That alone will not do anything for all the subsonic stuff below 8 Hz- it needs to be quite a high Q notch to avoid significant rolloff at 20 Hz.
Yes, but not much as all you can do is limit the value of C0 in the bottom of the RIAA feedback arm, which is what the much-despised IEC Amendment did.
I am suggesting the use of Inverse Chebyshev filters, which have a maximally-flat passband, a faster rolloff than Butterworth, and one or more notches in the stopband, as above. There will be some rapid phase shifts near 20 Hz, but who cares? You can't hear phase.
If anyone is aware of a good source of design info for Inverse Chebyshev filters, I'd be very glad to hear about it.
I can add to the above that preventing rumble from being generated, as much as possible, is just as important.
Using a pressure weight over the spindle and a rather heavy ring on the outer circumference resting on the LP are helping a great deal.
It is also to be expected that in this ever improving world, modern record players are suppressing rumble quite a bit better than those papers of some 40 years ago were telling us.
It would be nice to have a very recent update on the subject.
To support that idea even further, almost all high end Phono_Amps can switch their rumble filter off, a seemingly senseless feature if it would be impossible to use it.
And I have have never ever read a Stereophile test where they mentioned that is was not possible to play without rumble filter.
On the contrary, the impression is given that they always play without.
The only reason I can imagine why in my personal experience the Butterworth rumble filters (6db/oct or 12dB/oct) degrade sound, is because of LF phase distortion.
So phase distortion might not be so unimportant after all.
With my Quad ESL's the bass becomes "blubbery" when playing records with lot's of LF information.
Without rumble filter everything stays well in control, and that for a loudspeaker that does not even go that far downwards in Bass reproduction and playing with equipment that is fully DC coupled from start to end.
Hans
Quoted by rif:
It can't be both Butterworth and minimum phase unless it is 1-pole. So which is it? Minimum phase means Bessel, not Butterworth. Butterworth is maximally flat (and monotonic).
Yes. I am not blaming you!
Maybe someone could reverse engineer the circuit and determine what filter shape it provides. My main point is that a technical error in a manual could be just a slip-up, or a sign of misunderstanding or the result of marketing people embroidering the truth.
Maybe someone could reverse engineer the circuit and determine what filter shape it provides. My main point is that a technical error in a manual could be just a slip-up, or a sign of misunderstanding or the result of marketing people embroidering the truth.
yes, Hans, some records are warped much, while others are not. But if one plays vinyl over loudspeakers at high sound pressure levels, then one needs rumble filters, unless one wants to oversize diaphragm excursion by a factor of 10 only for the subsonic.
The side-to-side movement of the stylus carries the mono with the bass, which is the main band, while the up-down movement only carries stereo satellite, the add-on band. This is similar to FM radio and other stuff; there is a compatible main band, and there may be optional add-on bands. One could build a pickup, which outputs the main band (L+R with bass) and the add-on band (L-R without bass) instead of L and R, just by turning it by Pi/4.
The side-to-side movement of the stylus carries the mono with the bass, which is the main band, while the up-down movement only carries stereo satellite, the add-on band. This is similar to FM radio and other stuff; there is a compatible main band, and there may be optional add-on bands. One could build a pickup, which outputs the main band (L+R with bass) and the add-on band (L-R without bass) instead of L and R, just by turning it by Pi/4.
If you have not visited this page, might be a good read for you. At the end, its shows his 3rd order rumble LPF. I see in the pic of the pcb he is using 1uF caps but in the schematic he shows 1u5?
Low Noise Design Schematics
Low Noise Design Schematics
I don't understand why it is needed to add rumble filter. If rumble exists, chances are you have a bad designed table, unmatched arm and cartridge and ineffective anti vibration mechanism.
In fact, I am thinking of building a simple RC rumble filter for measuring azimuth on my iPad.
I have the feeling that we are trying to solve a problem without knowing the size of the problem and its cause and by introducing solutions in the blind, we might create new problems in the form of phase and amplitude distortion.
We can't rely on 40 years old papers describing the problem in a field where so many things have been changed over the years.
That LF is always recorded as a mono signal and that out of phase LF information has to be caused by rumble seems to me just another assumption.
And how do we define rumble, is 1) it the motor/mechanical noise of the turntable or 2) a turntable sitting on a stand that is not acoustically isolated from the floor or 3) is it the direct or indirect soundwave from the music hitting the element which may even lead to a "microphone" instability or is it 4) rumble in the LP itself.
The first 3 points should not be a problem with modern equipment and careful setup. Only in case of number 4) there could be a problem, but as mentioned in the beginning "how big is this problem"?
So before deciding on the need to designing a special gun to kill a dragon, we should first know the size of the dragon plus the information from which side he is coming.
Hans
We can't rely on 40 years old papers describing the problem in a field where so many things have been changed over the years.
That LF is always recorded as a mono signal and that out of phase LF information has to be caused by rumble seems to me just another assumption.
And how do we define rumble, is 1) it the motor/mechanical noise of the turntable or 2) a turntable sitting on a stand that is not acoustically isolated from the floor or 3) is it the direct or indirect soundwave from the music hitting the element which may even lead to a "microphone" instability or is it 4) rumble in the LP itself.
The first 3 points should not be a problem with modern equipment and careful setup. Only in case of number 4) there could be a problem, but as mentioned in the beginning "how big is this problem"?
So before deciding on the need to designing a special gun to kill a dragon, we should first know the size of the dragon plus the information from which side he is coming.
Hans
For any particular LP this is an assumption, unless the recording engineer has told you. For LPs in general it seems to be usually a matter of fact; true sufficiently often that assuming it is a wise thing to do. However, I understand that this thread is about rumble filters which do not exploit this.Hans Polak said:
I always thought rumble is any LF or subsonic signal from the cartridge which was not part of the original sound recording. With a decent system most of it will come from the LP itself.
The dragon exists, but he changes size from time to time. Maybe we need a flexible gun?
Here is one of probably many examples where someone (with a sub woofer) was experiencing huge rumble problems which seemed impossible to cure.
At the end the solution was found by placing the turntable on a different spot, problem gone.
It shows how important it is to isolate the turntable from its surrounding, mechanically, acoustically and from electric/magnetic inductions.
Low Frequecy Rumble - Turntable? | Steve Hoffman Music Forums
And Jan, you are absolutely right, the dragon might have more heads than the 4 that I have suggested.
So let's try to indentify them all and see if we can bring them back to acceptable portions, ultimately leading to possibly not having to use a rumble filter.
That was my whole point, assessing and removing where possible the cause of the rumble problems instead of taking the easy road by covering the problem with a big muddy blanket.
Hans
At the end the solution was found by placing the turntable on a different spot, problem gone.
It shows how important it is to isolate the turntable from its surrounding, mechanically, acoustically and from electric/magnetic inductions.
Low Frequecy Rumble - Turntable? | Steve Hoffman Music Forums
And Jan, you are absolutely right, the dragon might have more heads than the 4 that I have suggested.
So let's try to indentify them all and see if we can bring them back to acceptable portions, ultimately leading to possibly not having to use a rumble filter.
That was my whole point, assessing and removing where possible the cause of the rumble problems instead of taking the easy road by covering the problem with a big muddy blanket.
Hans
Last edited:
A good bearing, isolation and flat records gives us the option of having less electronics. But is a rumble filter with delay possible which can measure dynamic rumble and corrects/filters it.
Regards.
Regards.
Going into the archives of the JAES, I stumbled across this document, written april 1967 by Benjamin B. Bauer, Vice President Acoustics and Magnetics CBS laboratories, and former development engineer at Shure.
So he must have known what he was talking about when it came to rumble.
His definition of rumble was:
Reading the full article, the main rumble he measured was noticeable at 30Hz and 60Hz, plus all harmonics and restricted within the first 500Hz depending on the type of motor.
He even came to a nice definition called RRLL (Relative Rumble Loudness Level) telling how much rumble can be tolerated without interfering with the music.
Funy that old stuff, isn't it?
Now we are talking about rumble of unknown origin that seems to be at around 10Hz.
It is quite obvious that we are in desparate need for newer definitions.
Hans
So he must have known what he was talking about when it came to rumble.
His definition of rumble was:
Reading the full article, the main rumble he measured was noticeable at 30Hz and 60Hz, plus all harmonics and restricted within the first 500Hz depending on the type of motor.
He even came to a nice definition called RRLL (Relative Rumble Loudness Level) telling how much rumble can be tolerated without interfering with the music.
Funy that old stuff, isn't it?
Now we are talking about rumble of unknown origin that seems to be at around 10Hz.
It is quite obvious that we are in desparate need for newer definitions.
Hans
Excellent digging!
Back then, I wonder, if the lower limit of human hearing chosen to be 20 Hz as we do now?
Back then, I wonder, if the lower limit of human hearing chosen to be 20 Hz as we do now?
I have found a paper on low frequency oscillation, almost 40 years old, using a MM cartridge with a tracking force of 1 gram.
http://www.theanalogdept.com/images/spp6_pics/TT_Design/MechanicalResonances.pdf
The greatest cause of below 20Hz rumble according to the author is the Pick Up Arm.
I do not know how this all translates for an MC element, nor have I found a useful indication of how much of this type of rumble can be allowed.
But it LF rumble may cause IM distortion sidebands, as shown in several figures.
The question that rises immediately, if IM distortion has already been added to the signal, how could a rumble filter prevent this??
The harm has already been done !
Another nice article covering the subject is to be found in Linear Audio Vol 10, from Hannes Allmaier, using partly the information from the above article.
But he is also not giving answers to the amount of LF rumble still being acceptable and wether or not a rumble filter should be used in what situation.
A picture below shows the horizontal res frequency of my pickup with MC element:
Below the resonance frequency, falloff seems steeper and concave instead of convex as in the articles.
Is this the difference between MM and MC, I don't know, but frequencies at 0.5Hz and 1.2 Hz because of disk imperfections as shown in the article must be down by some 25 dB, will that be enough?
This seems like a very small step in finding some answers, but as usual, this leads to even more questions.
Hans
http://www.theanalogdept.com/images/spp6_pics/TT_Design/MechanicalResonances.pdf
The greatest cause of below 20Hz rumble according to the author is the Pick Up Arm.
I do not know how this all translates for an MC element, nor have I found a useful indication of how much of this type of rumble can be allowed.
But it LF rumble may cause IM distortion sidebands, as shown in several figures.
The question that rises immediately, if IM distortion has already been added to the signal, how could a rumble filter prevent this??
The harm has already been done !
Another nice article covering the subject is to be found in Linear Audio Vol 10, from Hannes Allmaier, using partly the information from the above article.
But he is also not giving answers to the amount of LF rumble still being acceptable and wether or not a rumble filter should be used in what situation.
A picture below shows the horizontal res frequency of my pickup with MC element:
Below the resonance frequency, falloff seems steeper and concave instead of convex as in the articles.
Is this the difference between MM and MC, I don't know, but frequencies at 0.5Hz and 1.2 Hz because of disk imperfections as shown in the article must be down by some 25 dB, will that be enough?
This seems like a very small step in finding some answers, but as usual, this leads to even more questions.
Hans
The concave falloff probably arises from using a linear frequency scale, instead of logarithmic. Merely a presentational issue.Hans Polak said:
- Status
- Not open for further replies.