[troutsnook]

Hey Dave,

I ran across your site while searching for information on copper foil inductors.  Very impressive speakers - I bet they sound as good as they look!  My question for you involves the use of Alpha Core foil inductors versus something like Solen Perfect-Lay or other simular designs.  I joined the open-baffle crowd last year with the purchase of a pair of Hawthorne Audio Silver Iris drivers.  The stock crossovers of their entry product are well built with what I would call medium-grade components to keep the costs down.  I recently made a new set of crossovers using Alpha Core copper foil inductors and Clarity PX grade caps.  The difference was rather larger than expected and in the case of the foil inductor paralleling the tweeter runs counter to some folks wisdom.  Empirically it works, but there is a lack of general information on the web supporting the use of conventional round wire versus foil in this application.  Opinions seem to be more common.  So, my question is along these lines.  Do you have any hard measurable benefits to using foil over wire in this application?  Listening tests count too.  When the hearing is backed up by science it is always nice.

I appreciate any time you may have to reply.

Thanks,

Robert Pace

[David Ellis]

Robert, this is a very good question that has been addressed on several occasions on speaker building forums.  The Madisound and Parts Express forums seem to get the most attention.  I will do my best to convey my general sentiments and welcome any comments from others on this subject too.  My word is grounded in some degree of experience, but this is mostly a subjective question for the experienced masses.

First, I have found absolutely no measurable differences using my computer based measurement gear (i.e. LspLab, Clio) and my humble meters.  However, I don't have "space-program" quality gear, and these instruments might actually reveal some minor differences.  Certainly there is theoretical substance regarding the shape of the conductors and the density of the windings that should affect changes.  Further, there are differences in the wire used and 1 gentlemen has a fairly good marketing diatribe regarding a low tension winding device that allegedly has an impact.

I find profoundly little difference in among the inductors I have subjectively tested.  I will offer specific comments on 3 inductor flavors. 

Generic ERSE & Madisound Inductors:  Very tolerable, but perhaps very slightly harsh and edgy.  Nonetheless, I could live with them.

Solen Air Core/Perfect Lay:  Very smooth and lush midrange from the midwoofer.  Yep, these are better.  I understand why people like these inductors and prefer them.  I believe they use an OFC copper wire and believe this is the root of the quality in these inductors.

Goertz Foil:  The midrange isn't lush, but remains very clean/grit-free.  The most important aspect is that these inductors seem to present slightly more detail.

Litz Inductors:  I have not tested the Litz inductors and have no desire to do this.  I auditioned some very well regarded litz speaker wire and it was awful.   I fully understand that specific litz wire configuration varies and that the Litz inductors might be tolerable.

I chose to use foil inductors because they are only marginally more expensive, and perhaps they are slightly better.  However, I would be hard-pressed to be 100% successful in discerning in a/b testing when comparing the Goertz inductors to Solen Air Core.  Further, even if I could always discern the difference between the Goertz and Solen inductors conveying a clear winner would be nearly impossible. 

I suspect that in your testing the propensity of impact is due to the capacitors.  IMO, there is a very significant difference in capacitor quality. However, I encourage you to be careful in this realm.  Dang, they can get expensive .

There are a handful of folks who actually know something about capacitors in the USA.  Most of them won't discuss matters with common consumers.  A few of them will chat briefly.  If you would like further eduction on these matters, I might suggest a telephone call with George Short, Chris VH, or Jeff Glowacki.  I have learned to trust Jeff Glowacki (www.soniccraft.com) over the years, but I certainly acknowledge that he is not the only person having accessible knowledge on these matters.  Jeff does have significantly better measuring equipment and has objectively quantified component differences.

Sincerely,

Dave

[jeffreybehr]

There are a handful of folks who actually know something about capacitors in the USA.  Most of them won't discuss matters with common consumers.  A few of them will chat briefly.  If you would like further eduction on these matters, I might suggest a telephone call with George Short, Chris VH, or Jeff Glowacki.  I have learned to trust Jeff Glowacki (www.soniccraft.com) over the years, but I certainly acknowledge that he is not the only person having accessible knowledge on these matters.  Jeff does have significantly better measuring equipment and has objectively quantified component differences.

After falling under the spell of a new local friend and golden-eared audiofool around 1980 and then buying and selling bunches of WonderCaps from Reliable and Styroflexes from Siemens, creating and running a parttime hi-end-audio store in my home in the '80s, replacing various-quality caps and resistors in lots of equipment over the years (decades, actually; I turned 65 yesterday     ), after reading so many positive things about Jeffrey Glowacki here and in the Asylums, I contacted him a few years ago.  I was instantly amazed at his level of knowledge and openness about virtually all things audio.  He has since become a friend, but I can honestly say that his advice has NEVER failed to be consistent with my own observations/opinions.  Said another way, what he says will work does work, and what he says won't work well does, indeed, not work very well.

TY, J-Glo, for all your help and TY, David, for reinforcing my thoughts about J-Glo.  Perhaps I can get him to share his thoughts on the original subject of inductors.

[richidoo]

This is a great topic, I look forward to learning more about inductors because I want to upgrade my crossovers. Thanks for sharing your informed opinion David.

Jantzen describes the advantage of foil inductors thus:
"No chain is stronger than its weakest link. This is especially so when we speak of loudspeakers, where there is often a cost reduction for crossover inductors. The result is distortion and power loss. High-end loudspeaker producers overcome this problem by using foil coils, where the copper is flat instead of round. If you take a 1.6 mm² copperwire and roll it flat, the copper surface area becomes 12 times larger. And why is this an advantage? We know that with rising frequency, the electrons will reach for the surface. This phenomenon is known as power distortion. And this is why we need as large a surface as we can get, so the electrons do not "get squeezed" along the way."

Goertz blurb about inductors:
http://www.madisound.com/manufacturers/alpha-core/goertzinductors.php

Happy Birthday Jeffrey!
Rich

[David Ellis]

(decades, actually; I turned 65 yesterday     ), 

  Congratulations!!!  Surely, the alternative is much worse  . 

We know that with rising frequency, the electrons will reach for the surface.

  Hmmm, I have never heard/seen it expressed in this fashion previously, but I suppose this is reasonable.  I have always seen and convey the inverse statement - that lower frequencies and higher power require a larger gauge cable due to increased saturation.  Hence, the abundance of electrons need a bigger wire. 

And, generally, I completely understand the need for fancy marketing phrases rooted in various scraps of truth. Some of this bothers me, but have grown accustomed to it.  At the end of day, I remain comfortable with my decisions & opinions regarding inductors.  When I turn-on the music it sounds soooo good.  Perhaps spending $10 more inductors is perfectly acceptable - even when the impact is arguably very very insignificant.

Perhaps I can get him to share his thoughts on the original subject of inductors.

  I endorse this avenue of approach, but must provide fair warning.  When you call Jeff, be certain you have a Snicker's Bar ready, because it's going to be a while .  I have never experienced  a short conversation with Jeff .  It is my belief that Jeff spends many nights experimenting with gear because he certainly spends many days on the telephone. 

The two people who have helped me the most in this hobby are Jeff Glowacki www.soniccraft.com  and Dennis Murphy www.murphyblaster.com .  Yep.

[troutsnook]

Hey Dave,

Thanks for your reply and also thanks to others that have replied.  You may be correct about the cap having a larger impact in this case.  One thing to mention would be the original crossover inductors were mounted at the proper 90 degree orientation about 3 inches apart.  My upgraded version ended up with about 9 inches between inductors - also at the 90 degree orientation.  Perhaps this could also be affecting things.  I used foil in both the notch filter feeding the mid-woofer and in the parallel path with the compression driver.  I had heard that there was a disadvantage to using foil inductors to bypass the tweeter although I cannot imagine why that would be the case.  If anything the foil should be better due to the much thinner copper - when compared to round conductors - in the plane that would be most affected by eddy currents.  Further comments are more than welcome.

Thanks,

Bob

[David Ellis]

One thing to mention would be the original crossover inductors were mounted at the proper 90 degree orientation about 3 inches apart.  My upgraded version ended up with about 9 inches between inductors - also at the 90 degree orientation.

I don’t think this is significant.

Several years ago I tried to MEASURE the signal transfer across inductors on the same plane. I found there was no measured cross-talk until the 2nd inductor was laying directly on top of the 1st inductor with the current flowing.

I used a 10 volt SIN wave signal over a 2mh 14ga inductor for the 1st inductor. I attempted to measure the signal using an A/C multimeter. I found the ideal frequency for signal transfer, then attempted to move the 2nd inductor away. It didn't get very far when the voltage dissipated completely.

I must offer that I did NOT test the proximity of the inductors audibly using music, and DO allow there could be something valid in this realm. I must also offer that I continue to mount my inductors in perpendicular fashion whenever feasible. I have nothing empirically that supports this, but the theory sounds good.

Also, I have found that bypass resistors can also be heard in the notch filter portion of the crossover.  I remain somewhat shocked regarding this experience  . 

For most folks, reading some commentary and doing a few experiments satisfies their natural curiosity.  Certainly there is some desire to fully empirically test everything, but then we would never get any speakers built.  I suppose I have tested a fair number of loudspeaker artifacts, but I certainly haven't tested all of them.  I still long to setup a jig to measure cabinet panel movement using an accelerometer.  Then some concreted brew testing would follow.  This is obviously involves hundreds of dedicated hours that I simply don't have.  For now, I think that I'll go read some Curious George to my boys .  It's bedtime.

Dave

Dave

[troutsnook]

Hey Dave,

I had come across this posting regarding crosstalk that pretty effectively shows the effect of proper orientation.  http://www.audioholics.com/education/loudspeaker-basics/inductor-coil-crosstalk-basics.  It appears to agree with you regarding distance - as long as the coils are oriented correctly.

I would like to thank you for taking the time to answer my questions.  It is most gracious of you - especially when you consider that I am not using one of your product offerings.

Bob

[Bob in St. Louis]

My apologies for the intrusion gentileman. But to conserve space would it be possible to shield the sensitive components with a metal box like the ones found covering the transformers on a tube amp?

If so, that would make it easy to make immediate audible AND/OR measureable "A/B" comparisons.

Bob

[Mike B.]

I shared a room at CES with Andy and associate from Rockport Tech several years back. I asked him about his crossovers, and remember asking him if he used foil inductors. He said no, he didn't like them. On the other hand, the pictures I have seen of Magico crossovers show foil inductors. My guess is the difference between copper aircore and foil is subtle?

[mgalusha]

Another page with some inductor interaction measurements: http://www.troelsgravesen.dk/coils.htm

[troutsnook]

Another page with some inductor interaction measurements: http://www.troelsgravesen.dk/coils.htm

Hey Mike,

Thank you for that link.  While I did consider signal coupling from inductor to inductor, I did not consider an inductance change from other coils in close proximity!  Nothing like shifting your crossover points around to ruin your crossover design.

Bob

[Jeff]

Hey Dave,

I ran across your site while searching for information on copper foil inductors.  Very impressive speakers - I bet they sound as good as they look!  My question for you involves the use of Alpha Core foil inductors versus something like Solen Perfect-Lay or other simular designs.  I joined the open-baffle crowd last year with the purchase of a pair of Hawthorne Audio Silver Iris drivers.  The stock crossovers of their entry product are well built with what I would call medium-grade components to keep the costs down.  I recently made a new set of crossovers using Alpha Core copper foil inductors and Clarity PX grade caps.  The difference was rather larger than expected and in the case of the foil inductor paralleling the tweeter runs counter to some folks wisdom.  Empirically it works, but there is a lack of general information on the web supporting the use of conventional round wire versus foil in this application.  Opinions seem to be more common.  So, my question is along these lines.  Do you have any hard measurable benefits to using foil over wire in this application?  Listening tests count too.  When the hearing is backed up by science it is always nice.

I appreciate any time you may have to reply.

Thanks,

Robert Pace

Dave/Jeffrey,

Thank you for your kind words!

Robert,

First, I suspect the cap change made up the bulk of the delta you heard.

1) Linearity.  The value of a coil is generally measured at an industry standard of 1kHz.  If one looks at the value of a CFC  (Copper Foil Coil)  vs. WWC (Wire Wound Coils) from 20Hz-20kHz, the CFC value is more linear.  I once had seen a plot comparing the two by John K.  The delta seen in his plot was only about 25-30% of that seen in mine.  The reason I never posted my findings is because I could not describe the exact "sonic" effect that this would have in every possible application.  There is little sense in pushing numbers if one can not directly prove or disprove the sonic benefits.

2) CFCs have lower DCR vs. WWCs in equivalent gauges.  This is due to a better stack/wind.  The foil allows a better geometry which actually reduces the length of conductor required for said inductance with a given geometry.  In a two-way woofer circuit series to the midbass, this might allow better damping for a tube amp or improve the Q of the driver.  Such that the driver plays lower with better control.

3) CFCs have lower FM distortion.  WWCs exercise quite a bit.  Many have heard capacitors "sing", but what about inductors?  The problem is that the freqs that work these coils the most are generally lower freq.  Wire in the WWCs does not allow audible reproduction of these lower freqs.  Just because the coil is not a very good low freq transducer, does not mean it is not vibrating.  While this is very small because of varnishing and zip ties, it is still greater than the tightly wound CFCs.

4) Thin foil does not have the skin effect seen in solid conductors larger than 23AWG in the audio band.

5) The characteristics mentioned above make the CFC a higher Q component.  This also means that it will increase the Q of any XO in which it is placed.

Now, what does all this mean?  First, it means CFCs and WWCs certainly operate a bit different.  Second, it means that either one may have a more desirable effect in one situation than in another.  I am in agreement to a degree with Dave's  assessment.  I tend to preffer the warmer sound of a Solen PL in series to most mids.  But, the CFC sounds superior more times than not in series to the midbass of a two-way loaded in a bass reflex situation.  Of course, these are broad generalizations.

As for hard numbers telling you when or if to use CFCs, there are none.  If you are lucky, you will find a good guide who will lead you in a good direction.  Read this as, you will have to buy less of your education.  Without hard numbers it is a bit of an alchemy.  This lack of hard numbers leads many calling snake oil.  Many do not believe there is an audible difference in capacitors, wire, resistors, etc...  That would mean the difference you heard yourself is the product of some mental issue, or you are as easily influenced as a thirteen yearold girl   However, I suspect many of these nay sayers may not actually be able to hear the difference (with their situation).  When the day comes that much of this is quantifiable and generally accepted, I wonder if they will suddenly be able to hear it then?

My apologies for the fast and dirty explanation (as I see it), but I better get back to work 

Best regards,
Jeff G.

[troutsnook]

Hey Jeff Glowacki,

Thanks for weighing in on this.  I am quite comfortable trusting my ears, but if scientific data is available to back up the ears - that is always preferred.  The crossover is below.



Both inductors in the new crossover were 16 gauge Goertz Copper Foil and the caps used were Clarity PX grade.  I was led to the PX caps by way of some cap shootouts I came across.  The PX grade was pretty close to the SA grade and not too far behind Mundorf's while being really economical in price.  One question I have for you - and this will pretty much kill any economy of the cap choice - is if the .1 uf sonicaps of yours - that Dave recomends - be of value in this crossover?

Thanks,

Bob

[David Ellis]

Robert,

Jeff (or anyone) is obviously welcome to respond too.

If this were my project, I wouldn't use the .1uf Sonicap Platinum capacitors.  They would completely blow the budget-orientation.  However, I completely understand and have embraced the upgrade process in many components.

My 2c.

[richidoo]

Thanks for the great explanation JeffG, and thanks for the cool link Mike!!

[Tieftoener]

Hello,

In response to Dave's comment...

Quote

    We know that with rising frequency, the electrons will reach for the surface.

  Hmmm, I have never heard/seen it expressed in this fashion previously, but I suppose this is reasonable.  I have always seen and convey the inverse statement - that lower frequencies and higher power require a larger gauge cable due to increased saturation.  Hence, the abundance of electrons need a bigger wire.

...Jeff covers the practicality here...

4) Thin foil does not have the skin effect seen in solid conductors larger than 23AWG in the audio band.

The physics revolve around "eddy currents" that essentially create a loss, thus a reduction in efficiency. It's analogous to the turbulent back currents on the edges of a paddle as you force it broad-side through water. The faster the change in current (higher the frequency), the more loss you have. This is largely why attenuation is greater as frequency gets higher (check any coax cable spec). The eddy currents push the electrons that actually "move" to the outside of the conductor (hence, the name, "skin effect"). This increases the impedance of the cable by effectively reducing the amount of electrons that can actually move - which then correlates to the equivalent of having a smaller wire (higher AWG), thus the higher impedance and attenuation.

As Jeff aptly points out, the math shows that it effectively should not affect typical applications for audio enthusiasts, as we commonly use larger sizes than 23 AWG.

But, as a caveat to that, I do want to state the following:
In the "mathematics" of measuring and calculating performance...
Say we take an amplifier with an incredibly complex feedback scheme, producing just 0.005% THD+N rating... It's well established that it sounds worse* (less realistic) than a simple Single-Ended design with 0.3% THD+N.

* by the overwhelming majority of "high end" reviewers and "ears that I trust"

God has given us INCREDIBLE instruments (our ears) that are capable of determining that 0.3% distortion sounds notably more "natural/realistic" than something that measures 1/60th of that.

For the record, I am not implying that THD is the be-all, end-all measurement to determine performance of a circuit (in fact, I believe my point above is quite clearly the opposite).

So, what I'm trying to get at... The fact that the "math" says that greater than 23 AWG speaker wire within the audio band should provide minimal to none skin effect, doesn't necessarily convince me that there isn't an audible effect. I'll leave it at that, but this is elaborated (indirectly) below.

As far as noted differences in inductors, I have the following comments.
1) I have done limited comparison of Foil wound vs. the "conventional" solid core cylindrical conductors. The comparisons I made were with Goertz foil and Solen Perfect-lay coils of matching wire AWG. The results were mixed. For lower frequencies, higher power applications, I prefer the Solen. For cone mids, depending on the driver, I usually prefer the conventional. For dome mids, I typically gave a slight edge of preference to the foil. It seems obvious, but this appears to be a frequency (and perhaps current load, in my opinion) dependency.
**2) At face value, efficiency would seem to be less for foil, as they typically use a tape insulation, which is notably thicker material than the polyimide extrusion over conventional coil wire. Granted, the lay and wraps are drastically different. In any case, DCR values seem to prove otherwise (that foil is in fact more efficient).
3) I do not prefer iron core inductors above "subwoofer" frequencies... I hear muddying of the mid-range already quite evident, not to mention tweeter ranges. This is usually done as a cost saving measure, in my experience. I do believe that a large size AWG air core is notably superior in SQ, particularly for midbass precision.
4) I typically use "perfect lay" Solen coils, in most applications. With the other air cores from Jantzen and Madisound and the like, I've found only minor differences in sound when comparing directly (matching inductance values, and matching AWG of the core wire). The differences I hear are very minor. One isn't necessarily better, just "different."
5) Comparisons of the same value with varying AWG wire, I find, make significant impact to sound. IMO, North Creek Music is where it's at for coils. Their large AWG (physical, not numerical) coils are my personal favorite.
5A) Budget affects us all. If I had my choice, I'd use 10 or 8 AWG coils on all woofer circuits. In reality, I generally don't use less than (again, size, not numerical) 14 AWG on any series network component.
5B) The larger the inductance, the more I prefer a larger size AWG wire. Unfortunately, this gets exponentially more expensive.
6) As for the comments in #1) above, I personally feel that the capacitors used in the mid and tweeter circuits have roughly 50-100 times the impact on performance. Yes, seriously. That may discount my comments altogether with a lot of folks... but I'm pretty sure at least Dave would agree with me here (not verified, a presumption).

** This is intended to be neither pro nor con - merely an observation.

Given the fact that I do indeed believe that larger size AWG coils sound better than smaller... I don't in any way claim that this proves that the cause is skin effect. But, I believe it at least brings light to the fact that it's possible that our ears are in fact capable of detecting skin effect anomalies. And that about wraps up my thoughts on the matter.

Also of note, in rereading/proofing of this post, I apparently really like to use "quotations."

Best regards,