Monthly Archives: March 2014

Electronic Maintenance Manuals

Traditionally, maintenance manuals have been printed and located away from the aircraft, causing mechanics time and effort to retrieve them. Maintenance manuals are increas­ingly being distributed electronically, and are often accessed via portable computers that the mechanic may bring onto the aircraft. Other means for making this data available on the aircraft (e. g., installation of computers containing this data on the aircraft) are expected to become more widely available. (See section on “Electronic Performance Support Tools’’ later in this article.)

Energy storage materials and architectures at the nanoscale

Nanotemplated materials have significant potential for applications in energy conversion and storage devices due to their unique physical properties. Nanostructured materials provide additional electrode surface area with short path lengths for electronic and ionic transport and thus the possibility of higher reaction rates. [112-114] Control of the active materials at the nanoscale is required for battery materials where solid-state ionic diffusion is a limiting factor in the electrode reactions. Template mediated fabrication is a potential method for ordered high density array fabrication of nanowire/nanotube on metallic current collector substrates. Early reports of nanotemplated materials focused on the use of polymer templates. [115-117] Subsequently, higher pore density anodic aluminium oxide (AAO) substrates (109 to 10u/cm2 by comparison with 108/cm2 for track etched polycarbonate) have been developed. [118] It is also possible to form pores with smaller diameter (to low 10’s of nm level) vertically aligned for more ordered channels and active materials. AAO processing on Si has been investigated [119-121] and is under development as a means to optimise the materials nanotemplating and integration of passive devices with silicon technology. [122,123] Other fabrication methods include vapour-liquid-solid (VLS) growth. [124] VLS growth combined with chemical vapour deposition (VLS-CVD) [125] and template mediated micelle deposition. [126] Optimising active materials and architectures not only requires assessment of the electrochemical prop­erties for the given application but also the electrical and mechanical characteristics during reaction.

1D nanowires or nanotubes have anisotropic morphologies and self-supported arrays grown directly on a current collector represent an attractive architecture for Li-ion batteries. Such arrays can ensure that each nanowire or nanotube participates in the electrochemical reaction with 1D electron transport pathways. Arrays of nanowires and nanotubes eliminate binders or additives that decrease power density by incorporating extra inactive materials in the total weight. Moreover, they can accomodate strain during charge and discharge. [127,128] Nanotubes function as electrolyte-filled channels for faster ionic and mass transport to the remote electrode surface. A variety of synthesis methods for the preparation of 1D nanowire or nanotubes for use as Li battery electrode materials have been investigated including sol-gel processing combined with template synthesis or hydrothermal treatment. The following is a brief review of recent progress in advanced anode and cathode materials for use in Li-ion batteries.


The balance equation can be used to implement numerous functions: modulation, demodulation, mean-square and root – mean-square extraction, power measurement, gain control; filterless demodulation, correlation (using the multiply-and – add feature), and many more nonlinear/linear operations, such as trigonometric-function synthesis, rms-dc conversion, and programmable filters. The availability of this Z interface is of great practical value, and is used in many of the applica­tion examples which follow (18), in most of which either the low-cost AD633 or the AD534 can be used.

Figure 18 shows the basic symbol for versatile multipliers of this class, and demonstrates how one may be connected for basic four-quadrant multiplication. With VW simply connected to VZ1, the balance equation (28) becomes

(Vx 1 – Vx 2 )(VY 1 – VY 2 ) — VU (VW – VZ2 ) (29)


Vw = ‘^rL+Vz 2 (30)


in this case, however, only a fraction of VW is returned to the Z interface, invoking familiar feedback techniques, which raises the output gain by the factor G = (R1 + R2)/R2, and


Figure 21. Vector rotation circuit.


thus the effective value of the denominator voltage is VUG. A voltage at the lower end of R2 adds directly to the output, with a scaling factor of unity.

The output may be taken in current-mode form from the VZ2 node by a resistor placed from VW to VZ2 since the product voltage VXVY/VU is forced across this resistor, whatever its value and for any load voltage (for both within a wide range); the output impedance is very high (megohms), being deter­mined by the common-mode rejection ratio of the Z interface. When this current is applied to a grounded capacitor, the time-integration function can be implemented. This is used in the rms-dc converter based on a difference-of-squares tech­nique (19), shown in Fig. 19. This scheme generates the prod­uct

V2- V*

" (Vin — Vout) —

the long-term average of which is forced to zero by the action of the loop including the integrator, which also serves to aver­age the signal at the multiplier output, VW. It follows that

Vout — ave(Vi2n)







Negative feedback



in other words, the rms value of Vin has been generated. Using an integration time constant CiRi of 16 ^s, the loop settles to within 1% of the final value after 123 ^s for a 10 V sine-wave input, and within 305 ^s for a 4 V input. With the capacitor omitted, the circuit provides the absolute-value function.

A further example of the utility of the summation feature is provided by the W-dimensional vector-summing circuit (Fig. 20), where this function is provided by daisy-chaining the multiplier outputs. The inputs V1 through VW may have either polarity, and the loop seeks to null the sum of the W indepen­dent difference-of-the-squares components. When the factor K is set to 1/VW, the overall scaling from the inputs is unity, that is

it is sometimes required to rotate a pair of signals repre­senting a two-dimensional vector, for example, in image pre­sentation systems. Figure 21 shows how just two multipliers may be used. The vector (u, v) is turned through an angle в controlled by the Ve input, to generate

u — u cos в + v sin в v’ — v cos в + u sin в


в ъ 2arctan(Ve/20)

Using AD534 or AD633 multipliers, the rotation scaling is 5.725 deg/V and is ±60° at Ve = ±11.55 V; the relative error is —2% at Ve = ±5 V (в = ±28°). The length of the vectors is unchanged. A related scheme for cathode-ray tube (CRT) ge­ometry and focus correction was described in Ref. 18.

in a more commonplace application, the Z interface is also used in the AM modulator shown in Fig. 22, where the carrier E sin wt is fed forward to the output (which responds as a voltage follower to the Z2 input) and thus added to the product of the modulation voltage VM and this carrier.

The Z interface is especially useful in analog division (Fig. 23), since it presents a high-impedance, fully differential in­put port. From Eq. (28), with Vx1 now being the output VW, we have

(VW – Vx2)(VY 1 – VY2) — VU (VZ1 – VZ2)


У —У У*1 I у

VW — VTT IT IT + v)




x 2

yw = vV2+V| + …+V£


The integration time, provided by the 10 кП resistor and the 100 pF capacitor, will normally be chosen to ensure stable loop operation, but when large enough to implement averag­ing, the circuit performs the root-mean-sum-of-squares oper­ation.

The use of normalized variables is often valuable in ana­lyzing and synthesizing multiplier applications. Thus:

x — Vx/VU; y — VY/VU; w — VW /VU ; z — VZ /VU

A high-impedance summing input is again available. To maintain negative feedback around the output amplifier, the denominator must be positive, but the numerator may have either polarity. The circuit therefore provides two-quadrant division, with the added advantage of differential signal pro­cessing available at both the numerator and denominator inputs.

The determination of the ratio of two voltages, both of which have indeterminate polarity, calls for a divider capable



Figure 25. Sine-function synthesis.

of accepting a bipolar denominator input. This is the four – quadrant division function. Commercial products that per­form this function are not in demand, but it is easily imple­mented using two of these general-purpose multipliers, as shown in Fig. 24. The behavior is benign through the singu­larity where the denominator is zero, though of course sig­nificant errors arise for small values. This circuit is unusual in being able to divide one sine wave by another. In a test in which a 400 Hz, 10 V amplitude sine wave was applied to both numerator and denominator inputs, the output was es­sentially constant at +10.03 V, with a ripple of only ±50 mV at the zero crossings.

These versatile multipliers allow the implementation of ar­cane and often complicated nonlinear functions. For example, the sine function can be approximated by

1.04680 – O.427802 1 – 0.26180

over the range 0 < в < пУ2. The theoretical error is ±0.4% of FS. While the synthesis requires considerable elaboration, Fig. 25 shows how simple the implementation becomes, using a single, low-cost multiplier. A better approximation, provid­ing a theoretical error of only ±0.01%, and <0.2% in practice, can be implemented with just two such multipliers and five resistors (18). Cosine synthesis needs only one multiplier and two resistors (18), with a peak error of ±2% at 22° and 73°, and the useful arctangent function may be approximated with a peak error of ±0.46° as shown in Fig. 26.

Many further examples could be provided. These general – purpose voltage-in, voltage-out translinear multipliers cover a very broad range of applications, and their structure is prac­ticable for operation from dc to at least 1 GHz. Contemporary wideband translinear multipliers, implemented using very fast complementary bipolar silicon-on-insulator (CB-SOI) pro­cesses and aided by advances in noise and distortion reduc­tion, calibrated to high accuracy using laser-wafer trimming and presented in tiny surface-mount packages, represent the epitome of the analog nonlinear art, while squarely meeting the pragmatic ongoing requirements for a wide variety of non­linear continuous-time signal-processing tasks, which go far beyond the basic task of multiplication.

All manner of special-purpose multiplier cells, of either transconductance or translinear types, have found their way into numerous IC systems: as gain-control elements, often im­plementing the AGC function; as power detectors and power controllers; in correlation applications; in analog-programma­ble filters; as modulators and demodulators, for example, syn­chronous detectors; and much else. Their simplicity, com­pleteness (no external components are usually required), very high speed combined with excellent accuracy, low supply volt­age requirements, and low power, coupled with low cost and very small size, ensure the continued use of these ubiquitous elements.



Operational Considerations in Diversity Systems

Operational Considerations in Diversity Systems

Figure 8. Rayleigh-fading statistics for 2-channel selection diversity assuming varying degrees of envelope correlation, p2. Fade distribu­tions are referenced to the SNR for a single (nondiversity) channel. (© 1994 IEEE.)

The performance gains estimated above for different types of diversity systems ignore several limitations that are con­fronted in practical applications. For example, impairments on the individual diversity channels may not be completely independent, and combining errors may introduce additional degradations in diversity system performance. Estimates of these degradations (1,2) are briefly illustrated here.

Mean SNR (dB)

Imperfect Channel Decorrelation. Prior analyses implicitly assumed independence of fading among the diversity chan­nels. In many environments, complete decorrelation is not achieved, and indeed is found to be unnecessary for successful diversity operation. General limits can immediately be placed on the behavior of the resulting statistical distributions: Com­plete independence of channel impairments yields results identical to analyses in the previous section, while complete correlation leads to Rayleigh fading statistics equivalent to a single channel without diversity. To investigate the effects of intermediate channel correlation, the complex correlation co­efficient, p, between the diversity signals must be taken into account, where p2 approximates the correlation function be­tween signal envelopes (7).

For selection diversity, analyzing more than two diversity channels is difficult, but for dual channels the probability dis­tribution is found to be (1)



P2(Ys) = 1 – e Ys/r[1 – Q(a, b) + Q(b, a)]

where Q(a, b) is expressed in terms of the zeroth-order modi­fied Bessel function, I0, as


Q(a, b) = I e~-a +x )/2/0(ax)xdx


in the control of the combiner. Degradations resulting from imperfect correlation between a pilot signal to control opera­tion of a maximal ratio combiner and the signals themselves are summarized here to indicate the magnitude of the antici­pated errors.

The output of an M-channel maximal ratio combiner that relies on a pilot for the reference control information is found to have the probability density (2)

with the parameters a and b given by



b =

a =

Г(1 + ІРІ2)





(yR > = I YRpM(yR )dYR = Г[1 + (M – 1)р2]





Figure 8 displays the cumulative distribution functions com­puted with these expressions (2). The curve for p2 = 0 corre­sponds to zero correlation, as in prior analyses, while for p2 = 1, no diversity advantage is conferred by switching between the two channels. However, substantial diversity gain is achieved even when the correlation between the two signal envelopes approaches 0.8, attesting to the efficacy of diversity operation for this environment. Similar results are obtained for other modes of diversity combining.

Switching and Combining Errors. No diversity switching or combining device is expected to operate perfectly, especially since randomly-fading signals supply much of the information used to control the switching or combining device. Errors in­troduced by imperfect operation degrade the performance of a diversity system. As already noted, a maximal ratio combiner must cophase and sum the diversity signals in proportion to the SNR in each channel, necessitating SNR estimates for each channel. In some systems, a continuous-wave pilot sig­nal is transmitted adjacent to the communication band to supply reference amplitude and phase information to assist

Г(1 – ІрІ2)


The first term after the summation is the binomial distribu­tion, and p is the correlation coefficient between the pilot and the adjacent signal channel. (Note that here correlation is de­sirable, unlike the case for envelope correlation.) The first mo­ment (mean) of yR is obtained by integrating with respect to the probability density in Eq. (24):

and is equivalent to the mean SNR. The probability distribu­tion obtained by integrating over the density function is

M-1 Im -1

1 – e-YR/Г ^ І І р2n (1 – р2)M-n-

n=0 n / k=0

which represents the statistics of the combiner output signal.


Pm(Yr ) = / pm (x) dx


Г(1 – р2)






Operational Considerations in Diversity Systems

Figure 9. Rayleigh-fading statistics for 4-channel maximal ratio di­versity combining with combiner errors (specified by p2) between ref­erence pilot and channel signal. Fade distributions are referenced to the SNR for a single (nondiversity) channel. (© 1994 IEEE.)

Mean SNR (dB)

If the correlation between the pilot and channel signal is perfect (p2 = 1), representing perfect operation of the diversity combiner, Eqs. (26) and (25) respectively give

^м(КЛ) = 1-^/ГЕ^^ {УЕ)=МГ (27)


which are equivalent to Eqs. (17) and (18), respectively. If there is no correlation between the pilot and channel signal (p2= 0), the resulting expressions are

Pm(Ул) = 1 – e-yR/г Ул) = Г (28)

showing that diversity operation provides no benefit for this case.

To illustrate the impact of such errors on the overall per­formance of a maximal ratio combiner, Figure 9 displays fad­ing statistics for 4-channel operation with varying degrees of correlation (2). The performance penalty imposed by com- biner-control errors is considerable, especially in the critical deep-fading portion of the distribution (for example, compare curves for p2 = 1.0 and p2 = 0.75). From Eq. (25), however, the degradation in mean received signal power is modest. If the correlation coefficient p2 decreases from 1 to 0.5, for in­stance, the mean SNR for the 4-channel combiner decreases from 4 Г to 2.5 Г, a loss of only 2.0 dB.

Other System Considerations. Even if diversity operation of­fers substantial improvement in overall circuit availability, other practical constraints may need to be taken into account. For example, site diversity significantly improves availability for earth-satellite paths subject to rain attenuation, as veri­fied in Fig. 3. However, in large earth stations that utilize wide bandwidths to serve many users, potential outages re­lated to switching among available diversity signals is a se­vere problem to be avoided. Therefore, the entire receive band for the diversity channel must be conditioned and synchro­nized with the main-station signal to support switching among channels with no loss of information.

However, on the uplink to the satellite, such synchroniza­tion is extremely difficult due to variations in radio path length (such as caused by satellite motion) between the earth stations and the satellite. Therefore, uplink site diversity is much less viable than downlink site diversity, except possibly in packet-switched applications where lost packets can be re­covered. A potential compromise solution for this case is to protect the downlink path with site diversity, but implement transmitter power control (27) to increase the availability of the uplink path.

Figure 10 shows a 14/11 GHz earth-space site diversity configuration (28), planned for the two sites represented in Fig. 1. In this system, the entire downlink receive band (500 MHz) from the secondary station is transported to the main station by the microwave Diversity Interconnect Link (DIL), buffered and synchronized with the main station receive sig­nal, and made available at the diversity switch. (Signal com­bining of the diversity signals is unlikely to be considered for this wideband application because of the difficulty in match­ing phase variations across the two 500 MHz receive bands.) Signal regeneration (demodulation and remodulation) is im­plemented in the DIL, not only to support frequency conver­sion, but also to preserve the quality of the transmissions. In this design, the uplink signal is also made available at both transmit sites, but this capability mainly increases the relia­bility of the overall system by enabling a redundant uplink signal transmission capability.

Coil Electrical Connections and Leads to the Outside World

Connections to the superconducting coil that come through the end ring should be mounted on copper bus bars that are electrically insulated from the end rings. These bus bars should be cooled in liquid helium in order to avoid heat from outside the coil being deposited directly into the supercon­ducting windings. Heat leaks down pulsed current leads, which are usually not gas cooled, can be particularly trouble­some. The cooling circuit used to cool bus bars at the ends of the coil should be part of the magnet helium cooling system. Since much of the cooling circuit is electrically grounded, in­line electrical insulators will be required in the cooling lines that cool the electrical bus bars connected to the supercon­ducting coil.

Most detector solenoids have gas-cooled electrical leads that are fed from a liquid helium pot located somewhere near the solenoid. The current buses between the lead pot and the coil are often cooled by conduction, a practice that has led to a number of failures. All current buses should be helium – cooled. The lead pot commonly used in detector magnets can be eliminated by using gas-cooled electrical leads that are attached to the ends of the coil structure. The helium used to cool these leads comes directly from the liquid helium cooling circuit. Gas-cooled leads attached to the end of the magnet are located within the cryostat vacuum, so these leads must be completely vacuum tight and they must withstand any in­crease in pressure that might occur in the cooling circuit dur­ing a quench (70). The bundled nested tube leads that were used on the PEP-4 experiment (71) and the g-2 solenoids (72) can be operated at any orientation within the cryostat vac­uum vessel. Properly designed gas-cooled leads are stable and they are capable of operating for more than 30 min without gas flow.