aplicação peltier

ThermoelectricTechnical Reference

Home >> Thermoelectric Technical Reference

Thermoelectric Technical Reference — Introduction to Thermoelectric Cooling

Ferrotec's Thermoelectric Technical Reference Guide is a comprehensive technical explanation ofthermoelectrics and thermoelectric technology.

1.0 Introduction to Thermoelectric Cooling

1.1 A thermoelectric (TE) cooler, sometimes called a thermoelectric module or Peltier cooler, is asemiconductor-based electronic component that functions as a small heat pump. By applying a low voltageDC power source to a TE module, heat will be moved through the module from one side to the other. Onemodule face, therefore, will be cooled while the opposite face simultaneously is heated. It is important to notethat this phenomenon may be reversed whereby a change in the polarity (plus and minus) of the applied DCvoltage will cause heat to be moved in the opposite direction. Consequently, a thermoelectric module may beused for both heating and cooling thereby making it highly suitable for precise temperature controlapplications.

1.1.1 To provide the new user with a general idea of a thermoelectric cooler's capabilities, it might be helpfulto offer this example. If a typical single-stage thermoelectric module was placed on a heat sink that wasmaintained at room temperature and the module was then connected to a suitable battery or other DC powersource, the "cold" side of the module would cool down to approximately -40°C. At this point, the modul e wouldbe pumping almost no heat and would have reached its maximum rated "DeltaT (DT)." If heat was graduallyadded to the module's cold side, the cold side temperature would increase progressively until it eventuallyequaled the heat sink temperature. At this point the TE cooler would have attained its maximum rated "heatpumping capacity" (Qmax).

1.2 Both thermoelectric coolers and mechanical refrigerators are governed by the same fundamental laws ofthermodynamics and both refrigeration systems, although considerably different in form, function inaccordance with the same principles.

In a mechanical refrigeration unit, a compressor raises the pressure of a liquid and circulates the refrigerantthrough the system. In the evaporator or "freezer" area the refrigerant boils and, in the process of changing toa vapor, the refrigerant absorbs heat causing the freezer to become cold. The heat absorbed in the freezerarea is moved to the condenser where it is transferred to the environment from the condensing refrigerant. Ina thermoelectric cooling system, a doped semiconductor material essentially takes the place of the liquidrefrigerant, the condenser is replaced by a finned heat sink, and the compressor is replaced by a DC powersource. The application of DC power to the thermoelectric module causes electrons to move through thesemiconductor material. At the cold end (or "freezer side") of the semiconductor material, heat is absorbed bythe electron movement, moved through the material, and expelled at the hot end. Since the hot end of thematerial is physically attached to a heat sink, the heat is passed from the material to the heat sink and then,in turn, transferred to the environment.

1.3 The physical principles upon which modern thermoelectric coolers are based actually date back to theearly 1800's, although commercial TE modules were not available until almost 1960. The first importantdiscovery relating to thermoelectricity occurred in 1821 when a German scientist, Thomas Seebeck, foundthat an electric current would flow continuously in a closed circuit made up of two dissimilar metals providedthat the junctions of the metals were maintained at two different temperatures. Seebeck did not actuallycomprehend the scientific basis for his discovery, however, and falsely assumed that flowing heat producedthe same effect as flowing electric current. In 1834, a French watchmaker and part time physicist, JeanPeltier, while investigating the "Seebeck Effect," found that there was an opposite phenomenon wherebythermal energy could be absorbed at one dissimilar metal junction and discharged at the other junction whenan electric current flowed within the closed circuit. Twenty years later, William Thomson (eventually known asLord Kelvin) issued a comprehensive explanation of the Seebeck and Peltier Effects and described theirinterrelationship. At the time, however, these phenomena were still considered to be mere laboratorycuriosities and were without practical application.

In the 1930's Russian scientists began studying some of the earlier thermoelectric work in an effort toconstruct power generators for use at remote locations throughout the country. This Russian interest inthermoelectricity eventually caught the attention of the rest of the world and inspired the development ofpractical thermoelectric modules. Today's thermoelectric coolers make use of modern semiconductortechnology whereby doped semiconductor material takes the place of dissimilar metals used in earlythermoelectric experiments.

1.4 The Seebeck, Peltier, and Thomson Effects, together with several other phenomena, form the basis offunctional thermoelectric modules. Without going into too much detail, we will examine some of thesefundamental thermoelectric effects.

1.4.1 SEEBECK EFFECT: To illustrate the Seebeck Effect let us look at a simple thermocouple circuit asshown in Figure (1.1). The thermocouple conductors are two dissimilar metals denoted as Material x andMaterial y.

Thermoelectric Technical Reference http://www.silram-cor.co.il/gp.asp?gpid=56

1 de 20 06-06-2012 01:22

In a typical temperature measurement application, thermocouple A is used as a "reference" and is maintainedat a relatively cool temperature of Tc. Thermocouple B is used to measure the temperature of interest (Th)which, in this example, is higher than temperature Tc. With heat applied to thermocouple B, a voltage willappear across terminals Tl and T2. This voltage (Vo), known as the Seebeck emf, can be expressed as: Vo =axy x (Th - Tc)

where:Vo is the output voltage in volts

axy is the differential Seebeck coefficient between the two materials, x and y, in volts/oK

Th and Tc are the hot and cold thermocouple temperatures, respectively, in oK

1.4.2 PELTIER EFFECT: If we modify our thermocouple circuit to obtain the configuration shown in Figure(1.2), it will be possible to observe an opposite phenomenon known as the Peltier Effect.

If a voltage (Vin) is applied to terminals Tl and T2 an electrical current (I) will flow in the circuit. As a result ofthe current flow, a slight cooling effect (Qc) will occur at thermocouple junction A where heat is absorbed anda heating effect (Qh) will occur at junction B where heat is expelled. Note that this effect may be reversedwhereby a change in the direction of electric current flow will reverse the direction of heat flow. The Peltiereffect can be expressed mathematically as:

Qc or Qh=pxy x I

Where: pxy is the differential Peltier coefficient between the two materials, x and y, in volts I is the electriccurrent flow in amperes Qc, Qh is the rate of cooling and heating, respectively, in watts

Joule heating, having a magnitude of I x R (where R is the electrical resistance), also occurs in theconductors as a result of current flow. This Joule heating effect acts in opposition to the Peltier effect andcauses a net reduction of the available cooling.

1.4.3 THOMSON EFFECT: When an electric current is passed through a conductor having a temperaturegradient over its length, heat will be either absorbed by or expelled from the conductor. Whether heat isabsorbed or expelled depends upon the direction of both the electric current and temperature gradient. Thisphenomenon, known as the Thomson Effect, is of interest in respect to the principles involved but plays anegligible role in the operation of practical thermoelectric modules. For this reason, it is ignored. Thermoelectric Technical Reference — Basic Principle s of Thermoelectric Materials

Ferrotec's Thermoelectric Technical Reference Guide is a comprehensive technical explanation ofthermoelectrics and thermoelectric technology.

2.0 Basic Principles of Thermoelectric Modules & Ma terials

2.1 THERMOELECTRIC MATERIALS: The thermoelectric semiconductor material most often used intoday's TE coolers is an alloy of Bismuth Telluride that has been suitably doped to provide individual blocksor elements having distinct "N" and "P" characteristics. Thermoelectric materials most often are fabricated byeither directional crystallization from a melt or pressed powder metallurgy. Each manufacturing method has itsown particular advantage, but directionally grown materials are most common. In addition to Bismuth Telluride(Bi2Te3), there are other thermoelectric materials including Lead Telluride (PbTe), Silicon Germanium (SiGe),and Bismuth-Antimony (Bi-Sb) alloys that may be used in specific situations. Figure (2.1) illustrates therelative performance or Figure-of-Merit of various materials over a range of temperatures. It can be seen fromthis graph that the performance of Bismuth Telluride peaks within a temperature range that is best suited formost cooling applications.

APPROXIMATE FIGURE-OF-MERIT(Z)FOR VARIOUS TE MATERI ALS


2 de 20 06-06-2012 01:22

Figure (2.1) Performance of Thermoelectric Materials at Various Temperatures

2.1.1 BISMUTH TELLURIDE MATERIAL: Crystalline Bismuth Telluride material has several characteristicsthat merit discussion. Due to the crystal structure, Bi2Te3 is highly anisotropic in nature. This results in thematerial's electrical resistivity being approximately four times greater parallel to the axis of crystal growth(C-axis) than in the perpendicular orientation. In addition, thermal conductivity is about two times greaterparallel to the C-axis than in the perpendicular direction. Since the anisotropic behavior of resistivity isgreater than that of thermal conductivity, the maximum performance or Figure-of-Merit occurs in the parallelorientation. Because of this anisotropy, thermoelectric elements must be assembled into a cooling module sothat the crystal growth axis is parallel to the length or height of each element and, therefore, perpendicular tothe ceramic substrates.

There is one other interesting characteristic of Bismuth Telluride that also is related to the material's crystalstructure. Bi2Te3 crystals are made up of hexagonal layers of similar atoms.

While layers of Bismuth and Tellurium are held together by strong covalent bonds, weak van der Waalsbonds link the adjoining [Te¹] layers. As a result, crystalline Bismuth Telluride cleaves readily along these[Te¹][Te¹] layers, with the behavior being very similar to that of Mica sheets. Fortunately, the cleavage planesgenerally run parallel to the C-axis and the material is quite strong when assembled into a thermoelectriccooling module.

2.1.2 Bismuth Telluride material, when produced by directional crystallization from a melt, typically isfabricated in ingot or boule form and then sliced into wafers of various thicknesses. After the wafer's surfaceshave been properly prepared, the wafer is then diced into blocks that may be assembled into thermoelectriccooling modules. The blocks of Bismuth Telluride material, which usually are called elements or dice, alsomay be manufactured by a pressed powder metallurgy process.

2.2 THERMOELECTRIC COOLING MODULES: A practical thermoelectric cooler consists of two or moreelements of semiconductor material that are connected electrically in series and thermally in parallel. Thesethermoelectric elements and their electrical interconnects typically are mounted between two ceramicsubstrates. The substrates serve to hold the overall structure together mechanically and to insulate theindividual elements electrically from one another and from external mounting surfaces. After integrating thevarious component parts into a module, thermoelectric modules ranging in size from approximately 2.5-50 mm(0.1 to 2.0 inches) square and 2.5-5mm (0.1 to 0.2 inches) in height may be constructed.

Figure (2.2) Schematic Diagram of a Typical Thermoelectric Cooler

2.2.1 Both N-type and P-type Bismuth Telluride thermoelectric materials are used in a thermoelectric cooler.This arrangement causes heat to move through the cooler in one direction only while the electrical currentmoves back and forth alternately between the top and bottom substrates through each N and P element.N-type material is doped so that it will have an excess of electrons (more electrons than needed to completea perfect molecular lattice structure) and P-type material is doped so that it will have a deficiency of electrons(fewer electrons than are necessary to complete a perfect lattice structure). The extra electrons in the Nmaterial and the "holes" resulting from the deficiency of electrons in the P material are the carriers whichmove the heat energy through the thermoelectric material. Figure (2.2) shows a typical thermoelectric coolerwith heat being moved as a result of an applied electrical current (I). Most thermoelectric cooling modules are


3 de 20 06-06-2012 01:22

fabricated with an equal number of N-type and P-type elements where one N and P element pair form athermoelectric "couple." The module illustrated in Figure (2.2) has two pairs of N and P elements and istermed a "two-couple module".

Heat flux (heat actively pumped through the thermoelectric module) is proportional to the magnitude of theapplied DC electric current. By varying the input current from zero to maximum, it is possible to adjust andcontrol the heat flow and temperature.

3.0 Applications for Thermoelectric Coolers

3.1 Applications for thermoelectric modules cover a wide spectrum of product areas. These includeequipment used by military, medical, industrial, consumer, scientific/laboratory, and telecommunicationsorganizations. Uses range from simple food and beverage coolers for an afternoon picnic to extremelysophisticated temperature control systems in missiles and space vehicles.

Unlike a simple heat sink, a thermoelectric cooler permits lowering the temperature of an object belowambient as well as stabilizing the temperature of objects which are subject to widely varying ambientconditions. A thermoelectric cooler is an active cooling module whereas a heat sink provides only passivecooling.

Thermoelectric coolers generally may be considered for applications that require heat removal ranging frommilliwatts up to several thousand watts. Most single-stage TE coolers, including both high and low currentmodules, are capable of pumping a maximum of 3 to 6 watts per square centimeter (20 to 40 watts per squareinch) of module surface area. Multiple modules mounted thermally in parallel may be used to increase totalheat pump performance. Large thermoelectric systems in the kilowatt range have been built in the past forspecialized applications such as cooling within submarines and railroad cars. Systems of this magnitude arenow proving quite valuable in applications such as semiconductor manufacturing lines.

3.2 Typical applications for thermoelectric modules include:

AvionicsBlack Box CoolingCalorimetersCCD (Charged Couple Devices)CID (Charge Induced Devices)Cold ChambersCold PlatesCompact Heat ExchangersConstant Temperature BathsDehumidifiersDew Point HygrometersElectronics Package CoolingElectrophoresis Cell CoolersEnvironmental AnalyzersHeat Density MeasurementIce Point ReferencesImmersion CoolersIntegrated Circuit CoolingInertial Guidance SystemsInfrared Calibration Sources and Black Body ReferencesInfrared DetectorsInfrared Seeking MissilesLaser CollimatorsLaser Diode CoolersLong Lasting Cooling DevicesLow Noise AmplifiersMicroprocessor CoolingMicrotome Stage CoolersNEMA EnclosuresNight Vision EquipmentOsmometersParametric AmplifiersPhotomultiplier Tube HousingPower Generators (small)Precision Device Cooling (Lasers and Microprocessors)Refrigerators and on-board refrigeration systems (Aircraft, Automobile, Boat, Hotel, Insulin,Portable/Picnic, Pharmaceutical, RV)Restaurant Portion DispenserSelf-Scanned Arrays SystemsSemiconductor Wafer ProbesStir CoolersThermal Viewers and Weapons SightsThermal Cycling Devices (DNA and Blood Analyzers)Thermostat Calibrating BathsTissue Preparation and StorageVidicon Tube CoolersWafer Thermal CharacterizationWater and Beverage CoolersWet Process Temperature ControllerWine Cabinets


4 de 20 06-06-2012 01:22

5.0 Heat Sink Considerations

5.1 Rather than being a heat absorber that consumes heat by magic, a thermoelectric cooler is a heat pumpwhich moves heat from one location to another. When electric power is applied to a TE module, one facebecomes cold while the other is heated. In accordance with the laws of thermodynamics, heat from the(warmer) area being cooled will pass from the cold face to the hot face. To complete the thermal system, thehot face of the TE cooler must be attached to a suitable heat sink that is capable of dissipating both the heatpumped by the module and Joule heat created as a result of supplying electrical power to the module.

A heat sink is an integral part of a thermoelectric cooling system and its importance to total systemperformance must be emphasized. Since all operational characteristics of TE devices are related to heat sinktemperature, heat sink selection and design should be considered carefully.

A perfect heat sink would be capable of absorbing an unlimited quantity of heat without exhibiting anyincrease in temperature. Since this is not possible in practice, the designer must select a heat sink that willhave an acceptable temperature rise while handling the total heat flow from the TE device(s). The definitionof an acceptable increase in heat sink temperature necessarily is dependent upon the specific application,but because a TE module's heat pumping capability decreases with increasing temperature differential, it ishighly desirable to minimize this value. A heat sink temperature rise of 5 to 15°C above ambient (or c oolingfluid) is typical for many thermoelectric applications.

Several types of heat sinks are available including natural convection, forced convection, and liquid-cooled.Natural convection heat sinks may prove satisfactory for very low power applications especially when usingsmall TE devices operating at 2 amperes or less. For the majority of applications, however, naturalconvection heat sinks will be unable to remove the required amount of heat from the system, and forcedconvection or liquid-cooled heat sinks will be needed.

Heat sink performance usually is specified in terms of thermal resistance (Q):

Qs=Ts - Ta

____________Q

where:

Qs = Thermal Resistance in Degrees C per Watt Ts = Heat Sink Temperature in Degrees C Ta =Ambient or Coolant Temperature in Degrees C Q = Heat Input to Heat Sink in Watts

5.2 Each thermoelectric cooling application will have a unique heat sink requirement and frequently there willbe various mechanical constraints that may complicate the overall design. Because each case is different, itis virtually impossible to suggest one heat sink configuration suitable for most situations. We have several offthe shelf heat sinks and liquid heat exchangers appropriate for many applications but encourage you tocontact our engineering department.

Note that when combining thermoelectric cooling modules and heat sinks into a total thermal system, itnormally is NOT necessary to take into account heat loss or temperature rise at the module to heat sinkjunctions. Module performance data presented herein already includes such losses based on the use ofthermal grease at both hot and cold interfaces. When using commercially available heat sinks forthermoelectric cooler applications, it is important to be aware that some off-the-shelf units do not haveadequate surface flatness. A flatness of 1mm/m (0.001 in/in) or better is recommended for satisfactorythermal performance and it may be necessary to perform an additional lapping, flycutting, or grindingoperation to meet this flatness specification.

5.2.1 NATURAL CONVECTION HEAT SINKS: Natural convection heat sinks normally are useful only forlow power applications where very little heat is involved. Although it is difficult to generalize, most naturalconvection heat sinks have a thermal resistance (Qs) greater than 0.5°C/watt and often exceeding 10°C/ watt.A natural convection heat sink should be positioned so that (a) the long dimension of the fins is in thedirection of normal air flow, vertical operation improves natural convection and (b) there are no significantphysical obstructions to impede air flow. It also is important to consider that other heat generatingcomponents located near the heat sink may increase the ambient air temperature, thereby affecting overallperformance.

5.2.2 FORCED CONVECTION HEAT SINKS: Probably the most common heat-sinkingmethod used with thermoelectric coolers is forced convection. When compared to natural convection heatsinks, substantially better performance can be realized. The thermal resistance of quality forced convectionsystems typically falls within a range of 0.02 to 0.5°C/watt. Many standard heat sink extrusions are a vailablethat, when coupled with a suitable fan, may be used to form the basis of a complete cooling assembly.Cooling air may be supplied from a fan or blower and may either be passed totally through the length of theheat sink or may be directed at the center of the fins and pass out both open ends. This second air flowpattern, illustrated in Figure (5.l), generally provides the best performance since the air blown into the face ofthe heat sink creates greater turbulence resulting in improved heat transfer. For optimum performance, thehousing of an axial fan should be mounted a distance of 8-20mm (0.31-0.75") from the fins. Otherconfigurations may be considered depending on the application.


5 de 20 06-06-2012 01:22

Figure (5.1) Forced Convection Heat Sink System Showing Preferred Air Flow

The thermal resistance of heat sink extrusions often is specified at an air flow rate stated in terms of velocitywhereas the output of most fans is given in terms of volume. The conversion from volume to velocity is:

Velocity = Volume / Cross-sectional Area of Air Passage or: Linear Feet per Minute = Cubic Feet per Minute / Area in Square Feet or: Linear Meters per Minute = Cubic Meters per Minute / Area in Square Meters

5.2.3 LIQUID COOLED HEAT SINKS: Liquid cooled heat sinks provide the highest thermal performance perunit volume and, when optimally designed, can exhibit a very low thermal resistance. Although there aremany exceptions, the thermal resistance of liquid cooled heat sinks typically falls between 0.01 and0.1°C/watt. Simple liquid heat sinks can be constru cted by soldering copper tubing onto a flat copper plate orby drilling holes in a metal block through which water may pass. With greater complexity (and greater thermalperformance), an elaborate serpentine water channel may be milled in a copper or aluminum block that lateris sealed-off with a cover plate. We offer several liquid-type heat sinks that may be used advantageously inthermoelectric systems. With other commercial heat sinks, always check the surface flatness prior toinstallation. While liquid cooling may be considered undesirable and/or unsatisfactory for many applications, itmay be the only viable approach in specific situations.

6.0 Installation of Thermoelectric Modules

This section of the technical reference guide explaines the techniques that can used to install or mount athermoelectric module or peltier cooler including:

» Clamping

» Bonding with Epoxy

» Soldering

» Mounting Pads and other Material

6.1 Important Installation Considerations

Techniques used to install thermoelectric modules in a cooling system are extremely important. Failure toobserve certain basic principles may result in unsatisfactory performance or reliability. Some of the factors tobe considered in system design and module installation include the following:

Thermoelectric modules have high mechanical strength in the compression mode but shear strengthis relatively low. As a result, a TE cooler should not be designed into a system where it serves as asignificant supporting member of the mechanical structure.

All interfaces between system components must be flat, parallel, and clean to minimize thermalresistance. High conductivity thermal interface material is often used to ensure good contact betweensurfaces.

The "hot" and "cold" sides of standard thermoelectric modules may be identified by the position of thewire leads. Wires are attached to the hot side of the module, which is the module face that is incontact with the heat sink. For modules having insulated wire leads, when the red and black leads areconnected to the respective positive and negative terminals of a DC power supply, heat will bepumped from the module's cold side, through the module, and into the heat sink. Note that for TEmodules having bare wire leads, the positive connection is on the right side and the negativeconnection is on the left when the leads are facing toward the viewer and the substrate with the leadsattached presented on the bottom.

When cooling below ambient, the object being cooled should be insulated as much as possible tominimize heat loss to the ambient air. To reduce convective losses, fans should not be positioned sothat air is blowing directly at the cooled object. Conductive losses also may be minimized by limitingdirect contact between the cooled object and external structural members.

When cooling below the dew point, moisture or frost will tend to form on exposed cooled surfaces. Toprevent moisture from entering a TE module and severely reducing its thermal performance, aneffective moisture seal should be installed. This seal should be formed between the heat sink andcooled object in the area surrounding the TE module(s). Flexible foam insulating tape or sheetmaterial and/or silicone rubber RTV are relatively easy to install and make an effective moisture seal.Several methods for mounting thermoelectric modules are available and the specific productapplication often dictates the method to be used. Possible mounting techniques are outlined in the


6 de 20 06-06-2012 01:22

following paragraphs.

6.1.1 HEIGHT TOLERANCE: Most thermoelectric cooling modules are available with two height tolerancevalues, +/-0.3mm (+/-0.010") and +/-0.03mm (0.001"). When only one module is used in a thermoelectricsubassembly, a +/-0.3mm tolerance module generally is preferable since it provides a slight cost advantageover a comparable tight-tolerance module. For applications where more than one module is to be mountedbetween the heat sink and cooled object, however, it is necessary to closely match the thickness of allmodules in the group to ensure good heat transfer. For this reason, +/-0.03mm (+/-0.001") tolerance modulesshould be used in all multiple-module configurations.

6.2 Clamping

The most common mounting method involves clamping the thermoelectric module(s) between a heat sink andflat surface of the article to be cooled. This approach, as illustrated in Figure (6.1), usually is recommendedfor most applications and may be applied as follows:

a) Machine or grind flat the mounting surfaces between which the TE module(s) will be located. To achieveoptimum thermal performance mounting surfaces should be flat to within 1mm/m (0.001 in/in).

b) If several TE modules are mounted between a given pair of mounting surfaces, all modules within thegroup must be matched in height/thickness so that the overall thickness variation does not exceed 0.06mm(0.002"). Module P/N with a "B" ending should be specified.

c) Mounting screws should be arranged in asymmetrical pattern relative to the module(s) soas to provide uniform pressure on the module(s)when the assembly is clamped together. Tominimize heat loss through the mounting screws,it is desirable to use the smallest size screw thatis practical for the mechanical system. For mostapplications, M3 or M3.5 (4-40 or 6-32) stainlesssteel screws will prove satisfactory. Alternately,nonmetallic fasteners can be used, e.g., nylon.Smaller screws may be used in conjunction withvery small mechanical assemblies. Bellevillespring washers or split lock-washers should beused under the head of each screw to maintaineven pressure during the normal thermalexpansion or contraction of system components.

d) Clean the module(s) and mounting surfaces to ensure that all burrs, dirt, etc., have been removed.

e) Coat the "hot" side of the module(s) with a thin layer (typically 0.02mm / 0.001" or less thickness) ofthermally conductive grease and place the module, hot side down, on the heat sink in the desired location.Gently push down on the module and apply a back and forth turning motion to squeeze out excess thermalgrease. Continue the combined downward pressure and turning motion until a slight resistance is detected.Ferrotec America recommends and stocks American Oil and Supply (AOS) type 400 product code 52032.

f) Coat the "cold" side of the module(s) with thermal grease as specified in step (e) above. Position and placethe object to be cooled in contact with the cold side of the module(s). Squeeze out the excess thermal greaseas previously described.

g) Bolt the heat sink and cooled object together using the stainless steel screws and spring washers. It isimportant to apply uniform pressure across the mounting surfaces so that good parallelism is maintained. Ifsignificantly uneven pressure is applied, thermal performance may be reduced, or worse, the TE module(s)may be damaged. To ensure that pressure is applied uniformly, first tighten all mounting screws finger tightstarting with the center screw (if any). Using a torque screwdriver, gradually tighten each screw by movingfrom screw to screw in a crosswise pattern and increase torque in small increments. Continue the tighteningprocedure until the proper torque value is reached. Typical mounting pressure ranges from 25 - 100 psidepending on the application. If a torque screwdriver is not available, the correct torque value may beapproximated by using the following procedure:

In a crosswise pattern, tighten the screws until they are "snug" but not actually tight. In the same crosswisepattern, tighten each screw approximately one quarter turn until the spring action of the washer can be felt.

h) A small additional amount of thermal grease normally is squeezed out soon after the assembly is firstclamped together. In order to insure that the proper screw torque is maintained, wait a minimum of one hourand recheck the torque by repeating step (g) above.

i) CAUTION : Over-tightening of the clamping screws may result in bending or bowing of either the heat sinkor cold object surface especially if these components are constructed of relatively thin material. Such bowingwill, at best, reduce thermal performance and in severe cases may cause physical damage to systemcomponents. Bowing may be minimized by positioning the clamping screws close to the thermoelectricmodule(s) and by using moderately thick materials. However, if hot and/or cold surfaces are constructed ofaluminum which is less than 6mm (0.25") thick or copper which is less than 3.3mm (0.13") thick, it may benecessary to apply screw torque of a lower value than specified in step (g) above.


7 de 20 06-06-2012 01:22

Figure (6.1)TE Module Installation Using the Clamping Method The proper bolt torque for TE module assemblies can

be determined by the following relationship:

T=((Sa x A)/N) x K x d

Where: T= torque on each bolt

Sa= cycling 25-50 psi, static 50-75 psi.

A= total surface area of module(s)

N= number of bolts used in assembly

K= torque coefficient (use K=0.2 for steel, K=0.15 for nylon)

d= nominal bolt diameter

For steel fasteners, we typically recommend either:

6-32 d=.138 in (.350 cm)4-40 d=.112 in (.284 cm)

The following recommended torque is calculated for nine 9500/065/018 modules held by four 4-40 steelfasteners:

T=((75 lbs/in.2 x (.44" x .48") x 9)/4)x 0.2 x .112 in. = 0.8 in-lbs.

6.3 BONDING WITH EPOXY

A second module mounting method that is useful for certain applications involves bonding the module(s) toone or both mounting surfaces using a special high thermal-conductivity epoxy adhesive. Since thecoefficients of expansion of the module's ceramics, heat sink and cooled object vary, we do not recommendbonding with epoxy for larger modules. Please consult your applications engineer for guidance. Note: Unlesssuitable procedures are used to prevent outgassing, epoxy bonding is not recommended if the TE coolingsystem is to be used in a vacuum. For module mounting with epoxy:

a) Machine or grind flat the mounting surfaces between which the TE module(s) will be located. Althoughsurface flatness is less critical when using epoxy, it is always desirable for mounting surfaces to be as flat aspossible.

b) Clean and degrease the module(s) and mounting surfaces to insure that all burrs, oil, dirt, etc., have beenremoved. Follow the epoxy manufacturer's recommendations regarding proper surface preparation.

c) Coat the hot side of the module with a thin layer of the thermally conductive epoxy and place the module,hot side down, on the heat sink in the desired location. Gently push down on the module and apply a backand forth turning motion to squeeze out excess epoxy. Continue the combined downward pressure andturning motion until a slight resistance is detected.

d) Weight or clamp the module in position until the epoxy has properly cured. Consult the epoxymanufacturer's data sheet for specific curing information. If an oven cure is specified, be sure that themaximum operating temperature of the TE module is not exceeded during the heating procedure. Note thatmost TE cooling modules manufactured by Ferrotec America have maximum operating temperatures of either150°C for the ValueTEC™ Series or 200°C for the Sup erTEC™ Series.

6.4 SOLDERING

Thermoelectric modules that have metallized external faces may be soldered into an assembly provided thatreasonable care is taken to prevent module overheating. Soldering to a rigid structural member of anassembly should be performed on one side of the module only (normally the hot side) in order to avoidexcessive mechanical stress on the module. Note that with a module's hot side soldered to a rigid body,however, a component or small electronic circuit may be soldered to the module's cold side provided that thecomponent or circuit is not rigidly coupled to the external structure. Good temperature control must bemaintained within the soldering system in order to prevent damage to the TE module due to overheating. Ourthermoelectric modules are rated for continuous operation at relatively high temperatures (150 or 200°C) sothey are suitable in most applications where soldering is desirable. Naturally these relative temperaturesshould not be exceeded in the process. Since the coefficients of expansion of the module ceramics, heat sinkand cooled object vary, we do not recommend soldering modules larger than 15 x 15 millimeters. Solderingshould not be considered in any thermal cycling application. For module mounting with solder, the followingsteps should be observed:

a) Machine or grind flat the mounting surface on which the module(s) will be located. Although surface


8 de 20 06-06-2012 01:22

flatness is not highly critical with the soldering method, it is always desirable for mounting surfaces to be asflat as possible. Obviously, the heat sink surface must be a solderable material such as copper or copperplated material.

b) Clean and degrease the heat sink surface and remove any heavy oxidation. Make sure that there are noburrs, chips, or other foreign material in the module mounting area.

c) Pre tin the heat sink surface in the module mounting area with the appropriate solder. The selected soldermust have a melting point that is less than or equal to the rated maximum operating temperature of the TEdevice being installed. When tinning the heat sink with solder, the heat sink's temperature should be just highenough so that the solder will melt but in no case should the temperature be raised more than the maximumvalue specified for the TE device.

d) Apply soldering flux to the TE module's hot side and place the module on the pre tinned area of the heatsink. Allow the module to "float" in the solder pool and apply a back and forth turning motion on the module tofacilitate solder tinning of the module surface. A tendency for the module to drag on the solder surface ratherthan to float is an indication that there is an insufficient amount of solder. In this event, remove the moduleand add more solder to the heat sink.

e) After several seconds the module surface should be tinned satisfactorily. Clamp or weight the module inthe desired position, remove the heat sink from the heat source, and allow the assembly to cool. Whensufficiently cooled, degrease the assembly to remove flux residue.

6.5 Mounting Pads And Other Material

There are a wide variety of products available designed to replace thermally conductive grease as aninterface material. Perhaps the most common are silicon-based mounting pads. Originally for use in mountingsemiconductor devices, these pads often exhibit excessive thermal resistance in thermoelectric applications.Since the pads allow for rapid production and eliminate cleanup time, they are popular in less demandingapplications. Leading manufacturers in this area include The Bergquist Company and the Chomerics Divisionof Parker Hannifin Corporation.

7.0 Power Supply Requirements

7.1 Thermoelectric coolers operate directly from DC power suitable power sources can range from batteriesto simple unregulated "brute force" DC power supplies to extremely sophisticated closed-loop temperaturecontrol systems. A thermoelectric cooling module is a low-impedance semiconductor device that presents aresistive load to its power source. Due to the nature of the Bismuth Telluride material, modules exhibit apositive resistance temperature coefficient of approximately 0.5 percent per degree C based on averagemodule temperature. For many noncritical applications, a lightly filtered conventional battery charger mayprovide adequate power for a TE cooler provided that the AC ripple is not excessive. Simple temperaturecontrol may be obtained through the use of a standard thermostat or by means of a variable-output DC powersupply used to adjust the input power level to the TE device. In applications where the thermal load isreasonably constant, a manually adjustable DC power supply often will provide temperature control on theorder of +/- 1°C over a period of several hours or more. Where precise temperature control is required, aclosed-loop (feedback) system generally is used whereby the input current level or duty cycle of thethermoelectric device is automatically controlled. With such a system, temperature control to +/- 0.1°C maybe readily achieved and much tighter control is not unusual.

7.2 Power supply ripple filtering normally is of less importance for thermoelectric devices than for typicalelectronic applications. However we recommend limiting power supply ripple to a maximum of 10 percent witha preferred value being < 5%.

7.2.1 Multistage cooling and low-level signal detection are two applications which may require lower values ofpower supply ripple. In the case of multistage thermoelectric devices, achieving a large temperaturedifferential is the typical goal, and a ripple component of less than two percent may be necessary to maximizemodule performance. In situations where very low level signals must be detected and/or measured, eventhough the TE module itself is electrically quiet, the presence of an AC ripple signal within the module andwire leads may be unsatisfactory. The acceptable level of power supply ripple for such applications will haveto be determined on a case-by-case basis.

7.3 Figure (7.1) illustrates a simple power supply capable of driving a 71-couple, 6-ampere module. Thiscircuit features a bridge rectifier configuration and capacitive-input filter. With suitable component changes, afull-wave-center-tap rectifier could be used and/or a filter choke added ahead of the capacitor. A switchingpower supply, having a size and weight advantage over a comparable linear unit, also is appropriate forpowering thermoelectric devices.

Figure (7.1)Simple Power Supply to Drive a 71-Couple, 6-Ampere TE Module


9 de 20 06-06-2012 01:22

7.4 A typical analog closed-loop temperature controller is illustrated in Figure (7.2). This system is capable ofclosely controlling and maintaining the temperature of an object and will automatically correct for temperaturevariations by means of the feedback loop. Many variations of this system are possible including adaptation todigital and/or computer control.

Figure (7.2)Block Diagram of a Typical Closed-Loop Temperature Controller

8.0 Thermal System Design Considerations

8.1 The first step in the design of a thermoelectric cooling system involves making an analysis of the system'soverall thermal characteristics. This analysis, which may be quite simple for some applications or highlycomplex in other cases, must never be neglected if a satisfactory and efficient design is to be realized. Someof the more important factors to be considered are discussed in the following paragraphs. Although we havemade certain simplifications that may horrify the pure thermodynamicist, the results obtained have been foundto satisfy all but those few applications that approach state-of-the-art limits.

Please note that design information contained in this manual is presented for the purpose of assisting thoseengineers and scientists who wish either to estimate cooling requirements or to actually develop their owncooling systems. For the many individuals who prefer not to become involved in the details of thethermoelectric design process, however, we encourage you to contact us for assistance. Ferrotec America iscommitted to providing strong customer technical support and our engineering staff is available to assist incomplex thermoelectric-related design activities.

8.2 ACTIVE HEAT LOAD: The active heat load is the actual heat generated by the component, "black box"or system to be cooled. For most applications, the active load will be equal to the electrical power input to thearticle being cooled (Watts = Volts x Amps) but in other situations the load may be more difficult to determine.Since the total electrical input power generally represents the worst case active heat load, we recommendthat you use this value for design purposes.

8.3 PASSIVE HEAT LOAD: The passive heat load (sometimes called heat leak or parasitic heat load) is thatheat energy which is lost or gained by the article being cooled due to conduction, convection, and/orradiation. Passive heat losses may occur through any heat-conductive path including air, insulation, andelectrical wiring. In applications where there is no active heat generation, the passive heat leak will representthe entire heat load on the thermoelectric cooler.

Determination of the total heat leak within a cooling system is a relatively complicated issue but a reasonableestimate of these losses often can be made by means of some basic heat transfer calculations. If there is anyuncertainty about heat losses in a given design, we highly recommend that you contact our engineering stafffor assistance and suggestions.

8.4 HEAT TRANSFER EQUATIONS: Several fundamental heat transfer equations are presented to assistthe engineer in evaluating some of the thermal aspects of a design or system.

8.4.1 HEAT CONDUCTION THROUGH A SOLID MATERIAL: The relationship that describes the transfer ofheat through a solid material was suggested by J.B. Fourier in the early 1800's. Thermal conduction isdependent upon the geometry and thermal conductivity of a given material plus the existing temperaturegradient through the material. Although thermal conductivity varies with temperature, the actual variation isquite small and can be neglected for our purposes. Mathematically, heat transfer by conduction may beexpressed as:

Q=(K)(DT)(A)x

Symbol DefinitionEnglish

UnitsMetric Units

QHeat Conducted Through theMaterial

BTU/hour watts

K Thermal conductivity of the material BTU/hour-ftoFwatts/meter-oCA Cross-sectional area of the material square feet square metersx Thickness of length of the materials feet meters

DTTemperature difference betweencold and hot sides of the material

Degrees F Degrees C

8.4.2 HEAT TRANSFER FROM AN EXPOSED SURFACE TO AMBI ENT BY CONVECTION: Heat leak toor from an uninsulated metal surface can constitute a significant heat load in a thermal system. Isaac Newtonproposed the relationship describing the transfer of heat when a cooled (or heated) surface is exposeddirectly to the ambient air. To account for the degree of thermal coupling between the surface andsurrounding air, a numerical value (h), called the Heat Transfer Coefficient, must be incorporated into theequation. Heat lost or gained in this manner may be expressed mathematically as: Q=(h)(A)(DT)

SymbolDefinition English Metric Units


10 de 20 06-06-2012 01:22

UnitsQ Heat transferred to or from ambient BTU/hour watts

hHeat transfer coefficient. For still air use a value of:For turbulent air use a value of:

BTU/hour-

ft2-oF4 to 5

15 to 20

watts/meter2-oC23 to 2885 to 113

A Area of the exposed surface square feet square meters

DTTemperature difference between theexposed surface and ambient

Degrees F Degrees C

8.4.3 HEAT TRANSFER THROUGH THE WALLS OF AN INSULAT ED ENCLOSURE: Heat leak to or froman insulated container combines an element of thermal conduction through the insulating material with anelement of convection loss at the external insulation surfaces. Heat lost from (or gained by) an insulatedenclosure may be expressed mathematically as:

Q = (A)(DT) x + 1K h

SymbolDefinition English Units Metric Units

QHeat conducted through theenclosure

BTU/hour watts

KThermal conductivity of theinsulation BTU/hour-ftoF watts/meter-oC

AExternal surface area of theenclosure

square feet square meters

x Thickness of the insulation feet meters

DTTemperature difference betweenthe inside and outside of theenclosure

Degrees F Degrees C

hHeat transfer coefficientFor still air use a value of:For turbulent air use a value of:

BTU/hour-

ft2-oF4 to 5

15 to 20

watts/meter2-oC23 to 2885 to 113

8.4.4 TIME NEEDED TO CHANGE THE TEMPERATURE OF AN O BJECT: Determination of the timerequired to thermoelectrically cool or heat a given object is a moderately complicated matter. For goodaccuracy, it would be necessary to make a detailed analysis of the entire thermal system including allcomponent parts and interfaces. By using the simplified method presented here, however, it is possible tomake a reasonable estimate of a system's thermal transient response.

t =(m)(Cp)(DT)

Q

Symbol Definition English Units Metric Units

t Time period for temperature change hours secondsm Weight of material pounds grams

Cp Specific heat of the material BTU/pound-oF calgram-oCDT Temperature change of the material Degrees F Degrees CQ Heat transferred to or from material BTU/hour cal/second

Note (1) : 1 Watt = 0.239 calories/second Note (2): Thermoelectric modules pump heat at a rate that is proportional to the temperature difference (DT)across the module. In order to approximate actual module performance, the average heat removal rate shouldbe used when determining the transient behavior of a thermal system. The average heat removal rate is:

Q = 0.5 (Qc at DTmin + Qc at DTmax)

Where: Qc at DTmin is the amount of heat a thermoelectric module is pumping at the initial objecttemperature when DC power is first applied to the module. The DT is zero at this time and the heat pumpingrate is at the highest point.

Qc at DTmax is the amount of heat a thermoelectric module is pumping when the object has cooled to thedesired temperature. The DT is higher at this time and the heat pumping rate is lower.

8.4.5 HEAT TRANSFER FROM A BODY BY RADIATION: Most thermoelectric cooling applications involverelatively moderate temperatures and small surface areas where radiation heat losses usually are negligible.Probably the only situation where thermal radiation may be of concern is that of a multistage cooler operatingat a very low temperature and close to its lower limit. For such applications, it sometimes is possible to attacha small radiation shield to one of the lower module stages. By fabricating this shield so that it surrounds theupper stage and cooled object, thermal radiation losses may be reduced substantially.

As an indication of the magnitude of heat leak due to thermal radiation, consider the following. A perfect

black-body having a surface area of 1.0 cm2 and operating at -100°C (173K) will receive approx imately 43milliwatts of heat from its 30°C (303K) surrounding s. Be aware that the accurate determination of radiationloss is a highly complicated issue and a suitable heat transfer textbook should be consulted for detailedinformation. A very simplified estimation of such losses may be made, however, by using the equation givenbelow.


11 de 20 06-06-2012 01:22

QR=(s)(A) (e) (Th4 – Tc

4)

SymbolDefinition English Units Metric UnitsQR Radiation heat loss BTU/hour watts

s Stefan-Boltzmann constant1.714 x 10-9

BTU/hour-

ft2-oR4

5.67 x 10-8

watts/meter2-K4

A Area of the exposed surface square feet square meterse Emissivity of exposed surfaces -- --

ThAbsolute temperature of warmersurface

Degrees R Degrees K

TcAbsolute temperature of coldersurface

Degrees R Degrees K

8.4.6 R-VALUE OF INSULATION: The R-value of an insulating material is a measure of the insulation'soverall effectiveness or resistance to heat flow. R-value is not a scientific term, per se, but the expression isused extensively within the building construction industry in the United States. The relationship betweenR-value, insulation thickness, and thermal conductivity may be expressed by the equation:

R = x 12K

where:x = Thickness of the insulation in inchesk = Thermal conductivity of the insulation in BTU/hr-ft-°F

Note : Insulation R-value normally is based on insulation thickness in inches. Since thermal conductivityvalues in Appendix B are expressed in feet, the k value in the equation's denominator has been multiplied by12.

8.5 THERMAL INSULATION CONSIDERATIONS: In order to maximize thermal performance and minimizecondensation, all cooled objects should be properly insulated. Insulation type and thickness often is governedby the application and it may not be possible to achieve the optimum insulation arrangement in all cases.Every effort should be made, however, to prevent ambient air from blowing directly on the cooled objectand/or thermoelectric device.

Figures (8.1) and (8.2) illustrate the relationship between the heat leak from an insulated surface and theinsulation thickness. It can be seen that even a small amount of insulation will provide a significant reductionin heat loss but, beyond a certain point, greater thicknesses give little benefit. The two heat leak graphs showheat loss in watts per square unit of surface area for a one degree temperature difference (DT) through theinsulation. Total heat leak (Qtot) in watts for other surface areas (SA) or DT's may be calculated by theexpression:

Qtot = Qleak x SA x DT

Figure (8.1)Heat Leak from an Insulated Surface in Metric Units


12 de 20 06-06-2012 01:22

Figure (8.2)Heat Leak from an Insulated Surface in English Units

9.0 Thermoelectric Module Selection

9.1 Selection of the proper TE Cooler for a specific application requires an evaluation of the total system inwhich the cooler will be used. For most applications it should be possible to use one of the standard moduleconfigurations while in certain cases a special design may be needed to meet stringent electrical,mechanical, or other requirements. Although we encourage the use of a standard device whenever possible,Ferrotec America specializes in the development and manufacture of custom TE modules and we will bepleased to quote on unique devices that will exactly meet your requirements.

The overall cooling system is dynamic in nature and system performance is a function of several interrelatedparameters. As a result, it usually is necessary to make a series of iterative calculations to "zero-in" on thecorrect operating parameters. If there is any uncertainty about which TE device would be most suitable for aparticular application, we highly recommend that you contact our engineering staff for assistance.

Before starting the actual TE module selection process, the designer should be prepared to answer thefollowing questions:

At what temperature must the cooled object be maintained?1.How much heat must be removed from the cooled object?2.Is thermal response time important? If yes, how quickly must the cooled object change temperatureafter DC power has been applied?

3.

What is the expected ambient temperature? Will the ambient temperature change significantly duringsystem operation?

4.

What is the extraneous heat input (heat leak) to the object as a result of conduction, convection,and/or radiation?

5.

How much space is available for the module and heat sink?6.What power is available?7.Does the temperature of the cooled object have to be controlled? If yes, to what precision?8.What is the expected approximate temperature of the heat sink during operation? Is it possible thatthe heat sink temperature will change significantly due to ambient fluctuations, etc.?

9.

Each application obviously will have its own set of requirements that likely will vary in level of importance.Based upon any critical requirements that can not be altered, the designer's job will be to select compatiblecomponents and operating parameters that ultimately will form an efficient and reliable cooling system. Adesign example is presented in section 9.5 to illustrate the concepts involved in the typical engineeringprocess.

9.2 USE OF TE MODULE PERFORMANCE GRAPHS: Before beginning any thermoelectric design activity itis necessary to have an understanding of basic module performance characteristics. Performance data ispresented graphically and is referenced to a specific heat sink base temperature. Most performance graphsare standardized at a heat sink temperature (Th) of +50°C and the resultant data is usable over a rang e ofapproximately 40°C to 60°C with only a slight error . Upon request, we can supply module performancegraphs referenced to any temperature within a range of -80°C to +200°C.

9.3 To demonstrate the use of these performance curves let us present a simple example. Suppose we havea small electronic "black box" that is dissipating 15 watts of heat. For the electronic unit to function properlyits temperature may not exceed 20°C. The room ambie nt temperature often rises well above the 20°C leve lthereby dictating the use of a thermoelectric cooler to reduce the unit's temperature. For the purpose of thisexample we will neglect the heat sink (we cannot do this in practice) other than to state that its temperaturecan be maintained at 50°C under worst-case conditio ns. We will investigate the use of a 71-couple, 6-amperemodule to provide the required cooling.

9.3.1 GRAPH: Qc vs. I This graph, shown in Figure (9.1), relates a module's heat pumping capacity (Qc)and temperature difference (DT) as a function of input current (I). In this example, established operatingparameters for the TE module include Th = 50°C, Tc = 20°C, and Qc = 15 watts. The required DT = Th-Tc =


13 de 20 06-06-2012 01:22

30°C.

It is necessary first to determine whether a single 71-couple, 6-ampere module is capable of providingsufficient heat removal to meet application requirements. We locate the DT=30 line and find that themaximum Qc value occurs at point A and with an input current of 6 amperes. By extending a line from point Ato the left y-axis, we can see that the module is capable of pumping approximately 18 watts while maintaininga Tc of 20°C. Since this Qc is slightly higher than necessary, we follow the DT=30 line downward until wereach a position (point B) that corresponds to a Qc of 15 watts. Point B is the operating point that satisfiesour thermal requirements. By extending a line downward from point B to the x-axis, we find that the properinput current is 4.0 amperes.

Figure (9.1)Heat Pumping Capacity Related to Temperature Differential as a Function of Input Current for a 71-Couple,6-Ampere Module

9.3.2 GRAPH: Vin vs. I This graph, shown in Figure (9.2), relates a module's input voltage (Vin) andtemperature difference (DT) as a function of input current (I). In this example, parameters for the TE moduleinclude Th = 50°C, DT = 30°C, and I = 4.0 amperes. We locate the DT=30 line and, at the 4.0 ampereintersection, mark point C. By extending a line from point C to the left y-axis, we can see that the requiredmodule input voltage (Vin) is approximately 6.7 volts.

Figure (9.2)Input Voltage Related to Temperature Differential as aFunction of Input Current for a 7I-Couple, 6-Ampere Module

9.3.3 GRAPH:COP vs. I This graph, shown in Figure (9.3), relates a module's coefficient of performance(COP) and temperature differential (DT) as a function of input current (I). In this example, parameters for theTE module include Th = 50°C, DT = 30°C, and I = 4.0 amperes.

We locate the DT=30 line and, at the 4.0 ampere intersection, mark point D. By extending a line from point Dto the left y-axis, we can see that the module's coefficient of performance is approximately 0.58.


14 de 20 06-06-2012 01:22

Figure (9.3)Coefficient of Performance Related to Temperature Differential as a Function of Input Current for a 71-Couple, 6-Ampere Module

Note that COP is a measure of a module's efficiency and it is always desirable to maximize COP wheneverpossible. COP may be calculated by:

9.4 An additional graph of Vin vs. Th, of the type shown in Figure (9.4), relates a module's input voltage (Vin)and input current (I) as a function of module hot side temperature (Th). Due to the Seebeck effect, inputvoltage at a given value of I and Th is lowest when DT=O and highest when DT is at its maximum point.Consequently, the graph of Vin vs. Th usually is presented for a DT=30 condition in order to provide theaverage value of Vin.

Figure (9.4)Input Voltage Related to Input Current as a Function ofHot Side Temperature for a 71-Couple, 6-Ampere Module

9.5 DESIGN EXAMPLE: To illustrate the typical design process let us present an example of a TE coolerapplication involving the temperature stabilization of a laser diode. The diode, along with related electronics,is to be mounted in a DIP Kovar housing and must be maintained at a temperature of 25°C. With the hous ingmounted on the system circuit board, tests show that the housing has a thermal resistance of 6°C/watt. Thelaser electronics dissipate a total of 0.5 watts and the design maximum ambient temperature is 35°C.

It is necessary to select a TE cooling module that not only will have sufficient cooling capacity to maintain theproper temperature, but also will meet the dimensional requirements imposed by the housing. An 18-couple,1.2 ampere TE cooler is chosen initially because it does have compatible dimensions and also appears tohave appropriate performance characteristics. Performance graphs for this module will be used to deriverelevant parameters for making mathematical calculations. To begin the design process we must firstevaluate the heat sink and make an estimate of the worst-case module hot side temperature (Th). For the TEcooler chosen, the maximum input power (Pin) can be determined from Figure (9.5) at point A.

Max. Module Input Power (Pin) = 1.2 amps x 2.4 volts = 2.9 wattsMax. Heat Input to the Housing = 2.9 watts + 0.5 watts = 3.4 wattsHousing Temperature Rise = 3.4 watts x 6°C/watt = 2 0.4°CMax. Housing Temperature (T,) = 35°C ambient + 20.4 °C rise = 55.4°C Since the hot sidetemperature (Th) of 55.4°C is reasonably close to t he available Tin = 50°C performance graphs,these graphs may be used to determine thermal performance with very little error.


15 de 20 06-06-2012 01:22

Figure (9.5)Vin vs. I Graph for an 18-Couple, I.2 Ampere Module

Now that we have established the worst-case Th value it is possible to assess module performance.

Module Temperature Differential (DT) = Th - Tc = 55.4 - 25 = 30°C

Figure (9.6)Qc vs. I Graph for an 18-Couple, 1.2 Ampere Module

From Figure (9.6) it can be seen that the maximum heat pumping rate (Qc) for DT=30 occurs at point B and isapproximately 0.9 watts. Since a Qc of only 0.5 watts is needed, we can follow the DT=30 line down until itintersects the 0.5 watt line marked as point C. By extending a line downward from point C to the x-axis, wecan see that an input current (I) of approximately 0.55 amperes will provide the required cooling performance.Referring back to the Vin vs. I graph in Figure (9.5), a current of 0.55 amperes, marked as point D, requires avoltage (Vin) of about 1.2 volts. We now have to repeat our analysis because the required input power isconsiderably lower than the value used for our initial calculation. The new power and temperature values willbe:

Max. Module Input Power (Pin) = 0.55 amps x 1.2 volts = 0.66 wattsMax. Heat Input to the Housing = 0.66 watts + 0.50 watts = 1.16 wattsHousing Temperature Rise = 1.16 watts x 6°C/watt = 7°CMax. Housing Temperature (Th) = 35°C ambient + 7°C rise = 42°C

Module Temperature Differential (DT) = Th-Tc = 42°C -25°C = 17°C

It can be seen that because we now have another new value for Th it will be necessary to continue repeatingthe steps outlined above until a stable condition is obtained. Note that calculations usually are repeated untilthe difference in the Th values from successive calculations is quite small (often less that 0.1°C for goodaccuracy). There is no reason to present the repetitive calculations here but we can conclude that theselected 18-couple TE module will function very well in this application. This analysis clearly shows theimportance of the heat sink in any thermoelectric cooling application.

9.6 USE OF MULTIPLE MODULES: Relatively large thermoelectric cooling applications may require the useof several individual modules in order to obtain the required rate of heat removal. For such applications, TEmodules normally are mounted thermally in parallel and connected electrically in series. An electrical series-parallel connection arrangement may also be used advantageously in certain instances. Because heat sinkperformance becomes increasingly important as power levels rise, be sure that the selected heat sink isadequate for the application.

10.0 Reliability of Thermoelectric Cooling Modules


16 de 20 06-06-2012 01:22

10.1 INTRODUCTION: Thermoelectric cooling modules are considered to be highly reliable components dueto their solid-state construction. For most applications they will provide long, trouble-free service. There havebeen many instances where TE modules have been used continuously for twenty or more years and the lifeof a module often exceeds the life of the associated equipment. The specific reliability of thermoelectricdevices tends to be difficult to define, however, because failure rates are highly dependent upon theparticular application. For applications involving relatively steady-state cooling where DC power is beingapplied to the module on more-or-less continuous and uniform basis, thermoelectric module reliability isextremely high. Mean Time Between Failures (MTBFs) in excess of 200,000 hours are not uncommon insuch cases and this MTBF value generally is considered to be an industry standard. On the other hand,applications involving thermal cycling show significantly worse MTBFs especially when TE modules arecycled up to a high temperature.

The publishing of thermoelectric module reliability data entails some risk because there are numerousapplication parameters and conditions that will affect the end result. Although reliability data is valid for theconditions under which a test was conducted, it is not necessarily applicable to other configurations. Moduleassembly and mounting methods, power supply and temperature control systems and techniques, andtemperature profiles, together with a host of external factors, can combine to produce failure rates rangingfrom extremely low to very high. In an effort to provide users with certain basic information aboutthermoelectric module life and to aid engineers in designing systems for optimum reliability, we institutedseveral test programs to acquire the necessary reliability data. Test results to date are presented for severalsituations that may be useful to end-users having similar or related applications. This data will be shared on acase-by-case basis depending on application and availability.

General requirements for the proper installation of thermoelectric modules may be found in Section 6 of thistechnical manual. It is important that modules are installed in accordance with these general requirements inorder to minimize the possibility of premature module failure due to faulty assembly techniques. Someinstallation related factors that can affect module reliability include:

a) Thermoelectric modules exhibit relatively high mechanical strength in a compression mode but shearstrength is comparatively low. A TE cooler should not be designed into a system where it serves as a majorsupporting member of the mechanical structure. Furthermore, in applications where severe shock andvibration will be present, a thermoelectric cooling module should be compression-mounted, i.e., installed bythe clamping method. When properly mounted, thermoelectric coolers have successfully met the shock andvibration requirements of aerospace, military, and similar environments.

b) Although the maximum recommended compression loading for thermoelectric modules is 15 kilograms persquare centimeter (200 pounds per square inch) of module surface area, tests have shown that well over 75kilograms per square centimeter (1000 pounds per square inch) compression normally can be applied to mostof our modules without causing failure. It is important to ensure that when modules are installed using theclamping method, sufficient pressure is maintained so that a module is not "loose" whereby it may easily bemoved by applying a small sideways or lateral force. Loose modules may be a particular problem whenseveral modules are grouped together in the same cooling assembly. In this situation, the lack of adequateclamping pressure may result in both reduced cooling performance and early module failure. When multiplemodules are mounted in an array, modules with a close height tolerance of +/- .03mm (.001") arerecommended. In all cases, clamping pressure must be applied uniformly and mating surfaces must be flat(see section 6 for Installation Guidelines).

c) A large unsupported mass should not be directly bonded to a module's cold surface to prevent the possiblefracture of module components when subjected to significant mechanical shock. Where a large mass isinvolved, thermoelectric modules should be clamped between the heat sink and either the mass itself or anintermediate "cold plate" on which the mass is mounted. In this arrangement, the clamping screws willeffectively increase shear strength of the overall mechanical system.

d) Moisture should not be allowed to enter the inside of a thermoelectric module in order to prevent both areduction in cooling performance and the possible corrosion of module materials through electro-chemicalaction or electrolysis. When cooling below the dew point, a moisture seal should be provided either on themodule itself or between the heat sink and cooled object in the area surrounding the TE module. Anelectronic-grade silicone rubber RTV may be used to directly seal a thermoelectric module. Flexibleclosed-cell foam insulating tape or sheet material, possibly combined with RTV to fill small gaps, may be usedfor a seal between the cold object and the heat sink.

e) When an application will involve large temperature changes or thermal cycling, thermoelectric modulesshould not be installed using solder or epoxy whereby an object is rigidly bonded to the module. Unless thethermal coefficients of expansion of all system components are similar, rigid bonding combined withtemperature cycling often will result in early module failure due to the induced thermal stresses. Rigidbonding to the module's hot side generally is less of a problem because the hot side temperature tends to berelatively constant during operation. When significant temperature variation or temperature cycling isinvolved, we strongly recommend that modules be mounted by clamping (compression) using a flexiblemounting material such as thermal grease or foils of graphite or indium. In addition, rigid mounting to bothsides of modules is not recommended for devices larger than about 15mm (5/8") square.

Temperature control methods also have an impact on thermoelectric module reliability. Linear or proportionalcontrol should always be chosen over ON/OFF techniques when prolong life of the module is required.

10.2 MODULE RELIABILITY RELATED TO HIGH TEMPERATURE EXPOSUREThermoelectric module failures typically may be classified into two groups: catastrophic failures anddegradation failures. Degradation failures tend to be long-term in nature and usually are caused by changesin semiconductor material parameters or increases in electrical contact resistance. High temperatureexposure may lead to material parameter changes and, therefore, reduced thermoelectric performance. A testwas conducted to study this effect. Several groups of ValueTEC™ Series (6300 Series) and SuperTEC™Series (9500 Series) TE modules were subjected to long-term, continuous exposure to an elevatedtemperature of 150°C in a normal air atmosphere. Du ring the test period, relevant module parameters were


17 de 20 06-06-2012 01:22

regularly measured and recorded. One parameter that is a good indicator of overall module performance isthe maximum temperature differential (DTmax). This parameter was tracked over a 42-month period with theaverage value being shown the graph of Figure (10.1). It can be seen that a small (2.5%) decline in DTmax,with a decreasing rate of change, occurred in the first 12 months of high temperature exposure. In theremaining 30 months, however, the additional reduction in DTmax was only about 1.3% as semiconductormaterial characteristics stabilized. Note that for this test, there was no measurable difference in resultsbetween the ValueTEC™ Series (150°C rated) or the S uperTEC™ Series (200°C rated) modules.

Figure (10.1)

10.3 MODULE RELIABILITY RELATED TO THERMAL CYCLINGThe continuous thermal cycling of thermoelectric modules over a wide temperature range effectivelyconstitutes a module "torture test," especially when the modules are raised to a relatively high temperature atone end of the cycle. Except for a few unusual applications, module failure rates are higher for this mode ofoperation than for any other operating condition. The basis for most thermal cycling failures is theunavoidable mismatch of thermal expansion coefficients of the various module components and materials.Such failures tend to be catastrophic in nature but some degradation normally may be observed prior tofailure.

It is necessary, at this point, to define thermal cycling. Many thermoelectric applications involve the periodicraising and lowering of the control temperature, sometimes over a fairly wide temperature range. Althoughthere often is not a well defined line between a cycling and noncycling application, thermal cycling usually isconsidered to be an operation where the temperature is regularly, and more or less continuously, raised andlowered over a long period of operation. A cycling application tends to suggest automatic or machine controlof the temperature excursion as opposed to manual control. If the temperature of an apparatus istemperature-cycled up and down a few times each day, this generally is not considered to be a temperature-cycling application for the purpose of this discussion. If you are uncertain about the status of your particularapplication, please do not hesitate to contact us for assistance.

At least four factors relate to failure rate in thermal cycling including (1) the total number of cycles, (2) thetotal temperature excursion over the cycle, (3) the upper temperature limit of the cycle, and (4) the rate oftemperature change. Highest reliability and module life is seen when the number of cycles is small, thetemperature excursion or range is narrow, the upper temperature limit is relatively low and the rate oftemperature change is minimalized. (Conversely, a large number of cycles over a wide temperature rangewith a rapid rate of change and a high temperature value on the up cycle results in significantly lower modulelife.) It is important to note that absolute module life is dependent upon the total number of cycles rather thatthe total time required to accrue those cycles. Consequently, when discussing thermal cycling, MTBF is beststated in terms of number of cycles instead of hours; we will take the liberty of using MTBF in this manner inthe following discussion.

The type of module used in thermal cycling applications also is important in respect to failure rate. Modulesrated at higher maximum operating temperatures provide substantially better life than do lower rated devices.This is true even though the upper temperature of the cycle is well below the maximum rated moduletemperature. In one application involving a two-stage thermoelectric assembly that was being cycled between-55°C and +125°C, a 150°C rated module provided a M TBF of 8100 cycles while a module rated at 200°Cexhibited a MTBF of 17,500 cycles. Modules rated at even lower maximum operating temperatures shouldonly be used for relatively low temperature cycling applications. In general, we recommend the SuperTECseries modules (rated for 200°C) be used for therma l cycling applications exceeding 90°C.

It should be mentioned that two other factors also may affect thermal cycling MTBFs. Physically smallermodules having fewer couples appear to provide improved life as do modules having larger elements or"dice." Sufficient data is available to suggest that modules having a size of 30mm (1.17") square or lessexhibit better reliability in thermal cycling applications than do physically larger modules. Thermally inducedmechanical stresses are greater in larger modules and such modules generally have a greater number ofcouples resulting in many more individual solder connections which may become fatigued by thermal stress.

In order to better define module failure rates under high temperature thermal cycling conditions, a test wasconducted involving the continuous cycling of SuperTEC™ Series modules between +30°C and +100°C.Modules were mounted on a forced convection heat sink and covered with an insulated aluminum plate.Polarity of the applied DC power was alternately reversed to provide active heating and cooling and thecover-plate temperature was measured to determine cycling limits. The total time period of the cycle was 5minutes (2.5 minutes from 30°C to 100°C and 2.5 min utes from 100°C to 30°C) resulting in 288 cycles pe rday or 2016 cycles per week. Module parameters were measured weekly and a failure was indicated by asharp rise in electrical resistance.


18 de 20 06-06-2012 01:22

Modules showed a slow and predictable rise in electrical resistance until reaching a point where a rapidresistance increase occurred indicating failure. All modules achieved a minimum of 25,000 cycles withoutfailure, see Figure (10.2), and the test was continued until 50% of the modules failed. MTBF of the modulegroup was calculated to be 68,000 cycles. Once again it is important to note that mounting methods, andoverall assembly details are important factors when the application involves thermal cycling. Someapplications have been tested between 5°C and 95°C exhibiting MTBF's over 100,000 cycles.

Figure (10.2) Before leaving the subject of thermal cycling it might be worthwhile to mention a practical usefor this process. Because of the resulting mechanical stresses within a thermoelectric module, thermal cyclinghas been shown to be an effective "burn-in" technique. By subjecting thermoelectric devices to a wellcontrolled cycling program, it is possible to identify potentially unsatisfactory modules thereby reducing thelikelihood of infant mortality failures. There obviously is some cost associated with this operation but it maybe useful when extremely high reliability is required.

10.4 MODULE RELIABILITY RELATED TO ON/OFF POWER CYC LINGAs discussed previously, the accepted industry standard for thermoelectric module MTBF is 200,000 hoursminimum. This MTBF value is based on relatively steady-state module operation where system power isoccasionally (typically a few times per day) turned on and off. In some applications power is turned on and offmore frequently especially where thermostatic temperature control is used. A test was conducted usingValueTEC™ Series modules to study the effects of ON/OFF power cycling at a relatively constanttemperature. Modules were mounted between a pair of forced convection cooled heat sinks using thermalgrease at the module/heat sink interfaces. Full rated current was supplied to the modules for a period of 7.5seconds followed by a 7.5 second "off" period that resulted in one complete ON/OFF cycle every 15 seconds.The input current to each module was monitored and a failure was indicated by an appreciable currentdecrease resulting from an increase in module electrical resistance. The test was run until an arbitrary total of25,000 hours or approximately 6 million cycles was accrued. For these test conditions, the calculated MTBF

was 125,000 hours or 3x107 on off cycles.

CAUTION: Most conventional thermostats inherently have moderately large open/close temperaturedifferentials. This condition may effectively set up a thermal cycling situation where the temperature of the TEmodule is continuously varying between the upper and lower differential limits. Since thermal cycling is knownto reduce the life of thermoelectric modules, the use of traditional ON/OFF thermostatic temperature controlschemes is not recommended for high-reliability applications.

10.5 ENVIRONMENTAL CONSIDERATIONSThermoelectric modules often are installed in systems that are subject to significant shock, vibration, and/orother potentially detrimental environmental conditions. As mentioned earlier in this report, modules willwithstand moderate compression forces but shear strength is relatively low. However, when thermoelectricmodules are properly mounted within a mechanical subassembly, they will withstand substantial mechanicalstress without failure.

Ferrotec's modules have been subjected to a number of environmental/mechanical test conditions and havesuccessfully met those conditions without failure. Such tests include:

High Temperature Operationsand Storage:

150°C for 30,000+ hours

Low Temperature Operationsand Storage:

-40°C for 1000+ hours

Thermal Shock: (a) 100°C (15 sec)/100°C (15 sec), 10 cycles(b) 150°C (5 min)/-65°C (5 min)/ 150°C, 10 cycles(c) MIL-STD-202, Method 107

Range for ValueTEC™ Series modules: -55oC to +85oC

Range for SuperTEC™ Series modules: -65oC to 150oC

Mechanical Shock: (a) 100G, 200G 2 6msec; 500G, 1000G @1 sec 3-axis, three shockseach axis(b) MIL-STD-202, Method 213, Test Condition I

Vibration: (a) 10/55/10 Hz,1 minute cycle, 9.1G, 3-axis, 2-hours each axis(b) MIL-STD-202, Method 204A, Test Condition B, 15G Peak


19 de 20 06-06-2012 01:22

10.6 STANDARD QUALITY CONTROL PROCEDURESThermoelectric device manufacturers have independently developed quality control and test procedures toinsure that products meet published specifications and exhibit acceptable standards of workmanship. Whilefew formal standards (Military Specifications, etc.) exist within the industry, there have emerged certainminimum recognized criteria to which most major thermoelectric manufacturers adhere. However, if usershave particular concerns about quality-related issues that may affect their specific application, it generally isdesirable for users to discuss their concerns with individual thermoelectric manufacturers.

Ferrotec America's test and quality program has evolved from many years of industry experience covering anextensive range of thermoelectric cooling applications. General aspects of this program include 100%electrical and mechanical testing/inspection of products prior to shipment; in-process testing and screeningusing either 100% inspection or sampling inspection as per MIL-STD-105; and the use of statistical processcontrol techniques on various critical operations. The overall quality assurance program is structured inaccordance with MIL-Q-9858A.

10.7 CONCLUDING REMARKSIn the foregoing discussion we have emphasized the great dependence of thermoelectric module reliability onapplication conditions. By following some basic guidelines, and with knowledge of how certain factors tend toaffect module life, it should be possible for designers to optimize system reliability. While some may wish toperform a comprehensive analysis and model all relevant parameters, many users having unusualrequirements or nontraditional configurations often turn to an empirical approach for determining the reliabilityof their specific application.

SILRAM-COR BLOG | HEAT SINKS | ALPHA FAN ACCESSORIES | ALUMINUM HEAT SINKS | CALIENTE FAN

ACCESSORIES | COOLER FANS | COOLING FANS | CPU COOLERS | DC MOTORS | FANS ACCESSORIES | HEAT

PIPES | Thermal Pads | Thermoelectic Coolers | Thermoelectric Technical Reference | NMB FANS | What Are Heat

Pipes | FANDIS THERMOELECTRIC | Thermoelectric Units | HEAT SINKS

Developed & Powered By: Pluto-Group LTD.


20 de 20 06-06-2012 01:22

aplicação peltier

Documents

Transcript of aplicação peltier