If you forget about "springs" for a minute and just consider a mass hung by a wire, then you have a simple pendulum and the period of oscillation can be calculated by a straightforward formula:
T ~ 2*pi*(L/g)^0.5
where T is the period in seconds, L is the length of the pendulum in meters, and g the gravitational constant 9.8 m/s^2. The period or frequency is independent of mass and only depends on the length of the wire.
There are many online calculators like this one that you can use:
https://www.omnicalculator.com/physics/simple-pendulum
For example, if L=10cm = 0.1m then T = 0.63 sec or f=1.58 Hz. This would be far below the passband of any driver, even a subwoofer, just using a 10cm/4 inch length of wire.
T ~ 2*pi*(L/g)^0.5
where T is the period in seconds, L is the length of the pendulum in meters, and g the gravitational constant 9.8 m/s^2. The period or frequency is independent of mass and only depends on the length of the wire.
There are many online calculators like this one that you can use:
https://www.omnicalculator.com/physics/simple-pendulum
For example, if L=10cm = 0.1m then T = 0.63 sec or f=1.58 Hz. This would be far below the passband of any driver, even a subwoofer, just using a 10cm/4 inch length of wire.
Last edited:
Thanks, I was momentarily unable to go do the homework (as you so graciously did in your reply) but thought that it might set a good baseline for Arthur Jackson, since he mentioned things such as guitars.
Are the speakers suspended by one wire? No. Is the pendulum emitting sound? No. A pendulum speaker clearly demonstrates Doppler effect.
Back to my point: There are two masses, the cone and the loudspeaker frame. There are four springs. So, if the cone goes forward the speaker frame and magnet go proportionally backwards while under tension from the springs. Then the cone and magnet will settle back to rest, but the energy added to the system will take a while to dissipate. As more frequencies and rhythms are added there is ringing. The magnet doesn't travel far, but it does move, and this changes the phase of the sound transmitted from the voice coil to the cone. If the signal is instantaneous the system will ring like a bell due to the lack of damping on the springs. Also, the springs themselves will ring and because the strings aren't exactly the same the frame will rock left to right, back and forth, and twist side to side.
If the springs are under tension it may bring the oscillations back into the passband. If the wires are tight enough they may act as a solid frame but warp the loudspeaker.
I submit that it is better to have the loudspeaker frame and magnet structure absolutely still, anchored, and not moving. Get rid of the springs and use damped high tension wires, I bet it sounds better.
All this horror movie can be stopped if the spring is elastic ( made of multiple strands of rubber).
The field hasn't been explored extensively. I just say that in vinyl replay, the cantilever is attached to the body with rubber.
The surround of a speaker is rubber.
The field hasn't been explored extensively. I just say that in vinyl replay, the cantilever is attached to the body with rubber.
The surround of a speaker is rubber.
A moving coil motor with a conical driver membrane has several elastic components which resonate - spider (7), surround (4), cap (5) and cone (3). Also the frame (2). Each has has more or less mass and one or more fundamental resonances and harmonics. Total mess... Some think that the magnet or the frame is the best part for firm attach, some say it should have elastic mounting... Nearfield measurements can tell the differences in response, distortion and decay.
A planar electrostatic or magnetostatic driver has very low mass membrane that is actuated differently, much less moving parts and inertia. But they can not produce high spl at low freq.
How so?
Ask yourself this: what are the movements of the driver, suspended by wires and what is causing that movement? Then compare that to what causes doppler distortion. These two things are not at all the same. I think I have clearly explained how the system moves, and there is no component that can cause doppler distortion unless you physically grab the speaker, pull it back, and let it "swing" (at its resonance frequency) while music is playing.
Here is another way to think about what is happening: you start with the speaker hanging there, motionless. No audio playback is happening and there is no motion/movement. Then you introduce the audio signal to the driver. There is nothing keeping the driver in place except gravity, because it is 100% freely hanging in air. There are two wires, one on each side of the frame, by which it hangs. So it can swing freely in the front-back direction. Because there is nothing keeping the driver in place, when the motor pushes on the voice coil this displaces the voice coil AND the rest of the driver in opposite directions. They move together and then apart along the axis of the voice coil. But since they are of very different masses, the force pushing them apart causes a small motion in one and a larger motion in the other. Because this is an AC signal with no DC component, they just follow the music signal while moving apart and together. The cone is still pushing on air (the "air load" is part of its mass) and so it still produces sound pressure waves. But there is no motion other than that what is following the music signal, and as a result there cannot be any doppler effect. The only source of motion is from the motor force acting on the voice coil, so the only source of motion is the music signal. Because the system is freely hanging there is no reactive component, and so there cannot be any phase change or modulation, e.g. the phase angle is zero between all the moving parts, always. THERE. IS. NO. DOPPLER. EFFECT.
Last edited:
There's no such thing as 'absolutely still', though, is there? (And would wires of any kind be the way to achieve it?)
@CharlieLaub Surely for your analysis to hold, the whole assembly would need to be suspended in a way that was unaffected by the differential motion of the two elements? But the wires are attached to one element, the frame/magnet assembly. So when that assembly moves at all, gravity will begin to act on it to produce pendulum motion, in addition to the differential motion of the cone and magnet/frame. No?
@ianbo I'm not quite following your wording, above. But let me give it a try. If you think about the motion of the "two elements" :
element 1 = what we normally call the moving parts of the driver, e.g. the cone, voice coil, and VC former
element 2 = the rest of the driver, e.g. frame, magnet, etc.
under the effect of an audio signal these are moving in such a way that the center of mass of the whole remains in exactly the same place at all times. Their movement follows the electrical signal but in opposite directions and with different magnitudes.
If what you are saying is that the wires are attached to element 2, which is moving, then yes you are correct but this is an AC motion with no DC component, so any motion in one direction has motion in the other direction about the original position. It must, because the music signal has a lower cutoff frequency (e.g. a highpass character to its passband) and therefore no DC component. As long as the resonant frequency of the whole driver-on-a-pendulum is much lower than the low end of the passband of the audio signal, there is no component of the music signal that can "energize" the pendulum swing because it is all happening at a much higher frequency. Thus it is all mechanically decoupled from whatever the pendulum is hung from (above) and the center of mass remains fixed.
Let's think about a driver screwed down to a cabinet which is stting on the floor. When an electrical signal flows through the voice coil the magnetic force pushes on the fixed magnetic field of the motor, which is attached to the frame, cabinet and ground. The attachments between motor and ground mean that the motor does not move but rather it is the cone that moves. But when the frame is hanging by wires freely, there is nothing to keep the motor in place. It is completely free to move back and forth along the axis of the voice coil in response to a force acting on it. So now when the electrical signal causes the motor to push on the voice coil the "pushing" causes the motor to move away from the voice coil, e.g. they move "apart". But both will move following the exact same signal, which is why there cannot be a doppler effect taking place. The doppler effect can only happen when there are two stimuli of different frequencies, one modulating the other.
element 1 = what we normally call the moving parts of the driver, e.g. the cone, voice coil, and VC former
element 2 = the rest of the driver, e.g. frame, magnet, etc.
under the effect of an audio signal these are moving in such a way that the center of mass of the whole remains in exactly the same place at all times. Their movement follows the electrical signal but in opposite directions and with different magnitudes.
If what you are saying is that the wires are attached to element 2, which is moving, then yes you are correct but this is an AC motion with no DC component, so any motion in one direction has motion in the other direction about the original position. It must, because the music signal has a lower cutoff frequency (e.g. a highpass character to its passband) and therefore no DC component. As long as the resonant frequency of the whole driver-on-a-pendulum is much lower than the low end of the passband of the audio signal, there is no component of the music signal that can "energize" the pendulum swing because it is all happening at a much higher frequency. Thus it is all mechanically decoupled from whatever the pendulum is hung from (above) and the center of mass remains fixed.
Let's think about a driver screwed down to a cabinet which is stting on the floor. When an electrical signal flows through the voice coil the magnetic force pushes on the fixed magnetic field of the motor, which is attached to the frame, cabinet and ground. The attachments between motor and ground mean that the motor does not move but rather it is the cone that moves. But when the frame is hanging by wires freely, there is nothing to keep the motor in place. It is completely free to move back and forth along the axis of the voice coil in response to a force acting on it. So now when the electrical signal causes the motor to push on the voice coil the "pushing" causes the motor to move away from the voice coil, e.g. they move "apart". But both will move following the exact same signal, which is why there cannot be a doppler effect taking place. The doppler effect can only happen when there are two stimuli of different frequencies, one modulating the other.
Last edited:
You've understood me correctly, Charlie. I''m still not sure I agree, though. 😀 I know that intuition is a poor guide to the laws of physics, but it strikes me that the only thing needed to energise a pendulum is displacement from the neutral position. No? Gravity then supplies the rest, and it wouldn't matter that the displacement resulted from an AC signal at a frequency above that of the pendulum. But perhaps my instinct on this is all wrong.
A single DC displacement (you poke the pendulum with your finger) would indeed do what you say but that is not the case for an AC motion with zero DC component. To make an analogy, this is why there is no net charge on a capacitor after one (or an integral number of) cycle(s) of AC that has no DC component even though there was current flowing.
If it helps, this physics question and answer describes the same sort of problem:
https://physics.stackexchange.com/q...sh-something-standing-on-a-frictionless-floor
Because the driver is hung by wire and it not touching anything, movement is "frictionless" (you can ignore friction with the air around it).
See second answer, containing the image below, as well as other answers down that page. You can think of the stick figure with mass "M" as the cone and VC, and the blue block with mass "m" as the rest of the driver. Note that m >> M.
https://physics.stackexchange.com/q...sh-something-standing-on-a-frictionless-floor
Because the driver is hung by wire and it not touching anything, movement is "frictionless" (you can ignore friction with the air around it).
See second answer, containing the image below, as well as other answers down that page. You can think of the stick figure with mass "M" as the cone and VC, and the blue block with mass "m" as the rest of the driver. Note that m >> M.
OK, I think I get it now. You're saying that the wire-suspended driver isn't going to act as a pendulum if it gets an AC impulse (well above the natural pendulum frequency), because the wires are floppy, essentially. I see that now. If they were stiff rods, it would be different, but the wire's not stiff..
^ Even with stiff rods it shouldn't start swinging like a pendulum. Natural frequency of such pendulum would likely be few hertz or less, and it doesn't get excited with higher frequency motion. You can go sit into a garden / children swing and try to shake it as fast as you can (few Hertz) and it won't start swinging. Only when you move your weight back and forth about at the natural frequency you get the swinging motion going. One could tune the natural frequency with weight and length of the wires, but it's hard to imagine any driver with any length wires to swing faster than say 10 times a second, 10Hz, which would already be quite violent swinging, still way below audible music and likely not excited at all by music.
Last edited:
Back to the midrange drivers 🙂
Per my last comment, while the MX19TX looks very good, the key advantage vs other top picks seems to be around the capability to be crossed lower but I won't be able to benefit from this, and being larger it tends to be limited to 2kHz for xo. So I'm deprioritizing it.
So I'm down to PTT6.5M08, M142, LM10n.
M142 is the most expensive by far and I haven't been able to come across info about it in open baffle, except for @profiguy comments above. So I'll focus on the other two, at least for now.
Glad to see @Juhazi comment above! It's good to have in the discussion someone who has chosen a planar midrange in OB.
I found Vance Dickason reviewed the PTT6.5M08 and LM10n at Voice Coil, which is great because he uses the same measurement approach for both and hence easier to compare results across drivers. I wish HiFi Compass had measured LM10n!!
To measure distortion, Dickason mounts the driver rigidly in free-air, sets it to play at 94dB SPL measured at 1m with pink noise. Interestingly, the PTT required 4.2V to geth there vs LM10n requiring 2.63V. Both shown below side by side.
Would appreciate your thoughts on this. 2nd harmonic seem to be very comparable, while PTT has lower 3rd harmonics than LM, although LM still shows very low 3rd harmonic level. Note the range I'm looking at is 400/500 to 2000/3000Hz.
Per my last comment, while the MX19TX looks very good, the key advantage vs other top picks seems to be around the capability to be crossed lower but I won't be able to benefit from this, and being larger it tends to be limited to 2kHz for xo. So I'm deprioritizing it.
So I'm down to PTT6.5M08, M142, LM10n.
M142 is the most expensive by far and I haven't been able to come across info about it in open baffle, except for @profiguy comments above. So I'll focus on the other two, at least for now.
Glad to see @Juhazi comment above! It's good to have in the discussion someone who has chosen a planar midrange in OB.
I found Vance Dickason reviewed the PTT6.5M08 and LM10n at Voice Coil, which is great because he uses the same measurement approach for both and hence easier to compare results across drivers. I wish HiFi Compass had measured LM10n!!
To measure distortion, Dickason mounts the driver rigidly in free-air, sets it to play at 94dB SPL measured at 1m with pink noise. Interestingly, the PTT required 4.2V to geth there vs LM10n requiring 2.63V. Both shown below side by side.
Would appreciate your thoughts on this. 2nd harmonic seem to be very comparable, while PTT has lower 3rd harmonics than LM, although LM still shows very low 3rd harmonic level. Note the range I'm looking at is 400/500 to 2000/3000Hz.
Apologies, @LewinskiH01, we have gone off topic. One final comment from me: my thanks to @CharlieLaub and @tmuikku - I've read up on the physics and see it clearer now - the pendulum really presents a high impedance to frequencies well above its natural frequency, so the reactive motion of the driver frame and magnet is actually impeded significantly. That's very interesting!
OK I'm done, back on topic.
OK I'm done, back on topic.
For your application, the measured differences seem minor. The planar driver with its symmetrical front & back directivity has to have the advantage, methinks.
Yes, I think LM10n is possibly better for dipole 400-3000Hz, since its off-axis is begin to drop fast after 3500Hz. So crossover point can be as high as 2500-3000Hz, which is better for majority of planar/AMT tweeter. Not to mention its front/back similarity.