Back to the basics: Using current-feedback op amps for high-speed designs - Embedded.com

Back to the basics: Using current-feedback op amps for high-speed designs

The current-feedback op-amp architecture has emerged to become adominant solution for many applications. Possessing a number ofstrengths, this amplifier architecture can be used in nearly anyapplication that calls for an op amp.Current-feedback amplifiers donot have a fundamental gain bandwidth product limitation.

This is often shown by a very small loss in bandwidth as signalamplitudes increase. Since large signals can be accommodated withminimal distortion, these amplifiers have good linearity at very highfrequencies.

While voltage-feedbackamplifiers lose bandwidth with increasing gain,current-feedbackamplifiers keep most of their bandwidth over awide range of gains. Thus, it's accurate to say that current feedbackamplifiers have no gain bandwidth product limitation.

Of course, current-feedback amplifiers are not infinitely fasteither. The slew rate islimited not by the internal bias currents, but by the speed limitationsof the transistors themselves. This allows larger slew rates for agiven bias current without usingpositivefeedback or other slew-boosting techniques that canoften hamper stability.

So how do we build one ofthesewonder circuits?
Current feedback op amps have an inputbuffer as opposed to a differentialpai r . The input buffer is most often an emitter follower orsomething very similar. The non-inverting input is high impedance,while the buffer's output is low impedance. In contrast, both inputs ofa voltage-feedback amplifier are high impedance.

The output of a current-feedback op amp is a voltage; it is relatedto the current that flows out of or into the inverting input of the opamp by a complex function called transimpedance,Z(s), which is a very high number at DC. Like avoltage-feedback op amp, it has a single pole roll off with increasingfrequency (Figure 1, below ).

Figure1. Voltage is related to the current that flows out of or into theinverting input of the op amp by a complex function calledtransimpedance Z(s).

The current-feedback op amp's key flexibility is adjustablebandwidth and stability. Because the feedback resistor value actuallychanges the AC loop dynamicsof the amplifier, it can impact both the bandwidth and stability.

Coupled with very high slew rates and adjustable bandwidth based onthe feedback resistor, you can get large signal bandwidths that arevery close to the small signal bandwidth of the device. Moreover, thisbandwidth is largely preserved over a wide range of gains. And becauseof the inherent linearity, you can get low distortion with largesignals at high frequencies too.

Finding the best RF
Having the AC characteristics of the amplifier partially dependent onthe feedback resistor enables us to tailor the amplifier for eachunique application. Lowering the value of the feedback resistor booststhe loop gain.

At low gains, the feedback resistor is set to a higher value topreserve stability and maximum bandwidth. As gain goes up, the loopgain is naturally reduced. This loop gain can be partially recovered byusing a smaller feedback resistor when a high gain is required. Figure 2 below illustrates whathappens to the bandwidth as you change the feedback resistor.

Figure2. The RF = 300 curve has excellent flatness and gain, and still goodbandwidth comparable to the peaked frequency response.

At the far right-hand curve where the RF = 147 ohm, the frequencyresponse has peaked quite substantially. This curve also has thehighest bandwidth. Decreasing the resistor too far below 147 ohmsresults in ringing on your pulse response, and it will actuallyoscillate.

The RF = 300 curve has excellent flatness and gain, and still goodbandwidth comparable to the peaked frequency response. Thus, we havegained a great deal of stability without giving up much of thebandwidth.

With a feedback resistor of 600 ohms, you can scale back yourfrequency response. If an application only needs 50MHz or 60MHz ofbandwidth and anything over that contributes to noise, you can tailorthe frequency response of your part with the feedback resistor. Thereason to use a fast amplifier with such a limited bandwidth is theexcellent signal fidelity it provides.

Figure 3 below is from thedatasheet of the same device and shows the suggested feedback resistorfor a given non-inverting gain.

Figure3. The recommendation for a gain of two is the 300 ohm resistor, withthe best combination of gain flatness, settling time and speed.

As might be expected, the recommendation for a gain of two is the300 ohm resistor, with the best combination of gain flatness, settlingtime and speed. Moreover, for a gain of one, you need a 600 ohmfeedback resistor to get optimum performance. This is because the loopgain is very high, and a bigger resistor value is required forstability.

This is a key difference from the voltage-feedback architecture. Acurrent-feedback amplifier cannot be used with the output shorted tothe inverting input.

The most common resistor specified on datasheets is for the gain oftwo. However, there is a good deal of flexibility in the actual valueyou could end up using (Figure 2 above ).The value recommended in a datasheet is the value used to generate thespecifications published in the performance tables and curves. For again of five, RF is down to 200 ohms (Figure3 above ).

The gain set resistor is only 50 ohm now, so we are getting to thepoint where the input buffer resistance and gain set resistance are ofsimilar values. This lowers the closed loop transimpedance of the opamp and will begin to limit the bandwidth as gain increases.

At a gain of eight, we are back up to a 275 ohms feedback resistor.Once the feedback resistor cannot be decreased for higher gain, thebandwidth is compromised, and the amplifier begins to behave like avoltage-feedback amplifier.

Board layout
One of the things to carefully consider with current-feedback op amps(or high-speed parts, in general) is the board layout. Thesurface-mount, ceramic supply bypassing capacitors need to be veryclose to the device, typically less than 3 millimeters.

If needed, the larger electrolytic capacitor can be placed fartheron the board. Often, there is an on-board voltage regulator. In thiscase, there is no need for additional electrolytic capacitors beyondthose recommended by the voltage-regulator vendor.

The small ceramic bypass capacitors that are placed near theamplifier provide energy for the high-frequency response of theamplifier. Depending on the amplifier's speed and the signal beingamplified, it may be desirable to use two ceramic capacitors withvalues about a decade apart. For example, a 400MHz amplifier could use0.01 microfarad and 1 nanofarad capacitors installed in parallel.

When purchasing capacitors, it is important to check theself-resonant frequency. The capacitors provide no benefit atfrequencies near or above this frequency.

Ground and power planes help provide low impedance paths for boththe ground and the power currents. Both the ground and power planesshould be removed from underneath the I/O pins of the amplifier and thefeedback resistor. This helps maintain amplifier stability by reducingundesired parasitic capacitance.

Always try to use surface mount components where possible. Theseprovide the best performance and take up minimal board space as well.PCB traces should be kept as short as possible and should be sized tominimize parasitic effects.

On power-supply traces, the worst parasitic characteristic is DCresistance and inductance. Thus, power traces should be as wide aspossible. On the other hand, I/O traces often carry very littlecurrent, so capacitive parasitics are the most harmful. For signalpaths of over 1 centimeter, it is best to use controlled impedance anddoubly terminated transmission lines.

Since small amounts of parasitic loading are unavoidable, thefeedback resistor of a current- feedback amplifier provides flexibilityfor tuning amplifier performance for the specific application.

Figure4. By including a resistor (Rout), nearly any amount of capacitance onthe output of an amplifier can be driven without stability problems.

With a really challenging board layout, though, even a very largefeedback resistor may not be sufficient. In this case, another optionis available. By including a resistor (Rout), nearly any amount ofcapacitance on the output of an amplifier can be driven withoutstability problems (Figure 4, above ).

This is a common technique with op amps that work on both voltage-and current-feedback amplifiers. This technique is especially usefulwhen driving a high-speed ADC. The Rout resistor is placed between theop amp and the capacitive load (the ADC). The resistor should be asclose to the amplifier as board space allows. The curve in Figure 5, below shows the suggestedresistor values of Rout based on the size of the capacitor.

Figure5. If RL is smaller in value, Rout can be smaller too.

It is based on a 1k-ohm resistive load. If RL is smaller in value,Rout can be smaller too. Another option is to put the Rout inside thefeedback loop. Instead of connecting RF between Rout and the amplifier,you can connect RF to the output side of the isolation resistor.

This will preserve gain accuracy, but you will still lose the sameamount of voltage swing in the isolation resistor as in the otherexample. It should be realized, however, that this technique has adrawback. Because the resistor and capacitor form a low pass filter,there is some bandwidth penalty for using this circuit.

Figure 6 below shows the response for the LMH6738. Regardless of theresistorvalue, higher capacitance is harder to drive and reduces bandwidthaccordingly.

Figure6. Regardless of the resistor value, higher capacitance is harder todrive and reduces bandwidth accordingly.

Reducing system noise
Minimizing noise is particularly important if you are building an IFamplifier or a low-frequency RF amplifier. Withcurrent-feedbackamplifiers (although it may seem counter-intuitive), increasing thefeedback resistor can often reduce system noise. This is because thefrequency response falls off faster than the resistor noise goes up.

To reduce noise to the portions of the circuit following theamplifier, it is very important to have only the bandwidth necessaryand no more. Besides using the best value of feedback resistor, you canadd additional filtering to the circuit.

The filter can often be incorporated right into the amplifierfeedback network using a Sallen-Keyfilter topology. If possible, AC coupling will help removelow-frequency noise that is often called 1/F noise. The goal is to cutout all the noise except in the band that you are amplifying.System-level considerations call for placing the lowest noise andhighest gain blocks early in the circuit.

The earlier you add gain, the less noise impacts your signal lateron. If possible, avoid large source resistances. Resistances addthermal noise proportional to the value of the resistor.

When you compare a current feedback op amp with a voltage feedbackop amp, there are a few areas where the voltage-feedback amplifier mayoffer an advantage.

With the current-feedback topology, the input bias currents are notsystematically matched. The non-inverting input—being a higherimpedance input than the inverting input—typically has a lower inputbias current.

The inverting input bias current will usually be larger, which canlead to input voltage offsets if the bias currents have to flow throughlarge valued resistors.

The offset voltage on a current feedback part can be matched andmade quite small, but they are not systematically zero. Thus, while thetypical offset voltage of a current-feedback op amp can be made verygood, it will tend to vary more with normal process lot variations andwith temperature variations. If very high precision is required for theinput offset voltage, a voltage-feedback amplifier is usually a betterchoice.

The buffer configuration of a current-feedback amplifier needs afeedback resistor, while a voltage- feedback amplifier can use a directshort. This is not normally a problem unless replacing an existingvoltage-feedback amplifier in an existing design.

And finally, capacitance in the feedback loop of a current-feedbackamplifier creates instability. Some common circuit topologies will notaccommodate a current feedback amplifier. For most of these circuits,there are substitute layouts that will work for a current feedbackamplifier.

Applications that benefit from the unique qualities of a current-feedback amplifier include: presentation-quality video line drivers androuters, ADC drivers, IF amplifiers and clock buffers. Current-feedbackamplifiers shine anywhere signal fidelity and high speed are theprimary goals.

Loren Siebert is an ApplicationsEngineer at National SemiconductorCorp.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.