NE5517 vs LM13600/13700

[Mark Hammer]

Looking at the datasheets for these, I see that there seems to be a more complex internal architecture to the 5517, compared to the 13600, despite their apparent pin-for-pin similarity.  Does anyone have a sense of what might be "better" about the NE5517, or in what sorts of applications it might be a better choice than a 13600?

[scratch]

Mark,

not a clue, but reading the following in the datasheet ...

Impedance Buffer
The upper limit of transconductance is defined by the
maximum value of IB (2.0 mA). The lowest value of IB for
which the amplifier will function therefore determines the
overall dynamic range. At low values of IB, a buffer with
very low input bias current is desired. A Darlington
amplifier with constant-current source (Q14, Q15, Q16, D7,
D8, and R1) suits the need.

seems to indicate that it might have a wider dynamic range than the 13600/13700, now where that might be useful

[Cliff Schecht]

I can think of a few uses for that actually .

[scratch]

well, it might be useful in some filter configurations ... voltage controlled ... hmmm ..

or maybe in some phaser circuits based on 13600/13700's ??

[JDoyle]

The only difference is the output buffer biasing. Though I'd go with the NE5517 if only because it is probably uses newer diffusion processes, as it is a newer chip, and I think will probably be quieter.

An important difference between both of these and the CA3080 is that they both have their Iabs port TWO diode drops above ground, whereas the CA3080's is only one.

[PRR]

> a wider dynamic range

In their context, this is not the same as "signal to noise+distortion".

Being a naked transistor, or transistor+diode, SNAD is God-given (as close as mortal foundries can come, which is remarkable close).

The output buffer limits the very-very-VERY low-current performance. I've never run the 36000 that low. Below 1uA is MIGHTY thin for audio systems; "small" stray capacitance really bites.

> it is probably uses newer diffusion processes, as it is a newer chip, and I think will probably be quieter.

You might think... and yet the specs are mostly identical. Many TYPs are the same, and MAX same or one hair better. Under-stated?



The specific claim is that the buffer stage does not give a thump as you pull the OTA up from near zero bias. That clearly does not matter for phase-shifter. It could matter for limiter, except when limiting solo guitar the source "thumps" (asymmetric plucked waveform rise) far more than the small transistor offset. And if you actually cared, a TL072 would be even better (wonder why nobody throws that onto a 36000 chassis?).

[JDoyle]

PRR -

I don't know about the bottom set of graphs (noise vs. Iabc) but for the top set I don't believe they ever COULD be different as long as an OTA is based upon a BJT diff. amp; those charts are the same becuase the physics are the same. You can see the formula in question on the top left corner of the 13600 graph.

Regards,

Jay Doyle