Summing amp?

[JohnRoberts]

Yes. A diagram plotting compensation capacitance vs. slew rate/unity gain bandwidth for the NE5534 would be nice though.

How much slew rate do you need? To save you some mental effort the rate of change for a sine wave is maximum at zero crossing. That rate of change is roughly 1.5x a similar frequency and amplitude triangle wave.

[TMI warning]... Back in the 70s-80s when I was still pondering such things, I considered coming up with a new amplifier metric that quantified linearity wrt slew rate. Based on the ASSumption that slew limiting is not a hard threshold but probably has a soft region of increased distortion on the edge of slew limiting. I even pondered ways to measure this but came to my senses... IMO the world did not need yet another audiophool specification when off the shelf op amps were several times faster than audio could ever need.

A lot of OK music was tracked through 0.5V/uSec paths, albeit using lower (-10dBV) 0VU nominal levels. Slew rate demands only became more apparent when people started multitrack recording and close miking the cymbals (audible LF IMD caused by too slow electronic path). Distant miking cymbals causes the extreme HF to attenuate while passing through the air.

JR

 

[living sounds]

How much slew rate do you need? To save you some mental effort the rate of change for a sine wave is maximum at zero crossing. That rate of change is roughly 1.5x a similar frequency and amplitude triangle wave.

Not very much probably, since the signal goes through an EQ section filled with RC4558 anyway. But I was surprised that upping the frequency of the low pass filter on the mix bus was very audible, since it didn't do much in the audio band.

 

[JohnRoberts]

Not very much probably, since the signal goes through an EQ section filled with RC4558 anyway. But I was surprised that upping the frequency of the low pass filter on the mix bus was very audible, since it didn't do much in the audio band.

Have you attempted to quantify what you were hearing?

Depending on the topology simple LPF can have small response impacts an octave lower. There are also known psycho acoustic effects where more HF noise is perceived as extended HF response.

JR

 

[living sounds]

Have you attempted to quantify what you were hearing?

Depending on the topology simple LPF can have small response impacts an octave lower. There are also known psycho acoustic effects where more HF noise is perceived as extended HF response.

JR

It's a combined LPF and output transformer driver. I didn't see any difference in the measurements within the audible range. But it might also have had something to do with (marginal) stability. I'll take another deep dive with the measuring equipment soon.

 

[Renzo]

The NE5532s are excellent, I used the Siliconix in the 90s for my mixer and for an MS matrix, presented at an AES conference as "professional do it yourself"