What is the SEMI F47 line sag spec all about?

We have all seen lights dim at home or at work and this is an indication that the AC line voltage has been reduced or sagged. Although an occasional dimming of lights can be tolerated, when it comes to factory automation equipment, line sag can be the source of a production shutdown, resulting in significant revenue losses. Since the production of semiconductors, including microprocessors, is a very precise and expensive process, back in 1999 the Semiconductor Equipment and Materials Institute (SEMI), established standards relative to AC line sag immunity. This specification is called the SEMI F47 Voltage Sag Immunity Standard and has been revised periodically. Because many other factory automation processes are equally critical, some of these production products need to comply with the SEMI F47 standard as well.


Basically, this standard requires that the AC-DC power supply that is used in semiconductor production, or in other factory automation equipment, continue to provide the required output voltage and current, even if the input voltage dips below its specified limits. As can be seen in the chart below, in the blue area, the basic specs require that the power supply perform normally even if the input voltage sags down to 50% of its nominal voltage for up to 200 ms, or sags to 70% for up to 500 ms, and sags to 80% for up to one second. Since this sag percentage refers to the nominal line voltage, this means for example that with a nominal 220VAC input, the AC voltage can sag down to 50% or 110VAC for up to 200 ms, down to 70% or 154VAC for 500 ms, and down to 80% or 176VAC for up to one second.


There are additional sag ride-through “recommendations” within the latest version of the standard, which is the SEMI F47-0706 (these recommendations are not requirements) that includes operation of the power supply with 0% input power (no power) for up to 20 ms. This recommendation can be accommodated by insuring the selected power supply has a “hold-up time” specification of 20 ms or longer. Other newly recommended thresholds within the SEMI F47-0706 include sags of 80% for 10 seconds, and continuous sags of 90%. Most power supplies that meet the previous versions of this standard, the SEMI F47-0200, and have a hold-up spec of 20 ms or greater, should be able to meet the new recommendations (not requirements) as well.

The simplest and lowest cost method of complying with the SEMI F47 standard is to use a power supply with a universal input, such as 90 to 264VAC, and operate it from a 220VAC or higher line input. In this way you automatically meet and exceed the standard since this type of power supply can operate down to 90VAC (even lower than the 50% line sag spec of 110VAC). Note that this method does apply to auto-strap power supplies.

Another way of meeting the SEMI F47 is to draw less power than the supply can normally provide (de-rate the supply). If you do this, always check with the manufacturer to confirm that your reduced load will allow the supply to fully meet the SEMI F47 standard. This may require extra testing to confirm compliance, either by the power supply manufacturer or the end-product OEM. Alternatively, the power supply manufacturer may be able to modify the supply to meet the SEMI F47 standard.

Some factory automation equipment require the use of SEMI F47 “certified” power supplies, which means the supplies were tested by an outside agency or laboratory and found to fully comply with the standard (similar to UL certification). If this is a requirement, always look for supplies that have existing certifications from a reliable manufacture, because the cost of getting this type of certification can amount to $2,000 or more. There is a grandfather clause in the updated standards that provides for equipment that was tested or certified under the previous versions of the standard to not require re-testing or re-certification.

Many industrial-type power supplies are designed and/or certified to meet the SEMI F47. These supplies may be a bit more expensive, but it will be the lowest cost solution, especially if you compare it to the cost of adding an external constant voltage transformer or UPS to the input of the power supply.

Power supply manufacturers such as TDK-Lambda offer supplies that are SEMI F47 certified and supplies that operate with a wide universal input of 90 to 264VAC. In addition, modified supplies can be provided that meet this and other prevailing power supply specifications.

Ripple & Noise Specs and Measurements

AC-DC power supply and DC-DC converter datasheets should always include output “Ripple & Noise” specifications. The Ripple & Noise spec is sometimes referred to as Periodic And Random Disturbances or PARD. The following drawing shows how ripple and noise may look when viewed on an oscilloscope that is attached to the output of a typical switchmode power supply.


The output “Ripple” frequency is primarily determined by the switching frequency of the power supply. The higher frequency “Noise” spikes are generated by the fast rise and fall times of the pulses associated with the switching and rectification components of the power supply. Typical ripple and noise specs are defined as peak-to-peak measurements in mV units.

Ripple & Noise Measurements
Unfortunately, there is no universally accepted method for measuring ripple and noise. It seems that each manufacture, and sometimes different products from the same manufacturer, may have varying methods for these measurements. In some cases the bandwidth of the test oscilloscope is defined as 20MHz or 100MHz. In addition, added components such as capacitors, resistors, twisted wires, and/or coax are sometimes required in the test set-ups that are defined by the manufacturer. In order to meet the power product’s specified ripple and noise specs, care must be taken to follow the manufacture’s defined test set-up. There are a few standardized methods for ripple and noise measurements; one of which is the JEITA-RC9131A standard.

Fig 1: JEITA-RC9131A Ripple & Noise Test Set-Up

The above drawing (Fig 1) shows the test set-up per JEITA-RC9131A. This standard defines a custom oscilloscope connection comprised of a length of 50 ohm coax that is connected to the output of the power supply with the other end terminated at the scope with a 50 ohm resistor in series with a 4700pF capacitor. Notice that the coax is attached to the output of the power supply within 150mm or 6 inches of the output terminals and has two added capacitors (22uF electrolytic and 0.47uF film type) soldered across those points. The 50 ohm coax should not exceed 1.5M or 5 feet in length. All coax pigtails and added component’s lead lengths should be kept to a minimum to prevent pick-up of radiated noise.

Other Measurement Precautions
Some ripple and noise measurements can be made with the use of a standard oscilloscope scope probe that has been modified by removing the plastic tip cover and ground clip wire and replacing the ground connection with a short length of bare copper wire that is wound around the probe’s ground ring. In this way the probe’s tip and ground connections are kept to a minimum length, thereby reducing the chance of the ground lead acting as an antenna and picking up radiated noise signals, which can result in out-of-spec measurements.

Figures (a), (b), and (c) below show incorrect set-ups for ripple and noise measurements.

When making ripple and noise measurements a standard load should be used. This precaution is to prevent any noise from the power supply’s normal system load, which may contain noisy digital or RF circuits, from feeding noise back to the output of the supply, which again can result in out-of-spec test measurements. In some cases, to reduce ground loops, it may be necessary to isolate or float the oscilloscope from the AC source by plugging it into an isolation transformer.

Unless otherwise stated, the ripple and noise specifications are usually based on measurements taken while operating the power supply with its nominal input voltage, at the rated output voltage and current load, and at or near room temperature (typically 72°F to 77°F).

Over Current Protection in Power Supplies & Converters

Most AC-DC power supplies and DC-DC converters have internal current-limiting circuits to protect the power device, and to some degree its load. The majority of over-current-protections include an automatic recovery feature. In practice, the current limit feature typically starts operating when the output current exceeds it maximum rating by 10 to 20%.

In many cases, should an overload (e.g., short circuit) be allowed to exist for a prolonged period, it can reduce the product’s field life by temperature stressing the electrolytic capacitors, and in extreme cases, it can damage the user’s printed circuit traces. Therefore, always check the power supply’s “Instruction Manual” to be sure you understand the precautions associated with the power product’s over-current-protection feature. Also, if the power product has an Output Good signal, this can be used as an indication that the power supply is either faulty or could be in an over-current mode.

There are a number of ways to implement over-current-protection (OCP), and below are descriptions of the most common methods.

Fold-Back Current Limiting: When this method is employed if an overload condition exists, the output voltage and current reduce to safe levels. As can be seen from the following curve, should an overload occur the supply will provide current up its current limit point (aka ‘knee’), and then the output current will fold-back to a lower value as the output voltage reduces towards zero.

This technique is employed in linear power supplies because it reduces the strain on the supply’s internal power devices to minimum. One drawback of fold-back current limiting is that if the supply turns on into a heavy capacitive load, it could latch-up at a reduced current before reaching its full output voltage. Depending upon the design, recovery from a fold-back current limit condition can be automatic, or after a built-in time delay when the overload condition is removed.

Fold-Back Current Limiting

Fold-Forward Current Limiting: In this method, when an overload is sensed the output voltage reduces towards zero, but the current increases. When driving motors, pumps, or highly capacitive loads, employing a fold-forward current feature can help overcome the electrical inertia of these loads. Recovery from a fold-forward current limiting situation is usually automatic when the overload is removed.

Fold-Forward Current Limit

Constant Current Limiting: In this method, should an overload occur, the output current stays at its limit point and the output voltage reduces towards zero in a somewhat linear fashion. This technique is used in many switchmode power supply designs. Typically, the supply will automatically return to its normal output voltage when the overload condition is no longer present.

Constant Current Limiting

Current Limit Shutdown: In some power supply designs, when an overload occurs the power supply will begin to go into a constant-current limit mode, but when the output reaches a preset reduced voltage, the supply will shutdown. Recovery from this condition can be automatic or require recycling of the input power.

Hiccup Mode Current Limiting: Some low power supplies have what is termed a hiccupcurrent-limit feature. As the name implies, if a current limit is sensed, the supply will reduce its output voltage to zero and then, after a short time, it will attempt to provide its normal voltage. These On-Off attempts at operation are referred to as a hiccup-mode. Should the overload condition be removed, the supply will again operate normally.

Peak-Current Power Supplies
It should be mentioned that some power supplies are designed specifically to provide large peak-currents, which can range from 200 to 300% of the maximum current rating for a short duration, without going into a current-limit condition. These are especially useful when powering loads that include electric motors such as computer hard drives, fans, actuators, pumps, etc. When using this type of power supply it is important to limit the “average power” that is delivered to load. More information about peak-current-rated supplies will be provided in a separate article.

Power Supplies with Wide Range Adjustable Outputs

For some power supply applications it is desirable to change the output voltage over a wide range. There are a number of ways to control the output voltage of power supplies that are designed to provide wide adjustment ranges. Remotely adjustable output voltages can be implemented by using one of the following methods.


Variable Voltage Control
In this case an external variable control voltage (e.g., 1-6V) is connected to the designated input of the power supply, sometimes called the PV input. As the input control voltage is varied it will cause the output voltage to change in a fairly linear fashion over a wide range (e.g., 20% to 120% of the nominal output voltage). For some applications this is a low cost method of providing a programmable power supply. Below are diagrams showing an example of this type of remote voltage adjustment for Lambda’s HWS/PV and SWS-L series of power supplies.

External Variable Voltage Control (1-6V)

Output Voltage Change (20-120%) with Ext. Variable Voltage Control (1-6V)

Variable Resistive Control
Some power supplies can be remotely adjusted via a variable resistive control (external potentiometer). This method has the advantage that an external voltage is not required since an internal Ref. voltage is provided by the supply. As the resistance changes, it will cause the output voltage to change in a non-linear fashion over a wide range (e.g., 20% to 120% of the nominal output voltage) as shown in the diagrams below (Lambda’s HWS/PV series). For some applications this is a low cost method of providing a programmable power supply.

External Variable Resistive Control (50k ohm pot.)

Output Voltage Change (20-120%) with Ext. Variable Resistive Control (50k ohm pot.)

Serial Digital Control
Programmable Power Supplies can be remotely controlled via a serial digital port such as RS232 or RS485. Both the output voltage and current can be controlled from zero to the maximum output ratings. In addition, alarm signals from the supplies can be sent back to the remote computer or controller via the same digital link. Programmable Power Supplies are more expensive than wide adjustable supplies mentioned above, but they have a large array of local and remote control features that are not found elsewhere. Lambda’s ZUP series is a good example of a feature-rich Programmable Power Supply.

Up to 31 ZUP Series Programmable Supplies can be Remotely Controlled via RS485 Interfaces

What is Remote Sensing?

Most medium to high power AC-DC power supplies and DC-DC converters have “Remote Sense” connection points (+/- Sense) that are used to regulate the supply’s output voltage at the load. Since the cables that connect a power supply’s output to its load have some resistance, as current flows it will cause a voltage drop in the cables. Since it is best to regulate the voltage at the load site, the use of the two Remote Sense wires connected from the supply to the load will compensate for these voltage drops.

Typical remote sensing circuits are capable of correcting from 0.3V to 1.0V of voltage-drop in the output cables. However, to be sure, always check your power supply’s instruction manual to determine the maximum remote sense compensating range. If the voltage drop across the cables exceed the range of the remote sense circuits, this can be remedied by either reducing the length of the cables or increasing the size of the cable’s conductors. The remote sense leads carry very little current, so light gauge wires can be used. Steps should be taken to ensure the remote sense wires do not pick up noise by either twisting the +/- Sense wires together and/or shielding the wires from noise. It is important to observe the correct polarities, i.e., the +Sense wire should connect at the load to the +V output cable and the –Sense wire should connect at the load to the –V output cable. Refer to Figure 1.

Fig. 1: Power Supply with Remote Sense Wires Connected at the Load

When not using the remote sense feature, Local Sense (LS) connections should be used. In this case the +/-Sense points should be connected to their corresponding output or local sense terminals at the power supply (+Sense to +V output or +LS and, –Sense to –V output or -LS). Most power supplies are shipped from the factory with these “Local Sense” connections in place.
Refer to Figure 2.

Fig. 2: Power Supply with Local Sense Jumpers Installed

Choosing an Input EMI / EMC Filter for a Power Supply

There are two primary functions that an input EMI filter can perform:

1. Minimize outgoing electrical noise to avoid interfering with neighboring equipment
2. Attenuate (reduce) incoming electrical noise that could damage the system

Regarding outgoing noise, although most power supplies meet the governmental regulations for EMI, noise is additive and if there are multiple power supplies or high speed processor boards, it can result in a failing grade.

If the noise is only slightly out of specification, then a (lower cost) single stage filter may suffice. If the noise is considerably out of specification then a higher performance two stage filter will be required.

An example of these would be Lambda’s RSEN (single stage) and RSHN (two stage) filters. Look for the terms “wideband” or “low frequency attenuation” in the features.

Incoming electrical noise is usually in the form of a spike or burst of energy. It can be generated from natural causes such as a lightning storm or man made by a large piece of industrial equipment.

This type of filter may have “high pulse attenuation” listed as a feature and will have internal values optimized to reduce these potentially harmful spikes from reaching the power supply. The filter will also have some outgoing noise attenuation, but may not be as effective. An example would be Lambda’s RSMN series of filters.

Droop Mode Current Share

If two power supplies are to be connected together to produce more power or share the load, then a parallel capable model should be selected. TDK-Lambda’s DPP100, 120, 240 and 480 models are all parallel capable. On the front of each power supply is a small black switch. For parallel operation this switch should be set to “parallel” (Fig. 1).


In single mode the load regulation (the amount the output voltages changes with load) is minimal, the difference being less than 0.24V from zero load to full load for a 24V output power supply.

In parallel mode that load regulation is artificially increased to 1.2V using internal circuitry (Fig. 2).

The extra voltage drop or “droop” is proportional to the load drawn, so that when two or more power supplies are connected in parallel the output load is shared between the power supplies. If one of the paralleled power supplies tries to provide more current, its output will droop slightly and the other supplies will balance.

For optimal performance, all power supplies should have their outputs set to the same voltage.

What is PFC and why do I need it?

Switchmode power supplies without Power Factor Correction (PFC) tend to draw the AC input current in short bursts or spikes relative to the line voltage, as shown in Fig. 1. The Power Factor of a power supply is technically the ratio of the real power consumed to the apparent power (Voltsrms x Ampsrms) and is a decimal between 0 and 1.0. If left uncorrected the Power Factor (PF) of switchmode supplies will generally be around 0.65 or less.

Figure 1. Input of switchmode power supplies without PFC. The voltage waveform is a sinewave and the current waveform is a pulse or spike. PF<1 span="">

The Power Factor can be improved by using PFC circuits. These circuits “smooth out” the pulsating AC current, improving the PF, and reducing the chances of a circuit breaker tripping prematurely. There are two basic types of PFC, passive and active. Passive PFC circuits are less expensive and typically can correct the PF to about 0.85. Active PFC circuits are the most popular, are built into the switchmode power supply and can increase the PF to 0.98 or higher. The closer the PF comes to being 1.0, the better the performance of the power supply. Ideally, we want to end up with the input voltage and current waveforms being sinusoidal and in phase with each other as shown in Fig. 2.

Figure 2. Voltage and Current waveforms are sinusoidal and in-phase. PF=1.

PFC is Required by International Regulations
An important reason to have PFC in your power supply is to comply with international regulations, especially if you intend to sell your equipment in Europe. Since 2001, the European Union (EU) established limits on harmonic currents that can appear on the mains (AC line) of switchmode power supplies. Today, the most important regulation is the “European Norm” EN61000-3-2. This regulation applies to power supplies with input power of 75 watts or greater, and that pull up to 16 amps off the mains. It sets severe limits on the harmonic currents up to the 39th, when measured at the input of switchmode power supplies.

For example, the first harmonic is the primary input frequency, typically 50 Hz for the EU countries. The third harmonic is 150 Hz, and the 39th harmonic is 1,950 Hz. These unwanted harmonic currents have a direct relationship to the Power Factor of switchmode power supplies. Therefore, power supplies that meet EN61000-3-2 inherently have high power factors that are typically 0.97 or higher.

PFC Increases the Supply’s Output Power Capability
The PF, much like the supply’s efficiency rating, determines the amount of useful power a switchmode power supply can draw from the AC line and then deliver to its output load. Specifically, the formula that determines this is:

VLrms x ILrms x PF x Eff = Pout

As an example, if a power supply is operating off of 120VAC line, which is protected by 15A circuit breaker, UL guidelines say you should not draw more than 12A. So, using the formula above, we can compare two power supply examples with different Power Factors, as follows:

Example A: No PFC, PF = 0.65, 85% Efficiency, 120VAC input, 12A max. current:
Therefore: 120VAC x 12A x 0.65 x 0.85 = 796 Watts Output Power

Example B: PFC used, PF=0.98, 85% Efficiency, 120VAC input, 12A max. current:
Therefore: 120VAC x 12A x 0.98 x 0.85 = 1200 Watts Output Power

As can be seen above, the power supply in Example B (with PFC) can deliver 404 Watts or 51% more power to its output load than the non-PFC supply, a significant increase.

Why do I need PFC?
A power supply with PFC can supply higher output load currents than those without PFC. PFC significantly reduces the AC current harmonics, leaving mainly the “fundamental” current frequency that is in-phase with the voltage waveform (Fig. 2). International regulations dictate the substantial reduction of harmonic currents. The vast majority of AC-DC power supplies manufactured by Lambda Americas has active PFC, is in accordance with EN61000-3-2 and provides typical power factors in the range of 0.97 to 0.99.

What does 1U, 2U or 3U mean?

Many rack-mounted power systems are specified as being 1U, 2U, 3U, etc. What does this mean? For electronic equipment racks (e.g., 19 or 23 inches wide), the term 1U is used to define one rack unit of height.

1U equals 1.75-inches (44.45mm) of rack height. Therefore, a 2U rack mount height would be 2 x 1.75", which equals 3.5-inches high. A 3U height would be 3 x 1.75" = 5.25-inches.

It should be noted that the 1U, 2U, 3U, etc., heights are maximum dimensions. In order to allow for mechanical tolerances and to provide some space between panels, typically, for each 1U of height manufacturers may deduct about 0.03" (see Photo #1). For example, a 2U panel, which has a nominal height of 3.50" may be only 3.44" high [3.50" – (2 x 0.03") = 3.44"].

Individual power supplies are sometimes mounted within rack-mounted enclosures that require integral power. In these cases, the power supply needs to be a bit shorter than the equipment’s overall height to allow for the top and bottom covers. So a 1U high enclosure-mountable power supply needs to be shorter than 1.75-inches; a 2U enclosure-mountable supply needs to be shorter than 3.5-inches, and so forth (see Photo #2).

Examples

Photo #1: This 19" rack-mountable power system can hold up to 3 plug-in, hot-swap and redundant power supplies. The enclosure with mounting ears is 1.72" high (= 1.75" minus 0.03") and is therefore considered 1U high.

Photo #2: This 1000-watt switch-mode power supply is 3.25” high and, therefore, can be mounted in a 2U rack-mountable enclosure, which can vary between 3.44" to 3.50" high.

Since we still live in an English and Metric measurement world, here are a couple of handy conversion factors: 1 inch = 25.4 millimeters (mm), 1 mm = 0.03937 inch

As a side note, Lambda ran a clever ad campaign that those who understand what “1U” or “2U” really means would appreciate. Here is a copy of that ad, which hopefully you will find humorous.