Did you try to replace your msh file with the one I generated? I mean the file ath.msh in your ABEC project directory.
And did you try to open directly the project I generated?
Last edited:
Hi,
Read this for food of thought https://www.edn.com/loudspeaker-operation-the-superiority-of-current-drive-over-voltage-drive/ just ignore the amplifier debate wording of it, and consider it's about impedance in series with the driver, amplifier has nothing to do with the phenomenon in sense that everything happens within the driver basically, driver emits it's own distortion and dampens itself depending what impedance load it has. Examples are for a direct radiator but I guess the same applies with horn attached although I have not imagined how the acoustic impedance part affects here in horn context.
In short, the short that low output impedance amplifier provides between driver terminals allows a driver dampen itself electrically maximal way. This only works around the main resonance, where mass of the driver is cancelled by spring so that excitation force of the motor goes directly into velocity of the moving assembly, and resulting back EMF makes current that is 180 deg out of phase of the excitation current so it dampens. On frequencies above the main resonance force needs to accelerate the mass so current from backEMF is perpendicular to the excitation current, so is the counteracting force, and does nothing "controlling the cone", so above main resonance these are just error currents with no useful purpose. So only excitation that makes the cone move about at the resonant frequency gets electrically damped with low source impedance, anything above doesn't.
I'm not sure how the acoustic loading works here, if it also cancels the mass somehow then electrical damping should work and as small of an impedance in series with the driver would be of benefit. Otherwise high impedance should help reduce effects of the movement. This you can test with the phone sine generator and changing the impedance in series. You'd likely want to have low impedance at the main resonance at least.
Read this for food of thought https://www.edn.com/loudspeaker-operation-the-superiority-of-current-drive-over-voltage-drive/ just ignore the amplifier debate wording of it, and consider it's about impedance in series with the driver, amplifier has nothing to do with the phenomenon in sense that everything happens within the driver basically, driver emits it's own distortion and dampens itself depending what impedance load it has. Examples are for a direct radiator but I guess the same applies with horn attached although I have not imagined how the acoustic impedance part affects here in horn context.
In short, the short that low output impedance amplifier provides between driver terminals allows a driver dampen itself electrically maximal way. This only works around the main resonance, where mass of the driver is cancelled by spring so that excitation force of the motor goes directly into velocity of the moving assembly, and resulting back EMF makes current that is 180 deg out of phase of the excitation current so it dampens. On frequencies above the main resonance force needs to accelerate the mass so current from backEMF is perpendicular to the excitation current, so is the counteracting force, and does nothing "controlling the cone", so above main resonance these are just error currents with no useful purpose. So only excitation that makes the cone move about at the resonant frequency gets electrically damped with low source impedance, anything above doesn't.
I'm not sure how the acoustic loading works here, if it also cancels the mass somehow then electrical damping should work and as small of an impedance in series with the driver would be of benefit. Otherwise high impedance should help reduce effects of the movement. This you can test with the phone sine generator and changing the impedance in series. You'd likely want to have low impedance at the main resonance at least.
Since I don't really understand a single word of what you're saying, I guess I will just pretend I have not seen those impedance graphs - for an active xover I don't need it anyway 🙂
😀 Yeah, I think the difference is quite small in practice, especially in home hifi use where SPL are low anyway, so this stuff is not among most important things, especially with active system. With a system that is ran around it's limits it might be more obvious, although haven't ever experimented with high SPL, also low SPL tests only few times.
Ok one more simple explanation to help you think it through 😀
What makes acoustic sound is current through voice coil. Since we have a voltage amplifier ( low output impedance ) connected directly to driver terminals, the current in this circuit, so the acoustic output, is dominated by impedance of the driver. Now, as the driver moves the impedance varies some, which results varying current in the circuit which means distortion in the acoustic output. The more you add non changing impedance in series with the driver the less the current varies due to movement of the driver, so that the driver doesn't dominate circuit impedance anymore, and the less there is distortion. Basically any movement of the driver, be it from excitation from the amplifier or some external excitation like microphone action, makes the circuit impedance vary.
Ok one more simple explanation to help you think it through 😀
What makes acoustic sound is current through voice coil. Since we have a voltage amplifier ( low output impedance ) connected directly to driver terminals, the current in this circuit, so the acoustic output, is dominated by impedance of the driver. Now, as the driver moves the impedance varies some, which results varying current in the circuit which means distortion in the acoustic output. The more you add non changing impedance in series with the driver the less the current varies due to movement of the driver, so that the driver doesn't dominate circuit impedance anymore, and the less there is distortion. Basically any movement of the driver, be it from excitation from the amplifier or some external excitation like microphone action, makes the circuit impedance vary.
Last edited:
So do I, at least, get correctly the notion that the external sound pressure acting on a diaphragm directly distorts the current through the voice coil, and by that also the radiated sound? This is all about distortion, right?
Well, distortion is relative to SPL so 😀 I simplified a lot there, but as said compression driver has quite small excursion compared to woofers so perhaps it's impedance and suspension parameters vary less than a woofer's, so perhaps not a concern. It's interesting phenomenon though, fun to think through 🙂 Loudspeaker driver are quite good microphones.
I've got prototype A460 and another shallower and bit smaller custom device, I try to find time to make simple tests with this stuff. Simplest would be just use ear in place of the driver and try to hear how the waveguide affects sound. Another is to record, make the system woofer play sounds and compare how much the horn picks it up as a microphone. Perhaps one can calculate excursion from the output, at least relative excursion, and try to figure out whether it matters or not. Likely not, but it would be fun test I think.
I've got prototype A460 and another shallower and bit smaller custom device, I try to find time to make simple tests with this stuff. Simplest would be just use ear in place of the driver and try to hear how the waveguide affects sound. Another is to record, make the system woofer play sounds and compare how much the horn picks it up as a microphone. Perhaps one can calculate excursion from the output, at least relative excursion, and try to figure out whether it matters or not. Likely not, but it would be fun test I think.
How exactly do you wire a compression driver as a microphone to record the picked-up signal?
Just hook it up to the mic input as a 2-terminal + and - microphone and record ;-)
Switch of any phantom power.
//
Switch of any phantom power.
//
You see, I'm really bad at this. Thanks for helping me, I'm not ashamed to ask 😀
I guess the horn will make a big difference.
I guess the horn will make a big difference.
If you put a horn to your ear you'll notice it's quite directional compared to no horn at ear situation 😀 Block the other ear with a finger to hear directivity the horn adds.
So, here is another way to "reduce microphonics" if one has to: do not point the horn directly towards a wall, whose sound would reflect right back in to the device (first specular reflection). If you point it so that direct reflections come at an angle, you'd basically help to randomize sound that enters the device reducing the ripple seen in measurements (outcome of microphonics). Another question is the woofer in the same system, directly below the device, and how much it has effect. That would be by far the loudest sound source always nearby no matter which direction the system points at. Also, from the directivity, I bet that a direct radiating tweeter would show less microphonics in sense that sound that enters it is more random (less directivity).
So, here is another way to "reduce microphonics" if one has to: do not point the horn directly towards a wall, whose sound would reflect right back in to the device (first specular reflection). If you point it so that direct reflections come at an angle, you'd basically help to randomize sound that enters the device reducing the ripple seen in measurements (outcome of microphonics). Another question is the woofer in the same system, directly below the device, and how much it has effect. That would be by far the loudest sound source always nearby no matter which direction the system points at. Also, from the directivity, I bet that a direct radiating tweeter would show less microphonics in sense that sound that enters it is more random (less directivity).
Last edited:
I have yet to measure the other horns. I almost haven't measured the electrical impedances until now.
I quickly tried the A400G2+DH450 last night and it's the same. It may also be the long adapter that is prone to this (i.e. high-efficiency midrange in general), that's to be tested. I have never seen this with a "regular" OS WG.
I quickly tried the A400G2+DH450 last night and it's the same. It may also be the long adapter that is prone to this (i.e. high-efficiency midrange in general), that's to be tested. I have never seen this with a "regular" OS WG.
^^ sheeple I'm not sure which measurements you refer to specifically? If that is so, perhaps the wiggle is from resonance of the device itself. This can be easily tested by damping the device by hands. Stand behind the system and grab from the lips, or at any other part of the device that deadens it.
It's easy to find out what parts of the device ring by handling test. Knock the driver / throat portion of the horn with a knuckle and use the other hand to find spots on the device that reduce the sound, if there is any. A horn like this can have crazy loud "handling sound", it amplifies any sound like when one touches the horn, so this is another fun test.
ps. I think the handling sound, sounds that travel through the system mechanically, are far louder than distortion from microphonics, especially listening at home SPL.
It's easy to find out what parts of the device ring by handling test. Knock the driver / throat portion of the horn with a knuckle and use the other hand to find spots on the device that reduce the sound, if there is any. A horn like this can have crazy loud "handling sound", it amplifies any sound like when one touches the horn, so this is another fun test.
ps. I think the handling sound, sounds that travel through the system mechanically, are far louder than distortion from microphonics, especially listening at home SPL.
Last edited:
To me the most striking thing is still how stable and repeatable are those small wiggles on the impedance curve. As the horn is turned around on its stand, each direction it faces has its own distinctive imprint.
You can repeat the measurement and you still get the same exact curve including all the wiggles - for that direction. For a different direction, you get slightly different wiggles, but again perfectly repeatable.
It's likely that the ripple is due to the first specular reflections, the loudest, through microphonics. What if you point the horn directly towards a wall, and change distance to the wall? the ripple should move but also change in amplitude if you are close to a wall.