I believe he is using leftovers from his early iteration. I may have been the one to supply starting point values.
dave
The original values were from an old Dayton 3-way crossover and a couple spare Solen 51uF caps I had laying around. Actually, there really wasn't anything if the way of values as I'll I could do was measure the impedance of the iron-cores off of those x-overs. The only known values were in fact the Solen caps and a pair of 10 ohm/10 watt resistors.
Dave did help me out a lot with the original project which I am definitely grateful for.
The current iteration that I am using is still the Solen 51uF caps and a pair of Erse 3mH air-core inductors, nothing more.
I really didn't have a choice but to rely on what others told me to use. I've been on this forum going on 12 years and have conversed with Dave a million times over the years, so I've come to trust him, especially when it comes to something like x-over design and what values to use.
I'm still trying to figure out x-over design and know that the calculators can sometimes be way off from what the values should really be, so they're really not all that useful. Plus the fact that there's not a whole lot of information out there about series x-overs, even though I keep getting told that the values are pretty much the same as a 1st order Butterworth parallel x-over.
Quite honestly, I'm to the point that I might just drop the entire series x-over thing altogether and just go with a standard design instead. There really aren't any advantages of one over the other performance-wise.
At this stage of the game, the only sure thing that I know how to figure out on my own is the baffle step (f3 = 4560/Wb). I know that's typically the best place for the x-over point. Well, that and figuring out the air space of an enclosure, but I've know that for years.
I think that if I give up on the series x-over and focus on the standard parallel x-over, I'll probably start grasping the concept... Hopefully.
Passive XOs are something i try to avoid, and if i am going to use them set everything up so they can be quite simple cause i am not very good at designing XOs
dave
I know as much as you guys keep pushing the whole active x-over deal, I just don't want to do it. I'd much rather go the passive route. As I've said before, I'm not willing to use up more money on cables, equipment and space. Just want to keep it traditional and keep it passive.
To help simplify the x-over, there's three things I can do with the enclosure to help...
Those will reduce the component count and complexity of the x-over quite a bit. I may have to run a resistor for when I upgrade to the 10.3 drivers, and I may need to insert a Zoble network for the woofer. The rest would be for the actual x-over point.
That should be it unless I'm missing something. I'm not sure if I need to have a notch filter on either driver. I don't think I do, but not sure.
To help simplify the x-over, there's three things I can do with the enclosure to help...
- I'll be making the baffle wide to lower the BS
- I'll either have the baffle angled or stepped to compensate for time alignment
- have the Alpair driver in a small midTL sub-enclosure to help with impedance peak
Those will reduce the component count and complexity of the x-over quite a bit. I may have to run a resistor for when I upgrade to the 10.3 drivers, and I may need to insert a Zoble network for the woofer. The rest would be for the actual x-over point.
That should be it unless I'm missing something. I'm not sure if I need to have a notch filter on either driver. I don't think I do, but not sure.
As long as the drivers C-C is 1/4 wavelength or less at the XO frequency that is already pretty much the case.
dave
If I go with a 15" baffle, that would get me down around 300 Hz which is where I want my x-over point to be. And with the drivers only a couple inches apart, I should be more than fine.
So you're saying that time alignment isn't an issue then in this situation?
As long as you stick with 1st order XO and wire the drivers in the same polarity.
Also keep in mind that the rule of thimb is that the XO ends up somewhere between root(2) to 1 x the BS -3dB. Juggling of values used at XO can be useful as they can cause a bump or dip at the XO which can be helped to blend things flat at the XO.
dave
Could you run that part by me again, please?
Does that mean the actual x-over point could range between 150 - 450 Hz?
So in other words, just because the baffle is 15" wide doesn't necessarily mean the BS is going to fall off right at 300 Hz? Or just because the BS occurs at 300 Hz doesn't necessarily mean the x-over point will sound best there?
Also, is a Zoble network needed for the woofer or Alpair? The Vifa/Scanspeak peaks around 20 ohms at 60 Hz, the Alpair around 50 ohms at 40 Hz. However, the Alpair peaks well below the x-over point. Then again, the midTL enclosure is going to help with the other resonance/impedance peak.
Baffle width formulas give a 1st approximation. Yes.
A zobel is for the top end of a driver. You are talking about a resonance trap. You don't even want to think about how big the parts would be to try to deal with the resonance peaks.
dave
Okay, so BS is roughly 300 Hz, but you tune to taste, at or somewhere below BS.
So if I were to add a Zobel, it would only be for the Alpair as it plays all the way out past 20 kHz where its impedance is rising. Since the Vifa will be rolled off between 200-300 Hz, it won't even get up to that point, no requiring a Zobel at all.
I don't know why I was referring to the resonance trap with the Zobel. I know what Zobels are for. Maybe lack of sleep. LOL
BTW, that resonance trap changes once in an enclosure, correct? Any way of coping with that?
Luckily, my amp could care less about the impedance swing. It'll take anything you throw at it. Not only that, like I've said many times before, the simpler I can keep the x-over network, the better.
On thing that still doesn't make sense to me is that my receiver still sees the main speakers as being out of phase from the other three when I run the Audyssey setup. Actually, it sees them as being in phase and the rest of the speakers out of phase. If I reverse the polarity of the main speakers, then it shows ALL of the speakers out of phase.
Is it because of the difference between the series x-overs vs the traditional parallel x-overs in the other speakers?
Keep in mind that I am only running a single cap and inductor per speaker now.
Well, since there's a ton more knowledge out there about parallel x-overs, that's the way I am going to go with this build. However, I still want to keep it as clean and basic as possible, meaning just a simple 1st order design with minimal to no compensation circuits required, again due to design aspects of the enclosure.
With the level of performance I'm getting from this cobbled together speaker system, I'm really excited to see what I can do with a proper enclosure, proper sub-enclosure for the fullrange driver, baffle width, x-over point, etc, etc.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.