’s usually true, unless that laptop is also connected to an AC adapter or an AC-powered printer, which typically attach to earth ground at the AC outlet.


Learn more: Circuit Grounding & Why it’s Important

Back to table of contents

Simplified Ground Loop Error via Shared Path to Ground

It’s important to recognize that ground loop interference can result from more than one earth ground connection, (see Figure 1). However, it can also occur from sharing a return path to earth ground or common, (see Figure 2).

Simplified Ground Loop Error via Shared Path to Ground - Why Do I Need Electrical Isolation? Acromag Whitepaper
Figure 2: Simplified Ground Loop Error via Shared Path to Ground

Both figures above demonstrate two loops of current combining along a shared circuit path to earth ground or common; or between earth grounds where individual circuit signals overlap and interfere. Figure 1 depicts the more common form of ground loop; created between two earth ground referenced circuits, linked together with their individual earth grounds at different potentials; driving current to flow in the interconnection between the devices, which produces signal offset error. In contrast, Figure 2 depicts another type of ground loop error with only one ground or common connection; but a shared path to ground or common used by both circuits.

Of Particular Concern with Ground Loop Error

In contrast, as depicted in Figures 1 and 2, the offset normally encountered in your measurement will not be a steady DC level. Rather, it usually carries a continuously variable AC component from 50-60Hz. This is because of the grounded AC-powered devices common to most system earth grounds. (A variable load of electric motors and machinery, florescent lighting, switching power supplies, personal computers, etc.) Thus, the resultant error voltage takes the form of a DC offset, combined with 50-60Hz of ripple; coupled into your measurement signal common-mode via their common connections to earth ground. The continuously variable nature of this offset makes it difficult to filter or calibrate it out of your measurement.

There are only two remedies for combating the negative effects of offset error contributed by earth ground loops:

Either take complicated measures to ensure all connections are made to the same potential; perhaps via a star grounding scheme applied to the multiple earth ground or common connections of your system (very difficult compared to the alternative). Or, you can simply add isolation to the circuit (recommended).

Isolating a signal path will segment the circuit formed by two earth ground connections at different potentials; or by separating a shared path to earth ground or common. Isolation allows both circuits to operate at different earth ground potentials, and without generating an intervening earth ground loop current. (The current has no conduction path across the isolation barrier.) Consequently, this is the simplest remedy for properly earth grounding a measurement system. Additionally, it avoids the difficulty of single-point grounding schemes like “star-grounding,” which are only effective in close proximity or over very short transmission distances.

Longer Circuit Extension = Greater Potential Difference

So, we say that adding isolation breaks the ground loop circuit that carries unwanted current flow along a circuit path. It’s most important not assume that earth ground in a circuit refers to zero potential and no difference in earth ground potentials means no current will flow. Contrarily, this is only true in an ideal world. In reality; for any circuit earth ground, especially earth grounded circuits that cover long distances, there’s a potential difference across that circuit. In short, the longer the circuit is extended, the greater the potential difference becomes.

It’s important to realize and be cognizant of this and recognize that all ground circuits have some resistance. In addition, any current flowing in this path reinforces and adds to the potential difference across the intervening path. This earth ground loop current and resultant voltage drop interfere with normal signal current and voltage; and will offset signal measurements (particularly troublesome for small signals). In fact, ground loops create electric shock hazards if earth “grounded” parts of equipment that encounter human contact are not at ground potential (0 volts). To summarize, we apply isolation to safely constrain our circuit (and its circulating currents) from other circuits, and visa-versa. Isolation blocks interference between circuits, and builds a barrier to protect one circuit from the high voltages/currents produced in the other circuit. Spikes, surges, and noise are stopped from propagating along a circuit’s conductive path at the isolation barrier.


A Common Misconception

Often, when we schematically represent an isolated I/O circuit, (like a transmitter or signal isolator), customers are confused by the connection to earth ground on each side of an isolation barrier. The appearance of the earth ground symbol in both circuits causes them to think that the isolation barrier is somehow being violated because of a common connection to earth ground on each side of the barrier. After all, earth ground is a conductor.

However, even though a path between earth ground points is present, no circuit can be completed between the circuits; because the isolation barrier broke the return conduction path. (i.e., there is no bidirectional circuit path.) In fact, the presence of the isolation barrier allows each side to connect to the same or different earth ground potentials (within the limits of the isolation rating) without causing problems. Thus, it helps to think of a conduction path in terms of a circuit loop, not a one-way street; the isolation barrier breaks this loop and consequently prevents the flow of current between the earth grounds within the circuit itself.


Back to table of contents

Safely Testing Isolation

DC Voltage Insulation Test

One common method of safe isolation testing applies a high DC voltage across an isolation barrier, then measures the amount of DC leakage current across that barrier. Generally, a common-mode voltage of approximately 100V DC is connected across each respective isolation barrier in a circuit; and the maximum current flow between any two isolated entities must be less than or equal to one microampere to prove isolation. It’s for this reason that Acromag normally does this test for all isolated products that we ship.


Learn more: Acromag Quality Control & Quality Management Certifications

High-Potential Testing

High-Potential (HIPOT) testing measures an isolation barrier’s strength (also known as Dielectric Withstand Test). In contrast to continuity tests, no continuity with a high-voltage potential applied across an isolation barrier is checked. HIPOT safety tests electrical insulation in appliances, cables and wired assemblies, printed circuit boards, electric motors, and across transformers. The amount of leakage current across the isolation barrier under HIPOT conditions benchmarks its isolation rating. Further, the HIPOT test voltage applied is typically twice the safety voltage rating, plus 1000V.

Generally, isolated instruments have an operating safety voltage rating of 250VAC; it’s doubled in the calculation for safety margin, then 1000VAC is added. Thus reflecting real-world line transients that may occur. The HIPOT test usually connects all electrical connections on one side of an isolation barrier to earth ground or neutral; as well as all electrical connections on the opposite side of the barrier, to the AC hot-wire. Then 1500VAC is applied across the barrier; current flow across that barrier should be negligible to prove isolation. Per IEC 60950, the duration of this application is one minute. In some production environments, it’s not always practical to test every barrier for one minute; so the HIPOT voltage can be increased by 10-20%, making application time 1-2 seconds. The amount of allowable leakage is related to the applicable standard. Acromag’s test limit range, for instance, is 1-5uA.

DC Voltage HIPOT

It’s also possible to substitute DC voltage HIPOT for AC voltage; then a DC level equal to the peak AC level will be used (i.e. 1500VAC*SQRT[2]=2121VDC). Limit currents can be set lower for DC voltage HIPOT than for AC HIPOT.

Acromag performs HIPOT testing for all transformers before board assembly, and during product qualification. It’s otherwise not repeated for all product; except occasionally for small samples to check conformance over time, if mandated by a customer, or if required by a safety agency.

 

Common-Mode Noise Rejection

Common-mode noise rejection refers to the ability of a differential input amplifier to amplify the “normal” mode signal and reject the “common” mode signal at its input. But the common-mode noise rejection of an isolated instrument is also an indirect measure of the ability of its isolation to block the feed-through of common-mode noise across its isolation barrier. Noise in industrial applications is commonly sourced from:

  • 50-60Hz AC powered devices
  • Differences in earth ground potentials between these devices
  • and even varying levels of common-mode voltage between signals; isolated instruments block this “noise” propagation

Thus, isolation enhances the instrument’s common-mode rejection specification and its ability to reject noise signals common to both of its input leads, inclusive of the 60Hz “line noise” sourced from AC-powered devices along the signal chain. Comparatively, the higher the common-mode rejection of your instrument, the better.

Common-Mode Rejection Ratio

The common-mode rejection ratio is computed as the ratio of signal gain to the gain applied to its common-mode signal as follows; CMRR = Vsignal_gain/Vcm_gain. Very high multiples are particularly desirable to accurately discern signal from noise. Often when measuring small signals in noisy environments (i.e. the resistance of an RTD or the voltage of a thermocouple), this noise will offset the real signal on both measurement leads. But other times a signal will have inherently large common mode offsets; such as that encountered in measuring the cell voltage of a large solar cell or hybrid battery. The CMRR of the instrument refers to the attenuation it applies to this common signal, present on both leads. Since CMRR is expressed in decibels and computed as 20*LOG [AVsignal/AVcm]; ideally it should be 100dB or better. Expressed in dB, this is computed as 20*LOG [CMRR]. Thus, 100dB of CMRR implies CMRR=LOG-1[100/20] = 100,000–a 100000:1 signal to noise ratio.

Testing AC Common-Mode Rejection

One way to test the AC common-mode rejection of an isolated I/O circuit; first apply a 60Hz voltage signal through the input to the instrument (powered), then across its isolation barrier (from input to isolated output). Then apply a common-mode voltage by placing the AC-neutral wire on the isolated output common lead, and the AC-hot wire separately on the input+ and input– leads. Lastly, these are linked by a 100Ω resistor; simulating source impedance as well as helping to make the 60Hz noise common to both leads.


!WARNING: DO NOT TRY THIS AT HOME! THIS TEST IS POTENTIALLY DANGEROUS AND IS FOR FOR INSTRUMENT MANUFACTURERS ONLY.
NEVER DIRECTLY CONNECT LINE VOLTAGE ACROSS YOUR CIRCUIT; ISOLATED OR NOT.

The instrument’s added I/O isolation expands its common-mode voltage range for this test. The amount of 60Hz signal leaking across the isolation barrier (when high common-mode is applied) is measured by examining the output with an oscilloscope; determining if any 60Hz ripple is present (measured p-p). That’s compared to the 60Hz ripple which may be present before applying the common-mode signal across barriers. Increases in 60Hz pp noise are direct results of small leakage current across isolation barriers. Often, since the common-mode rejection is very good, no difference can be discerned. In this case, if the instrument digitizes the signal; a noise signal equivalent to the least significant bit weight is assumed for calculation. (The smallest digital signal that can be discerned by the circuit.)

For true differential inputs and isolated I/O, the common-mode rejection is very high. So high, it’s often difficult to discern any 60Hz feed-through ripple on the output. Occasionally, the input unbalance resistor is increased from 100Ω to 1KΩ; thus raising the input’s sensitivity for noise pick-up relative to this small leakage current. Placing an unbalance resistor across the input converts some common-mode leakage to normal-mode noise (via the small IR drop). Additionally, the normal-mode noise signal is well-attenuated via normal-mode filtering by the circuit. However, a portion is passed along with the desired signal through the isolation barrier, as well as the common-mode noise; a direct result of “leakage” across the isolation barrier.

It’s hard to imagine an isolated signal chain, with no metal conduction path, allowing a small current to pass across the isolation barrier under these conditions.

No isolation barrier is perfect because a small amount of leakage takes place that’s usually related to the capacitance of the isolation barrier. In many modern applications, a small safety-rated isolation capacitor is also placed across an isolation barrier between isolated ground planes. This helps suppress radiated emissions of the circuit; Y types are used, which fail open if their voltage rating is exceeded. Additionally, this helps steer transient energy to earth ground on the secondary side when the primary side is missing a path to earth ground (fault condition). This added capacitance further reinforces the small capacitance already present in the isolation barrier. It’s kept small (less than 1000pF), thus keeping resulting 60Hz leakage small. If too much capacitance is added, then the device fails HIPOT testing; its leakage current limit is exceeded under a HIPOT difference applied across the isolation barrier.

Back to table of contents

Conclusion

At this point you should have a better understanding of the important role that isolation plays in signal transmission. To summarize, isolators, isolated transmitters, isolated signal conditioners, and other types of isolated equipment:

  • Protect both personnel and equipment from electric shock
  • Enable measurement with high common-mode voltage present
  • Block the transmission of both noise and interference
  • Prevent the formation of earth ground loops when linking circuits together

By and large, the isolation between circuits allows for operating each circuit at different ground potentials without creating a problem; as long as you keep the potential difference inside the limits of the isolation rating.

Contact Us

For more information about isolators or any Acromag product, click on the link below. Our sales team will be happy to clarify any questions you may have!


Click here to download your whitepaper: Why Do I Need Electrical Isolation?


More Signal Conditioning Technical Papers:


To view the industry’s best selection of process signal isolators, click here.