GT ViPre Rise Time

[George Necola]

I came here because of the schematics..

but I have to put in a brick ( ;D) for the GT VIPRE. I agree with butch vig and all the other VIPRE users, that it's one of the best Valvepres available. you have to dig into the sound.. kinda..  8)

 

[emrr]

Today's measurements show rise time is a shallow LPF of sorts, in any and all settings it looks the same. Kinda goofy.

 

[JohnRoberts]

I am not familiar with that SKU but recall an AES paper written by Dr Marshall Leach (RIP) in the context of audio power amp design where the input stage could be rise time limited. The benefit was an audio path that could not be slew limited. An arbitrarily fast edge square would be nicely smoothed like a one pole RC. In theory with a rise time limited circuit a 1V square wave and a 10V square wave would exhibit the same rise time.

JR

 

[abbey road d enfer]

Today's measurements show rise time is a shallow LPF of sorts, in any and all settings it looks the same. Kinda goofy.

IIRC, the LPF is in the open-loop path. When NFB is applied, it results in rise-time increase. The low-level HF response is almost unchanged because of NFB compensating, but as soon as the frequency/level combination results in slew-rate limiting, it can be legitimately be qualified as some sort of rise-time control.
I discovered that a long time ago, when my opamp of choice was the 748. Accidentally, the compensation cap (typically 30 pF) was replaced with a higher value in production (300 or 3000pF). Standard QC did not see that but final audible test was suspicious. It took some time to find out, because the effect is quite subtle in normal operating conditions, where energy tends to decrease at HF.
Had we done the test on cymbals or soprano voice, it would probably have been more obvious...
Finally, a full frequency sweep at high-level revealed the issue, with the output mutating from a standard sinewave to a near-perfect triangle.

 

[JohnRoberts]

IIRC, the LPF is in the open-loop path. When NFB is applied, it results in rise-time increase. The low-level HF response is almost unchanged because of NFB compensating, but as soon as the frequency/level combination results in slew-rate limiting, it can be legitimately be qualified as some sort of rise-time control.
I discovered that a long time ago, when my opamp of choice was the 748. Accidentally, the compensation cap (typically 30 pF) was replaced with a higher value in production (300 or 3000pF). Standard QC did not see that but final audible test was suspicious. It took some time to find out, because the effect is quite subtle in normal operating conditions, where energy tends to decrease at HF.
Had we done the test on cymbals or soprano voice, it would probably have been more obvious...
Finally, a full frequency sweep at high-level revealed the issue, with the output mutating from a standard sinewave to a near-perfect triangle.

Slew rate and rise time are two different ways to characterize signal rate of change. Since this is touted as a presumably good sounding feature it is unlikely to be simple slew rate limiting. When an audio path encounters slew rate limiting bad things happen. The all too familiar rectification where paths that are too slow to pass RF cleanly, decode the radio station program (not cleanly). Signals slower than RF but too fast for the audio path slew rate will turn cymbal crashed into coarse (grungy) intermodulation distortion.

As Leach taught in his AES paper a rise time limited audio path harmlessly attenuates above band content.

Making a rise time limit circuit variable could be a subtle LPF? Or not, I never experimented with rise times that infringed down into the audio bandpass.

JR

 

[abbey road d enfer]

Slew rate and rise time are two different ways to characterize signal rate of change.

True if the circuit is in its linear operation.

Since this is touted as a presumably good sounding feature it is unlikely to be simple slew rate limiting. When an audio path encounters slew rate limiting bad things happen.

Not necessarily. In the example I gave earlier, the person who commissioned the desk noted that it sounded very good although it didn't measure well when fed with a +20dBu 10k sine.

The all too familiar rectification where paths that are too slow to pass RF cleanly, decode the radio station program (not cleanly).

I'm not sure it happens when SR limiting happens in a single well defined place with a passive circuit.

As Leach taught in his AES paper a rise time limited audio path harmlessly attenuates above band content.

I can't open this document, but doesn't it seem to confirm my vision?

 

[JohnRoberts]

True if the circuit is in its linear operation.

Not necessarily. In the example I gave earlier, the person who commissioned the desk noted that it sounded very good although it didn't measure well when fed with a +20dBu 10k sine.

Because distorting a 10kHz sine wave generates inaudible even higher frequency distortion products. This is why I preferred HF IMD testing to THD for clean audio paths.

I'm not sure it happens when SR limiting happens in a single well defined place with a passive circuit.

Passive like an RC? That rise time limits not slew rate limits. Most variable slew rate circuits I've ever seen or messed with involve active circuitry (OTAs make nice intentional slew limited paths).

I can't open this document, but doesn't it seem to confirm my vision?

I discussed this paper as recently as one month ago.. leach paper

I was able to read the thumbnail... as I recall the entire paper was only a few pages long.

I let my AES membership lapse decades ago.

JR

 

[emrr]

It's funny what a 'late stage / modern' tube circuit this device is, to then have this rise time feature to give variable 'antique' sound. So many bells and whistles for a single channel device. Huge quality audio transformers in it, ruler flat at the fastest rise time from 5Hz to at least 44kHz, I didn't test at a higher rate.

 

[JohnRoberts]

A revealing test would be feeding in a wide band square wave. Properly rise time limited would look like the familiar RC waveform for any level square wave. A slew rate limited signal will clearly look different for smaller and larger square waves.

Not sure what an "antique sound" but a hard and fast rule for merchandising exotic products, it helps when claiming that a path sounds better it must at least sound different.

JR

 

[user 37518]

I am not familiar with that SKU but recall an AES paper written by Dr Marshall Leach (RIP) in the context of audio power amp design where the input stage could be rise time limited. The benefit was an audio path that could not be slew limited. An arbitrarily fast edge square would be nicely smoothed like a one pole RC. In theory with a rise time limited circuit a 1V square wave and a 10V square wave would exhibit the same rise time.

JR

If I'm not mistaken, Nelson Pass and Doug Self made some slew rate experiments with music and I was surprised that even at power amp voltages, a very low slew rate is required to reproduce music without distortion. The truth is that 80V peak-to-peak 20kHz sine waves just don't occur in practice.

 

[JohnRoberts]

If I'm not mistaken, Nelson Pass and Doug Self made some slew rate experiments with music and I was surprised that even at power amp voltages, a very low slew rate is required to reproduce music without distortion. The truth is that 80V peak-to-peak 20kHz sine waves just don't occur in practice.

it's pretty much math... an 80Vp-p triangle wave at 20kHz needs to slew 80v full scale in 25 uSec (1/2 a 20kHz period) so that's 3V and change per microsecond. A sine wave needs roughly 1.5x that slew rate for clean reproduction so still less than 5V/uSec. Of course we want to provide some headroom above that but it isn't heavy lifting to slew that fast with modern technology.

In multiway systems dedicated tweeter amps can be hit pretty hard, but amplifier speed has not been a problem for decades.

JR

 

[user 37518]

it's pretty much math... an 80Vp-p triangle wave at 20kHz needs to slew 80v full scale in 25 uSec (1/2 a 20kHz period) so that's 3V and change per microsecond. A sine wave needs roughly 1.5x that slew rate for clean reproduction so still less than 5V/uSec. Of course we want to provide some headroom above that but it isn't heavy lifting to slew that fast with modern technology.

In multiway systems dedicated tweeter amps can be hit pretty hard, but amplifier speed has not been a problem for decades.

JR

John, I know the math behind it, my point was that music doesn't have too much high amplitude and high frequency content, so 20kHz 80Vp-p is not a huge problem by the mere fact that it is not present at high amplitudes in music.

 

[JohnRoberts]

John, I know the math behind it, my point was that music doesn't have too much high amplitude and high frequency content, so 20kHz 80Vp-p is not a huge problem by the mere fact that it is not present at high amplitudes in music.

There are different ways of looking at this, designing to pass worst case signals, or to work most of the time with typical signal..

There are several designs that depend on this only modest amount of HF content to work. FM radio and tape NR HF pre-de emphasis. The HF gets boosted up before broadcast or printing to magnetic tape, then rolled off for play back while also rolling off channel noise.

Sorry, veers on the internet can keep veering...

JR