Crossover- impedance matching, level matching, resonance matching and harmonic balanc

[sudhirbhosale]

Hi all
I built speaker using peerless drivers. For crossover used only capacitors & inductor coil.

In recent DIY post by Hari sir, I saw below terms in crossover section:
Impedance matching, level matching, resonance matching and harmonic balance

So I want to know :
1. If I build only basic crossover with capacitor and inductor then what I would miss
2. What are these terms? impedance matching, level matching, resonance matching and harmonic balance
3. Out of these which are must and which are nice to have
4. Does all these needed for floor standers?

Thanks

 

[keith_correa]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

See this for information on:
Impedance
Level [in #6.0]

I have never heard of resonance matching and harmonic balance in crossovers - maybe others can chime in and help on this.

And every multi driver speaker needs to have a properly designed crossover - irrespective of the box being floor/shelf/stand/wall standing

BTW, you forgot "time alignment" and "Phase"

 

[saheb11123]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Does full range need a crossover too?
What would be ideal for a 4" full range speaker of local(decent) make which i know not the parameters of ? given it is enclosed in a rough box of about 17cm(D) * 22cm(H) * 15cm(B)




Again Im about to build a sub which is also purchased locally (some J audio) cost me around 1800rs i dont know parameters about, what shall be box volume given it should be in a sealed enclosure with the driver of an 8inch. I know of trial and error methods but where should i start? and also if vented how about we go that way?

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Harmonic balance is a concept where in the wave envolope preserves the wave shape during its attack, decay, sustain and release and there by have excellent timbre and trainsent response. This is done by focusing on the harmonic decay between two harmonic say 1KHz and 2KHz and how much the level at these frequencies can decay to make them sound at the same SPL without hearing fatigue and more openness.

You can check about harmonics and timber at these links

Timbre - Wikipedia, the free encyclopedia

Harmonic - Wikipedia, the free encyclopedia

 

[keith_correa]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Does full range need a crossover too?
What would be ideal for a 4" full range speaker of local(decent) make which i know not the parameters of ? given it is enclosed in a rough box of about 17cm(D) * 22cm(H) * 15cm(B)

Again Im about to build a sub which is also purchased locally (some J audio) cost me around 1800rs i dont know parameters about, what shall be box volume given it should be in a sealed enclosure with the driver of an 8inch. I know of trial and error methods but where should i start? and also if vented how about we go that way?

ONE and only full range driver does not need a x-over though it might need a BSC circuit.

If you do not know the TS params of a driver and build a box irrespective be prepared for nasty suprises. But then you might also like it. Stranger things have happened..;-)

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Harmonics is a very vast subject.

What are harmonics are multiples of fundamentals.

Timber ( spelled as Tamber )

The distinct behaviour of each instrument is all about how the harmonics are generated and how the decay between each harmonics are being decayed.

So when it comes to speakers even drivers have harmonic distortion. Which you cannot control using crossover believe me I am designing drivers and Im very clear why harmonic distortion occurs and what is it so important.

So the question is if you take 1khz and lets consider if its intensity is 90db then your second harmonic will be at 2khz which in cheap speakers it will be at 70db and the higher end you go this value will be low like 40db or even 30db as best ( but 30db is super rare )

So how do you measure it in general you need RTAs to do that its not easy to measure it properly. You need a professional equipment.

Total harmonic distortion is measured in % and mostly

Tweeters will be in the order of < .3%
Midranges will be in the order of 2 to .3%
Bass drivers will be in the order of 5 to .5%

so lower the distortion higher the cost of the drivers..

so I will tell u why some serious audiophiles love listening at lower volume because..

Lets take your noise floor in the room is at 40db which in general we have in most of our homes. So if your fundamental is at 90 db and if your 2nd Harmonic is at 50db then your can hear that 2khz tone which is distortion. so if you reduce your fundamental to 80db ( low volume listeners ) then your 2nd harmonic is under the noise floor which you cannot hear hence you feel great about the quality. This is what happens with low volume lovers the Harmonic distortion will be under the noise floor hence you love it.

So how do you make drivers to sound low harmonic distortion is different subject goes into the driver design it has absolutely nothing to do with the enclosure dimensions nor the crossover infact tweaking that for the box dimensions will soon make things worse.


2nd 3rd 4th 5th 6th 7th 8th 9th ......nth harmonic

So the harmonic distortion in drivers and harmonic dist in Amplifiers so in Class A amplifiers the 3rd harmonic distortion will be mostly under the noise floor thats why people love class A amplifiers.

Odd order harmonics are much easily perceived by the human ears so its much more important to consider the odd harmonics.

An ideal system should not generate any harmonics apart from what is being in the material.

Attack Delay Sustain Release is generally termed as ADSR which describes the instrument signature.

ADSR circuits are used more in Public address and studio equipment where they want to artificially control the tone.

The more number of components you use in the crossover the subtle sounds will be lost.
The principle of designing crossover is to consider that the best crossover is no crossover. If your drivers are really good then you can use very simple crossover. Its worst to take bad drivers and say i use 4th order LR crossover and get best results coz it doesnt work that way.

Impedance matching will introduce phase errors which is very much perceived by human ears.

the best way is to do is to select the drivers which has flat impedance response. That tells everything about the driver and preferred wide band drivers.

Lots and lots of parameters to consider before you get into crossover designing.

Alright...............

 

[keith_correa]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Harmonic balance is a concept where in the wave envolope preserves the wave shape during its attack, decay, sustain and release and there by have excellent timbre and trainsent response. This is done by focusing on the harmonic decay between two harmonic say 1KHz and 2KHz and how much the level at these frequencies can decay to make them sound at the same SPL without hearing fatigue and more openness.

You can check about harmonics and timber at these links

Timbre - Wikipedia, the free encyclopedia

Harmonic - Wikipedia, the free encyclopedia

Given that the "attack, decay, sustain and release" are variable, how do you design a x-over optimized for harmonic balance around this?

 

[keith_correa]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Time alignment: see: Time Alignment

I like what they say:

There are fads in marketing inferring that the words time-aligned alone, magically transform the whole sound quality, believing this is more important than using quality speakers. The term time-aligned can be used at random stamped on products similar to washing machines stamped with heavy duty. There are audiophiles who have obsessions with time alignment that go beyond having a sense of perspective. These obsessions are equivalent to painting a single grain of sand a different colour, throwing it back in the beach and expecting to find it when one is allready unable to find ones keys.

 

[kapvin]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Time alignment: see: Time Alignment

I like what they say:

just a couple of points.

time alignment is relevant most on the design axis of the speaker. (google this) once you go off design axis, less relevant.. also works best if the difference between first arrival of the soundwave and the reflected wave arrival is high both in terms of intensity as well as the time domain. otherwise the sound will be pretty smeared anyway, negating any benefits but the "feel good" one. having said that, in a well treated room, a time aligned speaker can make a big difference to image cohesiveness and the feeling of being there.

I am guessing that by resonance matching you mean impedance (peak) compensation This is useful, not just from the POV of presenting a benign load to amps, but also from the point of taming spikes in the speaker response

 

[kapvin]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Harmonics is a very vast subject.

What are harmonics are multiples of fundamentals.

Timber ( spelled as Tamber )

The distinct behaviour of each instrument is all about how the harmonics are generated and how the decay between each harmonics are being decayed.

So when it comes to speakers even drivers have harmonic distortion. Which you cannot control using crossover believe me I am designing drivers and Im very clear why harmonic distortion occurs and what is it so important.

So the question is if you take 1khz and lets consider if its intensity is 90db then your second harmonic will be at 2khz which in cheap speakers it will be at 70db and the higher end you go this value will be low like 40db or even 30db as best ( but 30db is super rare )

So how do you measure it in general you need RTAs to do that its not easy to measure it properly. You need a professional equipment.

Total harmonic distortion is measured in % and mostly

Tweeters will be in the order of < .3%
Midranges will be in the order of 2 to .3%
Bass drivers will be in the order of 5 to .5%

so lower the distortion higher the cost of the drivers..

so I will tell u why some serious audiophiles love listening at lower volume because..

Lets take your noise floor in the room is at 40db which in general we have in most of our homes. So if your fundamental is at 90 db and if your 2nd Harmonic is at 50db then your can hear that 2khz tone which is distortion. so if you reduce your fundamental to 80db ( low volume listeners ) then your 2nd harmonic is under the noise floor which you cannot hear hence you feel great about the quality. This is what happens with low volume lovers the Harmonic distortion will be under the noise floor hence you love it.

So how do you make drivers to sound low harmonic distortion is different subject goes into the driver design it has absolutely nothing to do with the enclosure dimensions nor the crossover infact tweaking that for the box dimensions will soon make things worse.


2nd 3rd 4th 5th 6th 7th 8th 9th ......nth harmonic

So the harmonic distortion in drivers and harmonic dist in Amplifiers so in Class A amplifiers the 3rd harmonic distortion will be mostly under the noise floor thats why people love class A amplifiers.

Odd order harmonics are much easily perceived by the human ears so its much more important to consider the odd harmonics.

An ideal system should not generate any harmonics apart from what is being in the material.

Attack Delay Sustain Release is generally termed as ADSR which describes the instrument signature.

ADSR circuits are used more in Public address and studio equipment where they want to artificially control the tone.

The more number of components you use in the crossover the subtle sounds will be lost.
The principle of designing crossover is to consider that the best crossover is no crossover. If your drivers are really good then you can use very simple crossover. Its worst to take bad drivers and say i use 4th order LR crossover and get best results coz it doesnt work that way.

Impedance matching will introduce phase errors which is very much perceived by human ears.

the best way is to do is to select the drivers which has flat impedance response. That tells everything about the driver and preferred wide band drivers.

Lots and lots of parameters to consider before you get into crossover designing.

Gyan ka bhandaar!!

Is this your knowledge or from some website?

If it's your own knowledge and you are true to your moniker, then you'd know that the statements you made, while sounding emperical, are actually not.

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Given that the "attack, decay, sustain and release" are variable, how do you design a x-over optimized for harmonic balance around this?

Using passive components. Come over for a listen and then we can talk. i am a bit lazy to pin down my points as i am a very very very poor writer.

 

[keith_correa]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

we can give you copper foil inductors as we are getting manufactured on defence standards by a company who manufactures for HAL, BHEL, ECIL and even Indian Airforce so the grade of construction will be very high as good as Duelund Inductors as the process is same what we make here.. Its all on custom order i design.

What do you mean by "give you"? Whom can you "give" and why?

Do you mean that you can sell foil inductors? If yes, then can you please provide indicative pricing? I know the price would depend on the gauge and value so let's say a 0.15mH 14 AWG - how much would this cost?

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

strip inductors

As far as i know strip inductors are used by high current transformer companies for line voltage supply. I have been associated with one company called KSH International (Pune) for 6 months who manufacture the strip copper wires for widing for transformers and have practically seen all types of strip copper wires right from raw material stage till the FG stage. But never thought of using them for speakers and i can very easily source the raw wires from them because atleast 5% of the orders goes as wastage and i also know their MD & CEO personally.

 

[keith_correa]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

As far as i know strip inductors are used by high current transformer companies for line voltage supply. I have been associated with one company called KSH International (Pune) for 6 months who manufacture the strip copper wires for widing for transformers and have practically seen all types of strip copper wires right from raw material stage till the FG stage. But never thought of using them for speakers and i can very easily source the raw wires from them because atleast 5% of the orders goes as wastage and i also know their MD & CEO personally.

Hari, supposedly, copper foil can be used to DIY speaker cable too. Since you say you might easily be able to get some, maybe you can try creating a pair of flat cables and let us know how that turns out.

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

flat copper wires for considering skin effect.

As far as my knowledge goes skin effect does not happen at audio frequencies. Its a phenomena that occurs at VHF, UHF & Microwave frequencies for very low / weak voltage / power signals. For audio freq < 20KHz skin effect at higher power does not apply. More over we use the inductor to filter the high freq for the woofer and hence it would not matter.

I prefer not to use any inductor for high freq even for a baffle step as this may cause some kind of non-linear distortion and would effect the HF quality. I even prefer using a metaloxide resistor if at all i am using a L-pad for the tweeter. Wirewound resistors can cause some inductitive HF distortion which is better avoided.

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

High-technology laminating makes it possible to manufacture inductors from high purity copper or silver foil wound onto a Lexan polycarbonate tube. Foil inductors are mounted by means of a single polycarbonate screw through the central bore, or a silicone elastomer, or a hot-melt adhesive.

Foil inductors have obvious advantages over wire-wound inductors in terms of improved electrical and sonic characteristics. This important new development has much to contribute to the design of state-of-the art loudspeaker crossover networks.

Negligible skin effect below 100 kHz many orders of magnitude below conventional wire-wound types.
High winding tension and vacuum fusing of faces provides high dimensional stability, effectively locking the conductors in place. Wire-wound inductors, which allow relative motion of conductors due to electromagnetic forces, will exhibit FM distortion by the process of reactance modulation.
Flat inductive reactance from 5 Hz to 50 kHz.
No saturation distortion due to air-core design.
Inductors contain no magnetic materials.
Reduced power loss.
Highest space factor (ratio of conductor cross area to total cross-section) means lower DC resistance within comparable dimensions. Negligible power loss due to skin-effect.
Extremely low stored charge.
Cool operation even during prolonged high output power operation.
The high space factor and winding density results in improved heat dissipation.

These high-end stuff may find use in individual DIY projects but will be difficult to be used if not priced appropriately. After all this will add to the price of the product which most customers will not be ready to pay. Customers here want the very best at the very lowest price. Hope you get what i mean.

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Metal oxide agreed a better ones but it will induce shot noise more which is more perceived in HF how do you solve these?

The output resistors anyway is metal oxide type in all audiophile amp. So whats the big deal here. Can you hear the noise is what matters even if you can measure it.

 

[kapvin]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

In that case DIY for many of them is to get best for what they are paying for rather than not as lowest price best quality which never works out. If someone thinks that DIY is a cheaper option i would rather say no for it. DIY is all about Value proposition.

some one pays 25k per pair of speakers in the market gets Peerlessindia based driver in it with cheaper components. If the same budget is spent on making their own stuff he might get imported drivers and very good quality crossover and even an veneer enclosure. But not using peerlessindia drivers and crossover with local components and eventually costing more or less than 15-18k so what makes much difference if he is able to get a professional finished product for 25k paying 7k more which will have more resale value isnt it?

this is quite OT to the thread but you make a very valid point. DIY is not about cheaper than mass market. (a DIYer can never better their sourcing muscle). But it is about disproportionate value for the cost once you get to genuine hifi.

also I am sure that peerless-mumbai has some really good drivers but clearly the ones available for sale in India are not among them. (though that dome midrange looks promising)

 

[kapvin]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

(though that dome midrange looks promising)

I have tried that 2 inch alu dome midrange its harsh and has generic metallic resonance with most of the metallic domes. I sold them off not that great man tried many crossover frequencies but that metal sound is very tough to eliminate.

As far as i remember of the response curve it has a breakup at about 13khz. should be easy to eliminate with a tank filter.

I have experience of the Dayton Rs52 (which is also made by peerless india) and it is really good if you know your way around crossovers and use them appropriately. the peerless interested me because the lower resonance frequency should allow for a bit more flexibility at the lower end.

your opinions seem very generic. every driver has breakups. for metal drivers the breakup can sound harsh, but that is precisely where crossover design comes in, you ensure no breakups in the passband and at least an octave above, and then notch out the primary breakup. in return you get pure pistonic behaviour in the passband and amazing articulation. all about trade offs.

 

[Hari Iyer]

Re: Crossover- impedance matching, level matching, resonance matching and harmonic ba

Alu tweeters do sound harsh and metallic and my experience with Peerless Alu dome tweeter has not been very good. It resonates at around 14KHz and will give a kind of fatigue if not controlled by a tank circuit. Even with the tank circuit which i had at 14KHz it was a bit fatigue prone compared to the Vifa silk dome.