1/f noise and pace

[[e-mail]]

Flicker 1/f Bass Noise and Pace
Op Amp noise can rise sharply in the bass frequencies below 1KHz. Taking a noise (µVrms) reading at 100Hz or at 50Hz, from the plot of noise by frequency, gives an indication of likely flicker 1/f noise.

The NE5532 IC op amp, used in the CD550c, has noise of 1.1Vrms at 100Hz (Figure 1 page 6 datasheet). That is an increase in noise, at 100Hz, of about 50% over its average figure of 0.7 Vrms and noise continues to rise as frequency falls. Bass instruments, with the NE5532/CD550c, could not be made out so well and were a bit lost in large ensemble pieces. There was, however, no noticeable problems with bass, with this op amp and player, playing solo instruments or small ensemble pieces and bass sounded clear and musical. The Exposure 2010S2 uses two https://highfidelity.pl/@main-97&lang=en NE5532s and https://www.whathifi.com/exposure/2010s2/review What Hi Fi? said of this player that the bass suffers from thinness and a lack of resolution. 1/f noise?

It needs to be said that you probably won't notice 1/f noise, if your system does not go much below 100Hz or if your system has a highier noise floor than the op amps or if you only listen to rhymically straight forward or single instrument bass.

Flicker 1/f Noise and Pace
What flicker noise means is that noise is not linear over the audio frequency range. And the bass, due to the 1/f noise, is heard, with some material, differently to the rest of the frequency band. This audio difference is thought to be interpreted as pace errors, but only with complex material.

Recently I have been experimenting with the 1/f compromised OPA1602 and this op amp had pace problems for about 5% of CDs played, even though it has a highish slew (20V/μs) value. For the OPA1602 noise increased from 0.27µVrms at 1kHz to 0.32 at 100Hz (18% increase) and to about 0.41µVrms (52% increase) at 50Hz (Figure 1 page 5 datasheet). However there is some doubt as to the gain compatibility of the OPA1602 in the host CD player (CD753) and this will have a performance impact. And there may well be other compatibility factors.

Julia Wolfe’s fast paced, complex and congested piece Lick (Bang on a Can) https://www.gramophone.co.uk/review/bang-on-a-can-industry is accurately presented using the high slew OP275 (22V/μs) in a CD753 but seems a bit muddled and had a perceived pace issue with the NE5532 (9V/μs) in a CA CD550c. The increase in 1/f bass noise of the NE5532 op amp below 100Hz, and the hypothesised effect of 1/f on noise linearity over frequency and on perceived pace, may have implications here. The OP275, in comparison, has no rise in 1/f noise. The NE5532/CD550c (9V/μs) excelled in playing rhythmically and bass simple but pacey opera.

Further, pace issues were identified in a shunt resistor based pre amplifiers, see https://www.tnt-audio.com/amplpreamp_shunt_comparison_e.html, and it is, perhaps, telling that resistors also have 1/f noise. The link between 1/f noise and percieved pace is, though, a hypothesis and needs further research and observation. Any thoughts?

[Peter Wieck]

Those op-amps may be purchased from Newark for less than $1.00.
Nearly US$1,000 for that player.
The full complement of Cirrus Logic DAC chips would cost about US$40 with shipping, onsie-twosie.

Not so sure it is worth all that.

Peter Wieck
Melrose Park, PA

[[e-mail]]

Not sure which datasheet the OP is using, but the TI NE5532 one claims 5 nV per
root Hz at 1 kHz and 8 nV per root Hz at 30 Hz, implying that i/F is not such a
serious problem. Its slew rate is kind of low. Flicker noise, called popcorn
noise when I was active, is totally different. Some sources claim that the
5532's bipolar input transistors have high popcorn noise, might be what OP is
hearing.

[[e-mail]]

Post by ***@gmail.com
Not sure which datasheet the OP is using, but the TI NE5532 one claims 5 nV per
root Hz at 1 kHz and 8 nV per root Hz at 30 Hz, implying that i/F is not such a
serious problem. Its slew rate is kind of low. Flicker noise, called popcorn
noise when I was active, is totally different. Some sources claim that the
5532's bipolar input transistors have high popcorn noise, might be what OP is
hearing.

Thanks for your response. I'm looking at Figure 1 page 6 Equivalent Input Noise Voltage vs Frequency https://www.ti.com/lit/ds/symlink/ne5532.pdf?ts=1665523375739&ref_url=https%253A%252F%252Fwww.google.com%252F. Table 7.6 page 5 gives the noise figure @ 30Hz of 8nV/√Hz and a noise figure @ 1kHz of 5nV/√Hz. 30Hz noise is more than a 50% increase over baseline1kHz. The increase in noise looks much worse in Figure 1 - and reading off noise at 30Hz gives a value of about 12 but the y axis labels this as nv - not a typical measure. Surprisingly the 5 nV/√Hz tabulated figure looks beliavable as an average over the audio frequency range in figure 1. converting 5 nV/√Hz to just nV gives a figure of 7 which doesn't really fit at all to figure 1. i think that the y axis of figure 1 is a typo and it should read nV/√Hz but then this would not agree with the tabulated figures unless they've used some averaging for the tabulated. datasheets can be opaque about noise. the popcorn noise yes i'll try and follow that up. clearly noise can meld instruments together and add a noise component giving an impression of an additional instrument. do you think the ear is sensitive to non linear noise?

[[e-mail]]

Post by ***@gmail.com
Not sure which datasheet the OP is using, but the TI NE5532 one claims 5 nV per
root Hz at 1 kHz and 8 nV per root Hz at 30 Hz, implying that i/F is not such a
serious problem. Its slew rate is kind of low. Flicker noise, called popcorn
noise when I was active, is totally different. Some sources claim that the
5532's bipolar input transistors have high popcorn noise, might be what OP is
hearing.

i'm looking at https://www.ti.com/lit/ds/symlink/ne5532.pdf?ts=1665523375739&ref_url=https%253A%252F%252Fwww.google.com%252F there is a mismatch between figure 1 and the tabulated figure with a possible averaging applied to the tabulated figure. The tabulated figure of 5nV/√Hz figure agrees with figure 1 as an average over the frequency. converting 5nV/√Hz to just nV gives a figure of 7nV and this doesn't really make sense in Figure 1. I think nV - y axis label in figure 1 - is a typo and should read nV/√Hz. yes i'll look into popcorn noise - thankyou for your contribution.

[j***@mich.com]

Post by ***@gmail.com
Not sure which datasheet the OP is using, but the TI NE5532 one claims 5 nV per
root Hz at 1 kHz and 8 nV per root Hz at 30 Hz, implying that i/F is not such a
serious problem. Its slew rate is kind of low. Flicker noise, called popcorn
noise when I was active, is totally different. Some sources claim that the
5532's bipolar input transistors have high popcorn noise, might be what OP is
hearing.

i'm looking at https://www.ti.com/lit/ds/symlink/ne5532.pdf?ts=1665523375739&ref_url=https%253A%252F%252Fwww.google.com%252F there is a mismatch between figure 1 and the tabulated figure with a possible averaging applied to the tabulated figure. The tabulated figure of 5nV/?Hz figure agrees with figure 1 as an average over the frequency. converting 5nV/?Hz to just nV gives a figure of 7nV and this doesn't really make sense in Figure 1. I think nV - y axis label in figure 1 - is a typo and should read nV/?Hz. yes i'll look into popcorn noise - thankyou for your contribution.

Have you considered the OPA1662 https://www.ti.com/lit/ds/symlink/opa1662.pdf?ts=1665661825192&ref_url=https%253A%252F%252Fwww.google.com%252F
I'm partial to it since I designed it, but it may fit your needs.If it doesn't, please tell what parameter disqualified it. .

[[e-mail]]

Post by ***@gmail.com
Not sure which datasheet the OP is using, but the TI NE5532 one claims 5 nV per
root Hz at 1 kHz and 8 nV per root Hz at 30 Hz, implying that i/F is not such a
serious problem. Its slew rate is kind of low. Flicker noise, called popcorn
noise when I was active, is totally different. Some sources claim that the
5532's bipolar input transistors have high popcorn noise, might be what OP is
hearing.

i'm looking at https://www.ti.com/lit/ds/symlink/ne5532.pdf?ts=1665523375739&ref_url=https%253A%252F%252Fwww.google.com%252F there is a mismatch between figure 1 and the tabulated figure with a possible averaging applied to the tabulated figure. The tabulated figure of 5nV/?Hz figure agrees with figure 1 as an average over the frequency. converting 5nV/?Hz to just nV gives a figure of 7nV and this doesn't really make sense in Figure 1. I think nV - y axis label in figure 1 - is a typo and should read nV/?Hz. yes i'll look into popcorn noise - thankyou for your contribution.

Have you considered the OPA1662 https://www.ti.com/lit/ds/symlink/opa1662.pdf?ts=1665661825192&ref_url=https%253A%252F%252Fwww.google.com%252F
I'm partial to it since I designed it, but it may fit your needs.If it doesn't, please tell what parameter disqualified it. .

Yes i'm listening to the OPA1662 now. It is impressive thankyou for the design. it combines low noise with a relative high CMRR of 96db in the top octave and a mid value slew. and no rise in 1/f noise though again there looks to be a mismatch between the tabulated (p4) and chart (p6) values. i took the chart values as the correct ones. I'll try and submit a piece on how i used the OPA1662 and with a comparison with other op amps and any use issues. Thanks for your response and a great design thanks.

Do you think that there is a link between 1/f noise and pace. i suppose it depends on how sensitive the ear is to non linear noise. but perhaps also the importance of bass is in contributing to percieved pace.

[Dick Pierce]

Post by ***@gmail.com
Do you think that there is a link between 1/f noise and pace.

Since there is no objectively agreed upon definition of "pace", indeed, it mayh well
be one of those normal words co-opted by the high-end high-priests as a completely
vague magic term, the answer is most decidely NO.

Post by ***@gmail.com
i suppose it depends on how sensitive the ear is to non linear noise. but perhaps
also the importance of bass is in contributing to percieved pace.

But, most importanly, it depends upon who is using the term "pace" to describe
what emotional response to which music under what crcumstances after
easting what food at which time of day after having what argument with their
spouse and whether they won it or not after a hard day at work and coming
home with a headache through rush hour traffic and greeted by their youngest
daughter telling them she's pregnant and running off to Morroco with her 57
year old boyfriend. And that's after getting yelled at by your son telling you
he's sick and tired of hearing claim the earth is flat.

[[e-mail]]

Post by Dick Pierce

Post by ***@gmail.com
Do you think that there is a link between 1/f noise and pace.

Since there is no objectively agreed upon definition of "pace", indeed, it mayh well
be one of those normal words co-opted by the high-end high-priests as a completely
vague magic term, the answer is most decidely NO.

Post by ***@gmail.com
i suppose it depends on how sensitive the ear is to non linear noise. but perhaps
also the importance of bass is in contributing to percieved pace.

But, most importanly, it depends upon who is using the term "pace" to describe
what emotional response to which music under what crcumstances after
easting what food at which time of day after having what argument with their
spouse and whether they won it or not after a hard day at work and coming
home with a headache through rush hour traffic and greeted by their youngest
daughter telling them she's pregnant and running off to Morroco with her 57
year old boyfriend. And that's after getting yelled at by your son telling you
he's sick and tired of hearing claim the earth is flat.

your right. and i cannot fault your argument. i did try and include an independent source (gramophone magazine) of how julia wolfe should sound. but there are sound presentation differences when switch between two CD players. and there are always going to be differences and all factors can not be accounted for and there is no independent measure of things like pace or tone. i suppose if it floats my boat go ahead. thanks for your response and i like your prose style.

[Peter Wieck]

your right. and i cannot fault your argument. i did try and include an independent source (gramophone magazine) of how julia wolfe should sound. but there are sound presentation differences when switch between two CD players. and there are always going to be differences and all factors can not be accounted for and there is no independent measure of things like pace or tone. i suppose if it floats my boat go ahead. thanks for your response and i like your prose style.

There is grave danger in accepting a third-party definition of how things 'should' sound. As that third party, typically, does not have access to your ears, listening area, equipment and specific preferences. ANd accepting 'received wisdom' in very nearly every case will lead you down the proverbial garden path, with potential quicksand a the end.

As you state, it floats your boat, and as your choice of options to investigate, entirely valid - for you!

Enjoy!

Peter Wieck
Melrose Park, PA

[[e-mail]]

Post by Peter Wieck

your right. and i cannot fault your argument. i did try and include an independent source (gramophone magazine) of how julia wolfe should sound. but there are sound presentation differences when switch between two CD players. and there are always going to be differences and all factors can not be accounted for and there is no independent measure of things like pace or tone. i suppose if it floats my boat go ahead. thanks for your response and i like your prose style.

There is grave danger in accepting a third-party definition of how things 'should' sound. As that third party, typically, does not have access to your ears, listening area, equipment and specific preferences. ANd accepting 'received wisdom' in very nearly every case will lead you down the proverbial garden path, with potential quicksand a the end.
As you state, it floats your boat, and as your choice of options to investigate, entirely valid - for you!
Enjoy!
Peter Wieck
Melrose Park, PA

thanks for this, the point is there are differences between op amps. if enough people make a comment about an op amp it must come from some common experience. we can not listen to every op amp and cd combiation playing every type of music. and op amp metrics offer some way to understanding how an op amp sounds or performs. i believe 1/f noise is one such metric. in the past i used to think slew gave an indication of pace. i now no longer accept this. 1/f noise and how we interpret non-linear noise seems a possible hypothesis for understanding pace.

[Dick Pierce]

Post by ***@gmail.com
thanks for this, the point is there are differences between op amps.

Indeed there are. There are a mayriad of objectively confirmable differences between op amps.
Some of them MAY be relevant, many are completel irrelevant. There are likely more significant
difference in how opamps are used than between opamps.

Post by ***@gmail.com
if enough people make a comment about an op amp it must come from some common experience.

Yes, BUT that "common experience could not only be objective differences in components resulting
in reliable detectable differences in sound, but those common experiences could also be shared
experiences ("yeah, I read that review, too", "high-end audio maven said such-and-such",) and so on.

Post by ***@gmail.com
we can not listen to every op amp and cd combiation playing every type of music. and op amp
metrics offer some way to understanding how an op amp sounds or performs.

For the most part, I would posi that except for the grossest of problems, they do not.

Post by ***@gmail.com
i believe 1/f noise is one such metric.

But we can, at least, put some objective thresholds in place. What is the total noise in the critical band?
What is the gain structure the system? What is thee sp[ectral content of the music you believe you
hear the prblen in and does this 1/f noise have ANY chance of audibility under IDEAL listening
conditions. There are all questions that have objectve answer.

Post by ***@gmail.com
1/f noise and how we interpret non-linear noise seems a possible hypothesis for understanding pace.

What on earth is "non-linear noise?" This seems to made-up term that could well be like "pace".

And, by the way, does there exist an agreed upon definition of "pace" that actually means something

[Peter Wieck]

One of the things that (for me) makes this hobby continually engaging is the soft pursuit of chimera, wild hares, and with the occasional red herring thrown in. A bit of background: I am an Architecture/Fine Arts major who worked his way through an Ivy League institution working as an electrician, and whose first 'real' job out of college was as a machinist. After doing that for a few years, I could afford to work as towards becoming an architect - and that job, I was making in a month what I *cleared* in a week as a machinist. Which led me to construction, engineering school, and ultimately property management. Not a transistor or vacuum tube along the way, but always an appreciation for music and the means to reproduce it 'at home'. Cutting to the chase, my eyes (ears) were always bigger than my wallet, so I learned the hard way the art of salvage, repair and restoration of cast-offs, garage-sale and flea-market finds, dumpster dives and similar. Lots of smoke and smells along that route. I do not have a 'technician' or 'engineering' attitude towards the hobby, but curiosity in great quantity, even now.

Now 50+ years in the hobby, I have learned a lot. And I have learned a lot about what does, and emphatically, what does not make a difference. Examples include, but are not limited to:

Cables/wires/connectors: After a basic level of quality, both build and material, it makes no difference under most conditions and circumstances. So, a 50' run is different from a 4' run - and appropriate gauges and materials apply for that 50' that may not apply to the 4'.

Within the realm of capacitors, again after that basic quality control is met, and the appropriate materials are specified, a cap is a cap is a cap is a cap.

And so forth.

Transducers are the last frontier of audio - that which changes the one sort of impulse to another. Noise to electrical, electrical to noise, vibration to electrical and so forth.
Amplifiers, tube and solid-state are done-and-dusted.
Pre-amps, tube or solid-state are done-and-dusted.
Tuners, tube or solid-state are done-and-dusted.
CD players are done-and-dusted (Virginia, for the record, there are but so many chipsets out there, similarly for DACs).

There is an entire subset of the audio industry that puts little green LEDs underneath their tube equipment for effect. And charge handsomely for them.
There is an entire subset of the industry that will soak various bits and pieces in liquid nitrogen, and charge handsomely for that.
Little catenary towers.
$5,000 receptacle blocks.

Now, if the purchaser of said 'stuff' is made happy thereby, can afford it, and feels pleasure in ownership of same - I have no problem with that at all. But only if that purchaser was treated honestly, fairly, and transparently throughout the transaction.

Keep in mind that this hobby, at any level, is mostly for a tiny subset of the population, clearly less than 1% if even 1/10th of that. There is enough legacy equipment out there, at every level of quality, to maintain the hobby, probably very nearly forever.

Peter Wieck
Melrose Park, PA

[[e-mail]]

Post by Peter Wieck
Within the realm of capacitors, again after that basic quality control is met, and the appropriate materials are specified, a cap is a cap is a cap is a cap.

I have to respectfully disagree. Capacitor dielectric materials have widely varying characteristics. One phenomenon which I struggled with in design of space electronics is dielectric absorption coefficient. This can range from thousandths of a percent to several percent. How it impacts a circuit is a function of design. The dielectric coefficient can also vary with applied voltage. Application and design are important. I would not state that a capacitor was better without examining its circuit diagram and consider its operating biases. In some cases, frightfully expensive ones, such as Teflon foil, are the only acceptable solution.

[Thomas Krueger]

Post by Peter Wieck
Within the realm of capacitors, again after that basic quality control is met, and the appropriate materials are specified, a cap is a cap is a cap is a cap.

I have to respectfully disagree. Capacitor dielectric materials have widely varying characteristics. One phenomenon which I struggled with in design of space electronics is dielectric absorption coefficient. This can range from thousandths of a percent to several percent. How it impacts a circuit is a function of design. The dielectric coefficient can also vary with applied voltage. Application and design are important. I would not state that a capacitor was better without examining its circuit diagram and consider its operating biases. In some cases, frightfully expensive ones, such as Teflon foil, are the only acceptable solution.

I tend to agree. In the realm of serious high-end audio, a fraction of a percent difference can sometimes be heard. When I was rebuilding stuff for my own amusement in college...

For those who don't remember or know me, when I created the group I was an electronics technician (student employee) at the University of Wisconsin---Milwaukee...

I would always invite friends over, tell them I made a change, and what does it sound like? Sometimes the weirdest oddball stuff made a difference. Good caps always helped. The very act of replacing all of the wire in a Stereo 70. Was it because of the wire I used? MIL-spec silver-plated with Teflon insulation? I doubt it. Probably just resoldering the joints.

One of the oddest upgrades that actually was noticeable was replacing generic 741 op amps with the LF351 in a New York Audio Labs tube phono preamp.

They were in the power supply.

I always have to laugh my butt off at guitar players who insist that the circuit board that Gibson now uses in their electric guitars is somehow crap compared to point-to-point wiring. You won't convince most of them otherwise.

Tom

[[e-mail]]

I looked at the figure and conclude that both axes are mislabeled. Should be nV per root Hz and the frequency axis is off by a factor of 10. Note that the "curve" is just a fitted line to discrete measurements. It looks as though the lowest frequency should be 1 Hz, reflecting the high popcorn noise caused by the input protection diodes. This op-amp is also compensated for unity gain, which always creates trade-offs.

[[e-mail]]

I looked at the figure and conclude that both axes are mislabeled. Should be nV per root Hz and the frequency axis is off by a factor of 10. Note that the "curve" is just a fitted line to discrete measurements. It looks as though the lowest frequency should be 1 Hz, reflecting the high popcorn noise caused by the input protection diodes. This op-amp is also compensated for unity gain, which always creates trade-offs.

thanks for the observations. looks like you're right. more things to think about - thanks

[[e-mail]]

Post by Peter Wieck
Those op-amps may be purchased from Newark for less than $1.00.

Post by Peter Wieck
Nearly US$1,000 for that player.

Post by Peter Wieck
The full complement of Cirrus Logic DAC chips would cost about US$40 with
shipping, onsie-twosie.

Post by Peter Wieck
Not so sure it is worth all that.

Post by Peter Wieck
Peter Wieck

Post by Peter Wieck
Melrose Park, PA

Do you think that the link between 1/f noise and pace is justifiable?