I just received the National Semiconductor monthly newsletter where they announce their new high-performance audio Op-Amp. The LM4562 boasts with impressive claims on various performance parameters. (0.00003% distortion and 2.7nV/sqHz noise).
See
formatting link
It seems if National has a challenge for other Op-Amp chip designers. Are there anybody out there that were able to test drive one of these? Any inside information on the National champion(s) involved in the chip design?
It seems we will have to upgrade our distortion measuring equipment.
On Fri, 29 Sep 2006 14:38:50 +0200, Gerhard wrote in Msg.
That's 1/3 the AD797's distortion, but 2-3 times the noise. Can't have everything!
But the National part is a dual, and probably cheaper. And what I like is that it can actually be had in a TO-99 metal can package, which is obviously geared towards the High-End community.
The distortion is likely not that much better than the AD797. AD notes in the 797 data sheet that the distortion graph is measurement-limited. I've measured AD797 distortion well below their data sheet value--similarly with the OPA627.
The LM4562 data sheet is a bit disappointing to me: a plethora of graphs, and none of noise voltage and current densities versus frequency. Plots of IMD that are characteristic at lower amplitudes not of IMD but of noise-limited measurements. Similarly, plots of THD+N that are noise-limited at all the lower amplitudes, when spectrally the harmonics would be very obvious above the narrow-band noise floor. Plots versus frequency that do not extend beyond 20kHz. (A low-distortion 55MHz GBW part should have uses beyond the assumed limit of the human audio band...)
In comparing noise, it's good to consider the input equivalent noise power, and not just the voltage, and of course what source resistance you'll be using to drive it. If your source resistance is lower than the optimal source resistance for the amplifiers under consideration, you'd pick the lowest voltage noise amp, but if it's higher, you'd pick the lowest current noise amp, and the LM4562 is a bit quieter in the current noise category. The optimal source resistance is about
1.7kohms for the LM4562, and about 450 ohms for the AD797. Noise voltage*current/Hz is 4.3e-21watts/Hz for the LM4562 and 1.8e-21 for the AD797. But an OPA627 checks in at about 1.3e-23, and is also very low distortion even with high source resistances. Transformers (good ones far more expensive than the op amp) may be used to match a source to the amplifier for optimal noise figure.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.