svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,253
FWIW, low frequency content of some selected CDs:
The authors note that environmental sounds like car engines and sounds under a bridge go down to 6 and 13 Hz.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Can amplifier speed and resolution be measured?
- Thread starter confucius_zero
- Start date
- Status
FWIW, low frequency content of some selected CDs:
The authors note that environmental sounds like car engines and sounds under a bridge go down to 6 and 13 Hz.
LF extension actually is SPEED RELATED ... here is the Nota Bene’s example:
He took a tone burst (form of the sound envelope is the square wave) with 50 Hz fundamental frequency and provided it’s spectrum of harmonics and subharmonics down to 1Hz. Then he has filtered this signal using 2nd order HPF tuned to 20Hz (blue line on the graph) and 50 Hz (red line on the graph).
The results shows how the transient process (sound envelop’s front) is delayed by limiting the lows - i.e. it is “slowed” by removing lows and infra-lows.
The source article is in Russian but if anyone wants to use some kind of an online translator here is the link to it http://www.reanimator-h.narod.ru/emos.html
Attachments
Not all older CDs are equal, I was talking about the historical trend. CDs on your list are more like an exception, not the common rule for that time.
The tone-burst used isn't a valid test signal as the start of each tone is instantaneous, which does not happen in any naturally occurring transient process. Consequently, claiming that the start is slowed, is a nonsense.
S.
- Joined
- Jun 19, 2018
- Messages
- 6,652
- Likes
- 9,403
The essence of that is correct theoretically at least, i.e. any non-linear-phase high-pass filter will cause group delay, i.e. a delaying of lower frequency content relative to higher frequency content.
It's going to be dwarfed by what the speaker is doing at those frequencies though, and even if it weren't, there's no evidence it would be audible except as an amplitude effect (the 2nd order high pass filters in the example, I mean).
20 Hz HPFs are not relevant to "speed." Nor, afaik, is it part of Red Book. And as others have pointed out, real-world transducers (both speakers and mics) are incapable of producing synthetic waveform like that, so as long as the f3 is a couple octaves below the cutoff point of the program material or transducers (nearly always the case), there's no significant alteration of the envelope.
The "no true Scotsman" argument at #243 isn't really a valid one.
Last edited:
It is just an example - the idea is obvious - subharmonics form the sound envelope transition. We can experiment with any spectrum of real musical instruments, like JEJ did with cymbals.
Just a simple example - why sometimes ported subwoofers sound slow comparing to the sealed ones? They have sharp roll off below the port frequency ... so they filter infra lows naturally. This effect is audible only listening to some music - why? Because our brain knows how it shall sound and “catches” imperfections.
Synthetic waveforms are not a real world, they just help to understand the processes, like math helps to understand how constructions will handle weight having nothing similar in the real world.
Last edited:
There are NO subharmonics from the gated tone. There is the envelope, which is at a much lower frequency, but that has nothing to do with subharmonics, but with group delay.
S
Do you mean that SpectraLab math is wrong or the window was too small? Levels of subharmonics are pretty high on the attached above spectrum screenshot.
Another question - did you ever experienced “slower” sounding of ported subwoofers comparing to the sealed ones (musical sources) while the measured GD in both cases was pretty good and below the borderline? If the answer is “yes”, do you have your own explanation of the effect? Thanks.
Last edited:
There are no subharmonics, those may look like subharmonics, but they are not. It's a function of modulation of the tones by the switching envelope. Think of it as sidebands, not subharmonics.
I don't understand 'slow' or 'fast' in terms of bass. Ported bass relies on energy storage, and so has group delay, which I do understand, but not the word 'slow' in this context.
S
Well, the definition of subharmonics “Of, relating to, or being a wave with a frequency that is a fraction of a fundamental frequency.” They stimulus spikes definitely look like subharmonics - fractions of the fundamental frequency 50Hz. But it doesn't really matter, let’s call them a lower part of sideband due to modulation, it doesn’t change the subject. Let’s rephrase it - filtering out the lower sideband changes the waveform of modulating signal.
You didn’t answer my question regarding your experience with ported and sealed designs, it could be related to speakers, and not subwoofers. “Slow” in this case means the feeling of delayed sounding for the musical signals. It might be subjective, of course, but why so many people have noticed it?
Last edited:
I've had sealed loudspeakers, ported and transmission lines, and none have sounded 'slow', so can't help with subjective impressions.
S
Ok, so as I understand, you didn’t hear or notice such imperfections. Thanks
restorer-john
Grand Contributor
Wikipedia is hardly a credible source. It is a crowd sourced and edited, group-think site.
Here is an excerpt (summary) from the Digital Audio Disc Study Group (DAD Convention), Joint Sessions W6.2,3. 17th November 1978 where the target frequency response was DC-20KHz (for Working Group 1-3) from an 'average' digital master tape.
35 Japanese manufacturers had met in September 1978 to move towards an optical disc standard. Sony, Mitsubishi and Hitachi had already demonstrated digital optical disc prototypes the previous year (September 1977) at the Audio Fair in Japan.
Source: Digital Audio Technology. 1st edition. Heitaro Nakajima, T Doi and J Fukuda.
Not only had they 'thought about' the frequency range below 20Hz, they planned, engineered and produced equipment for it.
solderdude
Grand Contributor
Was this ever sent to Amir ?
Thanks for your information, John. I shall take my words back.
However, majority of older CD records have frequencies cut below 20Hz, sometimes below 30 Hz.
I will provide more waterfalls a bit later. It is not the reason I spoke about, but it should be another reason for it.
Last edited:
IIRC a number of early CDs were dubbed from analog finals that probably included a low-cut (rumble) filter since they targeted LPs. At the same time marketing-speak was touting CD's ability to go to DC since DC-coupled amplifiers were in vogue. A trend I was glad to see go away, frankly, though I made a tidy sum repairing amps and the speakers they took out when they failed.
- Joined
- Jun 19, 2018
- Messages
- 6,652
- Likes
- 9,403
This is just a question of mixing/mastering decisions. In fact, for a lot of rock, pop and electronic music, and no doubt also some acoustic music, HPFs at 25-30Hz are standard practice at (preferably) the mixing stage, or if not then at the mastering stage.
The purposes may include:
- to remove any DC offset that may have crept in during recording or AD conversion, which if allowed to remain in the mix would reduce headroom
- to remove low frequency noise (rumble) that may have been picked up by the mics in the recording process
- to ensure that any compression (including limiting) works more effectively (i.e. as intended) on bass frequencies
- to prevent any low-frequency shelving EQ applied during mixing or mastering from amplifying inaudible or barely audible frequencies
Another underlying premise is that, for a human to hear a tone at say 25Hz, even in quiet and with no other maskers present, it must be around 70dB in level (refer for example to equal loudness contours). Very few systems are capable of these SPLs at these frequencies, and very few listening environments are suitably quiet. So to hear a tone at such frequencies in a recording in which other content is present and tending to mask any ultra-low frequency content (even if it's there intentionally) is a tall order, even if they are present on the recording and the system is capable of reproducing them at sufficiently loud levels.
Finally, many loudspeaker systems will not be capable of producing sub-30Hz content at audible levels, yet the presence of these frequencies in signal content will nevertheless exert strain on woofers.
Most mixing/mastering engineers therefore recognise that it's generally preferable, therefore, to remove such content from recordings.
Oh and a final point: a similar but often more aggressive practice also existed in the vinyl mastering days, since very low frequencies would be mechanically difficult in terms of vinyl cutting and playback - though I know less about this I must admit.
restorer-john
Grand Contributor
Yes, the DC to daylight amplifiers were in vogue alright. Practically every manufacturer had DC in their advertising and were at pains to say it wasn't just direct coupled stages, but "Pure DC" from end to end with "no capacitors" in the signal path etc.
My favourite for repairs was the Kenwood L07m/mk2 amplifiers and its matching preamplifier, the LO7c/mk2. DC-600KHz FR. Sanken/NEC T03 LAPTs, 6 per monoblock and stability issues from day 1. A low frequency DC protector prone to simply not work. Sansuis were also happy to bake voice coils or simply dislocate spiders. Rumble filter? No way- it wasn't 'pure' enough. Woofers wildly moving with infrasonics was an indication of how 'deep' your amp went.
I don't think I've ever seen a CD player without a coupling capacitor however. They don't really get to DC, but awfully close. My early test discs start at 5Hz for sweeps and I've never bothered to look lower.
When the holistic benefits of wideband amplifiers were implemented, but with a blocking cap on the outputs of the RIAA stage and the preamplifier output itself, the following stages could be direct coupled. I have a number of power amps with FET direct coupled (capacitorless) front ends like this:
Often the F3 of the 100% DC NFB RC network is set below 0.5Hz, so they rely on the effectiveness of the dedicated DC protector IC to disconnect the speakers...
I agree, there's heaps of early CDs with shelved off LF, often the ones where they make excuses (on the label) for originally being mastered on analog equipment. Was it the recording/mastering studios themselves? Did every piece in the chain of analog equipment pre-digital simply never exploit <20Hz, because there was no point? Was there a en-masse redesign/replacement of studio gear or did the 'engineers' simply zero sliders they previously cut?
Last edited:
- Status
Similar threads
