Site Loader
Rock Street, San Francisco

Abstraction: A This tutorial outlines the basic building, operation, and applications of parallel switches. On-resistance, two-dimensionality, charge-injection, and escape, each a performance-related specification, are defined. Features including ESD-protection, fault-protection, and force-sense capableness are explained. Application-specific picture, USB, HDMI, PCIe switches are outlined as good.

Besides see application note 4653, “ FAQs about parallel switches. ”

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Introduction:

First developed about 25 old ages ago, integrated parallel switches frequently form the interface between parallel signals and a digital accountant. This article provides an overview of the all-purpose and application specific parallel switches. The first subdivision presents the theoretical footing for parallel switches and describes common parametric quantities such as on-resistance, RON two-dimensionality, escape, charge-injection, and off-isolation.

In recent old ages, integrated parallel switches have offered better exchanging features, lower supply electromotive forces, smaller bundles, and integrated ESD protection. Specific application of parallel switches feature T-switches, Crosspoint switches, standardization multiplexers ( cal-MUXes ) , fault-protected switches, and force-sense switches. This application note discusses every bit good high-velocity USB, HDMI, and PCIe switches.

Because so many public presentation options and particular maps are available, the intelligent merchandise interior decorator can normally happen the right portion for a peculiar application.

Standard parallel Switch:

CMOS parallel switches are easy to utilize, so most interior decorators take them for granted. But one should non bury that these switches solve specific technology jobs. Conventional linear switches like the early ( 2 SPDT ) DG403 or the MAX383 are now offered by many semiconducting material makers ; their construction is shown in Figure 1a. Maxim besides offers devices such as the MAX14504, which offers better public presentation and smaller bundle size. The MAX14504 reduces board infinite over the older 2 SPDT configured devices ( MAX383 ) .

( a ) ( B )

Figure 1. The internal building of a typical parallel switch characteristics parallel n- and p-channel MOSFETs ( B ) and the gate-drive circuitry. ( B ) .

Basic parallel switch rules are linking an n-channel MOSFET in analogue with a p-channel MOSFET that allows signals to go through in either way with equal easiness. Whether the n- or the p-channel device carries more signal current depends on the ratio of input to end product electromotive force. Because the switch has no preferable way for current flow, it has no preferable input or end product. The two MOSFETs are switched on and off by internal inverting and non-inverting amplifiers.

These amplifiers level-shift the digital input signal as required, harmonizing to whether the signal is CMOS- or TTL-logic-compatible and whether the parallel supply electromotive force is individual or double. In CMOS-logic the control circuitry consists of far more constituents as logic “ NOT ” subdivision. It consists of a mention electromotive force phase, comparator and an inverting buffer ( Figure 1b ) as described in “ Analogschalter in der Praxis Teil 1 ” .

In some low electrical capacity applications merely n-channel MOSFETs with charge pump are used ( the MAX4887, for illustration ) as p-channel MOSFETs increase the parasitic electrical capacity of a switch. Charge pump switches allow input signal to transcend above/below the supply electromotive force. The MAX14504, for case integrates a charge pump to back up high output-signal scope ( A±VCC ) with negative signal capableness while extinguishing the demand for a negative supply. However, charge pump switches require extra power for extra circuitry.

Low-resistance switches:

Taking the p- and n-channel on-resistances ( RON ) in analogue ( merchandise over amount ) for each degree of VIN outputs a composite on-resistance feature for the parallel construction ( Figure 2 ) . This secret plan of RON versus VIN can be described as additive if you exclude the effects of temperature, power-supply electromotive force, and RON fluctuation with parallel input electromotive force. RON should be every bit low as possible in order to maintain signal loss and Propagation detain little, a RC clip changeless map of on-resistance and burden electrical capacity. By cut downing RON the Width/Length ( W/L ) ratio of a MOSFET must be increased, ensuing in higher parasitic electrical capacity and larger silicon country ; larger parasitic electrical capacity reduces the bandwidth. W and L are non the lone parametric quantities for RON, it is complex map of negatron and hole mobility ( Aµn and Aµp ) , oxide electrical capacity COX, threshold electromotive force VT, and VGS ( signal electromotive force VIN ) of N- and P-MOSFETs shown in Equation 1a/b. You should be cognizant that these effects represent disadvantages and that minimising them is frequently the primary intent of new merchandises. See Table 1.

Figure 2. The n-channel and p-channel on-resistances of Figure 1 signifier a low-valued composite on-resistance.

( Eq.1a )

( Eq.1b )

Table 1. Low-resistance switches

Part Number

Function

RDS ( ON ) ( I© , soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

RON Match ( I© , soap )

RON Flatness ( I© , soap )

tON/tOFF ( ns, soap )

Charge Injection ( personal computer, soap )

Supply Voltage Range ( V )

Pin-Package

MAX14535E

1 DPDT ; NO

0.35

10

0.05

0.001

90000/40000

+2.4 to +5.5

10-UTQFN

MAX4715/MAX4716

1 SPST ; NO/NC

0.4

1

0.09

18/12

20

+1.6 to +3.6

5-SC70

MAX4735

4 SPDT

0.4

100

0.03

0.75

200/180

100

+1.6 to +3.6

16-TQFN/TSSOP

MAX14504

2 SPDT ; Bi-Directonal

0.5

50

0.001

60000/3000

+2.3 to +5.5

12-WLP

MAX4626

1 SPST ; NO

0.5

2

0.1

50/30

40

+1.8 to +5.5

5-SOT

MAX4742

2 DPST ; NC

0.8

1

0.08

0.18

24/16

28

+1.6 to +3.6

8-AµDFN/AµMAX/SOT

MAX4754

4 DPDT

0.85

3

0.35

0.4

140/50

50

+1.8 to +5.5

16-TQFN/UCSP

MAX4758/MAX4759

4 DPDT/8 SPDT

0.85

5

0.35

0.45

140/50

40

+1.8 to +5.5

36-TQFN ; 32-UCSP/WLP

MAX4751/MAX4752

4 SPST ; NO/NC

0.9

2.5

0.12

0.1

30/25

21

+1.6 to +3.6

16-QFN/14-TSSOP

MAX4855

2 SPDT

1

2

0.12

0.275

60/40

8

+2 to +5.5

16-TQFN

MAX4783

3 SPDT

1

2

0.4

0.2

25/15

40

+1.6 to +3.6

16-QFN/TQFN/TSSOP

MAX4680/MAX4690/MAX4700

2 SPST ; NC/NO/NO-NC

1.25

0.5

0.3

0.3

275/175

550

A±4.5 to A±20

16-PDIP ( N ) /SOIC ( W ) /SSOP

MAX4677/MAX4678/MAX4679

4 SPST ; NC/NO/NO-NC

1.6

1

0.3

0.4

350/150

85

A±2.7 to A±5.5

16-PDIP ( N ) /TSSOP

MAX4688

1 SPDT

2.5

0.5

0.4

1

30/12

40

+1.8 to +5.5

6-UCSP

MAX4661/MAX4662/MAX4663

4 SPST ; NC/NO/NO-NC

2.5

0.5

0.5

0.5

275/175

300

A±4.5 to A±20

16-PDIP ( N ) /SOIC ( W ) /SSOP

MAX4667

2 SPST ; NC

2.5

0.5

0.4

0.4

275/175

450

A±4.5 to A±20

16-PDIP ( N ) /SOIC ( N )

MAX4706/MAX4707

1 SPST ; NC/NO

3

1

0.85

20/15

5

+1.8 to +5.5

6-AµDFN/SC70 ; 5-SC70

MAX4675/MAX4676

1 SPST ; NO/NC

3

1

0.7

300/110

87

A±2.7 to A±5.5

6-SOT

MAX4674

4 SPDT

4

0.5

0.4

0.8

18/66

10

+1.8 to +5.5

16-QSOP/SOIC ( N ) /TQFN/TSSOP

MAX4664/MAX4665/MAX4666

4 SPST ; NC//NO/NO-NC

4

0.5

0.5

0.5

275/175

300

A±4.5 to A±20

16-PDIP ( N ) /SOIC ( N )

MAX4739

4 SPST ; NO-NC

4.5

0.5

0.4

1.2

80/40

5

+1.8 to +5.5

14-TSSOP/UCSP

MAX4621/MAX4622/MAX4623

2 SPST/2 SPDT/2 DPST ; NO/-/NO

5

0.5

0.5

0.5

250/200

480

A±4.5 to A±20

16-PDIP ( N ) /SOIC ( N )

MAX4947/MAX4948

6 SPDT ; -/Bi-Directional

5.5

3

0.5

1

800/800

10

+1.8 to +5.5

24-TQFN/25-UCSP

MAX4729/MAX4730

1 SPDT

5.5

2

0.15/0.34

1.5/0.95

45/26

3

+1.8 to +5.5

6-AµDFN/SC70

MAX4670

8 SPDT ; NO-NC

9

0.01

0.15

0.18

400/200

8

+2.7 to +3.6

32-TQFN

MAX14756/MAX14757/MAX14758

4 SPST ; NC/NO/NO-NC

10

2.5

0.5

0.004

60000/3000

580

A±10 to A±35

16-TSSOP

The first parallel switches operated on A±20V supply electromotive forces and had several hundred ohms of RON. Recent merchandises ( the MAX4992, for illustration ) achieve 0.5I© maximal RON with a much lower supply electromotive force. Supply electromotive force and applied signal have a significant consequence on RON ( Figure 3a ) . The MAX4992 specifies signal and supply electromotive forces from 1.8V to 5.5V. As you can see, RON increases for lower supply electromotive forces. The soap RON is about 0.38I© at 1.8V, 0.3I© at 2.7V, 0.28I© at 3.3V and merely 0.25I© at 5V. Many new parallel switches ( the MAX4735, for illustration ) stipulate low-tension operation for supply electromotive forces down to 1.6V. The MAX4992 achieves really low on-resistance and RON two-dimensionality ( 1m I© ) with a single-supply. For 5V supplies, Figure 3b compares the Maxim switches with older types. Depending on the application Maxim offer many parallel switches that operate with single- or/and dual-supplies.

Figure 3a. Higher supply electromotive force causes lower on-resistance.

Figure 3b. At +5V supply electromotive force, later-generation parallel switches have lower on-resistance.

When choosing switches for single-supply systems, seek to take from those intended for single-supply usage. Such devices save one pin, because they do non necessitate separate V- and land pins. As a consequence, this economic system of pins enables a single-pole/double-throw ( SPDT ) switch to suit into a minuscule 6-pin SOT23 bundle. Similarly, low-tension dual-supply applications call for dual-supply switches. These switches require a V- pin in add-on to the land pin, and typically stipulate a logic interface with standard CMOS and TTL degrees. The SPST MAX4529, for illustration, is besides available in a 6-pin SOT23 bundle.

Many high-performance parallel systems still rely on higher-level bipolar supplies such as A±15V or A±12V. The interface to these electromotive forces requires an extra supply pin normally known as logic supply electromotive force ( shown in the MAX14756 data sheet ) . This pin ( VL ) connects to the system logic electromotive force, which is normally 5V or 3.3V. Having the input logic signals referenced to the existent logic degrees increases the noise border and prevents inordinate power dissipation.

Frequently misunderstood is the analog-switch construct refering to input logic degrees and their consequence on supply current. If the logic inputs are at land or VCC ( or VL when available ) , parallel switches have basically no supply current. Using TTL degrees to a 5V switch, nevertheless, can do the supply current to increase more than 1000 times. To avoid unneeded power ingestion, you should avoid TTL degrees, which are merely a bequest of the 1980s.

Signal Handling

Figure 3a besides shows the value of RON versus signal electromotive force. These curves fall within the specified scope of supply electromotive force, because parallel switches can merely manage analog-signal degrees between the supply electromotive forces. Under- or overvoltage inputs can for good damage a protected switch by bring forthing uncontrolled currents through internal rectifying tube webs. Normally, these rectifying tubes protect the switch against short-duration electrostatic discharge ( ESD ) every bit high as A±2kV.

RON for a typical CMOS parallel switch causes a additive decrease of signal electromotive force that is relative to current passing through the switch. This might non be a disadvantage for modest degrees of current or if the design accounts for RON effects. However, if you accept a certain degree of RON, so impart fiting and RON two-dimensionality can involvement you. Impart fiting describes the fluctuation of RON for the channels of one device ; RON two-dimensionality describes the fluctuation of RON versus signal scope for a individual channel. Typical values for these parametric quantities are below 0.1I© to 5I© , for really low RON switches smaller values of channel matching and RON two-dimensionality can be achieved. These parametric quantities for the MAX4992 are 3mI© and 1mI© , severally. The MAX14535E for illustration achieves typically 0.135I© RON and 0.3mI© RON two-dimensionality. This merchandise is ideal for portable devices in AC-coupled sound or picture, managing negative-rails down to -1.5V. The smaller the ratio of matching/RON or flatness/RON is, the more accurate the switch.

In most applications, you can avoid inordinate switch current by modifying the circuit design. To alter the addition of an op A by exchanging between different feedback oppositions, for illustration, take a constellation that places the switch in series with a high-impedance input ( Figure 4a ) . Because switch currents are undistinguished, you can disregard the value of RON and its temperature coefficient. Switch over current in the alternate design ( Figure 4b ) can be significant, because it depends on the end product electromotive force.

Figure 4a b. Gain-control circuits are good ( a ) or bad ( B ) depending on the sum of current through the switch.

A major public presentation demand in all audio systems is riddance of hearable chinks and dads. It does non count how good the audio public presentation of a device is. If it makes a noisy chink every clip the system turns on, the sensed quality of the merchandise is immediately degraded. Fortunately, there are solutions to take about every beginning of chink and dad jobs. Join us for a treatment about the definition of chink and dad, what causes it, and how to extinguish it.

Break-Before-Make

Turn-on and turn-off times ( short ton and toff ) for most parallel switches vary from below 15ns to every bit high as 1Aµs. For Maxim ‘s “ clickless ” sound switches ( Table 2 ) , tON and toff are in the msec scope to extinguish the hearable chinks otherwise present when exchanging audio signals. The MAX4992, for illustration characteristics a slow turn-on clip restricting clip-and-pop noise without excess constituents. Common beginnings of hearable chinks and dads are due to powering up/down audio beginnings and output-coupling capacitance. Another technique stamp downing chinks and dads is by utilizing internal shunt switches. The MAX4744, for case has internal shunt switches that discharges any unconnected end product terminuss to land. Extra Break-Before-Make characteristic ensures interrupting a connexion foremost before a new connexion is engaged. Maxim ‘s clickless switches use the combination of shunt switches and Break-Before-Make characteristic to avoid a measure DC electromotive force switched into the talker therefore cut downing chinks and dads.

The comparative magnitudes are besides of import: short ton & gt ; tOFF outputs break-before-make action, and toff & gt ; tON outputs make-before-break. This differentiation is critical for some applications as reference earlier. Figure 5a shows that you must take attention in exchanging between the two additions. One switch is usually closed in a typical make-before-break application. In altering addition you must avoid opening both switches at one time ; that is, the 2nd switch must shut before the first switch clears. Otherwise, the op amp applies open-loop addition and drives its end product to the tracks. The opposite constellation ( break-before-make ) is besides utile in exchanging among different input signals to a individual op A. To avoid short circuits between the input channels, a given connexion must be switched off before the following 1 is switched on.

Table 2. Clickless parallel switches

Part Number

Function

RDS ( ON ) ( I© , soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

RON Match ( I© , soap )

RON Flatness ( I© , soap )

tON/tOFF ( ns, soap )

Charge Injection ( personal computer, soap )

Supply Voltage Range ( V )

Pin-Package

MAX4992

2 SPDT ; Bi-Directional

0.5

100

0.003

0.001

150000/2000

+1.8 to +5.5

10-UTQFN

MAX4744/MAX4746H

2 SPDT

0.95

15

0.1

0.55

560/540

450

+1.8 to +5.5

10-AµDFN

MAX4910

4 SPDT

0.8

0.1

0.35

150/1000

300

+1.8 to +5.5

16-TQFN

MAX4764/MAX4765

2 SPDT

0.85

2

0.1

0.4

80/70

150

+1.8 to +5.5

10-TDFN-EP/UCSP

MAX4908/ MAX4930

2 SP3T

0.8

50

0.1

0.35

+1.8 to +5.5

14-TDFN-EP

MAX4901/MAX4902

2 SPST ; NO

1

6

0.25

0.5

100/100

125

+1.8 to +5.5

8-TDFN-EP ; 9-UCSP

MAX4571/MAX4573

11 SPST ; NO

35

0.2

3

6

8000/300

A

+2.7 to +5.25

28-QSOP/SOIC ( W ) /SSOP

MAX4572/MAX4574

2 SPST + 2 SPDT

35

0.2

3

6

8000/300

A

+2.7 to +5.25

28-QSOP/SOIC ( W ) /SSOP

MAX4562/MAX4563

2 SPST + 2 SPDT

20

1

5

5

12000/-

A

+2.7 to +5.5

16-QSOP

When a altering signal degree modulates the on-resistance, doing a fluctuation in the interpolation loss, parallel switches generate entire harmonic deformation ( THD ) . See a 100I© switch with 10I© RON two-dimensionality, for illustration. Loading this switch with a 600I© expiration produces 0.24 % of THD. THD can be critical in some application finding the quality or fidelity of a signal passing through a switch. Audio applications that require low RON two-dimensionality and THD specification, appropriate constituents choice and board layout design are critical undertaking. THD is defined as a ratio of a square root of all squared harmonic constituents divided by its cardinal harmonic constituent shown in Equation 2a. The maximal THD is calculated as shown in Equation 2b. The MAX4992, for case has a really low THD ( 0.004 % ) . Figure 5 shows some THD comparing for different switches.

( Eq.2a )

( Eq.2b )

FIGURE 5 MISSING

Charge-Injection Effectss

As mentioned above, low RON is non necessary in all applications. Lower RON requires greater bit country. The consequence is a greater input electrical capacity whose charge and discharge currents disperse more power in every shift rhythm. Based on the clip changeless T = RC, this bear downing clip depends on burden opposition ( R ) and electrical capacity ( C ) . It usually lasts a few 10s of nanoseconds, but low-RON switches have longer-duration on and off periods. High-RON switches are faster.

Maxim offers both types of switches, each with the same pinout in the same illumination SOT23 bundle. The MAX4501 and the MAX4502 stipulate higher on-resistance but shorter on/off times. The MAX4514 and the MAX4515 have lower on-resistance but longer exchanging times. Another negative effect of low on-resistance can be the higher charge injection caused by higher degrees of capacitive gate current. A certain sum of charge is added to or subtracted from the parallel channel with every on or off passage of the switch ( Figure 6a ) . For switches connected to high-impedance end products, this action can do important alterations in the expected end product signal. A little parasitic capacitance ( CL ) with no other burden adds a fluctuation of I”VOUT, so charge injection can be calculated as Q = I”VOUTCL.

A track/hold amplifier, which maintains a changeless parallel end product during transition by an A/D convertor, offers a good illustration of this ( Figure 6b ) . Closing S1 charges the little buffer capacitance ( C ) to the input electromotive force ( VS ) . The value of C is merely a few picofarads, and VS remains stored on C when S1 opens. The held electromotive force ( VH ) is applied to the buffer by shuting S2 at the beginning of a transition. The high-impedance buffer so maintains VH invariable over the ADC ‘s transition clip. For short acquisition times, the track/hold ‘s capacitance must be little and S1 ‘s on-resistance must be low. On the other manus, charge injection can do VH to alter by A±I”VOUT ( a few mVs ) , thereby impacting the truth of the undermentioned ADC.

Figure 6a. Charge injection from the switch-control signal causes a electromotive force mistake at the parallel end product.

Figure 6b. A typical track/hold map requires precise control of the parallel switches.

Escape

In add-on to RON mistake unwanted escape is another parametric quantity impacting the inactive ( ON-switch ) and dynamic ( Off-switch ) behavior. Figure 7a/b shows a simplified small-signal-model for the ON-/Off-state. In both exchanging provinces leakage current occurs on internal and external ( odd ) parasitic-diodes, increasing the mistake electromotive force. In On-state, if an input signal exceeds maximal supply electromotive force scope parasitic rectifying tubes will shoot current into the substrate and an increased current follows into the next channel. The ON-state mistake electromotive force is specified in Equation 3. In Off-state the mistake electromotive force is calculated by Vout=ILeakage A- RLOAD. A solution for escape is the n-channel “ Body-snatcher ” ( Q11, Q12 ) shown in Figure 1b. Q11 and Q12 FETs eliminate the escape. The Q11 FET ensures a changeless source-to-body electromotive force by linking its beginning to the Body of Q9, counterbalancing for the transition. The interior decorator should be cognizant of the Absolute Maximum Rating and must non transcend these bounds. Exceeding the maximal value can for good damage the device. The datasheet specifies leakage in worst instance scenario, when signal electromotive force approaches the supply electromotive force bounds. In add-on, the escape current is besides a map of temperature and doubles about every 10A°C.

Eq 3

Figure 7a.

Figure 7b.

Having reviewed these basicss, we now focus on advanced switches for particular applications.

T-Switches for Higher Frequencies

In Video signals trade-off between RON and parasitic electrical capacity is of import. Video switches with big RON demand excess gain-stages at the end product to counterbalance interpolation loss ( about 0.5dB ) , while low RON have big parasitic electrical capacity and reduces bandwidth therefore degrading video quality. Low RON switches require input buffers for continuing the bandwidth but increases constituents. The minimal and maximal picture frequences ( fmin, fmax ) are chiefly affected by big electrical capacity. The Bandwidth of a picture switch must run into the Nyquist ( 2 A- fmax ) demand, in pattern it is 3-6 times of fmax.

Using merely n-channel switches improves bandwidth as parasitic constituents and bundle size become smaller, leting more switches per unit country. However, N-channel switches suffer from a limited rail-to-rail operation up to a certain value of VINmax ( which is lower than VDD ) . When an applied picture signal exceeds VINmax the end product clinchs while falsifying the luminosity and chrominance constituents. When choosing a n-channel switch, guarantee that the signal scope is sufficient go throughing through the maximal input signal.

Video switches require multiplexing several picture signals. Often ( in security and surveillance system ) a signal proctor shows many picture beginnings, high off-isolation and XT are a cardinal parametric quantities for such application. In a turned off exchange the sum of feedthrough of an applied input signal find the Off-isolation. This parametric quantity increases about 20dB per decennary when operating in higher frequences. The T-switch topology is suited for picture and other frequences above 10MHz. It consists of two parallel switches in series, with a 3rd switch connected between land and their connection node. This agreement provides higher off-isolation than a individual switch. The capacitive XT for a T-switch turned off typically rises with frequence due to the parasitic electrical capacities in parallel with each of the series switches ( Figure 8a ) . The job in runing a high-frequency switch does non lie in turning it on, but in turning it off.

When the T-switch is turned on, S1 and S3 are closed and S2 is unfastened. In the off province, S1 and S3 are unfastened and S2 is closed. In that instance ( the off province ) the signal attempts to match through the off-capacitance of the series MOSFETs, but is shunted to land by S3. If you compare the off isolation at 10MHz for a picture T-switch ( MAX4545 ) and a standard parallel switch ( MAX312 ) , the consequence is dramatic: -80dB versus -36dB for the standard switch ( Figure 8b ) .

Maxim offer picture switches that are buffer or unbuffered. As mentioned criterion picture switches ( Passive picture switches ) may necessitate extra buffer/gain circuit ; the integrated attack ( Active picture switches ) combine switch and buffer in one bundle ( the MAX4310, for illustration ) . The integrated multiplexer-amplifiers have important off-isolation for most NTSC and PAL systems ( the MAX4310, for illustration ) .

( B )

Figure8. The T-switch constellation attenuates RF frequences that couple through the isolated electrical capacity between the beginning and the drain of an unfastened ( off ) switch.

Smaller Packages

Other advantages for CMOS parallel switches include little bundles, such as the bantam 6-bump UCSP, and no mechanical parts ( unlike reed relays ) . UCSP bundle reduces thermic dissipation by utilizing bumps soldering engineering than bundles with open tablets, but you should be cognizant that this bundle is automatically non every bit robust as other bundles. Mechanically more robust bundle is the SOT23. Maxim offers little low-voltage SPDT criterion switches ( for illustration the MAX4698 and MAX4688 ) . Both come in semen in 6-pin UCSP bundles and operate from supply electromotive force ranges in the 2V to 5.5V and 1.8V to 5.5V, severally. The MAX4698 and MAX4688 are the smallest SPDT parallel switches presently available with bantam bundle dimensions of 1.5mm2. See Table 3.

Table 3. Small bundles

Part Number

Function

RDS ( ON ) ( I© , soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

RON Flatness ( I© , soap )

tON/tOFF ( ns, soap )

Charge Injection ( personal computer, soap )

Off-Osolation ( dB soap & A ; typ ) /Freuqency ( MHz )

Supply Voltage Range ( V )

Pin-Package

Package Size ( mm2 )

MAX4698

1 SPDT

35

0.5

13

80/25

8

-75/0.1

+2 to +5.5

6-UCSP

1.5

MAX4688

1 SPDT

2.5

0.5

1

30/12

40

-90/0.1

+1.8 to +5.5

6-UCSP

1.5

MAX4594

1 SPST ; NO

10

0.5

1.5

35/40

5

-80/1

+2 to +5.5

6-AµDFN

1.6

MAX4706/MAX4707

1 SPST ; NC/NO

3

1

0.85

20/15

5

-82/1 ; -62/10

+1.8 to +5.5

6-AµDFN

1.6

MAX4729/MAX4730

1 SPDT

5.5

2

0.95

45/26

3

-67/1 ; -45/10

+1.8 to +5.5

6-AµDFN

1.6

MAX14508E/MAX14509AE/MAX14510E

1 DPDT ; Bi-Directional

5

10000

60000/5000

+2.7 to +5

10-UTQFN

2.5

MAX14535E/MAX14536E

1 DPDT ; NO

0.35

10

0.001

90000/40000

70/-

+2.4 to +5.5

10-UTQFN

2.5

MAX4992/MAX4993

2 SPDT/ 1 DPDT

0.5

100

150000/2000

-90/0.02

+1.8 to +5.5

10-UTQFN

2.5

MAX4719

2 SPDT

20

0.5

1.2

80/40

18

-80/1 ; -55/10

+1.8 to +5.5

10-UCSP

3.3

MAX14531E/MAX14532E

2 SP3T

2

2000

0.1

250000/6000

65/-

+2.7 to +5.5

12-WLP

3.3

MAX14504/MAX14505A

2 SPDT ; Bi-Directonal- NO

0.5

50

0.001

60000/3000

84/-

+2.3 to +5.5

12-WLP

3.3

MAX4906/MAX4906F

2 SPDT ; NO-NC

8

1000

1

60/30

6

-60/10 ; -26/500

+3 to +3.6

10-AµDFN

4.2

MAX4754

4 DPDT

0.85

3

0.4

140/50

50

-65/0.1

+1.8 to +5.5

16-UCSP

4.3

MAX4501/MAX4502

1 SPST ; NO/NC

250

1

75/10

10

-100/0.1

+2 to +12

5-SC70

5.3

MAX4624/MAX4625

1 SPDT

1

2

0.12

50/65

65

-57/1

+1.8 to +5.5

6-TSOT

8.3

MAX4514/MAX4515

1 SPST ; NO/NC

20

1

3

150/100

10

-90/0.1

+2 to +12

5-SOT

9

MAX14550E

2 SP3T

6.5

250

100000/5000

+2.8 to +5.5

10-TDFN-EP

9.6

MAX4908/MAX4930

2 SP3T

0.8

0.35

-80/0.02

+1.8 to +5.5

14-TDFN-EP

9.6

As mentioned earlier, Maxim offers many fluctuations of popular parallel switches like the DG411, including a household of 70V quad parallel switches ( MAX14756-MAX14758 ) . The MAX14756 is an betterment over the industry-standard 411, with higher analog-signal scope ( VSS to VDD ) and higher truth: channel matching to within 0.5I© maximal and channel two-dimensionality to 0.004I© typically. This household of parts offers three switch constellations, and their lower on-resistance ( & lt ; 10I© at A±20V ) suits industrial and battery direction applications. A bantam 16-pin TSSOP bundle solves the job of board infinite.

ESD-Protected Switchs

Change to newer parts?

As electrical overstress ( EOS ) and electrostatic discharge ( ESD ) can damage electronic circuits and constituents. Maxim offer many switches with A±15kV ESD protection bases on the IEC 6100-4-2 theoretical account ( Table 4 ) . All parallel inputs are ESD-tested utilizing the Human Body Model, every bit good as the Contact and Air-Gap Discharge methods specified in IEC 61000-4-2. The MAX4551/MAX4552/MAX4553 switches are pin-compatible with many standard quad-switch households such as the DG201/211 and the MAX391 types. To augment standard multiplexer households like the 74HC4051, MAX4581 and the MAX4927, Maxim besides released ESD-protected multiplexers. You no longer necessitate to utilize dearly-won TransZorbsA® to protect your parallel inputs.

Table 4. A±15k ESD per IEC 1000-4-2/ IEC 61000-4-2 criterion

Part Number

Function

Ron ( soap ) ( I© )

Icom ( off ) /ID ( off ) ( nA soap )

I”Ron ( soap ) ( I© )

Rflat ( on ) ( soap ) ( I© )

ton/toff ( ns soap )

Charge Injection ( personal computer typ )

Off-Isolation/Crosstalk ( dubnium )

Supply Voltage Range ( V )

MAX14535E/MAX14536E

1 DPDT ; NO

0.35

A±10

0.05

0.001

90000/40000

-70/-80 ( @ ? MHz )

+2.4 to +5.5

MAX4983E/MAX4984E

1 DPDT ; Bi-Directional

10

A±250

1

0.1

100000/5000

-48/-73 ( @ 10MHz )

+2.8 to +5.5

MAX4927

7 4:1 MUX ; NO

5.5

A±1000

1.5

0.01

50/50

-/-50 ( @ 25MHz )

+3 to +3.6

MAX4575/MAX4577

2 SPST ; NO/NO-NC

70

A±0.5

2

4

150/80

4

-75/-90 ( @ 1MHz )

+2 to +12

MAX4620

4 SPST ; NO

70

A±0.5

2

4

150/80

5

-75/-90 ( @ 1MHz )

+2 to +12

MAX4561

1 SPDT

70

A±0.5

2

4

150/80

17

75/- ( @ 1MHz )

+1.8 to +12

MAX4568/MAX4569

1SPST ; NO/NC

70

A±0.5

2

4

150/80

6

75/- ( @ 1MHz )

+1.8 to +12

MAX4558/MAX4559/MAX4560

1 8:1 MUX/ 2 4:2 MUX/3 SPDT

160

A±1

6

8

150/120

2.4

-96/-93 ( @ 0.1MHz )

A±2 to A±6 or +2 to +12

MAX4551/MAX4552/MAX4553

4 SPST ; NC/NO/NO-NC

120

A±1

4

8

110/90

2

-90/-90 ( @ 0.1MHz )

A±2 to A±6 or +2 to +12

Fault-Protected Switchs

As mentioned under “ Signal Handling ” above, the supply-voltage tracks for an parallel switch restrict the allowed scope for input signal electromotive force. Normally this limitation is non a job, but in some instances the supply electromotive force can be turned off with parallel signals still present. That status causes devices to latch-up and for good damage the switch, as can transients outside the normal scope of the power supply. Other causes of latch-up are improper supply electromotive force sequencing ( Vdd=Vss=0V ) and supply electromotive force transcending the absolute maximal evaluations. For improper sequencing the most positive electromotive force should be applied first followed by lower electromotive force and the most negative at last. Some switches do non necessitate power supply sequencing the multiplexer MAX14752, for illustration. This device can run with high-potential power supply ( 72V ) and internal rectifying tubes at the inputs protect the switch from over- and under-voltages. Additional mistake protection methods are using a series resistance at the input restricting the current flow into the rectifying tubes, and adding two excess rectifying tubes ( D1, D2 ) in series with the supplies. Disadvantage of D1 and D2 is restricting the input scope by the forward-bias electromotive force of the rectifying tubes ( Figure 9a ) . See application note “ Low-Voltage Fault Protection ” on more treatment. The MAX14752 is pin compatible with industry-standard DG408/DG409.

Figure 9a

Most Maxim ‘s fault-protected switches and multiplexers guarantee overvoltage protection of A±25V and power-down protection of A±40V, along with rail-to-rail signal handling and the low on-resistance of a normal switch ( Figure 9b ) . The input pin, furthermore, assumes a high electric resistance during mistake conditions irrespective of the switch province or burden opposition. Merely nanoamperes of escape current can flux from the beginning.

Figure 9b This internal construction shows the particular circuitry in a fault-protected parallel switch.

If the switch ( P2 or N2 ) is on, the COM end product is clamped to the supply by two internal ‘booster ‘ FETs. Thus, the COM end product remains within the supply tracks and delivers a upper limit of A±13mA depending on the burden, but without a important current at the NO/NC pin. The fault-protected switches, MAX4511/MAX4512/MAX4513, are pin-compatible with the DG411-DG413 and DG201/DG202/DG213 types ( Table 5 ) . Note that signals pass every bit good in either way through an ESD- and fault-protected switch, but these protections apply merely to the input side. Recent fault-protected Maxim switches have a flatter architecture, as these devices employ merely two parallel FETs ( the MAX4708, for illustration ) alternatively of the four FETs shown in Figure 9b. During a mistake status the COM end product of the MAX4708 is disconnected and becomes high electric resistance.

Table 5. Mistake Protection with rail-to-rail signal swings

Part Number

Function

RDS ( ON ) ( I© , soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

RON Match ( I© , soap )

Overvoltage Supplies ON/OFF ( V )

tON/tOFF ( ns, soap )

Charge Injection ( personal computer, soap )

Supply Voltage Range ( V )

Pin-Package

MAX9940

1 Line Protector

77.5

A

A±28

+2.2 to +5.5

5-SC70

MAX4505

1 Line Protector

100

A±0.5

A±36/A±40

+8 to +18 or A±9 to A±36

5-SOT ; 8-AµMAX

MAX4506

3Line Defender

100

A±0.5

A±36/A±40

+8 to +18 or A±9 to A±36

8-CDIP ( N ) /PDIP ( N ) /SOIC ( N )

MAX4507

8 Line Protector

100

A±0.5

A±36/A±40

+8 to +18 or A±9 to A±36

18-PDIP ( N ) /SOIC ( W ) ; 20-SSOP

MAX4708/MAX4709

1 8:1 MUX/2 4:1 MUX

400

A±0.5

15

A±25/A±40

275/200

0

+9 to +36 or A±4.5 to A±20

16-PDIP ( N ) /SOIC ( N )

MAX4534/MAX4535

1 2:1 MUX ; 2 4:1 MUX

400

A±2

10

A±25/A±40

275/200

10

+9 to +36 or A±4.5 to A±18

14-PDIP ( N ) /SOIC ( N ) /TSSOP

MAX4533

4 SPDT

175

A±0.5

6

A±25/A±40

250/150

1.5

+9 to +36 or A±4.5 to A±18

20-PDIP ( N ) /SOIC ( W ) /SSOP

MAX4508/MAX4509

1 8:1 MUX/2 4:1 MUX

400

A±0.5

15

A±25/A±40

275/200

10

+9 to +36 or A±4.5 to A±20

16-CDIP ( N ) /PDIP ( N ) /SOIC ( N )

MAX4632

2 SPDT

85

A±0.5

6

A±25/A±40

500/400

10

+9 to +36 or A±4.5 to A±18

16-PDIP ( N ) /SOIC ( N )

MAX4510/MAX4520

4 SPST ; NC/NO

75

A±0.5

A±36/A±40

500/175

5

+9 to +36 or A±4.5 to A±20

8-AµMAX ; 6-SOT

MAX4633

2 DPST ; NO

85

A±0.5

6

A±36/A±40

500/400

10

+9 to +36 or A±4.5 to A±18

16-PDIP ( N ) /SOIC ( N )

MAX4511/MAX4512/MAX4513

4 SPST ; NC/NC/NO-NC

160

A±0.5

6

A±36/A±40

500/400

5

+9 to +36 or A±4.5 to A±20

16-CDIP ( N ) /PDIP ( N ) /SOIC ( N )

MAX4711

4 SPST ; NC

25

A±0.5

1

A±7/A±12

125/80

25

+2.7 to +11 or A±2.7 to A±5.5

16-PDIP ( N ) /SOIC ( N ) /TSSOP

Force-Sense Switchs

Maxim offers a household of parallel switches with different switch types shacking in the same bundle. The MAX4554/MAX4555/MAX4556 devices, for case, are configured as force-sense switches for Kelvin feeling in machine-controlled trial equipment ( ATE ) . Each portion contains low-resistance high-current switches for coercing current and higher-resistance switches for feeling electromotive force or exchanging guard signals. On-resistance for the current switches is merely 6I© , and for the detection switches is 60I© at A±15V supply electromotive forces. The MAX4556 contains three SPDT switches with break-before-make action.

Typical force-sense applications are found in high-accuracy systems and in measurement systems that involve long distances ( Figure 10 ) . For 4-wire measurings, two wires force a electromotive force or current to the burden, and two other wires connected straight to the burden sense the burden electromotive force.

Figure 10. With the 4-wire technique, two wires force and two other wires sense the mensural electromotive force.

Alternatively, a 2-wire system senses load electromotive force at the terminals of the force wires opposite the burden. Load electromotive force is lower than the beginning electromotive force, because the forcing electromotive force or current causes a electromotive force bead along the wires. Longer distance between beginning and burden, larger burden current, and higher music director opposition all contribute for this debasement. The ensuing signal decrease can be overcome by utilizing a 4-wire technique in which the two extra voltage-sensing music directors carry negligible current.

Force-sense switches simplify many applications, such as exchanging between one beginning and two tonss in a 4-wire system. They are suited for usage in high-accuracy measuring systems, such as nanovoltmeters and femtoammeters, and for 8- or 12-wire force-and-sense measurings utilizing the guard wires of triax overseas telegrams. For more information, please see the MAX4554/MAX4555/MAX4556 data sheet.

Multiplexers

In add-on to switches, Maxim makes a figure of multiplexers ( MUXes ) . A MUX is a particular version of a switch in which two or more inputs are selectively connected to a individual end product. A MUX can be every bit simple as an SPDT switch or come in 2:1, 4:1, 8:1, 16:1 combinations for different channel of 1,2,4,7 and 8. The digital control for these higher order MUXes is similar to a binary decipherer with three digital inputs required to choose the appropriate channel.

A demultiplexer is fundamentally a MUX used backwards. That is, one input connects to two or more end products based on the decoded reference informations.

There are, eventually, cross-point switches that are employed in audio/video routing, picture on demand, security and surveillance systems. A cross-point switch is normally an M x N device, whereby any or all of M inputs may be connected to any or all of N end products ( and frailty versa ) .

For case, the buffered 32×16 crosspoint switch MAX4358 comes in a infinite salvaging 144-pin TQFP bundle, which is comparable with 512 T-switches. The MAX4358 is capable of implementing larger matrixes, an illustration of a 128×32 non-blocking matrix is shown in Figure 11a. The figure of ICs required are a map of figure of input channels, the figure of end product channels and whether the array is non-blocking or has exchanging restraints. The inputs of each MAX4358 in Bank 1 are parallel connected to Bank 2. The perpendicular end product of each Bank is a wired-OR constellation of all four MAX4358 devices. This wired-OR constellation is possible due to IC ‘s end product buffers that can be selected in handicapped or high-impedance end product province, while keeping low end product electrical capacity the inauspicious burden is minimized from handicapped end products.

This technique allows larger matrixes to be constructed by linking more devices ; nevertheless it requires linking many end products together. As a consequence, the end product node of each Bank sees two electric resistance tonss, the normal and handicapped electric resistance tonss. The handicapped electric resistance tonss of all the other end products has resistive and capacitive constituents. The MAX4358 end product buffer compensates for the resistive constituent but in some instances the electrical capacity additions ( & gt ; 30pF ) in larger matrixes, as the PC-board traces become longer. One solution is to cut down the figure wired-OR connexion of each end product node by increasing more crosspoint switches per Bank shown in Figure 11b. Another method is seting a little ( 5I©to 30I© ) resistance in series with the end product but drawback is a lowpass filter response with the parasitic electrical capacity. This is frequently a job in big systems as the cumulative consequence of many cascaded R-Cs signifier a roll-off at higher frequences doing “ softening ” of the image. Solution to this constrain is by routing the end product PC board hints in reiterating “ S ” constellation so that they exhibit some electric resistance. The Traces closest to each other exhibit common electric resistance that increases the entire induction. Series electric resistance increases the magnitude response at higher frequences, counterbalancing for the “ Softening ” consequence. The 2nd solution is adding a little inductance at the end product but optimal solution is a combination of both attacks.

Some critical crosspoint switch parametric quantities are cost, bundle size and XT between channels. Maxim offer a broad scope of crosspoint switches such as the 8×4 array MAX4360 a replacing merchandise for the MAX458 ; see Table 6 for some of the Maxim ‘s crosspoint switches.

( B )

Figure 1

Table 6. Crosspoint switches

Part Number

Function

RDS ( ON ) ( I© , soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

RON Match ( I© , soap )

RON Flatness ( I© , soap )

tON/tOFF ( ns, soap )

Off-Isolation ( dubnium )

Crosstalk ( dubnium )

-3dB Bandwidth ( MHz )

Supply Voltage Range ( V )

Pin-Package

Package Size ( mm2 )

MAX4989

2 2-of-4 Bidirectional switch

9

A±1

0.5

0.4

100000/6000

-43dB ( @ 10MHz )

-50dB ( @ 50MHz )

1000

+2.7 to +5.5

14-TDFN-EP

9.6

MAX4548/MAX4549

3 x 3:2

35

A±2

7

5

400/200

-72dB ( @ 10MHz ) /-85dB ( @ 20kHz )

-55dB ( @ 10MHz ) /-85dB ( @ 20kHz )

250

+2.7 to +5.5

36-SSOP

163.4

MAX4550/MAX4570

2 x 4:2

80

A±5

10

5

900/500

-78dB ( @ 4MHz )

-54dB ( @ 4MHz )

+2.7 to +5.5 or A±2.7 to A±5.5

28-SOIC ( W ) /SSOP

192.8

MAX9675

16×16

-110dB ( @ 6MHz )

-62dB ( @ 6MHz )

110

A±5

100-TQFP

262.4

MAX4355

16×16

-110dB ( @ 6MHz )

-62dB ( @ 6MHz )

110

+5 or A±3 or A±5

100-TQFP

262.4

MAX4357

32×16

-110dB ( @ 6MHz )

-62dB ( @ 6MHz )

110

+5 or A±3 or A±5

128-LQFP

359.6

MAX4356

16×16

-110dB ( @ 6MHz )

-62dB ( @ 6MHz )

110

+5 or A±3 or A±5

128-LQFP

359.6

MAX4358

32×16

-110dB ( @ 6MHz )

-62dB ( @ 6MHz )

110

+5 or A±3 or A±5

144-TQFP

492.8

MAX4359

4×4

-80 ( 5MHz )

-70 ( 5MHz )

35

A±5

24SOIC ( W ) /36-SSOP

163.4

MAX4360

8×4

-80 ( 5MHz )

-70 ( 5MHz )

35

A±5

36-SSOP

163.4

MAX4456

8×8

-80 ( 5MHz )

-70 ( 5MHz )

35

A±5

40-PDIP ( W ) /44-PLCC

311.5

Calibration Multiplexers

Calibration multiplexers ( cal-MUXes ) are used in preciseness ADCs and other self-monitoring systems. They combine different constituents in one bundle: parallel switches for bring forthing accurate electromotive force ratios from an input mention electromotive force ; internal preciseness resistor-dividers ; and a multiplexer for choosing between different inputs. Maxim introduced this combination of maps in a individual bundle.

Four of these devices ( MAX4539, MAX4540, MAX4578 and MAX4579 ) can equilibrate two major mistakes associated with an ADC system: beginning and addition mistake. Using the internal preciseness voltage-dividers, these devices step addition and offset in a few stairss, controlled trough the consecutive interface of a microcontroller. The mention ratios 15/4096 and 4081/4096 ( with regard to the external mention electromotive force ) are accurate to 15 spots. The ratios ( 5/8 ) ( V+ – V- ) and V+/2 are accurate to 8 spots.

The cal-MUX first applies one-half the supply electromotive force to verify that power is present. The system so measures zero beginning and addition mistake, and forms an equation to rectify the subsequent readings. Zero input electromotive force, for illustration, should bring forth a digital nothing end product. The cal-MUX calibrates for offset mistake by using a really little input electromotive force of 15/4096 referred to ( VEFHI – REFLO ) . For a 12-bit ADC with 4.096V mention, 15/4096 peers 15mV and besides 15 LSBs. The digital end product hence should be binary 000000001111. To mensurate countervail mistake, the microcontroller merely records the difference between binary 000000001111 and the ADC ‘s existent end product.

To mensurate addition mistake, the cal-MUX applies a electromotive force of 4081/4096 referred to ( VREFHI – VREFLO ) . The microcontroller so records the difference between binary 111111110000 and the ADC ‘s digital end product. Knowing the ADC ‘s beginning and addition mistake, the system package constructs standardization factors that adjust the subsequent end products to bring forth right readings. The cal-MUX so serves as a conventional multiplexer, but with the ability to recalibrate the system sporadically.

USB 2.0 Switches

A cosmopolitan consecutive coach ( USB ) is a high-velocity interface for hand-held devices to pass on with computing machines. Multiple USB devices can be connected to a computing machine, and parallel switches are used to route the USB signal to different devices. The USB 2.0 is a high-velocity signal that requires a high-bandwidth and low-capacitance parallel switch. The MAX14531E, for illustration provides low CON and RON and for highly high public presentation shift. Extra characteristics are ESD protection, negative signal capableness ( up to -2V ) and integrated Click-and-Pop switched shunt resistance. Some USB switches combine USB host informations and courser into one interface ( the MAX14550E, for illustration ) .

Maxim offers a good choice of USB 2.0-compliant switches ideal for USB 2.0 high-speed applications ( 480Mbs ) . Table 7 shows a few illustrations of USB 2.0 switches. Based on the success of Maxim ‘s 2.0 USB switches, the MAX14972 a 3.0 USB merchandise will be shortly available, leting informations rates up to 5GHz. Register in the Link to be notified when full information sheet is available.

Table 7. USB 2.0 switches

Part Number

Function

RDS ( ON ) ( I© , soap )

RON Match ( I© , soap )

RON Flatness ( I© , soap )

tON/tOFF ( ns, soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

Con / Coff ( pF, typ )

Charge Injection ( personal computer, soap )

BW ( MHz )

Supply Voltage Range ( V )

MAX14578E

2 SPST ; NO

+2.8 to +5.5

MAX14508E/MAX14509AE/HYPERLINK “ hypertext transfer protocol: //www.maxim-ic.com/datasheet/index.mvp/id/5886 ” MAX14510E

1 DPDT ; Bi-Directional

5

60000/5000

10000

8/8

950

+2.7 to +5

MAX14550E

2 SP3T

6.5

100000/5000

250

5.5/2

1000

+2.8 to +5.5

MAX14531E/MAX14532E

2 SP3T

2

0.1

250000/6000

2000

8/5

800

+2.7 to +5.5

MAX4999

8 8:1 MUX

12

0.8

10000/10000

1000

6/5

1200

+3 to +3.6

MAX4983E/MAX4984E

1 DPDT ; Bi-Directional

10

1

0.1

100000/5000

250

6.5/5.5

950

+2.8 to +5.5

MAX4906/MAX4906F

2 SPDT ; NO-NC

7

1.2

1

60/30

1000

6/2

5

1000

+3 to +3.6

MAX4907/MAX4907F

2 SPST ; NO

7

1.2

1

60/30

1000

4/2

5

1000

+3 to +3.6

MAX4906EF

2 SPDT ; NO-NC

5

0.8

0.5

1.4/35

1000

10/9

20

500

+3 to +3.6

MAX4899AE/MAX4899E

4:1 MUX/ 3:1 MUX

5

0.8

1.1

2800/3

1000

15/10.5

25

425

+2.7 to +3.6

HDMI Switches

A High-Definition Multimedia Interface ( HDMI ) is a high-velocity interface for all-digital audio/video signalling. In latest HDTVs, DVD participants and digital picture application HDMI criterion has replaced the VGA and component picture criterions.

Maxim offers good choice of HDMI switches for v1.3 and v.1.4 ( the MAX14886, for illustration ) . Table 8 shows a few illustrations of HDMI switches.

Table 8. HDMI switches

Part Number

Function

RDS ( ON ) ( I© , typ )

RON Match ( I© , typ )

RON Flatness ( I© , soap )

Off-Isolation ( dubnium )

Crosstalk ( dubnium )

BW ( MHz )

Supply Voltage Range ( V )

MAX14886

4 2:1 Switch ; NO-NC

0.28

5000

+3 to +3.6

MAX4814

1 2:4 Switch ; Bi-Directional

12

2.5

65 ( @ 1MHz )

75 ( @ 1MHz )

190

+4.5 to +5.5

MAX4929E

2 2:1 MUX ; NO-NC

10

2

13

70 ( @ 1MHz )

75 ( @ 1MHz )

40

+5 or A±5

MAX4886

4 2:1 Switch ; NO-NC

8

0.28

0.6

58 ( @ 50MHz )

-49 ( @ 50MHz )

2600

+3 to +3.6

High-Voltage Switchs

In ultrasound applications, high-voltage pulsations ( A±100V ) are applied to transducers to bring forth supersonic moving ridges. Analog switches are required for routing these high-potential signals between the transducers and the chief systems, so the switches must be able to manage high-potential signals.

Maxim offers a good choice of high-voltage parallel switches that are ideal for ultrasound medical applications. Table 9 shows a few illustrations.

Table 9. High-voltage switches

Part Number

Function

Single Vsupply ( min, V )

Single Vsupply ( soap, V )

Double Vsupply ( min, A±V )

Double Vsupply ( soap, A±V )

BW ( MHz )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

tON/tOFF ( ns, soap )

Con / Coff ( pF, typ )

MAX14802/MAX14803/MAX14803A

16 SPST ; NO

200

40

160

50

2000

5000/5000

36/11

MAX4800A/MAX4800B

8 SPST ; NO

40

200

40

100

20

2000

5000/5000

36/11

MAX4802A

8 SPST ; NO

40

200

40

100

50

2000

5000/5000

36/11

DisplayPort/PCIe switches

A Peripheral Component Interconnect Express ( PCIe ) is a point-to-point Input/Output interconnect coach criterion for higher information rates ( 10Gbps ) over the original PCI. The PCIe backplane coach is going popular for many system designs. PCIe switches are used in real-time Hot-Plug systems, in laptop enlargement card interfaces, Scaled Link Interface ( SLI ) and Crossfire. These switches facilitate high isolation and point-to-point connexion therefore cut downing system mistakes.

Maxim offers switches that support PCIe Gen I, Gen II and Gen III information rates utilizing CMOS engineering. The MAX4928A/MAX4928B are two hex double-pole/double-throw ( 6 DPDT ) parallel switches designed to manage PCIe/DisplayPort exchanging Table 10 shows more choices of PCIe switches.

Table 10. PCIe switches

Part Number

Function

RDS ( ON ) ( I© , soap )

ICOM ( OFF ) /ID ( OFF ) ( nA, soap )

RON Match ( I© , soap )

tON/tOFF ( ns, soap )

Off. Iso. ( dubnium )

Crosstalk ( dubnium )

BW ( MHz )

Supply Voltage Range ( V )

Pin-Package

Package Size ( mm2 )

MAX4888B/MAX4888C

2 1:2 MUX ; Bi -Directional

8

1000

0.5

12

30

8000

+3 to +3.6

28-TQFN

20.2

MAX4889B

1:2 Switch ; Bi -Directional

8.4

1000

0.5

12

30

5000

+3 to +3.6

42-TQFN

32.8

MAX4928A/MAX4928B

6 1:2 Switch ; Bi-Directional

8

2

120/50

22

40

10000

+3 to +3.6

56-TQFN

56.6

MAX4888A/MAX4889A

4 SPDT/8 SPDT ; Bi-Directional

8.4

1000

1

250/50

12

-56

5000

+1.6 to +3.6

28-TQFN

20.2

MAX4888/MAX4889

4 SPDT/8 SPDT ; NO-NC

7

1

250/50

26

32

1250

+1.6 to +3.6

28-TQFN

20.2

AµMAX is a registered hallmark of Maxim Integrated Products, Inc.

TRANSZORB is a registered hallmark of Vishay General Semiconductor, LLC.

UCSP is a hallmark of Maxim Integrated Products, Inc.

HDMI is a registered hallmark and registered service grade of HDMI Licensing LLC.

PCI Express is a registered service grade of PCI-SIG Corporation.

PCIe is a registered service grade of PCI-SIG Corporation.

SLI is a registered hallmark of NVIDIA.

Crossfire is a registered hallmark of AMD Graphics Product Group.

Post Author: admin

x

Hi!
I'm Gloria!

Would you like to get a custom essay? How about receiving a customized one?

Check it out