Announcement

Collapse
No announcement yet.

Anyone Built "The Knuckleheads" 3 Way Speakers Please?

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

  • Geoff Millar
    replied
    Thanks for your quick and helpful response Chris, that all makes sense.

    As most of the parts have to come from the spares cupboard, I'm limited to a DC160 or 6.5" Peerless SDS but I take your point about 'improvement'. Luckily I have most of the XO parts except the large value inductors.

    I'll be interested to compare the sound to the Classix, which I really like.

    Geoff
    Last edited by Geoff Millar; 01-07-2020, 01:53 AM.

    Leave a comment:


  • Chris Roemer
    replied
    If you know how sims work, all I did was recreate Dave's filter slopes. The "exact" summations would only be exactly the same if you had all the "z" axes the same. I've NO idea what those are. The woofer looks to be rear mounted - BUT - IT's baffle is "proud" of the TM baffle by maybe an inch?

    The shape of the baffle is NOT a deal-breaker in this design. As I said, Edge shows about the same diffraction signatures on a rectangular baffle approx. the same O.A. height and width of the Knuckleheads (I've included that in my recommended box model).

    For the nearest summation on a flat baffle, I'd probably surface mount the woofer, flush mount the mid, and rear mount the tweeter. None of this is as critical as it might seem (don't know if you'd consider the K.H. to be an "audiophile" speaker anyway - that's not really the point of them). By just listening to the drivers and flipping polarity, you should be able to get them all within 90* of each other ("quadrature" - that's how BW filters are designed anyhow). I've already covered that as well.

    I CAN tell you that a mid (on a 3-way using only a 6.5" woofer) won't offer nearly the same "improvement" over a two-way that it would if a larger woofer was used. Also, 3-way XOs usually cost more than double what a 2-way costs. Instead of an HP & LP filter, you've got TWO HP & TWO LP filters, and the parts cost more, mostly due to the LOWer Fc (in the 200-600Hz range).

    Leave a comment:


  • Geoff Millar
    replied
    Originally posted by Chris Roemer View Post
    I modeled Dave's "complex" baffle shape in Tolvan's "Edge" (not accounting for the cyl heads' rearward offset, OR the fact that I believe he rear mounted those drivers) and compared to a "boring" rect. box.


    My sim LOOKs like a "smiley face" (rather than the straight down-sloping BSC plot) running around 81dB in the mids (600-2k) w/+6dB BSC down to 100Hz, AND nearly +5dB rise from 2k-14k or so. Fcs are about the same, 600Hz and close to 2.5kHz. On a flat baffle, sim indicates a reverse polarity mid wiring. I'd hook up the woofer & mid and flip the mid polarity using your ears to tell you which way sounds best. Then hook up the tweeter and do the same. Good luck.
    Hi Chris, sorry if this is a stupid question: does your revised XO assume that the mid and tweeter are flush-mounted on the baffle, rather than recessed and time-aligned in the original design? Or should I still recess the mid and tweeter please?

    I did try and persuade my 'client' that a "Classix II" (same tweeter and woofer) would also do the trick and both be an easier build and possibly play louder, but he wants a three way!

    Thanks again

    Geoff

    Leave a comment:


  • Geoff Millar
    replied
    Originally posted by Drummer View Post
    This is the first time that I have seen that speaker. I love the visual aesthetics of it. The only thing missing is a knuckle rocker box cover! Glenn.
    Dave also did a speaker called "The Panheads" which used higher quality Dayton RS drivers, similar H-D theme and looked great too. There was also a design called "The Dragsters" in the shape of a V8 , not sure who did that one.

    Geoff

    Leave a comment:


  • Drummer
    replied
    This is the first time that I have seen that speaker. I love the visual aesthetics of it. The only thing missing is a knuckle rocker box cover! Glenn.

    Leave a comment:


  • Geoff Millar
    replied
    Originally posted by Chris Roemer View Post
    I modeled Dave's "complex" baffle shape in Tolvan's "Edge" (not accounting for the cyl heads' rearward offset, OR the fact that I believe he rear mounted those drivers) and compared to a "boring" rect. box. I believe that if you build a box (3/4" or 18mm walls - outside dims) that's 18"h x 10"w x 6-1/2"d that you can basically use about the same XO.
    That comes out to 0.42cf gross. I'd use a 5.5" long hunk of 6"id PVC pipe behind the mid (stuffed) for its chamber. It'd be equivalent to 0.10cf. The woofer ends up w/0.32cf.
    A 2"id x 11" long port tunes to the mid 40s with an F3 in the upper 40s. It can take 25wRMS to 40Hz before exceeding Xmax.

    I tried to recreate his summed FR curve and modded his XO values (a bit) using the SAME woofer (295-305), YOUR BC25TG dome (264-1040) and the Peerless TC9-8 3.5" (264-1062).
    Woofer LowPass: virtually the same. I'd use a "cored" (low DCR) 5.0mH series coil and a 50uF (npe) shunt cap (to gnd).
    HP (BC25): 3n(ohm) series resistor + 7uF series cap, shunt using a 1n resistor and a (#20) 0.20mH coil, then another 7uF series cap, and a 25n resistor (across the dome).
    BP on mid (the most changes in values): 7n series resistor + 1.0mH coil + 20uF series cap, then 2 shunts to grnd; a 15uF cap, and a 6mH coil. SEE NOTE on these coils...
    NOTE: the 1st coil in the RLC string models w/7ohms tot. series resistance. If you used a cored coil (low DCR, like 0.2n) then use a 7n resistor. OR use a 6n resistor witha small air core (#20, around 1.0n DCR). Same idea w/the giant shunt coil: 6n tot. resistance - PROB. a cored 6.0mH coil (.6n DCR?) and a 5n resistor. If you can get ANY cheaper aircore (w/higher DCR) then make the resistor that much smaller.

    My sim LOOKs like a "smiley face" (rather than the straight down-sloping BSC plot) running around 81dB in the mids (600-2k) w/+6dB BSC down to 100Hz, AND nearly +5dB rise from 2k-14k or so. Fcs are about the same, 600Hz and close to 2.5kHz. On a flat baffle, sim indicates a reverse polarity mid wiring. I'd hook up the woofer & mid and flip the mid polarity using your ears to tell you which way sounds best. Then hook up the tweeter and do the same. Good luck.
    Thank you very much for going to so much trouble Chris, that's great.

    I wish I had the skill and gear to make them as designed, but with something like that it just has to look good.

    Geoff

    Leave a comment:


  • Chris Roemer
    replied
    I modeled Dave's "complex" baffle shape in Tolvan's "Edge" (not accounting for the cyl heads' rearward offset, OR the fact that I believe he rear mounted those drivers) and compared to a "boring" rect. box. I believe that if you build a box (3/4" or 18mm walls - outside dims) that's 18"h x 10"w x 6-1/2"d that you can basically use about the same XO.
    That comes out to 0.42cf gross. I'd use a 5.5" long hunk of 6"id PVC pipe behind the mid (stuffed) for its chamber. It'd be equivalent to 0.10cf. The woofer ends up w/0.32cf.
    A 2"id x 11" long port tunes to the mid 40s with an F3 in the upper 40s. It can take 25wRMS to 40Hz before exceeding Xmax.

    I tried to recreate his summed FR curve and modded his XO values (a bit) using the SAME woofer (295-305), YOUR BC25TG dome (264-1040) and the Peerless TC9-8 3.5" (264-1062).
    Woofer LowPass: virtually the same. I'd use a "cored" (low DCR) 5.0mH series coil and a 50uF (npe) shunt cap (to gnd).
    HP (BC25): 3n(ohm) series resistor + 7uF series cap, shunt using a 1n resistor and a (#20) 0.20mH coil, then another 7uF series cap, and a 25n resistor (across the dome).
    BP on mid (the most changes in values): 7n series resistor + 1.0mH coil + 20uF series cap, then 2 shunts to grnd; a 15uF cap, and a 6mH coil. SEE NOTE on these coils...
    NOTE: the 1st coil in the RLC string models w/7ohms tot. series resistance. If you used a cored coil (low DCR, like 0.2n) then use a 7n resistor. OR use a 6n resistor witha small air core (#20, around 1.0n DCR). Same idea w/the giant shunt coil: 6n tot. resistance - PROB. a cored 6.0mH coil (.6n DCR?) and a 5n resistor. If you can get ANY cheaper aircore (w/higher DCR) then make the resistor that much smaller.

    My sim LOOKs like a "smiley face" (rather than the straight down-sloping BSC plot) running around 81dB in the mids (600-2k) w/+6dB BSC down to 100Hz, AND nearly +5dB rise from 2k-14k or so. Fcs are about the same, 600Hz and close to 2.5kHz. On a flat baffle, sim indicates a reverse polarity mid wiring. I'd hook up the woofer & mid and flip the mid polarity using your ears to tell you which way sounds best. Then hook up the tweeter and do the same. Good luck.

    Leave a comment:


  • Anyone Built "The Knuckleheads" 3 Way Speakers Please?

    Hello and Happy New Year from Australia

    I want to build a small-ish three way speaker for a family friend using mainly spare drivers from my stash and don't want to spend to many $$.

    "The Knuckleheads" were a very cool-looking 3 way speaker by Dave Pellegrene which used the DC160, a 3.5" Vifa mid and a Vifa tweeter. The cabinet was modelled on a Harley-Davidson engine and looked great.


    https://www.parts-express.com/project-gallery-speaker-project-the-knuckleheads

    The tweeter and mid are NLA, but it looks like the VIFA BC25TG and TC9 would be close enough for my purposes. I already have the DC160s and tweeters.

    I'd have to build them as a normal, boring box, but don't have the baffle dimensions such as the width and distances between the drivers; I contacted Dave, who very quickly responded but he doesn't have the speakers any more.

    So, as the question says, has anyone built these and if so, would you be willing to share these details please?

    Many thanks

    Geoff
Working...
X