Announcement

Collapse
No announcement yet.

3 Way Design - First Timer

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • mortron
    replied
    Yeah I have to admit that the idea of going quad is firmly rooted in active realm. Once I got the MTM playing nice and decide to employ a passive WM crossover I would have to go to two drivers per side.

    I could likely find use for the other 8's if need be ;)

    Leave a comment:


  • rpb
    replied
    Originally posted by mortron View Post
    I was planning on using the drivers I chose because they are on sale, well spec'd, and people have used them to some degree of success. If anything, I've"qualified" them. Not necessarily chosen them.

    The 830869 is an 8" woofer. I am not going to use a 12. I can buy almost 3 of the 8's for the price of the 12. Since I want to go narrower than usual the 8 is the logical choice. Using 4 per side should compensate for anything that is seen as a shortcoming.

    As for the 830991... It's quite smooth through the midrange and am planning on using it sealed. And it's on sale. I had considered the SB Acoustics PFC drivers as well due to their price.

    From an efficiency standpoint, four 8s would be too efficient for my mid and tweet. I just like the sensitivity and surface area of a quad vs dual 8s. Actively crossed it's a non-issue but would be a problem going full passive I suppose. The Neo 3 I have is a PDR. They're only 90.5db efficient come to think of it. I will look into the other Peerless driver mentioned above.

    It's frustrating what is and isn't available here. Free shipping and living stateside would make this a much more affordable hobby heh.
    Sorry for my confusion. I clicked on the 830669 rather than the 830869. I forgot that you were going active between the mid, and woofer. That makes everything a lot easier.






    Leave a comment:


  • mortron
    replied
    I was planning on using the drivers I chose because they are on sale, well spec'd, and people have used them to some degree of success. If anything, I've"qualified" them. Not necessarily chosen them.

    The 830869 is an 8" woofer. I am not going to use a 12. I can buy almost 3 of the 8's for the price of the 12. Since I want to go narrower than usual the 8 is the logical choice. Using 4 per side should compensate for anything that is seen as a shortcoming.

    As for the 830991... It's quite smooth through the midrange and am planning on using it sealed. And it's on sale. I had considered the SB Acoustics PFC drivers as well due to their price.

    From an efficiency standpoint, four 8s would be too efficient for my mid and tweet. I just like the sensitivity and surface area of a quad vs dual 8s. Actively crossed it's a non-issue but would be a problem going full passive I suppose. The Neo 3 I have is a PDR. They're only 90.5db efficient come to think of it. I will look into the other Peerless driver mentioned above.

    It's frustrating what is and isn't available here. Free shipping and living stateside would make this a much more affordable hobby heh.

    Leave a comment:


  • rpb
    replied
    Originally posted by Wolf View Post

    It would likely be fine at that point and slope. No cap required, this is a planar. However, I don't think it's sensitive enough to keep up with 4x 12" woofers.
    Later,
    Wolf
    Two mids will have a hard time as well. I did some sims on the 830874 driver, and it looks like it will be difficult to get 90dB output without the speaker impedance dipping pretty low. The 830874 is 87dB. Putting two in parallel gets the sensitivity to 93dB, but at 4 ohms. Adding a x-over appears to drop that a bit. I didn't do a full sim though. Just the mids.

    Leave a comment:


  • Wolf
    replied
    Originally posted by rpb View Post

    How would the Neo3 perform crossed 4th order at 2.2k? Power handling for tweeters confuses me sometimes. The room is big, and plan is for four 12" woofers.

    Also, does it need to be measured with a cap to protect the amplifier?
    It would likely be fine at that point and slope. No cap required, this is a planar. However, I don't think it's sensitive enough to keep up with 4x 12" woofers.
    Later,
    Wolf

    Leave a comment:


  • rpb
    replied
    Originally posted by Dave Bullet View Post
    Don't let the complexity put you off. My first ground up design was a 3-way, all metal drivers - nasty resonance modes and unconventional baffle geometry. It works quite well given the driver choices I had available (time and location).

    Use reliable input data for simulations. This means applying BSC and diffraction ripple from your intended baffle. The tricky part is impedance data which is best done in box with the intended drivers. You can get away with manufacturer ZMA for xo and be close - there are a couple of exceptions here I know of (targeting FS resonance peak compensation for example if your XO point is within 2 octaves for the driver / shallow slope).

    Post all your analysis / sims for critique.

    Finally - when buying crossover parts - go cheap first off (if shipping is not an issue). and multi-part your capacitors so you can go up / down in value where required. Also chaining inductors is ok if you allow for resistance. Get your XO nailed, then buy the fancy parts if you really want.
    There's also the "old school" way of doing a x-over. Trial and error may be an option. For example. Last night I had the urge to experiment. I had a 2-way with the x-over connected with jumpers on the floor. I set that box on top of another, and made a 3-way. Since the woofer was the same as the mid, I retained the tweeter portion of the x-over. I made a new filter for the mid that matched the tweeter, and a new filter for the woofer mid transition at 350hz. I was done in 3 hours. Nothing was simulated. I don't know what the impedance looks like, but the response is very smooth.

    Leave a comment:


  • Dave Bullet
    replied
    Don't let the complexity put you off. My first ground up design was a 3-way, all metal drivers - nasty resonance modes and unconventional baffle geometry. It works quite well given the driver choices I had available (time and location).

    Use reliable input data for simulations. This means applying BSC and diffraction ripple from your intended baffle. The tricky part is impedance data which is best done in box with the intended drivers. You can get away with manufacturer ZMA for xo and be close - there are a couple of exceptions here I know of (targeting FS resonance peak compensation for example if your XO point is within 2 octaves for the driver / shallow slope).

    Post all your analysis / sims for critique.

    Finally - when buying crossover parts - go cheap first off (if shipping is not an issue). and multi-part your capacitors so you can go up / down in value where required. Also chaining inductors is ok if you allow for resistance. Get your XO nailed, then buy the fancy parts if you really want.

    Leave a comment:


  • rpb
    replied
    Originally posted by Wolf View Post
    I've heard the 8" Exclusive, the 8" SLS x2, and the 10" SLS in the actual Tarkus, as well as higher end offerings from Peerless like Magico and others use in duality.
    Your best bet from above is likely the dual 10" SLS from an Sd/SPL standpoint. Your box will be considerably large though. Something like the HDS8 or even the SDF or XLS/XXLS might get you considerably cleaner performance.

    The 830991 has a reputation to knock out a lot of more costly midranges in terms of HD and response profile. It's a good sounding unit too!

    The BG Neo3 is a bit of a mixed bag, IMO. You can use it as low as 1.8k, but the response is difficult with tradeoffs being either sibilant or dull sound. If you are using the chambered unit, and it's just the flat-cap, glue some mass-damping to the rear of it. If you are making your own chamber, go the size of the element, 4" deep, and use either fiberglass, Ultratouch, or wool damping.

    Later,
    Wolf
    How would the Neo3 perform crossed 4th order at 2.2k? Power handling for tweeters confuses me sometimes. The room is big, and plan is for four 12" woofers.

    Also, does it need to be measured with a cap to protect the amplifier?

    Leave a comment:


  • rpb
    replied
    For mid-range duty, consider the Peerless 830874. I've used it in a 2-way crossed above 4k. A 3k x-over point would not be a problem for a tm, but I'm not sure about a mtm. The vertical response might not be ideal. I have the 830991 GFC Peerless as well. The larger 874 has lower distortion below about 400hz, or so. The 991 beats it slightly in spots above that. (I'm going from memory, so don't hold me to it. They are close over all. ) Since you plan to have an mtm portion, the vertical off axis might be better with the smaller 991. If you plan to really crank it, the 874 may be cleaner, if you cross much lower than 400hz. The 874 has a higher power rating. The 991 is on sale.

    Leave a comment:


  • mortron
    replied
    Sure. The way I look at it, I am building a 2 way MTM and some subs that I can actively cross over. I already own the tweeters, so all I am buying is mids for the MTM. While I am certain there are smaller cheaper options, there isn't really that much investment on the MTM part... About $100. Designing a crossover between the mids and tweet will be my focus.

    As for the woofers, I am just wanting to try something different using small drivers but still want the sensitivity and cleanliness of bigger woofers. Regardless of how they turn out, I can think of several uses for the drivers beyond this project.

    I'm normally the one to be the voice of reason but in this case, I am coming away from JBL L5 and quad 1858 Hframes. A small speaker won't do it for me... I broke myself lol

    Leave a comment:


  • Chris Roemer
    replied
    Wouldn't it be a much smaller investment to try something less ambitious (= expen$!ve), like a smaller 2-way, just to see how it goes?
    "The Space Shuttle wasn't built in a day." - cr

    Leave a comment:


  • mortron
    replied
    This will be my first complete passive speaker build. I've built subs, some open baffle speakesr, a few PLLXOs etc. I have a related background.

    I realize it's hard.

    Leave a comment:


  • Chris Roemer
    replied
    Have you ever designed/built a speaker or crossover before?

    Leave a comment:


  • mortron
    replied
    I am contemplating making it WWMTMWW and having 3 enclosures per speaker to make it more manageable. Then it's two 4.5cubic ft bass cabs. Looking at like a 6-7ft tall speaker in end if I go that way.

    Leave a comment:


  • rpb
    replied
    Originally posted by mortron View Post
    It would be like 9cubic feet or something... just throwing a number there. Dunno if that's right, but it's probably in that neighborhood.
    Just making sure. That's a monster sized speaker. Make sure your work table is very sturdy

    Leave a comment:

Working...
X