[Chewbacca]

Hello Friends!

So this has been something on my mind from before I even ordered my NX-Otica kit, and something I should have addressed before I did my kit... But here we are I know this has been asked before, but I haven't seen any responses:

Is it worthy to upgrade parts that are in parallel on the crossover? AKA parts that are not In-Line, or directly in the signal path.

See example of my crossover diagram for the tweeter below:



You will see there is a resistor, and a capacitor In-Line (in the signal path). Then there is an inductor in parallel (bridging the positive and negative) then another capacitor, inductor, and resistor in parallel. Are these parts worth upgrading to Copper Foil inductors compared to the Copper Erse air-coil inductors (still VERY high quality), Sonicaps (or better) for the Erse cap, and a better resistor? This would be about a $400 upgrade for the tweeter circuit alone... So, for me, not chump change!

Initially, I found it interesting that on the Oticas the Foil inductors were reserved for the mid circuit, but upon looking at it, these are In-Line (in the signal path), so it makes sense to have these fully upgraded.

SO, how much of a difference (if any) is there upgrading parts not directly in the signal path/in parallel? If I'm not using the proper terms, please correct me!

[Doublej]

Danny has a YouTube video where his says all parts matter. I am not a crossover expert but I don't think it whether or not the component is 'inline' or not but rather what the component's purpose is in the crossover.

For example when Danny does crossover reviews of stock crossovers he says iron core inductors are fine when used for low frequencies in a circuit but not good when used for high frequencies.

[Chewbacca]

Hi, Doublej! Thanks for the response!

I've seen all of Danny's videos at least once, and I would thoroughly agree that all parts matter in the signal path, but from some posts of what I read, parallel is "technically" not in the signal path. Or the direct path the audio is going through. If that statement is true (I don't know if it is) then it would lead me to believe that these items in the crossover are doing some sort of feedback to the signal path, but not directly adulterating the actual signal/path - if I'm explaining myself correctly lol.

[Tyson]

With a speaker as good as the NX-Otica, ALL the parts matter.  I'd upgrade everything to the highest level you can (including the internal wiring).  After a lot of trial and error, there's a few parts I've settled on as pretty reliable as upgrades for most gear. 

Inductors - the Jantzen film/wax/foil inductors are easily the best I've used
Main Capacitors - Miflex copper or Vcap ODAM
Bypass capacitors - Duelund JDM (copper or silver foil)
Resistors - Path Audio
Wire - Nimak Pure silver (beeswax treated)

[Chewbacca]

Hey Tyson, thanks for the response! I think in a few years I'm going to go through and completely upgrade the crossover with all these of parts, and maybe rewire as well... which is so sad to think about removing all the beautiful wiring lol!

BUT if anyone could still give any insight on the conundrum of In-Line vs Parallel, I'd greatly appreciate it!

I just think Danny must feel the same way, or be onto something given all the "non-upgraded" parts in his kits (at least on the Oticas) are the items that are in parallel: the non foil inductors, and erse cap.

[Tyson]

Yes, the things in parallel should be upgraded, IME.  Maybe not 'quite' to the level as the in-line stuff. 

For example, in my Super 7 crossovers I upgraded the Erse cap to some nicer film/foil caps and then bypassed those new caps with a .1 Miflex copper cap.

[Chewbacca]

Understood! So, most likely I'll probably upgrade what's currently in parallel with the quality of what's in-line (foil inductors, and sonicaps with miflex bypass caps), then upgrade the in-line to the "best-ish" you can get, alike what you have listed.

Currently, this is more for educational purposes, as I am just absolutely head over heels in love with the current version of my Oticas (with all the GR upgrades), and don't feel like anything is even CLOSE to missing! - I do believe Danny nailed the best price to performance ratio with the items chosen though, and above this point you probably really begin to split hairs.

I just foresee the upgrade bug hitting me in a few years to try and take them from a 10 to an 11 With a speaker this good it would almost be a disservice to them not to squeeze everything out of them!

If anyone else has any other insight or personal experience with this topic, I'm all ears!

[nrenter]

All elements in a crossover are in the signal path. Yes, I understand what you're trying to convey, but it's not conceptually correct.

When a signal hits a parallel junction, some of the signal travels through one of the forks, and some of the signal travels through the other fork. The amount of signal traveling through a given fork depends on 1) what's going on in the given fork and 2) what's going on in the other fork. So, what you claim is NOT in the signal path affects what you claim IS in the signal path.

This is why all parts matter.

[Chewbacca]

Hi, nrenter! Thanks for the reply! That makes total sense to me, I'm just going off of what I've read and the deep diver in me had to confirm or deny the statement

It appears there is a common trend that these items don't have the same amount of impact on the signal itself, compared to in-line. But DO have an impact. Of course, I'm not comparing a dental floss iron core to a foil.

So I guess to greater convey what I was asking or kind of assuming when thinking about it, that with the forks the signals go down: One fork leads directly to the driver itself and will be played (in-line) and the other leads back to negative in-line wire (parallel). So... is the parallel actually played/heard? Do these pieces actually produce sound? Or are they just feedback, resistance, or idk? to the other signal? or does it take away certain aspects of the in-line signal, and just pushes it back to negative to not be used? - Since it's path never actually reaches the drivers.

Does that make sense? I apologize if my density is showing, and I am ABSOLUTELY not arguing with anyone here, and hold everyone on this forum with the utmost respect. My curiosity of truly trying to understand crossovers is getting the better of me

[Chewbacca]

I'm too long winded in my questions...

Basically - Since parallel leads back to negative, will the signal that actually reaches the drivers and be played ever be any part of the parallel part of the circuit?

So, are these items in parallel simply more value based (specific uF, mH, & ohm) compared to what the quality of parts are? If these items aren't actually part of the in-line signal to the drivers, and for example the signal will never pass through the erse cap then to the drivers - is the only real thing it needs is the correct value?

Dang it.. did the long winded again

[Doublej]

https://www.youtube.com/watch?v=RrpIBQ7gW5Q

Watch starting at approximately 8:15. Adding the capacitor in parallel to the driver terminals increases the output of the woofer. How can something not in the signal path do this? Or is it in the signal path?

[Chewbacca]

https://www.youtube.com/watch?v=RrpIBQ7gW5Q

Watch starting at approximately 8:15. Adding the capacitor in parallel to the driver terminals increases the output of the woofer. How can something not in the signal path do this? Or is it in the signal path?

I guess the theory I'm diving into is that these items never reach the drivers themselves. They go from the positive lead to the negative without passing thru the driver. Bad illustration below:

AC Current Positive feed:



AC Current Negative feed:



I guess I just don't know how that a part in shunt (parallel) as Danny calls it can reach the driver. If the current goes in the opposite direction (AC) wouldn't it just go the opposite direction? go back from the negative up to the positive in parallel, and back to the amp? how can it now go back up the signal path, when everything else is coming back down? Wouldn't it just reverse direction? not change direction and go against another current?

So this is what Danny is saying?



How's that possible?

I'm absolutely not saying it doesn't have an effect on the signal... otherwise why on earth would it be there?? I just don't see how these pieces will ever have the direct signal that goes to the drivers go through them. Aren't they more complementary to the in-line signal?

[Chewbacca]

Duplicate post

[richidoo]

Mute circuit (short to ground) is parallel, but it definitely does affect the signal when engaged! LOL
Likewise, any "parallel" element in a circuit bleeds some current to ground, so it affects the signal directly.
Caps on tweeters and coils on woofers are the most influential on sound quality and happen to be in series with their load, but the parallel caps on woofers and parallel coils on tweeters are equally important. The voice coil itself is a parallel element and its resistance and inductance have major effect on sound quality.  Everything matters, so use the best quality parts you can afford.

[nrenter]

No disrespect intended, but it would behoove you to research basic circuit design. It’s not that complicated, but no need to reinvent that wheel in this forum.

But I’ll leave you with this: each element in a circuit has an effect on the signal (and some, different effects on different frequencies). I can tell by your questioning that you think there’s a “load” (aka the driver) and then a bunch of other stuff in the circuit. All of that stuff, in combination with the driver, present the load, with each element playing a specific role, shaping the signal for presentment at the driver.

That’s about as detailed answer you’ll get without walking through each element of the circuit and explaining the role in the crossover.

[Chewbacca]

Mute circuit (short to ground) is parallel, but it definitely does affect the signal when engaged! LOL

Hi richidoo, thanks for the time with your post.

If you read above you'll see that I never debated the fact that the items in parallel have a direct impact on the signal... I think no one debates that this is obvious. My first question was how much does it have an affect on the signal, and how much should you spend on this part of the crossover. It is clear Danny feels this is not where you should spend your money on the crossover, otherwise he would offer upgrades for these items as well. So I'm trying to gain knowledge on this fact - which I feel has been fairly adequately answered.

Second question that arose in this thread, and THIS is what I have been asking recently, and it's comically obvious I have not made myself clear, or I am screwing up some terminology. When I say in the signal path, I don't believe this is understood for how I understand it and was talking about it above. I am not asking if an item in parallel has an impact on the signal. Again, this is obvious. In the signal path to me, when used while talking about a crossover, meant the electrons in the signal path will literally travel through the driver and excite it to vibrations and create sound.

So. As simply as I can possibly ask:

Please explain how an electron traveling through a capacitor in parallel/shunt/short to ground will end up at the driver.

Because, to the best of my knowledge, this is basically what Danny was saying in his video. Or referring to the last picture I posted above. How can the signal go both directions at the same time? This is not how electricity works in my little brain. I have always held the theory of the path of least resistance. There cannot be two directions of one path. Of course the path can be split, and balanced. Thus a crossover.

I'm trying to gain understanding on how this works, and how it's possible these electrons that are flowing in parallel endup at the driver? If it's just "bleeds some current to ground", I don't think theoretically these items are in the signal path of electrons that will pass through the driver. To me bleeding current is not part of a design. Unless this is the design of parallel.

[Chewbacca]

No disrespect intended, but it would behoove you to research basic circuit design. It’s not that complicated, but no need to reinvent that wheel in this forum.

But I’ll leave you with this: each element in a circuit has an effect on the signal (and some, different effects on different frequencies). I can tell by your questioning that you think there’s a “load” (aka the driver) and then a bunch of other stuff in the circuit. All of that stuff, in combination with the driver, present the load, with each element playing a specific role, shaping the signal for presentment at the driver.

That’s about as detailed answer you’ll get without walking through each element of the circuit and explaining the role in the crossover.

Understood - Reply #2 this is basically what I asked. I'm just getting lost when people are basically saying the electrons flowing through an item in parallel will directly pass through the driver. How?

[Jaytor]

Circuits in parallel will affect the signal delivered to the load as much as circuits in series. Think about it this way - if a parallel circuit introduces distortion A for the signal flowing through it, this distortion is being subtracted from the signal being delivered to the load, which means that the load will see the negative of distortion A which is essentially the same thing.

Now, the less current that flows through the parallel circuit compared to the load, the less affect any generated distortion will have on the load, so any distortion created by the LRC notch filter will have less affect than the parallel inductor or the series capacitor and resistor.

[Chewbacca]

Hi, Jaytor! and thanks for the response.

I 100% thoroughly agree, and have understood this to be the case!

SO, this is why I'm so confused when someone would say these items are directly IN the signal that will pass through the driver. They directly have an AFFECT on the signal! NO DOUBT! They add/load frequencies, take away/unload frequencies, affect phase/timing, I get it... BUT the electrons passing through the parallel circuit, that are being taken away from the signal path, will NEVER reach the driver... is that incorrect somehow??

Danny in his video made it very clear he feels this way, or he incorrectly explained this to be the case. Am I wrong in this? How can electrons travel in both directions at once? They can't.

I think this is a very simple question:

How is this possible to happen at the same time?



Start the video at 9 minutes.

https://www.youtube.com/watch?v=RrpIBQ7gW5Q

My quick theory would be you're not pushing these electrons back into the driver itself, you're loading it after the driver on the negative lead, then the positive, back and forth (AC), which will create a "bottleneck" of additional electrons on this path, which will then load everything behind it in the circuit, aka the driver itself. Thus the additional output of the driver.

[Norman Tracy]

Chewbacca askes: “Please explain how an electron traveling through a capacitor in parallel/shunt/short to ground will end up at the driver.” Be careful what you ask for, Norman’s in a TLDR mood this morning. While I wrote the following Jaytor posted the short-n-sweet version.

Ok. First let’s in the interest of keeping the discussion manageable purposely ignore the physics issue of is the signal moving electronics through the wire or traveling as electromagnetic waves. I believe we can set that aside just like most electrical engineers do most of the time.

The first thing is to dissuade oneself of the notion the ground side of the circuit is a black hole into which after the signal has done its job moving the driver or is split down a parallel circuit path it disappears into without a trace. The ‘ground’ side of the circuit is just another conductor. In these speaker crossover circuits Danny has used the convention of labeling it ‘-‘ minus and the other side ‘+’ plus. The sound we hear from the tweeter (or mid or woofer) is due to the difference between plus and minus nodes at the drivers terminals inducing a current to flow thereby exciting the driver into moving and we have a sound emitted.

Next consider that in a discussion on upgrading quality of Rs, Ls, and Cs we are past the simple one dimensional analysis based on the parts’ stated value; 10 ohms, 5 mH, or 0.1 uF. We are considering secondary, tertiary, and onto Nth level characteristics. At this level of the game the Rs, Ls, and Cs and the wires and connectors in the minus conductors all have complex impedances made up of capacitance and inductance and resistance that vary with frequency. Therefore the terminal of the tweeter connected to minus and the parallel divided shunt node connected to minus encounter an impedance on that minus conductor. Considered in the AC time domain (rather that static DC domain) that impedance when current flows into it results in a voltage.

With that in mind let’s run through one mechanism as it plays out in the real world. Consider the dielectric absorption (DA) of the cap in the parallel leg of the tweeter crossover circuit. DA is one aspect of how we measure the imperfect real world behavior of the cap vs. its textbook ideal. DA is the primary reason why plastic film caps are preferred over aluminum electrolytic capacitors in high end audio application. If I charge a cap to 50 volts DC then connect it to the 3 Mohm impedance probe of my oscilloscope it will read 50 VDC. Next I short the leads of the cap together using a heavy gauge wire and voltage drops to 0 volts. Then remove the shorting wire and the ‘scope reads 0.0 VDC, no wait now its reading 0.001, no its 0.002, no now its 0.015 and I’m just watching the voltage slowly rise. What that test observes is some of the charge I stored in the cap when I charged it to 50 volts was absorbed into the dielectric remaining there when I shorted the part. Now reconnected to a high impedance that current has time to slowly leak back out of the dielectric and into the connected impedance where we can measure it. Electrolytic caps exhibit orders of magnitude higher DA compared to film caps. Within film caps Mylar is OK, polypropylene much better, Teflon and other exotics at the top. The specifics of the construction of the cap also play a big role. When the 50 VDC of my lab power supply is replaced with an AC signal derived from a recording of Beethoven’s 9th symphony 4th movement the voltage released from the DA effect lags in time behind the desired signal. So the minus conductor has the desired voltage analog of 9th symphony 4th movement and (curses, foiled again) that pesky time delayed noise voltage off the cap in the parallel leg of the circuit. Finally visualize that solder joint where the parallel leg of the crossover circuit joins the minus conductor. In one direction we have 8 inches of wire to the tweeter’s terminal. In the other direction we have 8 feet (20 vs 244 cm in metric) of wire until the minus leg gets to the amplifier terminal. So in AC World where hifi lives there is ~1/10 the impedance between the DA caused noise/distortion voltage and the tweeter terminal vs the amp terminal. Ah rats, it’s another voltage divider hidden in plain sight. In the real world that simple line in the NX-Otica tweeter crossover schematic shown above actually contains two resistors to represent the impedances of the conductors as I add here. With the 10:1 difference in minus leg impedances now revealed we know 90% of the DA caused noise/distortion voltage ends up on the tweeter’s ‘+’ terminal and 10% on the amp’s ‘-‘ terminal. Finally consider the resistors and inductors and wiring all have their own non-ideal characteristics potential sources of noise/distortion voltages.