I would say that mixing with different (or not) materials, the one with the higher (and louder) resonant frequencies is going to sound worse.😀
The disagreement here is based on not asking the right question, if you follow.
RAART asked how to brace his cabinet better. In fact KEF have already done a pretty good job AND USED THE SAME MATERIAL, which is what I was suggesting.
So that's an end to it, I'd say. 😎
I'd think tuantran has got it exactly right on resistors. Higher rated ones will sound better. 10W better than 3W on linearity with increasing power. Hefty 650V polypropylene capacitors do sound better/smoother than 250V types. All pleasantly simple and obvious really.
Damping is not one size fits all. Reflex and closed box cabinets of whatever material both benefit from damping pads which are really very BBC, but wadding is more of a closed box/acoustic suspension thing designed to absorb energy.
Steen Duelund would quite likely put damping pads on the OUTSIDE of the box, but then he is quite unconventional. Also quite brilliant at times. Here's one of his woofers with a flower arrangement of reinforcement on the cone and some treatment of the chassis, and one of his closed-box cabinets. Enjoy.
RAART asked how to brace his cabinet better. In fact KEF have already done a pretty good job AND USED THE SAME MATERIAL, which is what I was suggesting.
An externally hosted image should be here but it was not working when we last tested it.
So that's an end to it, I'd say. 😎
I'd think tuantran has got it exactly right on resistors. Higher rated ones will sound better. 10W better than 3W on linearity with increasing power. Hefty 650V polypropylene capacitors do sound better/smoother than 250V types. All pleasantly simple and obvious really.
Damping is not one size fits all. Reflex and closed box cabinets of whatever material both benefit from damping pads which are really very BBC, but wadding is more of a closed box/acoustic suspension thing designed to absorb energy.
Steen Duelund would quite likely put damping pads on the OUTSIDE of the box, but then he is quite unconventional. Also quite brilliant at times. Here's one of his woofers with a flower arrangement of reinforcement on the cone and some treatment of the chassis, and one of his closed-box cabinets. Enjoy.
Attachments
It depends, if the material resonates/flexes at a frequency above what the mid/bass will produce then it shouldn't be a problem. This is the reason why people use large amounts of cross bracing, to split the panels up into many small sections that will only resonate at high, out of band, frequencies.
Obviously the point of a laminate is so that when one of the materials wants to resonate, the other does not and visa versa, thus being superior to a single panel made of one material.
If you were going to use a single material though some will be superior to others.
Higher rated ones have the potential to sound better, but this is only if the currently fitted resistors are in any way undersized. Fitting resistors that are too small would, in the presence of MDF/BAF wadding, perhaps present something of a fire hazard, so it's unlikely that KEF would have in anyway undersized them for typical listening levels.
Saying that hefty higher voltage rated caps sound better then lower voltage parts is in no way simple, nor obvious. Certainly using parts that are run close to their max rated spec could bring about some changes in linearity that you'd rather were not there. A 250V cap though is far above any voltage that you're likely to see in a domestic environment so it is perfectly adequate for the job.
Caps also get harder to manufacture as the voltage goes up, they get physically larger and as a result the price increases too. I'd much rather (if possible) spend the extra on a lower voltage cap with the intention of spending that extra money on a part with closer tolerances or perhaps a better dielectric. Besides I don't think I've ever seen a properly constructed test in which anyone could readily tell one decent film cap apart from another decent film cap. SY has his 'Bastard box' and I think the consensus on that one is that no one can tell.
The role of wadding is exactly the same, be it in a sealed or vented box. To absorb and scatter energy around inside the box as to prevent resonances from occurring and to help prevent sound energy from being able to re-radiate back through the cone. This is highly desirable in both sealed and vented boxes, it just so happens though that adding too much wadding (as far as I am aware) has a detrimental effect on the port resonance.
You can see an example of this roughly 1/4 of the way down on this page.
Zaph|Audio
Clearly the resonance effect with the port has been altered slightly by the addition of all of the wadding, but I'd much rather have a cleaner/clearer midrange as a result of a small trade off in bass performance.
Damping pads also do the same thing regardless of enclosure type, attempt to prevent the walls of the enclosure from resonating, which is not the same as preventing resonances from occurring within the enclosure walls. As far as I am aware, a large contribution to the vibration felt in the walls of an enclosure is caused by the driver being rigidly mounted to the cabinet itself. Wadding will not help with this, damping pads though should.
I hate to be too technical, but you know that it's not true. You can go read Wayne Parham wattage for the crossover components and how to dimension their value, "Component Rating – Power and Voltage" pag. 62.
PiSpeakers: π Speaker Crossover Document analysis of circuits used in waveguide and constant directivity loudspeakers
Also if you use a cap in the direct current environment is not the same and "safe" as altern current. Very much similar mistakes are made in manufacturer crossovers for the "public" with low wattage resistors. You know this is a real problem, don't try to hide it. I also agree that's some low voltage caps like some 100W are better because they were tested as such (as you very well mention). It's dangerous to generalize very much and everything like you do even if other members like it. You have a lot more experience than I and that's a fact, I'm not trying to compare myself, but I have to mention some problems/mistakes I find, now and then, and appear (by mistake) in forums like this one.😱
It may well be that capacitor ratings are generally less for an AC environment compared to a DC one, but this is usually (afaik) because the AC values given are RMS. The overall maximum potential that the capacitor can deal with is still the same however. IE a cap standard rated at 63VRMS AC would be suitable for a DC voltage of 100 volts, 63*1.41 equalling 88volts etc. This might be a bit of a simplification though.
Reading Wayne's article it is obvious that I should have known better as I am aware of voltage peaking that can occur in inductive/capacitive circuits, but hadn't given it enough thought, so guilty as charged 😱. Still for normal domestic use I don't think the situation changes by very much, a 250V cap should be more then adequately rated for anything other then very high power set-ups.
In the case of the KEFs though, the drive units would probably cook far before any 250V cap would fail, even under compromised circumstances such as having another component fail before the cap does. My comment was more aimed however at the comment made by system7 saying that higher voltage caps just sound better and that that is supposed to be obvious, where in fact it's anything but.
Thank you for the 'link' to Wayne's work though.
Why does a reflex not need wadding, but closed box does? Very simple really. Reflex gets it's damping and control from the strong magnet of the driver and the voltage amplifier's low output impedance. Reflex is essentially a tuned pipe which is spoiled by anything that restricts airflow and the cone's back energy comes out of the port with a phase change. Closed box needs an internal method of dissipating back energy, and wadding is a help. KEF have thrown a bit of wadding in because their reflex has an option to block the port with a supplied plug, and turn it into closed box. Damping panels is a slightly separate thing. The object there is to reduce the midrange energy radiated by the box.
Power and voltage ratings of components is really a bit peripheral unless it causes catastrophic problems:
With resistors, if they get hot, they end up working like the well-known lightbulb limiter circuit used in pro-audio PAs. Not good to compress a signal, IMO. A 10W resistor is going to have less problems than a 3W, all things being equal. But common-sense applies with budget considerations here.
Lastly, the capacitor thing. I take it most people know that electrolytics have a worse damping factor (DF) than film types, along with some considerable non-linearity when no voltage bias is applied. The physics of it all is quite a big subject, but 630V capacitors are of heftier construction that 250V types, and will have thicker films and metallised layers, and any tendency to "sing" is going to be less. They sound noticeably brighter to my ears:
http://www.diyaudio.com/forums/multi-way/205836-do-someone-compared-these-mundorf-caps-3.html#post2894551
Whether you then need to buy hugely expensive silver or something is debateable. Lynn Olson pointed out that silver foil has a more benign oxidised layer than aluminium, but I'm not too convinced myself.
Which brings us back to BALANCE. The man who was fretting about replacing these little wire links with 2" of exotic cable was clearly unaware that his speaker box contains about 10m of copper cabling already! He wasn't listening. 🙄
Power and voltage ratings of components is really a bit peripheral unless it causes catastrophic problems:
An externally hosted image should be here but it was not working when we last tested it.
With resistors, if they get hot, they end up working like the well-known lightbulb limiter circuit used in pro-audio PAs. Not good to compress a signal, IMO. A 10W resistor is going to have less problems than a 3W, all things being equal. But common-sense applies with budget considerations here.
Lastly, the capacitor thing. I take it most people know that electrolytics have a worse damping factor (DF) than film types, along with some considerable non-linearity when no voltage bias is applied. The physics of it all is quite a big subject, but 630V capacitors are of heftier construction that 250V types, and will have thicker films and metallised layers, and any tendency to "sing" is going to be less. They sound noticeably brighter to my ears:
http://www.diyaudio.com/forums/multi-way/205836-do-someone-compared-these-mundorf-caps-3.html#post2894551
Whether you then need to buy hugely expensive silver or something is debateable. Lynn Olson pointed out that silver foil has a more benign oxidised layer than aluminium, but I'm not too convinced myself.
Which brings us back to BALANCE. The man who was fretting about replacing these little wire links with 2" of exotic cable was clearly unaware that his speaker box contains about 10m of copper cabling already! He wasn't listening. 🙄
An externally hosted image should be here but it was not working when we last tested it.
Yes except that this is only true around resonance/tuning frequency, which is only a very small % of the total bandwidth that the driver will reproduce. Above resonance though what box type you've got becomes largely irrelevant. The drivers behaviour is then controlled entirely by the compliance of the suspension and the drivers motor, in other words the box contributes very little and its main purpose is then to keep the rear wave from meeting with the front wave.
Above resonance they both need a way of dissipating back energy and controlling what the rear wave does inside the cabinet. This is the point of the previous post that shows some of Zaph's measurements on the subject. With only a small amount of wadding a standing wave occurs at a specific frequency well above the tuned resonance of the ported enclosure. Wadding is then inserted and the resonance issue goes away. Both sealed and ported enclosures require wadding to make sure problems like this are kept to a minimum.
Yes, wadding can and does often limit the effectiveness of the port in a reflex design, but this is seen as an acceptable trade off for the gained benefits throughout the midrange.
Resistors generally have to get pretty darned hot before their value changes by an amount that would exceed the given +-% tolerance, with values of a good resistor between 50-500ppm/degree C.
Lets take the worst case scenario with a 500ppm resistor and a 10 ohm resistor +-10%. This, in other words, means that our 10 ohm resistor will increase by a maximum of 5mOhms per degree. In other words given a 100 degree rise in temperature our 10 ohm resistor will have risen at most to 10.5 ohms, this is still well within its 10% tolerance band. We'd need a rise of over 200 degrees for the resistance to change by what could be considered a significant amount.
Given an average ambient temperature of 25 degrees, the resistor would be at 225+ degrees and would melt standard lead solder. Given it's location it would definitely be considered unsafe.
You can also easily get wire-wound power resistors with temperature coefficients of 50ppm/degree C meaning we'd need a rise of 2000 degrees to get a significant change in the resistors resistance.
Even allowing for transient heating of just the resistive element itself, during musical peaks, it is clear that the resistor isn't really going to do much in the way of power compression.
Yes a 10 watt resistor will, on the whole, be superior to a 3 watt resistor on this. But really the only thing you need to be concerned about is whether or not the lower wattage resistor will get hot enough to present itself as a fire hazard. Or, if the resistor is going to be capable of surviving transients well above its rated power handling, some resistors are built specifically to allow for this and some are not.
On the whole I would advocate using 10 watt resistors simply to help ensure that your loudspeaker wont become a fire hazard. But in terms of replacing components in crossovers from an already built commercial loudspeaker, I would expect the manufacturer, especially one as technically renowned as KEF, to have already made sure that the resistors in their loudspeaker are adequate as to be safe. Also if KEF have used 3 watt 50ppm/deg C resistors and you were to blindly replace them with 10 watt 500ppm/deg C resistors you would have in fact gone backwards in terms of any audible or not issues surrounding power compression.
In other words, leave the resistors alone.
Yes except that we're not discussing electrolytic capacitors here, we're discussing going from a 250V polyprop film cap to a 630V polyprop film cap. Neither of which should show any signs of resonance at audio frequencies, nor any signs of microphony in the application. They should both also have a low enough ESR such that it also can be considered negligible in the given application.
If the cap sounds brighter to your ears then it should be easily measurable in the frequency response and if it is then you're not really comparing the difference between a 250V cap and a 630V cap any more.
Like the resistors the caps voltage rating should really be picked to ensure that the capacitor will be safe in the given application. If it is, then I'd rather spend more money on tighter tolerances then unnecessarily spending more money on a part rated for a higher voltage that it will never see.
Exactly, buy a decent film cap suited to the application. If a 250V part will suffice then there's no need to spend extra for a 630V part. Spend that extra, if you must, on something else, like a nice bottle of wine.
I think that I will make soon a decision already in terms of capacitors, resistors and inductors... I thought first to change the components by myself but I did get a reasonable priced offer.
This what will happen probably, a custom made new x-overs:
1. Capacitors will be exchanged for Sonicap Gen 1
2. Resistors will be exchanged for Mills non-inductive wire wound
3. Inductors will be partially exchanged; The larger one will be exchanged for Erse Foil Q Inductor and the smaller one will be reused (still thinking to change both of them). Need to get price quote on this too.
4. All wires will be exchanged for Electra Cable hook-up wire. The conductors are solid core high purity Copper (99.98%) with a 30 mil polyethylene insulator.
5. Pure point to point connections (no PCB traces), custom 60 durometer board stand-offs, and SPAX mounting screws.
Now I have a few questions...
Should I exchange both inductors or to keep air-cored one?
Resistors, MRA-5 or MRA-12?
Is pure point to point connection better then PCB soldered ones?
The second option is that new x-over will have a totally different values then this ones, very close to what KEF uses in their Reference series. Still waiting on more info on this. While I will get the service manual for KEF iQ30 then all info will be there for constructing new x-overs.
The option above probably will cost me not more then 400USD for parts. labor and shipping. Too much??? what you guys thinking about? Opinions please!?
In terms of cabinets I found the following info:
"The cabinet mod requires fiberglass resin, hardener and a large carton or 2 of carbon pellets(like what you would use for a fish tank filter or air purifier), and a can or 2 of spray on V-blok. Mix the pellets with the resin and hardener and allpy to the entire inside of the cabinet walls, once dry and hardened, spray on V-blok. This will make the cabinet alot more inert, and add some weight to the speaker. The carbon pellets, resin and v-blok help to reduce outside cabinet vibrations and absorb and diffuse the backwaves of the speaker on top of the generic poly fill used stock by the factory."
What do you guys thinking of this?
All the best!
This what will happen probably, a custom made new x-overs:
1. Capacitors will be exchanged for Sonicap Gen 1
2. Resistors will be exchanged for Mills non-inductive wire wound
3. Inductors will be partially exchanged; The larger one will be exchanged for Erse Foil Q Inductor and the smaller one will be reused (still thinking to change both of them). Need to get price quote on this too.
4. All wires will be exchanged for Electra Cable hook-up wire. The conductors are solid core high purity Copper (99.98%) with a 30 mil polyethylene insulator.
5. Pure point to point connections (no PCB traces), custom 60 durometer board stand-offs, and SPAX mounting screws.
Now I have a few questions...
Should I exchange both inductors or to keep air-cored one?
Resistors, MRA-5 or MRA-12?
Is pure point to point connection better then PCB soldered ones?
The second option is that new x-over will have a totally different values then this ones, very close to what KEF uses in their Reference series. Still waiting on more info on this. While I will get the service manual for KEF iQ30 then all info will be there for constructing new x-overs.
The option above probably will cost me not more then 400USD for parts. labor and shipping. Too much??? what you guys thinking about? Opinions please!?
In terms of cabinets I found the following info:
"The cabinet mod requires fiberglass resin, hardener and a large carton or 2 of carbon pellets(like what you would use for a fish tank filter or air purifier), and a can or 2 of spray on V-blok. Mix the pellets with the resin and hardener and allpy to the entire inside of the cabinet walls, once dry and hardened, spray on V-blok. This will make the cabinet alot more inert, and add some weight to the speaker. The carbon pellets, resin and v-blok help to reduce outside cabinet vibrations and absorb and diffuse the backwaves of the speaker on top of the generic poly fill used stock by the factory."
What do you guys thinking of this?
All the best!
Last edited:
and I think that I will paint the speakers in piano gloss red... not bright red but also not the dark too... Maybe Ferrari Red! 🙄
Personally I think you're crazy. Spending more on supposed 'up-grades' then the pair of loudspeakers cost when new?
There is nothing wrong with either the inductors or the resistors in this design. KEF have made the right design decisions already by using a cored inductor for the mid/bass as to keep its DCR down as to minimise losses. The tweeter shunt inductor is small enough in value so that using an air core makes sense. The resistors will have been sized accordingly so as to be safe. None of these need replacing.
The only parts that could really benefit from being upgraded are the electrolytic caps, be warned though that the ESR of a decent film cap will be considerably lower then that of the lytics. This will have an effect on the frequency response and should have been factored into the design by KEF. If you change the caps there's no need to go overboard, a standard Solen Fast Cap will be fine, but I would buy some small value resistors too. Changing both the series pass caps in the tweeter leg could result in a slight rise in output level of the tweeter due to the lower ESR of the film caps. You may need to compensate for this, by adding ~0.5ohm extra to the tweeter resistor.
You do not need to change the internal wiring, the PCB if suitable for the new parts also does not need changing. Do NOT change the values of the components, KEF have chosen these specifically to work with the Uni-Q driver in the iQ30.
There is nothing wrong with either the inductors or the resistors in this design. KEF have made the right design decisions already by using a cored inductor for the mid/bass as to keep its DCR down as to minimise losses. The tweeter shunt inductor is small enough in value so that using an air core makes sense. The resistors will have been sized accordingly so as to be safe. None of these need replacing.
The only parts that could really benefit from being upgraded are the electrolytic caps, be warned though that the ESR of a decent film cap will be considerably lower then that of the lytics. This will have an effect on the frequency response and should have been factored into the design by KEF. If you change the caps there's no need to go overboard, a standard Solen Fast Cap will be fine, but I would buy some small value resistors too. Changing both the series pass caps in the tweeter leg could result in a slight rise in output level of the tweeter due to the lower ESR of the film caps. You may need to compensate for this, by adding ~0.5ohm extra to the tweeter resistor.
You do not need to change the internal wiring, the PCB if suitable for the new parts also does not need changing. Do NOT change the values of the components, KEF have chosen these specifically to work with the Uni-Q driver in the iQ30.
I might be... 😉
but I do like to such things like this. Having some fun it does always cost. If it is worth, maybe yes but this is up to each individual to decide. Same with cars... say some people used the car to bring them form point A to point B but some like to have some fun with them too. Even my wife agrees to this and was always telling me that I should keep my old Audi TT... A masterpiece of design and master of fun...
But back to speakers... I think that overall improvement will be greater then cost. I am still not decided if I should keep inductors or not. 95% I am tending to change just capacitors and resistors which will cost me around 80USD per speaker.
While some people swear that point to point assembly does have acoustic difference I am just skeptic, and think that PCB has the same properties.
Internal cables I do think that they can have some mild improvement overall in terms of gauge and copper purity.
For capacitors I think that everyone agrees about sonic properties of those and there it will be the biggest improvement.
Resistors! This where I am absolutely clueless. But it does not cost much to change them 20USD or less for both speakers and mostly people disagree with each other on this. I still think that say around 65% will say change it.
Everything else is known.
Where I am struggling is the cabinets. While I definitely prefer real wood cabinets and do not like mdf at all but KEF made it that way and I can not change this at all unless I build the new real wood cabinets and I am not a cabinet maker. I am wondering if suggestion a posted prior will really do improvement. The gentleman who did it says, yes! Mild but improvement. Wading will be the same and I do not plan to change it while it is ported box design.
With all those "small" improvements this speaker will outshine even more expensive ones and while someone did it already, measured everything and confirmed that it is a big improvement but he went for different route with external x-overs which I do not want.
To be, or not to be, that is the question... 🙄
Last edited:
What maybe you can do, preserving old original x-overs aside keeping their intrinsic value and for latter comparison, and building you a new maybe external x-over (working trough the ports) that you will adapt for next couple of months, with new parts.
You have to understand, and not denying, that tiny resistance (internal DCR) of iron cores is somewhat unattainable by most air-cored inductors. Unless you buy 100$ toroids or "Zero Ohm Coil" exotics that are very expensive parts. So when changing specs always ask advise from members for each part. You can always build another, new, better speaker from diyAudio or other kits, similar or not to yours, as an option to spend a lot of money you have from selling your car.😀
AUDIOXCEL
You have to understand, and not denying, that tiny resistance (internal DCR) of iron cores is somewhat unattainable by most air-cored inductors. Unless you buy 100$ toroids or "Zero Ohm Coil" exotics that are very expensive parts. So when changing specs always ask advise from members for each part. You can always build another, new, better speaker from diyAudio or other kits, similar or not to yours, as an option to spend a lot of money you have from selling your car.😀
AUDIOXCEL
Last edited:
1. Capacitors will be exchanged for Sonicap Gen 1
2. Resistors will be exchanged for Mills non-inductive wire wound
3. Inductors will be partially exchanged; The larger one will be exchanged for Erse Foil Q Inductor
and the smaller one will be reused (still thinking to change both of them). Need to get price quote on this too.
4. All wires will be exchanged for Electra Cable hook-up wire. The conductors are solid core high purity Copper (99.98%) with a 30 mil polyethylene insulator.😎
5. Pure point to point connections (no PCB traces), custom 60 durometer board stand-offs, and SPAX mounting screws.😎
Exchange both inductors
or to keep air-cored one?
Resistors, MRA-5
or MRA-12
?
Is pure point to point connection better then PCB soldered ones😎?
2. Resistors will be exchanged for Mills non-inductive wire wound
3. Inductors will be partially exchanged; The larger one will be exchanged for Erse Foil Q Inductor
and the smaller one will be reused (still thinking to change both of them). Need to get price quote on this too. 4. All wires will be exchanged for Electra Cable hook-up wire. The conductors are solid core high purity Copper (99.98%) with a 30 mil polyethylene insulator.😎
5. Pure point to point connections (no PCB traces), custom 60 durometer board stand-offs, and SPAX mounting screws.😎
Exchange both inductors
or to keep air-cored one?Resistors, MRA-5
or MRA-12?Is pure point to point connection better then PCB soldered ones😎?
Last edited:
I do understand that is the reason I still hesitating otherwise I will order the parts already... I remember when I was in college we have to go for three months to work in the factory (as part of school program) and I went to factory which has manufacturing toroidal transformers, inductors and some other interesting stuff. That was long time ago... and that were I started to love DIY. My first project was to change regular AC/DC transformer for toroidal one with very high specs on my amp and sonic improvement was noticeable immediately. Anyway, back to topic...
The gentleman who submitted offer to me was doing already a x-over mods for KEF iQ30 and he knows what he is talking about. While we agreed mostly on all choice of parts (Sonicap, Mills...) I am just still not sure about inductors and internal cabling but tend to do just cabling and leave inductors untouched. 🙄
What's confuses me is option "point to point" x-over... Does is this superior to regular PCB or not? 😕
Also, the choice of Mills resistors... MRA-5 or MRA-12 while it is just a few $$ difference I am just really confused there... 😕
Thank you! Own you a beer if I ever come to Ottawa (and they have pretty famous beer pub there I just forget the name, so many choices of beer)...Resistors, MRA-5
or if you ever come to Toronto...
tuantran - I have non of topic question for you... Do you know where I can get Low-Inductance hook-up wires in Canada. Want to try this "DIY Speaker Cables"
RAART,
What do you use hook-up wires for? To connect components of a crossover? I use solid pure copper 16 or 18AWG to connect components of a crossover.Because the wires are short and no loops, there is no or very small inductance. I see some good hook-up wire at www.partsconnexion.com. There are silver, silver/gold(very expensive), and pure copper hook-up wires. I think pure copper solid wires are good enough.There is another website which sales speakers and speaker parts www.solen.ca but I don't know if they sale hook-up wires.
What do you use hook-up wires for? To connect components of a crossover? I use solid pure copper 16 or 18AWG to connect components of a crossover.Because the wires are short and no loops, there is no or very small inductance. I see some good hook-up wire at www.partsconnexion.com. There are silver, silver/gold(very expensive), and pure copper hook-up wires. I think pure copper solid wires are good enough.There is another website which sales speakers and speaker parts www.solen.ca but I don't know if they sale hook-up wires.
- Status
- Not open for further replies.