First time DIY XO build, random questions

[ryanosaur]

Hi all.
So I have these plans for an upgrade. They come from a reputable source. (Due to the potential for these to be offered as a kit, I am not going to divulge the source/project. Suffice it to say that I will be replacing a tweeter in an existing 2-way speaker and building new XOs according to his plans.)
This is my very first time converting schematics to a parts list and shopping for the parts.

I will likely be back with more questions... this is just where I'm beginning. Thanks in advance for any help and advice!!!

I know I can wire multiple Caps together, in parallel, to get to the capacitance value I need.
Using the value of 16.5uF as the example, is it better to simply use a 15uF and a 1.5uF Cap? OR:
Would it be better to wire multiple caps that are the same value/closer in value than those in the first ex, ie 5ea 3.3uF caps?

Pros and cons of Film vs Metallized Poly?

Is there a benefit to higher VDC, say 400 with ±5% tolerance OR 250 with ±1%tolerance?

Again, my thanks!

 

[Russdawg1]

You want lower tolerance pieces, as they will stay closer to the specced crossover.

So in your case, 250 VDC with 1% tolerance will be much better than 400 VDC with 5% tolerance.

VDC is basically the power handling of the component, I’m not sure what the conversion from VDC to VAC is, but I’m sure 250 VDC is fine.

Regarding meeting a 16.5uF capacitor with multiple components, it doesn’t really matter. As long as you have low tolerance components, it won’t have any significant effect. Every one you add in parallel does reduce any possible resistance and increases power handling by some amount, again doesn’t matter.

Enjoy your project!

 

[ryanosaur]

You want lower tolerance pieces, as they will stay closer to the specced crossover.

So in your case, 250 VDC with 1% tolerance will be much better than 400 VDC with 5% tolerance.

VDC is basically the power handling of the component, I’m not sure what the conversion from VDC to VAC is, but I’m sure 250 VDC is fine.

Regarding meeting a 16.5uF capacitor with multiple components, it doesn’t really matter. As long as you have low tolerance components, it won’t have any significant effect. Every one you add in parallel does reduce any possible resistance and increases power handling by some amount, again doesn’t matter.

Enjoy your project!

Good to hear from you Buddy! Thanks!!!

 

[TLS Guy]

Hi all.
So I have these plans for an upgrade. They come from a reputable source. (Due to the potential for these to be offered as a kit, I am not going to divulge the source/project. Suffice it to say that I will be replacing a tweeter in an existing 2-way speaker and building new XOs according to his plans.)
This is my very first time converting schematics to a parts list and shopping for the parts.

I will likely be back with more questions... this is just where I'm beginning. Thanks in advance for any help and advice!!!

I know I can wire multiple Caps together, in parallel, to get to the capacitance value I need.
Using the value of 16.5uF as the example, is it better to simply use a 15uF and a 1.5uF Cap? OR:
Would it be better to wire multiple caps that are the same value/closer in value than those in the first ex, ie 5ea 3.3uF caps?

Pros and cons of Film vs Metallized Poly?

Is there a benefit to higher VDC, say 400 with ±5% tolerance OR 250 with ±1%tolerance?

Again, my thanks!

15 +1.5 uF is fine. I use Solen caps they are good quality and reasonable priced. You can get them from Madisound. How you lay it out is important to avoid mutual inductance.

 

[Swerd]

And here's my 2¢:

I know I can wire multiple Caps together, in parallel, to get to the capacitance value I need.
Using the value of 16.5uF as the example, is it better to simply use a 15uF and a 1.5uF Cap? OR:
Would it be better to wire multiple caps that are the same value/closer in value than those in the first ex, ie 5ea 3.3uF caps?

I don't see any real difference. Do what costs you the least. Remember, if the crossover design calls for 16.5 µF, anything ±10% of that value should work. 16.5 µF ±10% = 14.9 to 18.1 µF. So a single Solen cap at 16 µF would work, but it costs $9.66. You have some latitude. If you insist on getting as close as possible, how about two Dayton 5% MPP 8.2 µF caps at $3.40 each?

Pros and cons of Film vs Metallized Poly?

The differences are primarily cost. I think inexpensive metallized poly caps, such as those sold by Parts Express or Madisound are excellent. In my hands, Dayton, Bennic, Audyn Q4, or Solen are all very good. Dennis likes Audyn Q4 because the leads are sturdier, and they're RED. That was most important to his customers who liked to admire the XO boards. For them, red caps were much better than black .

Is there a benefit to higher VDC, say 400 with ±5% tolerance OR 250 with ±1%tolerance?

No benefit. 250 VDC is plenty, 400 makes for a physically larger cap. The rated failure voltage at AC is usually lower than the DC voltage. Look at the specs to see both the VDC and VAC ratings if you can find them. If I recall, a 250 VDC rated cap has about a 130 160 VAC rating. That should be plenty for passive audio crossover use.

I see no benefit for paying extra for caps with 1% tolerance vs. 5%. Dayton sells both. Every Dayton 5% cap I've measured had the exact value as printed on the label. I wonder if the only difference between the Dayton 1% and 5% caps is their level of inspection.

For arranging parts on a crossover board, pay close attention to the direction and orientation of the electromagnetic fields of the inductors. See this link http://www.troelsgravesen.dk/coils.htm. In contrast, arranging resistors and capacitors matter little if any. Put them where they best fit.

 

[ryanosaur]

Thanks guys.
I was already keyed into the Inductor alignment issue... @Swerd, thanks for the image... that is super useful as a reference!!!

The inductor conversation and your "range of ±10%" in terms of hitting the spec'd values brings up another question:
Does that hold true for Inductors and Resistors, too? (Or just the Caps?)

Best,
R

 

[everettT]

Thanks guys.
I was already keyed into the Inductor alignment issue... @Swerd, thanks for the image... that is super useful as a reference!!!

The inductor conversation and your "range of ±10%" in terms of hitting the spec'd values brings up another question:
Does that hold true for Inductors and Resistors, too? (Or just the Caps?)

Best,
R

I'll just note that @Swerd has always been good at helping with crossover layout, definitely trust what he recommends as it's never been off.

 

[Swerd]

The inductor conversation and your "range of ±10%" in terms of hitting the spec'd values brings up another question:
Does that hold true for Inductors and Resistors, too? (Or just the Caps?)

It holds for all crossover components. I try to get as close as reasonably possible, without extra cost or OCD-like heroics.

That ±10% variation is what I learned from Dennis Murphy & other DIY speaker builders years ago. Ask Dennis if he no longer follows that 'rule of thumb' now.

I do have an inexpensive LC (inductance/capacitance) meter, such as this. It is helpful to check caps, and if I need a specific inductor value that isn't available as a standard size, I order the next larger size, and unwind wire until the inductance is what I want.

 

[Swerd]

I will be replacing a tweeter in an existing 2-way speaker and building new XOs according to his plans.

If so, this means you will build a new crossover board and install it inside an existing speaker cabinet – most likely through the woofer hole.

Before you assemble & solder, test fit a suitably sized board and the largest item that will sit on it. That usually is the largest inductor sitting on its edge. Can the board plus the coil go through the woofer hole? If not, trim the board until it does go through.

If you use air core inductors, such as sold by Madisound, you can trim some plastic off the bobbin so it fits better and/or stands on edge without rolling. In the photo, on those 16g inductors, I had to trim the white bobbin in the back and the black bobbin in the front:

 

[Alex2507]

I use Solen caps they are good quality and reasonable priced. How you lay it out is important to avoid mutual inductance.

Dear Dr. Carter,

I've seen Solen caps recommended often enough to consider them to be the way to go. Parts Express also carries Solen along with Dayton caps and a few other brands. I liked the 250 VDC Daytons for their smaller size and lower price. IIRC I went with a 400 VDC Audyn cap for a particular value because it turned out to be cheaper than the Dayton 250 VDC equivalent. PE had a better selection than Madisound, providing a single source for the xo parts I needed. That was more important than obtaining the absolute lowest price for any one particular component.

My question is about capacitor layout to avoid mutual inductance. I didn't know this was a thing.

TIA,
Alex

 

[TLS Guy]

Dear Dr. Carter,

I've seen Solen caps recommended often enough to consider them to be the way to go. Parts Express also carries Solen along with Dayton caps and a few other brands. I liked the 250 VDC Daytons for their smaller size and lower price. IIRC I went with a 400 VDC Audyn cap for a particular value because it turned out to be cheaper than the Dayton 250 VDC equivalent. PE had a better selection than Madisound, providing a single source for the xo parts I needed. That was more important than obtaining the absolute lowest price for any one particular component.

My question is about capacitor layout to avoid mutual inductance. I didn't know this was a thing.

TIA,
Alex

Its not, I was referring to inductor layout.

 

[highfigh]

Is there a benefit to higher VDC, say 400 with ±5% tolerance OR 250 with ±1%tolerance?

If you ever slam speakers with 250V or 400V, run. Fast.

100W into an 8 Ohm load requires components that can handle 50V- 250V and 400V caps are for power supplies, usually in tube equipment, to supply the plate voltages needed for the tubes to work. Crossovers use non-polar caps, so make sure the ones you get aren't polarized with a + and - at each end.

You should be fine with 50V to 100V caps but the voltage depends on the power amp's output- 250W needs at least 50V but using 100W caps will ensure a longer life through more headroom.

 

[Pogre]

I'm not much help to you, but tagging this thread to watch your progress and hopefully learn a little myself.

 

[Swerd]

More info on the subject of capacitor voltage ratings:

Bennic caps – metalized polypropylene (MPP) caps made in Taiwan & sold by Madisound

Rated voltage: 250 VDC and 160 VAC
Life Test: 150% of rated working voltage at 105°C for 1,000 hours


Solen MPP caps

Test Voltage: 1.5 × rated Voltage for 5 sec.
Rated Voltage: 400 VDC and 250 VAC

 

[highfigh]

More info on the subject of capacitor voltage ratings:

Bennic caps – metalized polypropylene (MPP) caps made in Taiwan & sold by Madisound

View attachment 33122
Rated voltage: 250 VDC and 160 VAC
Life Test: 150% of rated working voltage at 105°C for 1,000 hours


Solen MPP caps
View attachment 33124

Test Voltage: 1.5 × rated Voltage for 5 sec.
Rated Voltage: 400 VDC and 250 VAC

A lot of major speaker manufacturers have used Bennic caps over the decades,

 

[Swerd]

FWIW, the Dayton house brand MPP caps, are said to be made in Taiwan by Bennic, under contract to Parts Express.

I've always treated Dayton and Bennic MPP caps as interchangeable equivalent products.

 

[Swerd]

I have one final thing to say about capacitors in loudspeaker passive crossovers.

Capacitors with the correct value (±10%) according to the crossover design, that haven't undergone drying, leakage, or failure as some electrolytic caps have been known to do, have no effect on speaker sound quality in any major way. In fact, they may not have any audible effect on sound quality at all.

There is a lot of capacitor Kool Aid on the internet – that different types and brands contribute noticeably different sound qualities to speakers. It's all nonsense. In fact, it's about as suspect as claims about speaker wires of exotic makes, materials, and especially prices. The few scientifically valid studies done have never established or supported these dubious claims.

There is one particular web page – I will not name it or identify it – where numerous different capacitors are "compared" and their "sound qualities" discussed at length. The method of testing was never discussed at all. I can only assume that the writer listened to speakers with these various caps and wrote about his impressions of their "sound qualities". He unfortunately assumed that capacitors do contribute different sound qualities, and that he could hear them all. He never established the he or anyone else could indeed hear such differences.

Please do not fall for this nonsense.

 

[ryanosaur]

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!

 

[Pogre]

I have one final thing to say about capacitors in loudspeaker passive crossovers.

Capacitors with the correct value (±10%) according to the crossover design, that haven't undergone drying, leakage, or failure as some electrolytic caps have been known to do, have no effect on speaker sound quality in any major way. In fact, they may not have any audible effect on sound quality at all.

There is a lot of capacitor Kool Aid on the internet – that different types and brands contribute noticeably different sound qualities to speakers. It's all nonsense. In fact, it's about as suspect as claims about speaker wires of exotic makes, materials, and especially prices. The few scientifically valid studies done have never established or supported these dubious claims.

There is one particular web page – I will not name it or identify it – where numerous different capacitors are "compared" and their "sound qualities" discussed at length. The method of testing was never discussed at all. I can only assume that the writer listened to speakers with these various caps and wrote about his impressions of their "sound qualities". He unfortunately assumed that capacitors do contribute different sound qualities, and that he could hear them all. He never established the he or anyone else could indeed hear such differences.

Please do not fall for this nonsense.

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!

 

[Verdinut]

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!

With a passive crossover, since a low frequency filter requires a higher value inductance than is the case for high frequencies, we have to ensure that the DCR of the series inductor connected to the woofer is sufficiently low not to adversely increase the driver Q which would result in a poor transient response.

For a good passive design, the DCR of the series inductor is usually limited to 5% of the driver load impedance. For example, with an 8 ohm woofer the DCR is limited to 0.4 ohm, but for a 4 ohm driver it shouldn't be higher than 0.2 ohm.

Low resistance air core inductors required for filter frequencies below 400 Hz are big, heavy and rather expensive. Most speaker manufacturers use metal core inductors which have lower DCR, are much lighter and more affordable for marketing purposes, but they tend to saturate at high SPL and cause distortion.

Active bi-amping is an attractive alternative to costly large air core inductors and possible sound deterioration when DCR would be too high.