Dont all comment at once guys....
Any takers?
Or is it not about subwoofers, thus boring?
Any takers?
Or is it not about subwoofers, thus boring?
-
Get an exclusive sneak peek into our new project. >>> Click Here
Crossover Signal to Ground. Do we really need high quality parts there?
Hi all,
As the title says.
Lets say I have a crossover schematic, and some pretty high value capacitors for signal to ground. Primarily the one in question C18 220uF on the woofer circuit... This is to cure a huge 100hz hump which occurs when you bend the response down to roll off at 300hz.
The final speaker sounds awesome, but now ive made another for another person, and will also possibly end up making a couple more in the future.
![]()
The crossover was very expensive, and that 220uF part was the size of a beer can and about 1/5 the cost of the whole crossover each, I do have another in the mid circuit, but that one has signal passing through it which would end up actually reaching the woofer and our ears so I am OK with that one.
So the question is, for components which are going to ground in a crossover, does it really matter the quality of the parts? I could chuck a far smaller, far cheaper cap in there, I am wondering if there would be audible difference in doing so. Lets say, apart from the variance in tolerance on hitting the actual values.
Nothing is passing through that part to actually reach our ears correct?
As the title says.
Lets say I have a crossover schematic, and some pretty high value capacitors for signal to ground. Primarily the one in question C18 220uF on the woofer circuit... This is to cure a huge 100hz hump which occurs when you bend the response down to roll off at 300hz.
The final speaker sounds awesome, but now ive made another for another person, and will also possibly end up making a couple more in the future.
The crossover was very expensive, and that 220uF part was the size of a beer can and about 1/5 the cost of the whole crossover each, I do have another in the mid circuit, but that one has signal passing through it which would end up actually reaching the woofer and our ears so I am OK with that one.
So the question is, for components which are going to ground in a crossover, does it really matter the quality of the parts? I could chuck a far smaller, far cheaper cap in there, I am wondering if there would be audible difference in doing so. Lets say, apart from the variance in tolerance on hitting the actual values.
Nothing is passing through that part to actually reach our ears correct?
1 - 17 of 17 Posts
If it is still rated for the amount of voltage and amperes required to flow through it, then I don't think the quality would matter.
That said, there usually isn't a ground in an AC output system; it is fed back to the transistors in the amplifier.
Those transistors are fed by some DC rails (likely), but often well-above and well-below ground potential on both sides at all times (usually).
i.e. Neither AC nor DC requires a ground to function.
In any case... I don't think the power supply or output transistors will really "care", since any power not converted to radiation in the voicecoil and XO's and speaker cables is always returned to the same source regardless.
Technically... there is no such thing as particles. Only waves and fields of potential energy with state information (which interact with each other).
Relativistically, they are at all-possible potential states relative to an observer in a different space-time frame of reference. i.e. the EM waves and electrons took all paths through the circuit at the same time, instantly (from its perspective or without observation.)
In this case: based on the mechanical properties of the crossover, the information system collapses into a single state based on the electrical properties of the system, and any given electron only takes one path in the parallel circuit (especially so, when we try to observe it...) [The EM fields take all paths, but are attenuated or manipulated accordingly.]
Electrons have a charge / polarity, and they move in a closed loop around the nuclei, based on fields-of-force (strong-force, weak-force etc). But at no point were they particles.
When these fields of charge move through a conductor, EM fields are generated, which in the case of a woofer oppose the magnetic fields of the permanent magnet (which is a solid, with electron-locked polarities in a cluster).
Thus the cone moves away from these opposing fields and air is pumped, making sound.
A "solid" is merely a (tightly packed) temperature-locked cluster of fields and waves. Nothing more.
Everything you were taught about particles is mostly a lie, and dark matter theory proves the ineptitude of their classic model.
The universe doesn't operate on a million different oversimplified equations. That's where they went wrong!
It ONLY operates on a single unified field equation that takes ALL relevant information-states and field-types into account SIMULTANEOUSLY (including spooky-action and all the laws of motion, heat, energy-transfer and gravity within-and-without a blackhole [and any other: place, time or "thing"....])
Nobody said it was "gonna be easy" or "had to be easy" to function the way it clearly does.
The behavior at the quantum level IS different from the macro level. But not in the way you'd expect / were told.
The observed complexity INCREASES as the scale DOES, it doesn't get simpler. LOL That's again where they are WRONG. It is the SAME equation but expressing ADDITIONAL levels of complexity at-scale.
A cluster of electrons have a different resultant than a single unit does.
Such as friction, and spring effect; those aren't found [easily] at smaller scales, but the SAME laws/equation is at play in all cases. (i.e. Consistency)
-My 2 cents
That said, there usually isn't a ground in an AC output system; it is fed back to the transistors in the amplifier.
Those transistors are fed by some DC rails (likely), but often well-above and well-below ground potential on both sides at all times (usually).
i.e. Neither AC nor DC requires a ground to function.
In any case... I don't think the power supply or output transistors will really "care", since any power not converted to radiation in the voicecoil and XO's and speaker cables is always returned to the same source regardless.
Technically... there is no such thing as particles. Only waves and fields of potential energy with state information (which interact with each other).
Relativistically, they are at all-possible potential states relative to an observer in a different space-time frame of reference. i.e. the EM waves and electrons took all paths through the circuit at the same time, instantly (from its perspective or without observation.)
In this case: based on the mechanical properties of the crossover, the information system collapses into a single state based on the electrical properties of the system, and any given electron only takes one path in the parallel circuit (especially so, when we try to observe it...) [The EM fields take all paths, but are attenuated or manipulated accordingly.]
Electrons have a charge / polarity, and they move in a closed loop around the nuclei, based on fields-of-force (strong-force, weak-force etc). But at no point were they particles.
When these fields of charge move through a conductor, EM fields are generated, which in the case of a woofer oppose the magnetic fields of the permanent magnet (which is a solid, with electron-locked polarities in a cluster).
Thus the cone moves away from these opposing fields and air is pumped, making sound.
A "solid" is merely a (tightly packed) temperature-locked cluster of fields and waves. Nothing more.
Everything you were taught about particles is mostly a lie, and dark matter theory proves the ineptitude of their classic model.
The universe doesn't operate on a million different oversimplified equations. That's where they went wrong!
It ONLY operates on a single unified field equation that takes ALL relevant information-states and field-types into account SIMULTANEOUSLY (including spooky-action and all the laws of motion, heat, energy-transfer and gravity within-and-without a blackhole [and any other: place, time or "thing"....])
Nobody said it was "gonna be easy" or "had to be easy" to function the way it clearly does.
The behavior at the quantum level IS different from the macro level. But not in the way you'd expect / were told.
The observed complexity INCREASES as the scale DOES, it doesn't get simpler. LOL That's again where they are WRONG. It is the SAME equation but expressing ADDITIONAL levels of complexity at-scale.
A cluster of electrons have a different resultant than a single unit does.
Such as friction, and spring effect; those aren't found [easily] at smaller scales, but the SAME laws/equation is at play in all cases. (i.e. Consistency
)-My 2 cents
Well, while we wait for more answers. Here is a harder math question:
How many random particle collisions does it take to form a single FULLY-functional self-replicating dna-based nanobot cluster (aka human-meatbag)?
All the, combinatorial-probabilities of a 3 billion wide chain of human-genome base-pairs-alignment, are belong...
What is 3 billion factorial?
What is 100 factorial even?
All the time and particles are belong...
Don't flip over a human, you'll see a "Made By Superior Intelligence" sticker underneath.
OK so I did some subsequent digging on the audio forums on other websites and it seems capacitors in parallel are less of a contributor to sound quality which actually hits your ears vs capacitors in series.
As long as the value and voltage etc is within tolerance and its accounted for in the crossover, it sounds like its fine.
Long story short, I will use Electrolytic caps in parallel in future and save a bucketload of money. Its also not just money, the 220uF caps I have used are freaking huge. The Electro caps are 10x smaller.
In the higher frequencies specifically where audio actually passes though the cap and reaches the woofer/tweeter then you want to use at the very least a poly cap. Which makes sense.
If it is still rated for the amount of voltage and amperes required to flow through it, then I don't think the quality would matter.
That said, there usually isn't a ground in an AC output system; it is fed back to the transistors in the amplifier.
Those transistors are fed by some DC rails (likely), but often well-above and well-below ground potential on both sides at all times (usually).
i.e. Neither AC nor DC requires a ground to function.
In any case... I don't think the power supply or output transistors will really "care", since any power not converted to radiation in the voicecoil and XO's and speaker cables is always returned to the same source regardless.
Technically... there is no such thing as particles. Only waves and fields of potential energy with state information (which interact with each other).
Relativistically, they are at all-possible potential states relative to an observer in a different space-time frame of reference. i.e. the EM waves and electrons took all paths through the circuit at the same time, instantly (from its perspective or without observation.)
In this case: based on the mechanical properties of the crossover, the information system collapses into a single state based on the electrical properties of the system, and any given electron only takes one path in the parallel circuit (especially so, when we try to observe it...) [The EM fields take all paths, but are attenuated or manipulated accordingly.]
Electrons have a charge / polarity, and they move in a closed loop around the nuclei, based on fields-of-force (strong-force, weak-force etc). But at no point were they particles.
When these fields of charge move through a conductor, EM fields are generated, which in the case of a woofer oppose the magnetic fields of the permanent magnet (which is a solid, with electron-locked polarities in a cluster).
Thus the cone moves away from these opposing fields and air is pumped, making sound.
A "solid" is merely a (tightly packed) temperature-locked cluster of fields and waves. Nothing more.
Everything you were taught about particles is mostly a lie, and dark matter theory proves the ineptitude of their classic model.
The universe doesn't operate on a million different oversimplified equations. That's where they went wrong!
It ONLY operates on a single unified field equation that takes ALL relevant information-states and field-types into account SIMULTANEOUSLY (including spooky-action and all the laws of motion, heat, energy-transfer and gravity within-and-without a blackhole [and any other: place, time or "thing"....])
Nobody said it was "gonna be easy" or "had to be easy" to function the way it clearly does.
The behavior at the quantum level IS different from the macro level. But not in the way you'd expect / were told.
The observed complexity INCREASES as the scale DOES, it doesn't get simpler. LOL That's again where they are WRONG. It is the SAME equation but expressing ADDITIONAL levels of complexity at-scale.
A cluster of electrons have a different resultant than a single unit does.
Such as friction, and spring effect; those aren't found [easily] at smaller scales, but the SAME laws/equation is at play in all cases. (i.e. Consistency
-My 2 cents
Well, while we wait for more answers. Here is a harder math question:
How many random particle collisions does it take to form a single FULLY-functional self-replicating dna-based nanobot cluster (aka human-meatbag)?
All the, combinatorial-probabilities of a 3 billion wide chain of human-genome base-pairs-alignment, are belong...
What is 3 billion factorial?
What is 100 factorial even?
All the time and particles are belong...
Don't flip over a human, you'll see a "Made By Superior Intelligence" sticker underneath.
Not boring. Good question. I read it yesterday but wanted to think about it.
The reason we put good caps in the signal path is to reduce distortion. Different capacitors have more or less qualities that result in non-linearities that result in distortion. By agreeing that "good" caps should be placed in series with the load (the signal goes through them) then I think you (and all of us) can agree that some caps are better then others for audio use.
Focusing on bypass, shunt, or "to ground" caps, I think it "depends". A capacitor placed in the circuit that bypasses the load or shunts to ground will still have more or less non-linearities based upon the capacitor chosen. If the bypassed, or returned, current experiences non-linearities because of the type of capacitor chosen, then it will still present those problems to the load (driver). If current gets returned to ground and it's messed up (non-ideal), then what's left over that goes to the load is also messed up.
Here's where I think "it depends". If the reactance of the bypass capacitor is sufficiently high compared to the impedance (reactance and resistance) of the series path that goes through the load, then variations (non-linearities) in the bypass path aren't going to affect the series path very much. If the bypass reactance is high, bypassed current to ground will be low and any variations just won't affect the load as much. If the bypass reactance gets lower (less) with respect to the impedance in series with the load, then variations are going to have a more substantial effect on what the load sees.
If it were me, I'd say that it's my crossover for my speaker and I'm going to use good caps everywhere. But I can certainly agree that if you've analyzed the circuit enough (in whatever circuit sim you prefer) and proved to yourself that the bypass path just isn't going to affect the load that much, then I think it's clever and good engineering practice to reduce the requirements on that bypass cap (and save some money).
In other words, I think you're right under certain conditions and you're smart to want to optimize cost vs performance. However, I'm lazy and would just spend the money
(I'm also cheap and would have to consider it depending on how much money could be saved).
One big example of something I am working on now is my Atmos speakers which will be in ceiling, they are 8" Coax. And the box is only about 30x30cm so there is only so much real estate inside to even fit a crossover. The design calls for 1 x 200uf (I can sneak that down to 150uf without too much problem though 220 is better) and 1 x 68uf caps both on the woofer circuit in parallel.Not boring. Good question. I read it yesterday but wanted to think about it.
The reason we put good caps in the signal path is to reduce distortion. Different capacitors have more or less qualities that result in non-linearities that result in distortion. By agreeing that "good" caps should be placed in series with the load (the signal goes through them) then I think you (and all of us) can agree that some caps are better then others for audio use.
Focusing on bypass, shunt, or "to ground" caps, I think it "depends". A capacitor placed in the circuit that bypasses the load or shunts to ground will still have more or less non-linearities based upon the capacitor chosen. If the bypassed, or returned, current experiences non-linearities because of the type of capacitor chosen, then it will still present those problems to the load (driver). If current gets returned to ground and it's messed up (non-ideal), then what's left over that goes to the load is also messed up.
Here's where I think "it depends". If the reactance of the bypass capacitor is sufficiently high compared to the impedance (reactance and resistance) of the series path that goes through the load, then variations (non-linearities) in the bypass path aren't going to affect the series path very much. If the bypass reactance is high, bypassed current to ground will be low and any variations just won't affect the load as much. If the bypass reactance gets lower (less) with respect to the impedance in series with the load, then variations are going to have a more substantial effect on what the load sees.
If it were me, I'd say that it's my crossover for my speaker and I'm going to use good caps everywhere. But I can certainly agree that if you've analyzed the circuit enough (in whatever circuit sim you prefer) and proved to yourself that the bypass path just isn't going to affect the load that much, then I think it's clever and good engineering practice to reduce the requirements on that bypass cap (and save some money).
In other words, I think you're right under certain conditions and you're smart to want to optimize cost vs performance. However, I'm lazy and would just spend the money
The cost of those, the cheapest I can find on PE is ~$20 for the 68uf and roughly ~$50 for 2x 100uf to make 200uf.
So those two parts for each speaker, of which I am making four, is already $70 isolated. Which is $280 for all four... Which brings total crossover price on all four to $500 when I add the other parts in!
Or, by using the electro's in place I would be looking at $193 for all four crossovers total including every part. Just those two parts changing, the price overall was 2.5 greater.
This same phenomenon extends to other designs I have made in the past, and have just used the more expensive parts (which are also colossal in size!). I should probably test this one day with a bigger speaker project. Those cheap Electro's are only dollars so I guess I have no excuse not to test it. But then if I dont hear any difference, I would be pissed I spend up to $100 on the expensive Poly's... Catch 22.
1 - 17 of 17 Posts
