Category Archives: PCB

Effective cooling of high-powered CPUs on dense server boards

By Norman Quesnel
Senior Member of Marketing Staff
Advanced Thermal Solutions, Inc. (ATS)

The main goal of electronics thermal management is to efficiently remove enough heat from a device’s active region so that it stays within its rated temperature. Providing effective cooling presents different design challenges, not all of which involve the chip itself. Some thermal challenges are related to the system in which the chip resides. A common example is cooling a device positioned on a crowded printed circuit board (PCB). The congestion of components restricts airflow and space, which makes the use of many conventional cooling devices difficult.

Dense Server boards

Figure 1. A dense motherboard from Gigabyte Technologies featuring an Intel P55 chipset. [1]

Optimizing PCB for thermal management has been shown to ensure reliability, speed time to market and reduce overall costs. With proper design, all semiconductor devices on a PCB will be maintained at or below their maximum rated temperature. Applying thermal management can sometimes be problematic for dense boards employing fine pitch devices. (Pitch is the space between the center of one BGA ball to the center of the next one.)

But if certain layout guidelines are not followed and considerations are not given to a PCB’s thermal performance, the device and the overall system can suffer from sub-par performance and reliability in the field. [2]

Today’s circuit boards are often assembled with increasing density with the goal of making smaller, lighter systems, or to provide more processing power in demanding applications such as data centers and IoT (Internet of Things) applications. PCB designers must use proven layout techniques to ensure effective thermal performance for the board and its components.

Figure 2. Crowded boards have limited space from where chip cooling air can be drawn. (Wikimedia Commons)

Part of the trend toward higher density boards is related to the industry’s adoption of increased server density. This means increasing the power of the chips, putting more chips per rack unit, and filling up the racks as much as possible. Rack power has transitioned from a normal of about 4 kilowatts to 70 kilowatts per rack.

High current electronic components like microcontrollers can generate a significant amount of heat. To keep the board temperature lower, it is usually best to mount these components near the center of the board. Heat can diffuse throughout the board and the temperature of the board will be lower.

Many components in this situation, such as GPU, will require a dedicated cooling system, such as a fan sink. But simply installing a fan sink on top may not provide the needed level of cooling. It is good practice to quantify system flow bypass on the fan sink, and to also consider the proximity of components neighboring the fan sink. The mass airflow rate is the true measure of available coolant, along with the air velocity.

Obstructions in the intake or exhaust of the fan (e.g. neighboring components) must be carefully considered as their presence will impact the performance of the fan sink. The size and position of adjacent components can impact the fan’s performance. [3]

Figure 3. The QuadFlow CPU cooler draws air from four sides, passing it through cooling fins and expelling warm air. (Advanced Thermal Solutions, Inc.)

One new and effective solution for cooling hot components on congested PCBs is the QuadFlow CPU cooler from Advanced Thermal Solutions, Inc. (ATS). The liquid-free cooler features a high-power blower that draws in air from four different directions. So, while proximate components may block local air in a couple of directions, the QuadFlow fin fields will pull in air from the other directions to make sure that the component is being cooled.

QuadFlow coolers are just 29 mm tall, so they will fit into standard 1-U racks and there are several options for base material (aluminum, copper, or vapor chamber) depending on performance, weight, or cost requirements. [4]

Before applying any thermal management hardware, the smartest engineering activity may be investing is various PCB design services. These include CFD studies on boards at the CAD stage to wind-tunnel testing of actual or dummy boards in conditions that simulate air distribution in real-world applications. Services are available for characterizing boards using research-quality instruments, heat and air velocity sensors, and PCs.

Figure 4. FloTHERM image reveals hotter and cooler regions on a PCB. (Advanced Thermal Solutions, Inc.)

Dummy or working PCBs can be tested in isolation or installed in their own packaging domain. Computational simulations can be made of engineered designs using computational fluid design packages such as 6SigmaET, FloTHERM and CFDesign.

These services are available from ATS, whose engineers can design board layouts to improve cooling airflow in dense systems. Natural airflow can be enhanced to individual hot components and to active cooling systems that rely on airflow for effective performance. Often, these studies head off more expensive cooling solutions by showing that minor changes to component layouts or to the volume of airflow will resolve thermal problems. [5]

References
1. https://www.techpowerup.com/103375/gigabyte-unwraps-latest-p55-series-motherboards
2. https://www.embedded.com/design/configurable-systems/4395845/Ultra-fine-pitch-devices-pose-new-PCB-design-issues
3. https://www.hpe.com/us/en/insights/articles/why-youll-be-using-liquid-cooling-in-five-years-1710.html
4. https://www.qats.com/cms/2013/06/21/how-system-flow-affects-fan-sink-performance/
5. https://www.qats.com/Consulting/PCB-Board-Layout

Advanced Thermal Solutions, Inc. (ATS) is hosting a series of monthly, online webinars covering different aspects of the thermal management of electronics. On Thursday, Jan. 29 from 2-3 p.m. ET the webinar will cover “Methodologies for Fan Characterization and Deployment within a System.” Learn more and register at https://qats.com/Training/Webinars.

For more information about Advanced Thermal Solutions, Inc. (ATS) thermal management consulting and design services, visit https://www.qats.com/Consulting/Custom-Cooling-Solutions or contact ATS at 781.769.2800 or ats-hq@qats.com.

Picking the Right Heat Sink Attachment to Avoid Costly PCB Damage

The design of a printed circuit board (PCB) is a complicated process that requires engineers to consider a number of different issues before the board is ready to move beyond prototype and into production. Engineers must think about the physical constraints of a board on component size and placement, the electrical interaction between components, the signal loss through wires and traces, and the thermal management of each component and the system as a whole. [1]

Heat Sink Attachment

ATS maxiFLOW heat sink with superGRIP attachment on a PCB. (Advanced Thermal Solutions, Inc.)

With all of that to consider, it is no wonder that many designs go through several iterations before moving into the production stage. Since the process is already complex and there is a certain amount of trial-and-error in designing a PCB, engineers will look for ways to avoid unnecessary rework that will add significant cost to the project in terms of both time and money.

As noted in a previous article, the type of heat sink attachment technology that an engineer chooses will impact the ease with which a design can be reworked and the amount of damage to the board that will be caused if a change needs to be made.

Push pins, threaded standoffs and z-clips require holes or anchors be drilled into a board, which leaves permanent damage if a component needs to be moved to a new location and could also impact signal routing. There is even the possibility of a short in installation, which also would damage the board. [2]

Non-mechanical attachments such as thermally conductive tape and epoxy are not guaranteed to provide the optimal thermal management because there is “risk of die damage and poor thermal performance due to uneven heat sink placement,” according to a case study from the Altera Corporation. [3]

The case study also said that thermal tape and epoxy have “high risk of damaging the device or PCB” when compared to mechanical attachment technology coupled with thermal interface material (TIM) or phase change material (PCM). In fact, to remove a heat sink attached with epoxy requires an even temperature of 115-120°C.

As the video below shows, removing thermal tape from a heat sink (even one that is not attached to a board) requires a lot of work and tools. If the heat sink is attached to a component, the process to remove it could damage the board or other devices in the vicinity:

A recent chart from NEMI (National Electronics Manufacturing Initiative) indicated that the cost of assembly can be very high per I/O (input/output) on the PCB – considering some of the new BGAs have hundreds of I/O and there are dozens of BGAs on the board, the cost can be prohibitively expensive to put together a board irrespective of the product sector. [4] Obviously, full reworks necessitated by the use of damaging heat sink attachments raise those costs exponentially.

Heat Sink Attachment

Board assembly roadmap from NEMI showing the conversion costs by product sector. [4]

Advanced Thermal Solutions, Inc. (ATS) has created a mechanical attachment technology that makes rework easy and allows engineers to make changes to the design without damaging the PCB or the components. superGRIP™ is a two-part attachment system with a plastic frame clip that fastens around the edge of the component and a metal spring clip that fits between the fins of the heat sink and quickly and easily attaches to the frame.

As the video below demonstrates, superGRIP™ can be installed and removed with common household tools and will provide a steady, firm pressure to ensure optimal thermal performance of the heat sink and the reliability of the device:

The advantage of superGRIP™ is not limited to its ease of use and the time and money that will be saved in reworking a PCB design. The pressure strength and security of the superGRIP™ attachment system allows the use of high-performance phase change materials that can improve heat transfer by as much as 20 times over standard thermal tapes. [4]

superGRIP™ comes with Chomerics Thermflow T-766, a foil PCM with a thickness of 0.0035 millimeters that has an operating range of -55°C to 125°C. According to Chomerics, the T-766 and other traditional non-silicone thermal interface pads “completely fill interfacial air gaps and voids. They also displace entrapped air between power dissipating electronic components. Phase-change materials are designed to maximize heat sink performance and improve component reliability.” [5]

Chomerics added, “Upon reaching the required melt temperature, the pad will fully change phase and attain minimum bond-line thickness (MBLT) – less than 0.001 inch or 0.0254 mm, and maximum surface wetting. This results in practically no thermal contact resistance due to a very small thermal resistance path.”

The combination of frame and spring clip provides uniform force over the heat sink and ensures no movement to optimize the impact of the PCM, while not damaging the solder holding the BGA component in place on the board. ATS engineers designed the attachment technology so that the in-plane and normal forces of both the frame and the spring clip hold the heat sink without stressing the solder even through NEBS (Network Equipment Building Systems) shock and vibration testing. [6]

Save time, save money, and avoid unnecessary headaches during the design phase by using ATS superGRIP™ technology.

References
[1] http://www.electronicdesign.com/boards/11-myths-about-pcb-layout
[2] “How the maxiGRIP™ attachment system impacts component mechanical behavior,” Qpedia Thermal eMagazine], May 2008.
[3] https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/an/an657.pdf
[4] http://thor.inemi.org/webdownload/newsroom/Articles/CA0599.pdf
[5] https://www.qats.com/cpanel/UploadedPdf/ATS_superGRIP_Launch_Release_
FINAL_with_Photo_0427092.pdf

[6] http://vendor.parker.com/852568C80043FA7A/468ea5de5ac341d385257d39005641c7/
9A63F6EE5B922F278525787600620419/$FILE/Phase_Change_Excerpt-5-08.pdf

[7] “How the maxiGRIP™ attachment system impacts component mechanical behavior,” Qpedia Thermal eMagazine, May 2008.

For more information about Advanced Thermal Solutions, Inc. (ATS) thermal management consulting and design services, visit www.qats.com or contact ATS at 781.769.2800 or ats-hq@qats.com.

Heat Sink Design: ATS Engineers Bring Ideas to Life

Marketing Communications Specialist Josh Perry sat down with Product Engineering Manager Greg Wong to discuss the process that Advanced Thermal Solutions, Inc. (ATS) engineers go through to create a heat sink and find a thermal solution for customers.

Watch the full conversation in the video below and scroll down to read the transcript of the interview.

JP: Greg, thanks again for joining us here in marketing to explain what it is that goes into designing a heat sink for a customer. So, how does that process begin?
GW: We usually start with a few basic parameters; we call them boundary conditions. So, we start with a few boundary conditions, basics like how much airflow we have, how much space constraint we have around a heat sink, and how much power we’re dissipating, as well as the ambient temperature of the air coming into the heat sink.

So, those are the real basic parts that we need to start out with and sometimes the customer has that information and they give it to us, and usually we double-check too, and then other times the customer has parts of the information, like they know what fan they want to use and they know what kind of chassis they’re putting it in and we take that information and we come up with some rough calculations so we can arrive at those things like air flow and stuff like that.

JP: When you get the data from the customer, how do you determine what the problem is, so that way you can move forward?
GW: We usually start out with an analytical analysis. So, we put pen to paper and we start out with basic principles of heat transfer and thermal resistance and stuff like that so we can understand if what we’re trying to achieve is even feasible and we can come up with some basic parameters just using that analytical analysis.

Like we can calculate what kind of heat sink thermal resistance we need or we can calculate how much air flow we need or, if we have several components in a row, we can calculate what the rough air temperature rise is going to be along that chain of parts. So, there’s a lot we can do when we get the basic information from the customer just on pen and paper.

JP: What’s the next step beyond analytical?
GW: Well, we can do some lab testing or a lot of times we also use CFD simulations and, if our customer has a model they can supply us, we can plug that into the CFD simulations and we can come up with an initial heat sink design and we can put that into the simulations as well and then we set those up and run them.

The great thing, having done these analytical analyses beforehand, we know what to expect from CFD simulations. So that way, if the simulations don’t run quite right, we already have an understanding of the problem, we know what to expect, because CFD is not 100 percent reliable.

I mean, you can go and plug all this stuff in there but you really have to understand the problem to know if the CFD is giving you a good result. So, oftentimes that’s the next stage of the process and from there we can actually produce low-volume prototypes right here in Norwood (Mass.), in our factory. We have CNC machines and manual milling machines, lathes, all that kind of stuff, and we can produce the prototypes and test them out here in our labs.

JP: How much of a benefit is it to be able to create a prototype and to be able to turn one around quickly like that?
GW: Oh, it’s great. I mean, if we had to wait to get parts from China it will take weeks to get. We can turn them around here in a few days and the great thing about that is we can test them in our labs and, you know, when it comes to getting results nothing beats the testing.

I mean, you can do analytical analysis, you can do CFD simulations, but when you actually test the part in a situation that is similar to what the actual thing is going to be that’s where the real meat comes down.

Heat Sink Design

ATS engineers take customer data and using analytical modeling and CFD simulations can design the right cooling solution to meet the customer’s specific thermal needs. (Advanced Thermal Solutions, Inc.)

JP: So, we test the prototypes before sending them out to the customer? We do the testing here or do we send it to them first?
GW: It all depends on what the customer requires. Sometimes the customer has a chassis that we really can’t simulate in our labs, so we might send the prototype heat sinks to the customer and the customer will actually put them into their system to test them out.

Other times, a customer might have a concept and they don’t actually have a product yet, so we’ll mock something up in our labs and we’ll test it and it all just depends what the customer needs and also how complex the problem is.

If it’s a simple heat sink and pretty simple airflow, we might not need to test that because we understand that pretty well, but the more complex the chassis is and how the airflow bends and stuff like that, the greater benefits we get out of lab testing.

JP: Well, I appreciate it Greg. Thank you for taking us through the process of making a heat sink and solving thermal problems for our customers.
GW: Sure Josh. We love seeing new thermal challenges and coming up with ways of keeping stuff cool.

For more information about Advanced Thermal Solutions, Inc. thermal management consulting and design services, visit www.qats.com or contact ATS at 781.769.2800 or ats-hq@qats.com.

Analysis of Fan Curves and Fan Laws in Thermal Management of Electronics

This is the second installment in a two-part series examining the use of fans in the thermal management of electronics. Part one, which can be found at https://www.qats.com/cms/2017/03/06/utilizing-fans-thermal-management-electronics-systems, took a closer look at the common types of fans and blowers and the factors that engineers should consider when picking a fan.

In part two, basic fan laws will be explored, as well as using fan curves to analyze fan performance in a system. These standard calculations can help engineers establish boundary conditions for air velocity and pressure drop and ensure that these will meet the thermal requirements (e.g. ambient and junction temperature) of the system.

Fan Laws

CFD simulations of air velocity in a system with fans drawing air across high-powered components. Utilizing fan curves and fan laws enabled ATS engineers to establish the parameters for a successful use of fans for cooling this system. (Advanced Thermal Solutions, Inc.)

FAN LAWS

As noted by Mike Turner of Comair Rotron in “All You Need to Know About Fans,” the primary principle for determining whether or not a fan work within a particular system is that “any given fan can only deliver one flow at one pressure in a particular system.” Each fan has a specific operating point that can be discovered on the fan curve at the intersection of fan static pressure curve and the system pressure curve. Turner advises, “It is best to select a fan that will give an operating point being toward the high flow, low pressure end of the performance curve to maintain propeller efficiency and to avoid propeller stall.”

Before getting to the fan curve though, engineers must run through basic calculations to understand the conditions of the systems in which the fans will be placed. The three basic fan laws, according to Eldridge USA, are as follows:

Fan Laws

While those fan laws will tell you about the specific fans, it is also critical to examine the system in which the fans will be operating. Among the equations that can be used to characterize a system are Volumetric Flow Rate, Mass Flow Rate, Pressure, Power, and Sound (equations are listed below).

Fan Laws

A Qpedia Thermal eMagazine article entitled, “How to Use Fan Curves and Laws in Thermal Design,” added:

“Published fan laws apply to applications where a fan’s air flow rate and pressure are independent of the Reynolds number. In some applications, however, fan performance is not independent and thus the change in Reynolds number should be incorporated into the equation. To determine if the Reynolds number needs to be considered, it must first be calculated.

“According to AMCA specifications, an axial fan’s minimum Reynolds number is 2.0×106. When the calculated Reynolds number is above this value, its effects can be ignored.”

The equation to calculate the Reynolds number is as follows:

Fan Law

In an “Engineering Letter” from The New York Blower Company, it was explained that fan laws only work “within a fixed system with no change in the aerodynamics or airflow characteristics of the system.” In the case of electronics cooling, in which the system requirements will be mostly consistent (with margins for error in case of max power usage), these laws will govern the capabilities of the fans to provide the necessary forced convection cooling for the components in the system.

The Engineering Letter continued, “During the process of system design, the fan laws can be helpful in determining the alternate performance criteria or in developing a maximum/minimum range.” A Qpedia article entitled, “Designing Efficient Fans for Electronics Cooling Applications,” added, “As a general rule, fan efficiency increases with blade diameter and rotational speed.”

There are tools that can assist engineers in the calculation of these basic fan laws, including fan calculators, such as the one provided by Twin City Fans & Blowers.

ANALYZING FAN CURVES AND FAN PERFORMANCE

The aerodynamics of a fan can be charted in a fan curve, which displays the static pressure of the system dependent on the amount of air flow. As Turner noted, fan curves are read from right to left, beginning “with healthy aerodynamic flow and follow it through to aerodynamic stall.” Turner continued, “It is best to select a fan that will give an operating point being toward the high flow, low pressure end of the performance curve to maintain propeller efficiency and to avoid propeller stall.”

Fan Laws

An example of a basic fan curve with static pressure on the Y-axis and airflow on the X-axis. Fan curves are read from right to left beginning with healthy airflow.

There are means for testing fan curves, such as the FCM-100 Fan Characterization Module (pictured below) from Advanced Thermal Solutions, Inc. (ATS). The FCM-100 is specially designed with flow restriction plates that allow the user to control pressure drop across the system during testing. Used in conjunction with pressure and velocity measurement equipment, it verifies manufacturer performance data.

Fan Laws

The ATS FCM-100 Fan Characterization Module is a specialized unit designed to test and characterize fans of various sizes and performance outputs. (Advanced Thermal Solutions, Inc.)

The Qpedia article on fan curves explained, “During a typical fan test, a dozen or more operating points are plotted for pressure and flow rate, and from this data a fan curve is constructed.”

Once a fan curve is determined, it is possible to examine the data and find the operating range for the fans that will meet the thermal requirements of a system. It is also important to note a section in the fan curve, often referred to as the knee of the curve in which the relationship between flow rate and static pressure is no longer easy to predict. There is no longer an easily recognizable, calculable relationship between how a change in one will affect the other.

ATS field application engineer Vineet Barot explained how he analyzed fan curve data, particularly the knee of the curve, in a recent project:

“This is flow rate versus pressure. The more pressure you have in front of a fan, the slower it can pump out the air and this is the curve that determines that.

Fan Laws

Fan operating points on the board, determined by CFD simulations. (Advanced Thermal Solutions, Inc.)

“This little area here is sometime called the knee of the fan curve. Let’s say we’re in this area, the flow rate and pressure is relatively linear, so if I increase my pressure, if I put my hand in front of the fan, the flow rate goes down. If I have no pressure, I have my maximum flow rate. If I increase my pressure then the flow rate goes down. What happens in this part? The same thing. In the knee, a slight increase in pressure, so from .59 to .63, reduces the flow rate quite a bit.

Stratix 10 FPGA

CFD simulations showed that the fans were operating in the “knee” where it is hard to judge the impact of pressure changes on flow rate and vice versa. (Advanced Thermal Solutions, Inc.)

“So, for a 0.1 difference in flow rate (in cubic meters per second) it took 0.4 inches of water pressure difference, whereas here for a 0.1 difference in flow rate it only took a .04 increase in pressure. That’s why there’s a circle there. It’s a danger area because if you’re in that range it gets harder to predict what the flow will be because any pressure-change, any dust build-up, any change in estimated open area might change your flow rate.

Fan Laws

CFD analysis of flow vectors across high-powered components on a PCB. This simulation was part of an examination of fan performance in a system. (Advanced Thermal Solutions, Inc.)

CONCLUSION

While it is important to know the types of fans on the market and manufacturers provide data about the power and operating ranges of each product, it is important for there to be a basic understanding of the laws that govern how fans operate in a system and an ability to examine fan curve data in order to optimize performance.

“Bulk testing of electronics chassis provides the relationship between air flow and pressure drop and determines the fan performance needed to cool a given power load. The fan rating is often a misunderstood issue and published ratings can be somewhat misleading. Knowledge of fan performance curves, and how they are obtained, allows for a more informed decision when selecting a fan. Continued and ever shortening product design cycles demand a ‘get it right the first time’ approach. The upfront use of system curves, fan curves and fan laws can help meet this goal.”

Read more and see examples of fan laws and curves in practice at https://www.qats.com/cms/2013/07/24/how-to-use-fan-curves-and-laws-in-thermal-design.

CLICK HERE FOR PART I

To learn more about Advanced Thermal Solutions, Inc. consulting services, visit www.qats.com or contact ATS at 781.769.2800 or ats-hq@qats.com.

Utilizing Fans in Thermal Management of Electronics Systems

Fans in Thermal Management

There are different types of fans that are used in thermal management of electronics with tube axial fans being the most common. (Wikimedia Commons)


The ongoing trend in the electronics industry is for increasingly high-powered components to meet the ever-growing demands of consumers. Coupled with greater component-density in smaller packages, thermal management is more and more of a priority to ensure performance and reliability over the life of an electronics system.

As thermal needs have grown, engineers have sought out different cooling methods to supplement convection cooling. While options such as liquid cooling have grown in popularity in recent years, still one of the most common techniques is to add fans to a system.

Through the years, fan designs have improved. Fan blades have been streamlined to produce great flow rate with less noise and fans have become more power-efficient to meet the desires of customers trying to use less resources and save costs.

While much has changed in the presentation of fans, there are many basic concepts that engineers must consider when deciding how to implement fans in a project.

This is part one of a two-part series on how to select the best fan for a project. Part one will cover the types of fans that can be used. Part two, which can be found at https://www.qats.com/cms/2017/03/10/analysis-of-fan-curves-and-fan-laws-in-thermal-management-electronics, will cover fan laws and analyzing fan curves.

COMMON TYPES OF FANS AND BLOWERS

As described by Mike Turner of Comair Rotron in an article for Electronics Cooling Magazine, “All You Need to Know About Fans,” fans are essentially low pressure air pumps that take power from a motor to “output a volumetric flow of air at a given pressure.” He continued, “A propeller converts torque from the motor to increase static pressure across the fan rotor and to increase the kinetic energy of the air particles.”

In a white paper from Advanced Thermal Solutions, Inc. (ATS) entitled, “Performance Difference Between Fans and Blowers and Their Implementation,” it was added that fans are at their core, dynamic pumps. The article added, that in dynamic pumps “the fluid increases momentum while moving through open passages and then converts its high velocity to a pressure increase by exiting into a diffuser section.”

The biggest difference between a fan and a blower is the direction in which the air is delivered. Fans push air in a direction that is parallel to the fan blade axis, while blowers move air perpendicular to the blower axis. Turner noted that fans “can be designed to deliver a high flow rate, but tend to work against low pressure” and blowers move air at a “relatively low flow rate, but against high pressure.”

The three types of fans are centrifugal, propeller, tube axial, and vane axial:

• In centrifugal fans, the air flows into the housing and turns 90 degrees while accelerating due to centrifugal forces before being flowing out of the fan blades and exiting the housing.
• Propeller fans are the simplest form of a fan with only a motor and propellers and no housing.
• Tube axial fans, according to Turner, are similar to a propeller fan but “also has a venture around the propeller to reduce the vortices.”
• Vane axial fans have vanes trailing behind the propeller to straighten the swirling air as it is accelerated.

The most common fans used in electronics cooling are tube axial fans and there are a number of manufacturers creating options for engineers. A quick search of Digi-Key Electronics, offered options such as Sunon, Orion Fans, Sanyo Denki, NMB Technologies, Delta Electronics, Jameco Electronics, and several more.

Fans in Thermal Management

A fan is added to a heat sink on a PCB in order to increase the air flow and heat dissipation from the board component. (Advanced Thermal Solutions, Inc.)

FACTORS TO CONSIDER WHEN PICKING A FAN

When selecting a fan, engineers must consider the specific requirements of the system in which they are working, including factors such as the necessary airflow and the size restrictions of the board or the chassis. These basic factors will allow engineers to search through the many available options to find a fan that fits his or her needs.

In addition, engineers may look towards combining multiple fans in parallel or in a series to increase the flow rate across the components without increasing the size of the package or the diameter of the fan.

Parallel operation means having two or more fans side-by-side. When two fans are working in parallel, then the volume flow rate will be increased, even doubled when the fans are operating at maximum. Turner added. “The best results for parallel fans are achieved in systems with low resistance.”

In a series, the fans are stacked on top of each other and results in increased static pressure. Unlike parallel operations, fans in a series work best in a system with high resistance.

The ATS white paper noted, “In real situations, the fans may interfere with each other. The end results is a lower than expected performance.” Turner warns that in either parallel or series configurations there is a point in the combined performance curve that should be avoided because it creates unstable and unpredictable performance, but analyzing fan performance and fan curves will be covered in more detail in part two of the blog.

Efficiency is a major factor when selecting a fan. As noted in an article from Qpedia Thermal eMagazine, “A large data center contains about 400,000 servers and consumes 250 MW of power. It has been estimated that about 20% of the total power supplied to a high end server is consumed by fans.”

Clearly, finding a fan that can work efficiently with lower power will save a considerable about of resources. The article details several methods for creating efficiency in designing a system that includes fans:

“Fan power consumption is traditionally reduced by controlling the motor speed to produce only the airflow required for adequate cooling, rather than operating continuously at full speed. Significant energy savings can be achieved beyond this technique through fan efficiency increase. Optimizing the motor and electronic driver, increasing fan aerodynamic efficiency through careful redesign, and optimizing fan-system integration are three ways of achieving this.”

Read more about the techniques for achieving efficiency at https://www.qats.com/cms/wp-content/uploads/2015/03/Designing_Efficient_Fans_for_Electronics_Cooling
_Applications.pdf
.

CLICK HERE FOR PART II.

To learn more about Advanced Thermal Solutions, Inc. consulting services, visit www.qats.com or contact ATS at 781.769.2800 or ats-hq@qats.com.