Inductors/resistors/capacitors. Which ones for the preferred DIY speaker and why?

[treitz3]

Those that have done the DIY route, would you go for an iron core inductor within a speaker or a regular inductor regardless of size? Space is not of concern. I will be going external with the crossover. I'm looking for the utmost of performance and reproduction...

[EDIT:] The discussion has progressed beyond that of inductors. The thread header has been edited to reflect the changing discussion.

Tom

 

[DonH50]

If by "regular" you mean air-core, that is what I would use for a speaker crossover. It will be larger and require more (and heavier gauge) wire, but you avoid the hysteresis and potential for exceeding the flux capacity of the iron-core types.

If iron-core, you need one that has the capacity to handle peak current flow, normally derated by at least 2x and often 4x or more for critical (low-distortion) applications.

Have you considered using a line-level crossover (active or passive) and separate amps for each driver?

FWIWFM - Don

 

[ack]

Absolutely air-core and Mundorf CFC or Duelund.

 

[treitz3]

Good morning, Don and thank you for your response. I am referring to an air core inductor. The current inductor is iron core and after a bit of discussion with the gentlemen that will be building the crossover in an external chassis, he had mentioned the challenges and benefits of the types of inductors. We will have to see where in the circuit the inductor is. That will have much to do with which one but from what I have heard, an air core is the way to go. It will be a large air core inductor and the placement of it within the circuit will definitely have a factor. It is my understanding that air core is the way to go for optimal sound and this is what we will try to achieve.

I will have to read up on hysteresis and potential for exceeding the flux capacity of the iron-core types. To be bluntly honest, I have heard of both of these but I have no understanding of them yet.

Have you considered using a line-level crossover (active or passive) and separate amps for each driver?

Absolutely. I have some friends around the nation that have taken this route on their systems and they are pleased with the results. The problem is, I'm not. There is something about the end result as to what hits my ears that just doesn't sound right and they all seem to constantly fiddle with the settings. I don't want that. I'm more of a set-it and forget it type of guy. I also have to add to the fact that I have other things in life that I need to budget for and with my choices of amplifiers and cables, it quickly becomes cost prohibitive to do so. Especially when you add in the cost of a new rack, isolation devices for the new gear and the added room for all of the gear. For me, it all boils down to me not liking the end result but it is these other factors seal the deal for me.

Now with that said, I have heard what upgrading the components and going external can do to the little brother of my Tyler speakers. The same gentlemen who will be doing the crossover upgrades for me had done this to his set. He had the Tyler Super Towers, which is one set down the lineup of Tyler speakers [Not the Decade Series]. I have the Linbrook Signature Reference. Before the crossover upgrade on his speakers, I had always gone home preferring my speakers by a longshot. After the break in of his going external and upgrading the crossover components, I would go home and hold my head down in shame. At one point, I didn't even listen to my speakers for three weeks after hearing his. The difference was awe-inspiring.

He wasn't there for this but I did tell him about what happened when I sat down to listen to his broken in and upgraded speakers with one of my reference songs. I played a Telarc version of "Olympic Fanfare" by Kunzel/Cincinnati Pops. As the song played, goose bumps took over my body like I had just been freshly plucked of feathers and the speakers sounded so good that they lasted 3 minutes after the song had ended. I was there with another friend of mine when this happened and when the song was over, both of us without saying a word, stood up and offered a standing ovation to his system. This was the only time in my life I had ever done this and while it may sound silly to you without actually experiencing this, it was beyond warranted.

Tom

 

[microstrip]

Tom,

You are in a dangerous zone - upgrading the components of an existing passive crossover will change the sound quality of the speakers. But not all choices will result in improvements - you can easily create an unbalanced sound. A good friend of mine rebuilt the whole very complex crossover of the big Dynaudio Consequences, but as they already had air core coils, only had to replace the capacitors and resistors. After a long burn-in phase he is now very pleased with the result. As you seem to know someone who has large experience and some success carrying these upgrades probably the best advice would be to stick with his advice.

But just in case you want to play the lottery I give my very biased advice - use Mundorf capacitors and Jantzen coils - Sonus Faber uses them with great success in their new speaker lines. But I would avoid mixing components from too many brands - do not ask me why, but I do not like this option.

 

[DonH50]

Hi Tom,

These links might help:

http://en.wikipedia.org/wiki/Hysteresis_loop
http://en.wikipedia.org/wiki/Inductor

As to using an active crossover, or one of any type before the power amps, should provide the best sound but requires the most care in design. Because you can fiddle with something does not mean you should; I suspect too many folk head down the tweak path of life when they should dial it in and then sit back and listen. It does require enough knowledge to decide upon the crossover parameters to use in the line-level design (which may be the knowledge of a techie friend), and willingness to bypass the speaker's crossovers (which may involve modifying the guts of the speakers). It is hard to describe how good a lot of speakers can sound when the drivers are driven directly by the amplifiers, and a good crossover that matches everything in time and frequency is used that also minimizes interaction among drivers (IMO one of the major failings of typical crossovers is that they are too low in order to keep from overlapping in frequency drivers more than they should be, and higher-order means more loss in general and a loss of speaker control by the amplifier).

IME replacing the inductors with big (high-current, low-DCR) low loss cores provides the greatest benefit, although others have commented upon the capacitor changes as well. Not sure if the stock caps have high ESR or what; I know a lot of stock inductors have high DCR that affects speaker control. years ago I used to see electrolytics used in crossovers and there might be some benefit in replacing those; I am not sure they are common anymore.

I do hear comments about long break-in time after replacing speaker crossover components but do not understand why that should be the case. A few minutes of operation at most should "break in" the capacitors and inductors in a crossover. I personally suspect it is more a case of us adapting to the change.

HTH - Don

p.s. DCR = DC resistance, the resistance of the wire. ESR = equivalent series resistance, a combination of the wire and other loss factors

 

[garylkoh]

Welcome to the loudspeaker designer purgatory

While iron core inductors exhibit saturation and hysteresis, sometimes they can be used to give a nice "single-ended triode" or "golden" sound to the midrange. You can also use them to soften the sound of some brittle-sounding drivers.

Different inductors sound different, and the better specified ones and more expensive ones may not be better in every respect.

 

[Odyno]

Correct me if I am wrong but the main reason for using core inductor is low DCR that can only be obtained with foil inductor such as Goertz with air core.
If you want to play in your cross over, start by changing resistors. You will be amazed by what you get with better than ceramic resistors.
Don’t forget, ceramic resistors change value with heat...
A local company charge a large chunk of money for the latest incarnation of one of their bigger seller by adding a heat sinked caddock resistor in the crossover

 

[garylkoh]

Actually, that's because those white ceramic resistors are usually wirewound resistors with significant inductance. The good caddock ones have Ayrton-Perry windings which have very low inductance.

 

[Odyno]

As a pro with knoledge can you develop on your finding about crossover parts

 

[garylkoh]

Odyno, the reason I called it speaker designer purgatory is that there is no one "correct finding". I could build a crossover using all the "best components" and still might not like it. There are compromises everywhere. In one spot, a non-inductive wire wound might be the best, in another, a metal film might be better, in a third, a metal foil or even carbon composition. There is even one spot where I use an inductive wire wound because that is the best component for that place.

And that's just a choice of resistors. Capacitors and inductors are two more complexities. And then, you have to add paralleling capacitors.... you can also series/parallel capacitors to give a different flavor of sound.

The problem is that the usual measurements - FR, phase, impulse, etc. on the crossover won't be able to detect the difference between a carbon composition and a metal film resistor for example. I still end up listening to the final result.

 

[DonH50]

I am getting confused...

DCR is determined by the effective cross-sectional area and material used to make the inductor's coils, air or iron (or whatever) core material does not matter.

Ceramic is an insulator, so the resistive material is something else... I assume wire wound on a ceramic substrate is what you mean by "ceramic resistor"? Or, a thin-film or thick-film resistor on a ceramic substrate?

Tempco depends upon the material used to make the resistors (or inductors). You can get 0-Tc resistors though they get pricey, especially for high-power applications.

Carbon film resistors usually have lower noise than metal film, and carbon comp the highest noise. Whether that matters in a speaker crossover I tend to doubt.

Gary, I wonder if part of the reason for the differences you hear are related to the tempcos causing resistance changes during operation?

 

[garylkoh]

I am getting confused...

DCR is determined by the effective cross-sectional area and material used to make the inductor's coils, air or iron (or whatever) core material does not matter.

Yes - DCR doesn't change with core material - hysteresis does, and so does saturation. I wasn't referring to DCR of inductors - Odyno brought up different resistors.

Ceramic is an insulator, so the resistive material is something else... I assume wire wound on a ceramic substrate is what you mean by "ceramic resistor"? Or, a thin-film or thick-film resistor on a ceramic substrate?

Usually, when a loudspeaker designer refers to "ceramic resistor" I assume they refer to what I call "white coffin"-type wire wound on a ceramic substrate.

Tempco depends upon the material used to make the resistors (or inductors). You can get 0-Tc resistors though they get pricey, especially for high-power applications.

Carbon film resistors usually have lower noise than metal film, and carbon comp the highest noise. Whether that matters in a speaker crossover I tend to doubt.

Gary, I wonder if part of the reason for the differences you hear are related to the tempcos causing resistance changes during operation?

Possible - but we over-specify resistors by such a large extent that I doubt tempco makes much of a difference. But I do hear the differences between resistors of different construction even with the same tempco.

 

[DonH50]

Sorry Gary, my post was a catch-all, not targeting any single poster.

Your findings with resistors in speaker crossovers I find perplexing but should probably be the subject of another thread or off-line discussion. I'd love to nail down the cause. Unfortunately, I got a last-minute gig that is going to suck up roughly 143% of my spare time this week...

 

[treitz3]

Hello, gentlemen. Please feel free to discuss the findings with regards to resistors here. I can always change the title of the thread to reflect the changing conversation. Just an FYI, everything within the crossover network will be upgraded. I will still have the stock crossover network to fall back on if in the unlikely event, I do not like what the end result is. Part of the decision for going external with the crossover is so I can quickly change out individual components within the crossover without having to tear the speakers apart. It started out with me just wanting to isolate the crossover network but the more I thought about it, having the crossover be external for the ease of swapping out parts just makes sense.

For resistors, I'm currently thinking of Deuland Graphite and Mills. For capacitors, I'm looking at the higher end Clarity Caps, Mundorf caps or something along those lines. The inductors will most likely be Goertz or Madisound air core. Some other things I'm planning on doing are the following;

External Crossovers, Non-Magnetic Enclosure, isolated from heat and/or vibrations
External Enclosure featuring solid Ribbon Mahogany to match my speakers
Heavy Duty, 99.99% Copper, Gold Plated Insulated Binding Post
Internally wired with DH Labs T-14, twisted with the shortest possible path
Enclosures lined with No-Rez (GR Research)
Tweeter surrounded with Diffraction Be Gone
Umbilical Cable from speakers to crossovers DH Labs Q-10
and then I'll finally get around to upgrading my spikes and spiking the hardwood plinths.

At least that's the plan and that plan is subject to change in a moment's notice. It depends on what I learn, what my ears tell me sounds the best and I'm not gonna rush this. I'm planning on sometime in mid-February taking them out of the system and having the guts extracted to make a schematic and start ordering the parts. I expect to be without the speakers for a period of about 1 month. I know it will probably only take about two weeks to get all of the parts in and have everything else done but I'm planning on having delays, just in case. I'm a very patient man when it comes to my audio. Good things do come to those who wait.

I'd like to thank those of you involved with this thread for taking the time out of your lives to comment. It is much appreciated.

Tom

 

[DonH50]

Not sure non-magnetic enclosure matters so long as the coils are on stand-offs. Isolation from vibration can help as the inductors' glue (holding the windings together) does tend to dry out and they can get microphonic over time. I would not twist (+) and (-) together to reduce capacitance and so you have a little more flex in the wires; be sure to leave a loop or at least a little slack for stress relief on all wires (inside the crossover, crossover to amp/speaker, inside the speaker).

Have to think on the resistors but gotta' run to rehearsal... - Don

 

[mbskeam]

Tom,

will your living room look like Angello has been there....

 

[treitz3]

No, sir! My living room shall look exactly as it is....only with isolated external X-overs. The WAF with where I have my system dictates this at the moment. You know this. It will never be like Angelo's pad. No offense to him whatsoever but I also have my standards. Been there, done that. Not for me.

My apologies for those who do not know who I speak of. Mike and Angelo are good friends of mine.

Tom

 

[treitz3]

Actually, that's because those white ceramic resistors are usually wirewound resistors with significant inductance. The good caddock ones have Ayrton-Perry windings which have very low inductance.

Hello, Gary. Which one would, in your experience, offer a better end result? I'm expecting you to say that it is situation and circuit dependent but I thought that it wouldn't hurt to ask. I'll be honest with you, we haven't opened up the speakers yet to see what we are dealing with so I don't even know if the Tyler's even have one like this but I'd like to be prepared when and if the time comes.

Tom

 

[treitz3]

Good evening, gentlemen. I figure that it was about time for an update.

Here's what I'm looking at right now for the crossover upgrades. Once again, this could change at a moments notice but here's what I have so far.

Tweeter Circuit -
Clarity Cap MR
Duelund Graphite Resistors
Madisound air core Inductors

Midbass Circuit along with the Bass Circuit-
Clarity Cap SA *maybe MR (undecided)
Mills MRC50
Goertz air core Inductors and *possibly Jantzen

I have decided that they will most certainly be external crossovers in a non-magnetic aluminum chassis that will have wood sides [Hopefully a hand rubbed Ribbon Mahogany to match the speakers]
I have also decided to go with a custom heavy duty 8-pin wound short distance umbilical cord with complementing aluminum binding post plates with 99.99% copper connections and DH Labs Q-10 wiring. The speaker will be internally wired with DH Labs T-14 and will be twisted with the shortest possible path. The speaker enclosure will be lined with GR Research No-Rez and just the tweeter will be lined with Diffraction-Be-Gone

Here's a really bad picture of the crossover enclosure that I will be using. Not the exact one as mine will have to be custom made but it should provide everybody with what I have in mind. Mine should be quite a bit bigger than this one as we will have 3 separate circuits along with the layout and real estate issues associated within.

Wow, that picture really is bad. The finish of the aluminum has a "swirl" pattern that the photo doesn't reveal that is rather nice eye candy. The expected dimensions will actually be quite a bit larger than the one pictured and I will offer a better picture as soon as I find one. I somehow lost the original link to the company that sells them.

If you are unfamiliar with my speakers, they feature all Seas drivers and here's a stock photo of them if you are interested...

I will be taking the speakers out of my rig next Sunday and expect to [as previously mentioned] be without them for approximately one month. One of the things I would like to have done is to take the tweeter network and much like Wilson Audio does, use a temporary mechanical connection only for the resistor leading to the positive leg of the tweeter itself. We will be getting some lower cost resistors at various values so that I can experiment and find the closest value to fit these speakers and the upgrades to the room. After that, the temporary mechanical connection will be removed and a permanent mechanical and electrical connection will be installed with the preferred value and upgraded resistor. Most likely a Deuland but I'm still open to options and researching.

Still deciding on what type of footing or isolation I will be using for the crossover enclosure but I figure that they will more than likely be spiked and I'm toying around with the concept of an actual wall mount with custom isolation implementations. More on that later.

So, I've got less than 7 days to enjoy what I currently listen too. Then they will be off to be recorded in a studio in stock form the following Tuesday. Another recording will be made right after the initial upgrades have been performed [using stock values] and a final recording in the same studio with the same exact parameters [placement, equipment, volume, etc...] after the final resistor is choosen for the tweeter network and the speakers have had a chance to be broken in and settle down a little bit.

I'd like to thank those who have responded in this thread for their opinions and advice. It is certainly most appreciated and I have to say.....this sure is a lot of fun!

Welcome to the loudspeaker designer purgatory

Glad to be here!

Tom