Design Article
Op amps in small-signal audio design - Part 1: Op amp history, properties
Douglas Self
5/25/2011 12:51 PM EDT
There is no point in regurgitating manufacturers' data sheets, especially since they are readily available on the internet. Here I have simply ranked the op-amps most commonly used for audio in order of voltage noise (Table 4.1).
The great divide is between JFET input op-amps and BJT input op-amps. The JFET op-amps have more voltage noise but less current noise than bipolar input op-amps, the TL072 being particularly noisy. If you want the lowest voltage noise, it has to be a bipolar input. The difference, however, between a modern JFET-input op-amp such as the OPA2134 and the old faithful 5532 is only 4 dB, but the JFET part is a good deal more costly.
The bipolar AD797 seems to be out on its own here, but it is a specialized and expensive part. The LT1028 is not suitable for audio use for reasons described later. The LM741, which is included in this chapter for purely historical reasons, is omitted from Table 4.1 because there are no noise specs on its data sheets.
Op-amps with bias-cancellation circuitry are normally unsuitable for audio use due to the extra noise this creates. The amount depends on circuit impedances, and is not taken into account in Table 4.1. The general noise behavior of op-amps in circuits is dealt with in Chapter 1.
Op-Amp Properties: Slew Rate
Slew rates vary more than most parameters; a range of 100:1 is shown in Table 4.2. The slowest is the 741, which is the only type not fast enough to give full output over the audio band. There are faster ways to handle a signal, such as current-feedback architectures, but they usually fall down on linearity. In any case, a maximum slew rate greatly in excess of what is required appears to confer no benefits whatever.
The 5532 slew rate is typically ±9 V/µs. This version is internally compensated for unity-gain stability, not least because there are no spare pins for compensation when you put two op-amps in an eight-pin dual package. The single-amp version, the 5534, can afford a couple of compensation pins, and so is made to be stable only for gains of 3× or more. The basic slew rate is therefore higher at ±13 V/µs.
Compared with power-amplifier specs, which often quote 100 V/µs or more, these speeds may appear rather sluggish. In fact they are not; even ±9 V/µs is more than fast enough.
Assume you are running your op-amp from±18V rails, and that it can give a±17V swing on its output. For most op-amps this is distinctly optimistic, but never mind. To produce a full-amplitude 20 kHz sine wave you only need 2.1 V/µs, so even in the worst case there is a safety margin of at least four times. Such signals do not of course occur in actual use, as opposed to testing. More information on slew limiting is given in the section on op-amp distortion.
Op-Amp Properties: Common-Mode Range
This is simply the range over which the inputs can be expected to work as proper differential inputs. It usually covers most of the range between the rail voltages, with one notable exception. The data sheet for the TL072 shows a common-mode (CM) range that looks a bit curtailed at -12 V. This bland figure hides the deadly trap this IC contains for the unwary.
Most op-amps, when they hit their CM limits, simply show some sort of clipping. The TL072, however, when it hits its negative limit, promptly inverts its phase, so your circuit either latches up, or shows nightmare clipping behavior with the output bouncing between the two supply rails. The positive CM limit is, in contrast, trouble-free. This behavior can be especially troublesome when TL072s are used in high-pass Sallen-and-Key filters.
Op-Amp Properties: Input Offset Voltage
A perfect op-amp would have its output at 0 V when the two inputs were exactly at the same voltage. Real op-amps are not perfect and a small voltage difference – usually a few millivolts – is required to zero the output. These voltages are large enough to cause switches to click and pots to rustle, and DC blocking is often required to keep them in their place. This issue is examined in depth in Chapter 11.
The typical offset voltage for the 5532A is ±0.5 mV typical, ±4 mV maximum at 25°C; the 5534A has the same typical spec but a lower maximum at ±2 mV. The input offset voltage of the new LM4562 is only ±0.1 mV typical, ±4 mV maximum at 25°C.
|
|
|
|
|
Bert22306
5/25/2011 6:50 PM EDT
Very interesting article. Many years ago, I designed a simple audio preamp for use in my home stereo setup. I used it for many years, and finally gave it up to get something with remote control. But it served me flawlessly and sounded great. Based on the TL082 dual JFET input opamp, similar to the TL072 mentioned in the article.
Two very important lessons I learned.
The first one was, beware of capacitance in the load! The thing started hissing loudly after being powered up for a few seconds, and would not work. Turns out, the capacitance of the shielded output cable was enough to put the opamp into oscillation (it shifts the phase of the feedback signal). All it took was a few hundred ohms at the output pins, in series with each output cable, to stabilize the circuit. I used 470 ohms per output. Probably 220 ohms would have been enough.
The second was, read up on thermal noise! The combined resistance to the + and - input pins creates thermal noise. For example, the feedback resistor and ground resistor connected to the - inpout pin. Keep that parallel combination low.
Thermal noise manifests itself as a constant background hiss. The idea is, if possible, design the circuit so the thermal noise caused by the various input resistors is less than the noise inherent in the opamp itself.
Sign in to Reply
agk
5/26/2011 4:26 AM EDT
By reading this artiicle i call back year 1975 my design of a low noise pre amplifier for my casette player's head with NAB equalization. I tried many practically found producing a hiss noise in the gaps of silence or the fading music. I tried fnally LM381 dual low noise amplifier by NS in the single ended input configuration which gave better results. Also i designed a 2 transistor preamp with equalization given me still better results.
Sign in to Reply
David Ashton
5/26/2011 8:22 AM EDT
Excellent and very interesting article. But could you clarify if the 5532 is a true dual 5534 - the characteristics in both the tables given (Slew rate / Noise) are different?? Thanks.
Sign in to Reply
bcarso
5/26/2011 12:09 PM EDT
Bert, the ability of the TL series to source/sink current is limited, hence the strategy to lower the impedance of the feedback network to reduce noise is similarly limited (however, bad news/good news, the intrinsic voltage noise of the amp is high enough that you can indeed make the feedback contribution fairly small by comparison).
David, Douglas mentions the distinctions between the 5534 single and the 5532 dual in his book in a later chapter. They are indeed different, albeit fairly close. The 5534 is slightly superior and also allows input offset trimming.
As testimony to the longevity of the 5534, I just noticed a phono preamp in the latest Stereophile (June 2011) that uses one per channel as the sole active signal chain components! And they still want 1300US :)
Sign in to Reply
Rick_Hille
5/26/2011 1:30 PM EDT
I recall using LM1458's and LM358's extensively in telecom circuits (tone generators, filters, voice coupling, etc.). Not Hi-Fi, but low enough noise and distortion to meet telecom requirements of the time. We encountered significant performance variations between different manufacturer's devices of the same part number, as well as some date code dependencies. One trick that still sticks in my mind was having to add a "forced class A" modification (output pulldown or pullup resistor) in small signal stages to tame output crossover distortion from certain makers.
Sign in to Reply
bcarso
5/26/2011 7:58 PM EDT
I have a tube of "GL358s" which are actually not unity-gain stable! A very good technician kept reporting that a very-low-frequency twin-t notch filter (with feedback to greatly increase the Q) was oscillating, and I assured him that he must have wired something up wrong! Nope!
The plain 358 (or as a quad, the 324) almost always has severe crossover distortion, except at very low output currents, and some companies actually treated that strategy of forced class A as a trade secret!
For years I never saw a voltage noise spec, but recently noticed that it is now listed as 40nV/rtHz.
Sign in to Reply
rpell2
5/28/2011 11:19 AM EDT
I first saw this technique ("forced Class A") used in Walt Jung's "POOGE" modifications for the Philips DAC960 DAC unit in the early 90s. It has since become de rigueur in the DIY/audiophile community to "bias op amps into Class A", where it seems any op amp in a circuit is an excuse for an accompanying pull-up/down resistor or CCS - even op amps driving high-input-impedance buffers:
http://tangentsoft.net/audio/opamp-bias.html
Sign in to Reply
Nic Cohen
7/27/2012 4:20 AM EDT
Does that work that well in comparison?
Regards
Nic
www.kdweb.co.uk
Sign in to Reply
Robotics Developer
5/26/2011 2:29 PM EDT
Really nice article with great details and information! Makes me want to read the book (BSEE - digital guy, but I like analog audiophile applications). Thanks!!! I always look forward to Doug Self's stuff..
Sign in to Reply
Work to Ride comma Ride to Work
5/26/2011 3:03 PM EDT
Gasp. Reminds me of the pre-amp I made using a pair of 741's for my new Technics turntable to go into my ancient receiver that didn't have the phono inputs, only line inputs. Ok, this was 1981 vintage, freshman EE cobbling, but it worked just fine. I saw it in the basement piles of old "junque" a few weeks ago.
Sign in to Reply
Bert22306
5/26/2011 3:40 PM EDT
Bcarso, right you are. The TL082 noise spec is 16 nV/SQR(Hz), and my resistor combinations at the + and - input terminals created 5.59 nV/SQR(Hz). But in my first attempt, where I used higher value resistors, the resistor noise was clearly audible when the circuit was idle. I lowered the value in two stages, and noticed the deminishing returns of lowering resistor values.
If I'd used very low noise opamps, with noise spec around 3 nV/SQR(Hz), it may not have been as easy to lower the resistors enough to make a negligible contribution. Because of interfacing issues with other components.
Just thought I'd mention those two items, load capacitance and theral noise, because I didn't think the article mentioned them explicitly.
Sign in to Reply
jewilson
6/1/2011 1:26 PM EDT
I have found the performance levels of the TL072/82 and the 5534 to be sub standards when compared to the current generation of high performance OP AMPs like the AD797 and AD8597/AD8599. In addition, the uA739 was superior in performance to 741 and other early opamps designs, excluding the 5534.
The AD797 is a very low noise, low distortion operational amplifier ideal for use as a preamplifier. The low noise of 0.9 nV(root)Hz and low total harmonic distortion of -120 dB at audio bandwidths give the AD797 the wide dynamic range necessary for preamps in microphones and mixing consoles.
Furthermore, the AD797's excellent slew rate of 20 V/µs and 110 MHz gain bandwidth make it highly suitable for low frequency ultrasound applications.
The AD8597/AD8599 are very low noise, low distortion operational amplifiers ideal for use as preamplifiers. The low noise of 1.1 nV/√Hz and low harmonic distortion of −120 dB (or better) at audio bandwidths give the AD8597/AD8599 the wide dynamic range necessary for preamplifiers in audio, medical, and instru-mentation applications. The excellent slew rate of 14 V/μs and 10 MHz gain bandwidth make them highly suitable for medical applications. The low distortion and fast settling time make them ideal for buffering of high resolution data converters.
Sign in to Reply
Laser Man
6/2/2011 10:02 AM EDT
I will never buy another book authored by Douglas Self. I have one of his books on audio circuits - they are simply reprints of his articles from a magazine and they contain no design details at all. Poor material for engineers.
Sign in to Reply
sinsinsin49
6/4/2011 12:00 PM EDT
I agree with the opinion that "performance levels of the TL072/82 and the 5534 to be sub standards when compared to the current generation of high performance OP AMPs. My top three devices for serious analog audio work are LME49990, OPA1611 and OPA211. LME49990 is the only one capable of true 24-bit analog performance when powered with +/- 15 V supply (S(N = 159 dB,THD = 146 dB). As the THD of LT1028 is 96 dB, it is "only" good for 16-bit systems. NE5534A is even less linear, THD = 93 dB.
Sign in to Reply