Updated: 4 April 2004

CONTENTS.

• Introduction.
• Specs.
• Fig a. Shunt feedback: THD versus level, no loading.
• Fig b. Series feedback: THD versus level, no loading.
• Fig c. Shunt feedback: THD versus loading.
• Fig d. Series feedback: THD vs supply rail volts.

INTRODUCTION.
Here is a look at a more recent opamp design, for comparison with the old faithfuls.
The OPA2134 is a Burr-Brown product, the dual version of the OPA134. There is a quad version called the OPA4134. The manufacturer claims the OPA2134 has superior sound quality, due to its FET input stage. Not a particle of evidence is given to back up this assertion, which is most unfortunate.

The two THD plots show the device working at a gain of 3x in both shunt and series feedback modes. It is obvious that a problem emerges in the series plot, where the THD is higher by about three times at 5 Vrms and 10 kHz. This distortion increases with level, which immediately suggests common-mode distortion in the input stage. This is ironical since this input stage is supposed to be a selling point.

Input noise voltage is typically 8nV/rootHz.
Slewrate is typically +/-20 V/us, but the minimum is rather lower at +/-15 V/us

This is a relatively modern and sophisticated opamp. When you need FET inputs (usually because significant input bias currents would be a problem) this definitely beats the TL072.

SPECS.
Here are the vital statistics: All typical values, for +/-15V supply rails.

Supply voltage	+/-18V abs max
Output range	+/-17V typ (2K load)
CM range	+/-13V
en	8 nV/rtHz typ 1 kHz
in	3 fA/rtHz typ
Ibias	5 pA typ
Slew rate:	20 V/us
Supply current	8 mA total
Unity gain stable	YES
Cost	114p RS Jan 2001

Above: The 2134 working in shunt feedback mode. The THD is below the noise until frequency reaches 10 kHz; it appears to be lower at 5 Vrms simply because the noise floor is relatively lower. (2134opa1.gif)

Above: The 2134 in series feedback mode. Note much higher distortion at HF. (2134opa2.gif)

Above: The 2134 in shunt feedback mode, (to remove input CM distortion) and with varying loads on the output. As usual, more loading makes linearity worse. 5 Vrms out, Gain= 3.3x (2134opa3.gif)

THE COMMON-MODE PROBLEM.
What can be done about this unwelcome common-mode distortion? Obviously choosing a circuit configuration that minimises or eliminates the CM signal is sensible. However, if this is not possible, a slight improvement may be gained by running the opamp off the highest supply rails permitted. The THD at 10 kHz is reduced from 0.0044% to 0.0030% on moving from 15V to 20V rails. See plot below:

Above: CM distortion reduces as you increase the supply rail to the maximum permitted. (and 2 Volts beyond!) Signal level was Vrms, non-inverting configuration.