First time DIY XO build, random questions

[Alex2507]

Please do not fall for this nonsense.

But whatever happens, keep those wires off the floor for higher lows.
That's 60% of your sound right there.

http://imgur.com/bNJhQb2

 

[Verdinut]

But whatever happens, keep those wires off the floor for higher lows.
That's 60% of your sound right there.

http://imgur.com/bNJhQb2

There's another advantage with those. The mice can pass under them !

 

[Russdawg1]

I know the site... exactly.
People like that need to learn more... humility. (I think that is the word which suits that site best.)

I'm taking notes here, guys! Don't think I dropped off the face of the earth.

OK:
Please discuss with me the importance of Inductor AWG. My understanding is that in the Woofer Circuit, you want Lower AWG, especially for the main Inductor. Again, as I understand things, the Inductors in the Tweeter Circuit can be 20awg, and maybe also any Shunts (proper usage?) in the Woofer circuit.

Thank you, All!

Hi Ryan, me again.

May I suggest downloading a copy of VirtuixCAD onto a windows PC of yours?

You can model crossovers with driver responses, even polar responses, directivity, etc. it’s useful for designing crossovers from scratch, which I know is what you aren’t doing, yet But it would be useful for you to play around with values so, for example, you can see what the effect of DCR in inductors can do with the response shaping. And there’s plenty more stuff you can learn!

I have found that I learn more, doing, than listening or watching.

You’d be able to learn quite a bit from it IMO, and it’s a good tool to get under your belt ASAP, since I know your DIY projects are going to begin very soon after this little mod you are doing right now

Enjoy, and have fun!

But to answer your question, low gauge for high power (generally woofer circuits), high gauge for not as high power (tweeter and circuits). This is a generalization obviously Higher power usually goes to the lower efficiency driver, generally the woofer lol.

 

[ryanosaur]

a windows PC of yours?

What is this of which you speak?
A windows PC?

 

[Russdawg1]

What is this of which you speak?
A windows PC?
View attachment 33148

A PC that runs the Windows OS?

Lol. I think there’s supposed to be a joke that I missed sorry.

 

[TLS Guy]

I know the site... exactly.
People like that need to learn more... humility. (I think that is the word which suits that site best.)

I'm taking notes here, guys! Don't think I dropped off the face of the earth.

OK:
Please discuss with me the importance of Inductor AWG. My understanding is that in the Woofer Circuit, you want Lower AWG, especially for the main Inductor. Again, as I understand things, the Inductors in the Tweeter Circuit can be 20awg, and maybe also any Shunts (proper usage?) in the Woofer circuit.

Thank you, All!

This is where crossover design gets interesting. You do need a low resistance for an inductor in series with a woofer. If the resistance is high you will increase the Qts of the driver and upset tuning. The inductors going to ground will increase the Q of the filter as resistance increases. Now this may be something you want to do on occasions and this can save the cost of a resistor. You should always add the resistance of the components into your crossover model.

The next issue is component size and wire gauge and resistance in low crossover points in three ways. As the crossover point goes down then size of components and cost increases greatly. For the inductors if you get to a point where the size of an air core inductor gets out of hand and the series resistance is too high. Then you have to accept the core hysteresis of a laminated iron core inductor. Do not use a solid iron core, the hysteresis distortion is too high.

Really the lower limit of a passive crossover is about 400 Hz. Below that the design is going to be compromised severely by the above issues. With some driver combinations this may happen above 400 Hz. When component values are getting out of hand it is time to switch to electronic crossover and bi and tri-amping.

I personally don't trust non polarizing caps to be truly linear and avoid them. If the cap values become impossibly high without resorting to them it is time for active electronic crossovers.

If you must use a non polarizing electrolytic caps, then best practice require they be by passed with non electrolytic cap to increase linearity.

If an electrolytic cap is leaking it needs replacing period. There is zero justification for leaving a leaking cap in circuit.

Lastly I don't think it is acceptable to have violent swings of impedance and phase angles. If there are there is no guarantee you don't have an amp buster on your hands. Certainly you can not be certain that an amp will behave like it did on the test bench. There are techniques for smoothing this out if needs be, but this gets into complicated territory. Peter Cooke founder of KEF was a great advocate for impedance corrected designs. This was a much bigger issue back then as it was the tube era and those amps have a high source resistance and will follow an impedance curve.

Lastly if you have a dip in impedance below the DC resistance of the drivers, it is back to the drawing board. That design is then certainly, and not may be, resonant. It will sound bad and be a real amp buster. There are actual clowns who put this sort of rubbish on the market.

 

[ryanosaur]

I ran out of time today to pull the XO from the Speaker I have and compare it to the schematics for the original and the revision. This will have to wait until later in the week... Hopefully I can Pop the Woof on Wednesday and see what I have to work with... if any parts are able to be salvaged and moved over to my XO build. Looks like the entire woofer circuit is a match though; no change to the XO point which iirc is at 1900Hz.
It is supposed to be a 6ohm speaker and his designs are generally considered friendly as far as impedance and such is concerned.

Eager to see what parts were used in the original vs what is available to me now. I started a test cart at PE to see how much this is running...
Ain't pretty right now.

 

[Swerd]

Please discuss with me the importance of Inductor AWG. My understanding is that in the Woofer Circuit, you want Lower AWG, especially for the main Inductor. Again, as I understand things, the Inductors in the Tweeter Circuit can be 20awg, and maybe also any Shunts (proper usage?) in the Woofer circuit.

I agree in general with what Verdinut and TLS Guy said about inductor AWG.

I especially agree with the idea that using a passive crossover designed to cross at 400 Hz or lower, is impractical. The inductor becomes huge, expensive, and is a significant source of heat.

In practice, here's what I like to use:

For woofer coils in 2-way speakers, I prefer using 16 AWG air core inductors sold under the Sidewinder name by Madisound. They seem to be well made and usually are lower priced than similar products sold by other competitors. There are 12 and 14 AWG air core inductors made with flat copper foil instead of wire, but I find they are much too expensive. And I wonder if the flat foil makes for an audibly better sound that standard wire.

For smaller mH values, I use smaller 18, 19, or 20 AWG air core inductors, also sold by Madisound.

For larger woofers, such as in 3-way speakers, I have no problem using 15 AWG Steel Laminate inductors sold by Madisound under the Sledgehammer name. They are inexpensive, develop very low DCR, and are an effective alternative to large air core inductors. I am told that steel laminate core inductors don't suffer from audible electromagnetic saturation problems, unlike the similar looking iron core inductors.

 

[Swerd]

But whatever happens, keep those wires off the floor for higher lows.
That's 60% of your sound right there.

http://imgur.com/bNJhQb2

Those risers made with large grain spruce or pine will produce poorer sound quality than risers made with denser grained woods such as maple . Even the finish makes a difference!

Similar Kool Aid is spouted about the flooring materials, especially how nylon carpets sound worse than natural wool or silk carpets . I know nothing about marble or terrazzo floors.

 

[ryanosaur]

I don't think we gotta worry about Ryan slipping down that rabbit hole!

Tho I get why you posted it. It's good info to have out there for folks looking into the subject!

Took me a while to find this:

https://www.mundorf.com/pdf/Mundorf_Fidelity_Components_Catalog.pdf

Reading some of the descriptions on pgs 7-8 is kind of fun.

And I know I really need those Supreme Evo Silver Gold Oil Caps to get the best sound of that $20 Dayton Woofer!

 

[Verdinut]

I agree in general with what Verdinut and TLS Guy said about inductor AWG.
The inductor becomes huge, expensive, and is a significant source of heat.

I doubt that a huge air core inductor would become a significant source of heat. For example, Solen make 8AWG air core inductors which would never heat significantly even at high SPL in any practical situation:

404 - Solen

solen.ca

 

[Alex2507]

Took me a while to find this:

https://www.mundorf.com/pdf/Mundorf_Fidelity_Components_Catalog.pdf

Reading some of the descriptions on pgs 7-8 is kind of fun.

And I know I really need those Supreme Evo Silver Gold Oil Caps to get the best sound of that $20 Dayton Woofer!

Man, that pic you see upon clicking that link makes me want to do so much better. In the DIY world, I do alright looks wise. Of course we expect better from you. Good luck,

http://imgur.com/8ne6xay

 

[ryanosaur]

Man, that pic you see upon clicking that link makes me want to do so much better. In the DIY world, I do alright looks wise. Of course we expect better from you. Good luck,

http://imgur.com/8ne6xay

Ya, I really like looking at the XOs that Troels builds:
Like this:

ATS-4

Or this:

Illuminator-5

www.troelsgravesen.dk

That last one is fun because you can see his "working XO during the build and test phase... then down at the bottom, his finished XOs.

Hopefully when I'm done, it'll be all "oohs" and- "ahs!"

 

[Verdinut]

Ya, I really like looking at the XOs that Troels builds:
Like this:

ATS-4

Or this:

Illuminator-5

www.troelsgravesen.dk

That last one is fun because you can see his "working XO during the build and test phase... then down at the bottom, his finished XOs.

Hopefully when I'm done, it'll be all "oohs" and- "ahs!"

But don't make the same mistake as Troels with some of his crossovers. Two inductors too close one from the other will affect the overall results. Inductors generate their own magnetic fields. It is recommended that they be mounted at 90° angles to each other and that they be at least 3 inches apart.
Also, as drivers can generate strong magnetic fields, the inductors should not be installed too close.

 

[Swerd]

I doubt that a huge air core inductor would become a significant source of heat. For example, Solen make 8AWG air core inductors which would never heat significantly even at high SPL in any practical situation:

404 - Solen

solen.ca

Are you really recommending a 15 mH, 8 AWG inductor that costs $462.97 (Canadian)?

Or, are you saying that such a monstrosity would not overheat? Overheat, no – Overkill, yes.

It has a DCR of 0.53 ohms. In comparison, a Madisound Steel Laminate 15 mH, 15 AVG inductor has a DCR of 0.505 ohms, and costs $37.70 (US).

 

[TLS Guy]

I ran out of time today to pull the XO from the Speaker I have and compare it to the schematics for the original and the revision. This will have to wait until later in the week... Hopefully I can Pop the Woof on Wednesday and see what I have to work with... if any parts are able to be salvaged and moved over to my XO build. Looks like the entire woofer circuit is a match though; no change to the XO point which iirc is at 1900Hz.
It is supposed to be a 6ohm speaker and his designs are generally considered friendly as far as impedance and such is concerned.

Eager to see what parts were used in the original vs what is available to me now. I started a test cart at PE to see how much this is running...
Ain't pretty right now.

If cost is high it suggests component count is high.

I have to tell you that the real trick of good speaker crossover design is driver selection. The trick is to match drivers where 50% or more of the crossover is done and dusted before you turn to component values. There does tend to be a definite trend to lower quality and high part count in the crossover.

This is not always true but definitely trends that way. So when you state that the cost of the part count is not pretty, my first reaction is that the speaker may be a poor design.

The best crossover is no crossover. Unfortunately we have precious few full rangers. Crossover are basically there as cover for driver deficiencies and limitations.

The two way in wall speakers in my living room only use five pretty cheap components in each crossover. They are really good speakers and among my best efforts ever.

 

[Verdinut]

Are you really recommending a 15 mH, 8 AWG inductor that costs $462.97 (Canadian)?

Or, are you saying that such a monstrosity would not overheat? Overheat, no – Overkill, yes.

It has a DCR of 0.53 ohms. In comparison, a Madisound Steel Laminate 15 mH, 15 AVG inductor has a DCR of 0.505 ohms, and costs $37.70 (US).

No, I would never recommend that huge coil which, apart from being ridiculously expensive, would most likely weigh more than all the contained drivers in an enclosure.

Instead of the Madisound steel laminate inductor, I would recommend the Jantzen Toroidal coil which costs about double that of the steel laminate, but can handle a lot more power without saturation and it has a lower DCR. It's definitely more affordable than the Solen air core:

C-Coil - Toroidal Core - Jantzen-audio.com

C-Coil is wound around a sandwich core, and it has extremely low resistance and can shift a large amount of power without getting over heated.

www.jantzen-audio.com

Parts-Express carries it, and should a value or AWG size not show on their website, it probably is still possible to ask Parts-Express to order it from Jantzen. Several years ago, I was in contact with a Jantzen rep, and he had mentioned that possibility:

Home

Jantzen Audio 16mH 14 AWG C-Coil Toroidal InductorIntroducing new Jantzen Audio C-Coil inductors, specially engineered for bass, sub-woofer and amplifiers. These inductors utilize a toroidal iron core design providing extremely low resistance and high power handling. The core is epoxy coated to...

www.parts-express.com

This 16 mH coil has a DCR of only 0.16 ohm +/- 10%.

 

[ryanosaur]

I'm Back!
Woofers are popped, XO boards pulled, parts confirmed and cross checked to utilize what I can from the original boards... Parts are ordered through PE and Madisound.
I did get to speak with the designer some and confirmed a few things:
The 2mH coil in the Woofer Circuit is an 18AWG Erse Steel Laminate. (Recycling from original XO.) Remaining coils are 20AWG.
The 100uF Cap is an electrolytic, non-polarized, used for size and cost concerns primarily. I decided to stick with this since it would have been part of the build had I bought the higher-tier upgrade to begin with, rather than version I did. No biggy... this is about learning to construct the XO from a schematic, so... I win! And I got the cost of the XO parts down by half!

The original unmodified XO is 8 elements. The first Upgrade utilizing the kit's original drivers is 12 elements (on a 5x7.5" pegboard). The second upgrade is LR4 with 14 elements and premium tweeters.
To view the original and final FR plots is pretty stunning in that the original is said to be "±2dB from 200-12000Hz" (but doesn't look it), with an F3 at 43hz. The final mod I will be installing brings the complete FR to ±3dB from the F3 of 48Hz-20kHz.
The Woofer in this build produces usable bass past 34Hz in my current version, as well as the Final upgraded version. This speaker makes Audyssey think they are full range!

Other than that, I'm hoping to get a chance to play with the layout this weekend. I know there are some random connectors I will need to find in local stores.

Pics or it didn't happen!

 

[ryanosaur]

Sent from a handheld device using a silly little app.

 

[ryanosaur]

Just in from Madisound:

And what’s this?!!!

Woo-hoo!
I won the golden paperclip! Does that mean I get a vip tour of the facility?
Thanks, Madisound!!!




Sent from a handheld device using a silly little app.