If I put my notes here, I might be able to find them again later!
Audio op amp comparison NE5534A vs. OPA134
Finally got around to some more comparison listening with the Sapphire headphone amplifier. To recap: the circuit has an open loop diamond buffer output, so the op amp is just providing voltage gain. It configured for a non-inverting gain of 21 dB to match my 300 ohm HD600 headphones. Pretty much textbook operating conditions.
The op amp inputs are impedance balanced at about 1 kohm. This is about the crossover point where you start thinking about using FET input stages, but BJTs should still be fine.
I'm interested to see if there is a definite signature to a FET-input opamp. The original build called for an OPA134, which is a JFET input circuit. I tried the OPA27, which is a low-noise, high-input-current BJT design, and last night I tried the NE5534A, a classic general purpose audio opamp with bipolar inputs.
I've long been in agreement with Douglas Self on the NE5532/NE5534 : anyone who reports these op amps sound bad is either not using them correctly, or is just being a snob. As cheap and generic as this IC is, it also happens to sound just fine. Really good, actually.
So yes, I could distinguish a difference in the sound of the three op amps I tested: OPA134, OPA27, and NE5534A.
First, the OPA27, which I did not feel sounded as good as the other two here. Music sounded a bit constrained, deadened sense of dynamics, very little air. A bit rough even.
Next the OPA5534A, which I thought was the most neutral and precise of the three. Solid, clean, perhaps a bit "bloodless", but never anything less than faithfull.
That leaves the OPA134, the FET op amp that I am still undecided on. I will say this: I love the way it sounds. It's airy, deep, hugely 3D... but there is this nagging doubt that it's adding a little extra in there. In direct comparison with the NE5534A it seems very slightly sloppy, less solid, and a bit over-accentuated at the top end.
The op amp inputs are impedance balanced at about 1 kohm. This is about the crossover point where you start thinking about using FET input stages, but BJTs should still be fine.
I'm interested to see if there is a definite signature to a FET-input opamp. The original build called for an OPA134, which is a JFET input circuit. I tried the OPA27, which is a low-noise, high-input-current BJT design, and last night I tried the NE5534A, a classic general purpose audio opamp with bipolar inputs.
I've long been in agreement with Douglas Self on the NE5532/NE5534 : anyone who reports these op amps sound bad is either not using them correctly, or is just being a snob. As cheap and generic as this IC is, it also happens to sound just fine. Really good, actually.
So yes, I could distinguish a difference in the sound of the three op amps I tested: OPA134, OPA27, and NE5534A.
First, the OPA27, which I did not feel sounded as good as the other two here. Music sounded a bit constrained, deadened sense of dynamics, very little air. A bit rough even.
Next the OPA5534A, which I thought was the most neutral and precise of the three. Solid, clean, perhaps a bit "bloodless", but never anything less than faithfull.
That leaves the OPA134, the FET op amp that I am still undecided on. I will say this: I love the way it sounds. It's airy, deep, hugely 3D... but there is this nagging doubt that it's adding a little extra in there. In direct comparison with the NE5534A it seems very slightly sloppy, less solid, and a bit over-accentuated at the top end.
Total Comments 1
Comments
-
The 5532/5534 is an excellent sounding and very durable work horse; the OPA134 series can be adequate too. However, if you want the best possible stereo imaging and resolution with regard to phase critical stereo information, much lower distortion is in order. For example: OPA2132 or THS4032 in appropriately designed circuits.
Transducers are more critical than electronic components in this regard, so if your speakers or headphones are not up to the task, then you may not be able to hear a difference in IC's.
The condition of one's ears is a huge factor, e.g. natural ability, ear training, age, or even hearing damage.
If you cannot hear the difference between a OPA2132 in a low distortion circuit, and say, some of the chips you refer to like OPA134 or NE5534, then you might at least consider that it isn't just your ears or ear training, that it might be that your transducers, your listening distance from them, or some environmental - acoustical limitations are involved...
Some people are capable of hearing qualitative difference in sound between signal chains and set - ups, that are difficult to quantify, or that are not yet well described by psycho-acoustic or electronics theory. Don't let that bother you, unless you want to use it as inspiration to do some research! Have fun, experiment, and use what sounds good to you! :-)Posted 8th September 2012 at 12:22 AM by skysound
