Bottlehead Forum

General Category => General Discussion => Topic started by: John Roman on July 13, 2012, 06:22:45 AM

Title: signal path
Post by: John Roman on July 13, 2012, 06:22:45 AM
I'm curious about the potential effects of signal degradation within a speaker cabinet. Specifically, what effect can a large magnet (sub driver) have on internal wiring? I've installed some binding posts at the bottom of the cabinet and the internal wire leading to both the sub crossover (located inside the cabinet) and mid/tweet crossovers  is:

Jantzen Audio 6N Solid Core loudspeaker cable - 99,9999% pure 6N oxygen free solid core copper 2x 1,00mm2 / polyethylene insulation / twisted cable design / cable core wrapped in polyester tape  This is the blue wire shown in the pix.
I was considering using this: AURIC 18 awg Black Stranded Copper Hook-up Wire. Stranded or solid core, which is best?

Does this seem appropriate, overkill or what? The crossover wiring is all internal and along the back wall of the speaker under the acoustic padding/foam.
Please see pix.
Title: Re: signal path
Post by: Jim R. on July 13, 2012, 07:43:36 AM

No real effect from the magnets as long as the wires can't move in the magnetic field.  If I were you, the thing that would cause me the most concern is having the crossovers themselves inside the box -- vibration can have an effect on the caps and inductors.  Is it going to and will that effect be audible, I have no idea and no good way to predict -- I'd think the layout of inductors on the crossover board would be far more important, but without experiments it's hard to say and you'll probably be fine just by leaving it all alone.  Generally I prefer solid to stranded wire, but again, the type of wire and dielectric is probably more important.

There are a zillion variables here, so probably best to just go with what you have and then try changing one thing at a time to dial the final sound in.

I've always been a fan of small gauge, high-purity solid copper wire, but Clark now has me rewiring my Nagas with huge (10 gauge) stranded wire -- though I will strip off all the insulation and put a natural, unbleached cotton dielectric on them and then retwist them.  Will that do me for the long term?  I have no idea but it's cheap and easy to do.


Title: Re: signal path
Post by: earwaxxer on July 13, 2012, 02:11:22 PM
Hi John - a couple of thoughts on crossovers. Dont worry about the magnetic fields. What you want to do is hack out all the crap that you dont need. No crossover is always better. The closer you get to that, the happier you will be. The first culprit is the inductor. It looks like you have some good film alpha air core's, that a good start. Not sure about that iron core in there? Better to get rid of them altogether. You may have to tweak your cap values after you do that. If your woofer/mids produce crappy highs then you are stuck with an inductor. I would recommend an 8 or 10awg from North Creek. I have (2) .75mH that I had in my MMG's. If you want to buy them I can give you a good price. They sound better than 12awg alpha air coils.  I obviously took them out. Not sure about the resistors. Try getting rid of them. Better caps could eliminate the need for them.

Play with it. Just make sure you dont feed the tweeter with lows that will bend it out of shape! What I did find, though, was that my MMG's sound MUCH better with the tweeter pseudo-ribbon crossed over at about 1800hz vs. 3300hz stock. You cant harm the woofer with high frequencies so its easy to take the inductor out and see how it sounds. 
Title: Re: signal path
Post by: John Roman on July 13, 2012, 03:21:45 PM
Thanks Jim and Eric,
Eric, when I get these playing some tunes I'll let you know. 
Title: Re: signal path
Post by: John Roman on July 22, 2012, 07:04:57 AM
Hey Eric,
Essentially the crossover design has been thoroughly tested as is. Specifically, the wiring of the crossover has been determined by just adhering to a schematic used by the original designer. I'm trying to refine it and as you say "get rid of as much crap" as possible. To me that includes any, what I would consider, additional wire. I do believe there's a lot to be said for the "less is more" philosophy, especially as it relates to crossover refinement. I've put so much time and effort into every aspect of this design that it's rather over the top. My current goal is to get them complete before some new tubes I ordered arrive, Psvane 6SN7's. Once they get here I know the pace will quicken. I just need to finish them and quit asking questions and let the cards fall, make some choices and see what fly's. So the question remains, do the negative wires for all drivers have to lead to the binding post separately? I'm thinking not but that choice makes a significant difference in how the exposed mid and tweet portions of the crossover will look and perhaps function. Can you say obsessed!!! Looking back I should have just gone with active crossovers and 4 more amps. A pricey step but I believe it's in the works anyway.
best regards,
Title: Re: signal path
Post by: corndog71 on July 22, 2012, 01:13:01 PM
I don't think it matters as long as you're still following the schematic. 
Title: Re: signal path
Post by: John Roman on July 22, 2012, 03:16:37 PM
That's the direction I'm leaning
Title: Re: signal path
Post by: 2wo on July 22, 2012, 04:49:19 PM
I think your ground plan looks OK. I would try to keep the ground buss as short as possible and have the elements connect as close together as practical.

 For instance, RE,your second photo. that small yellow cap could be moved to the right  side of the top inductor. Allowing the ground side to connect to the same node as the the big inductor. And not mess up the clean look...John         
Title: Re: signal path
Post by: earwaxxer on July 22, 2012, 05:52:41 PM
Hey John - I wouldnt worry too much about the wires. I use pure solid core silver that I get from Homegrown Audio. The best idea is to get a good mental grasp on what each component in the crossover does. Experiment with taking it out, changing the value and brand etc. See what seems to give you what you are looking for. There may be limits based on other pieces in the system. Try to figure out what they are. That is the hard part that we all face. For my money and time, I need things to be as simple as possible. There should be ways to simplify just about anything. As far as active crossovers, that seems like a nightmare of complexity to me. You have to take what you read with a grain of salt.
Title: Re: signal path
Post by: John Roman on July 23, 2012, 07:07:03 AM
I'm already on information overload. So thanks for the input everyone and now on with the show. The Psvane's will be here today so time to button this up. Again thanks so much for the help!
Title: Re: signal path
Post by: corndog71 on July 23, 2012, 09:23:31 AM
Consider this example.  This is the crossover for a 3-way speaker.  All 3 sections share the common/negative input.  I've seen versions of this where the 3 sections were divided to enable biamping/ biwiring. 


I also agree wires should be short as possible.  I tend to keep the leads of caps and resistors long if I think they might be used elsewhere but if I'm committing to a given project then I'll keep the leads short and tight.  Those foil leads can be trimmed down but be careful bending them as you don't want to break them off at the coil.

I like to use solid core 16awg wire with PE insulation which I can get pretty cheap from GR Research.  For line level stuff I like Kimber and HGA.

If you think you won't change or upgrade any further then feel free to secure the parts and wiring with silicone or even hot glue.

Here's another example of a 2-way crossover I put together.

Title: Re: signal path
Post by: RayP on July 23, 2012, 10:38:13 PM
I've been thinking about some of the comments about keeping the length of wire, leads etc as short as possible. Out of curiosity, has anybody actually listened and heard differences when you shorten leads.

I understand why shortened leads, star grounding and short buses are very important with active devices in amplifiers etc, particularly for hum reduction, but a crossover is essentially a passive device. Hum is not an issue.

My understanding of inductor placement is that placing at right angles as shown in the photo above is of prime importance, but separation also plays a part. (

In the above link I found the comment about crosstalk interesting.

I doubt that the slight change in resistance due to longer leads and wires really is audible, or even measurable on a standard multimeter.

As one of the posters has suggested, I suspect that there is a lot of leeway in crossover design and you really have to play around with it to get what works for you. I suppose that is part of the appeal of full-range speakers.

Title: Re: signal path
Post by: 4krow on July 24, 2012, 05:13:36 AM
I'm gonna shoot in the dark with a few thoughts. First, if vibration is a concern to you, locating the crossover outside the speaker box is a good idea as was recommended in an earlier post. If that is not possible, then SoundCoat makes a vibration dampening sheet that beats all others. I got mine from Parts Connxion. Secondly, If wire length were an issue, we would all be screwed by use of wire wound inductors in the first place, and no, I haven't ever heard any effects of wire length in the first place. Third, I have used sheets of ERS to help 'contain the waves' of inductors. or to protect those components that could be affected. BTW, pretty impressive x-overs you have there.
Title: Re: signal path
Post by: John Roman on July 24, 2012, 06:40:41 AM
Hello Ray,
The site you refer is what I've modeled my inductor placement after. I consider crosstalk a key consideration although it's difficult to say just how audible an effect it has. I agree that even a decent VOM would not be able to measure it. It amazes me how commonly inductors are arranged in a manner that seems to ignore crosstalk. I do not know for sure but I'd be willing to bet this is a common practice even in "high end" speakers. That said it leads me to think the overall effect is not so discernible to most people, and most certainly not to these "older" ears. But that does not prevent me from chasing these details down, go figure. The nature of DIY?   Full range speakers would seem to put a premium on crossover arrangement but again how audible is it? The DIY community seems to appreciate these details more. For me it leads to many hours spent, enjoyably so, reading and hopefully learning about this intriguing hobby. Alas, at times it does lead to paralysis and progress tends to grind to a halt. Does that make sense? In the end it sure does give me an appreciation for what it takes to develop a product. I tip my cap to folks like Doc and PJ  for their ability to develop designs that represent a great value. Now off to getting those crossovers completed.......
kind regards,
Title: Re: signal path
Post by: John Roman on July 24, 2012, 06:46:16 AM
Thanks bro, now if I can just get them finished!
Title: Re: signal path
Post by: John Roman on July 25, 2012, 04:30:13 AM
Hello Rob,
Thanks for the pix, they are always helpful. I'm surprised at how compact the crossovers you've assembled are. I've seen this in many rather high end systems. To me this suggests that inductor separation is not an audible topology, at least not for most folks. There are many ways to "skin the cat" and audio is no exception. Realizing how complex some of the details are gives me a new appreciation.
kind regards,
Title: Re: signal path
Post by: Grainger49 on July 25, 2012, 04:48:44 AM

Inductor separation, I am chewing on that.  The inductance is lower than a power supply choke or a plate choke.  The voltage and current in the speaker circuit is the highest seen in the signal path.  That is what gives me pause.  The magnetic field is proportional to the current.  Therefore it seems to me that the induced current from those magnetic fields will have some effect.  Probably the effect will be some muddying of the sound.  

I would put the inductors apart and axis at 90 degrees to each other.

I'm not knocking Rob's layout.  But with space not being a problem, and you are not assembling many crossovers to be assembled one after another, take the space.
Title: Re: signal path
Post by: Paul Joppa on July 25, 2012, 04:59:11 AM
The main thing about crossover inductors is that the good ones do not have a core, so most of their field extends outside the inductor. With an iron, or even just a ferrite core, most of the field is confined to the core.

For the most part you won't realize you are hearing the coupling unless you actually build two versions with different spacing and actually listen to the difference. Hardly anybody does that!
Title: Re: signal path
Post by: John Roman on July 25, 2012, 08:23:04 AM
Grainger and Paul,
Thanks for the helpful input!