Honestly I went a bit overboard with this one and thought to myself, "why not have a stupid capable center channel that can run full range?". So its inspired by a Troels build only I am using a W15 midrange, Excel tweeter, and W22 8" woofers. Its just turned up a few notches. Midrange is sealed off at 0.15 cubic feet and the woofers are seeing 3.15 cubic feet tuned to 30 HZ with a 4 inch precision port. Hope you enjoy the build pics..
Attachments
Thank you for posting. It still is a work in progress. I will admit the baffle width and height have not been really all that ideal for the specific layout of this particular drivers with respect to the midrange and tweeter integration. I switched from the 5.5 neodymium magnet driver over to the regular 5.5 with strontium magnet. Troels even discussed how tricky the neo magnet w15 was to get right on a baffle in order to get a good measurement. I traded off some efficiency (around 2-3 db) by switching to the standard w15 without neo magnets and it measures a bit better also with this baffle.
This was always for me going to be just an active DSP filtered speaker and honestly I think these days anything over a two way can be more cost prohibitive to do than designing and building what could be very expensive passive crossover for a 3 way. Amps and decent clean power are getting more affordable vs some huge value capacitors and other "magical" components marketed to folks with deep pockets that I sometimes just shake my head and laugh about.
The tweeter and mid are crossed at 2.2kh 12db LR. Both in phase. The mid and woofers are crossed at 350 12 db LR but the woofers are wired out of phase. I am getting way more energy from the tweeter mounted to this baffle than the mid so it is generously attenuated down some -12 db. DSP makes this easy without using a great deal of costly resistors.
It measures pretty good with these settings but the mid from time to time exhibits some resonances and colorations I feel that are due to the mid enclosure not having enough mass. It was a small sealed compartment at around .2 cubic feet that I honestly felt would be plenty inert at building with .75 inch plywood. However, it needs more mass so out will eventually come the foam lining and in will go some non drying modeling clay that I use for adding mass to the walls of this sort of thing.
On my tower speakers I used the same .75 inch plywood but knew more mass would be needed to combat resonances so I lined the insides of those cabinets with .5 inch mdf and that has worked nicely. MDF has mass whereas plywood has rigidity with low mass.
Thanks again for posting!
This was always for me going to be just an active DSP filtered speaker and honestly I think these days anything over a two way can be more cost prohibitive to do than designing and building what could be very expensive passive crossover for a 3 way. Amps and decent clean power are getting more affordable vs some huge value capacitors and other "magical" components marketed to folks with deep pockets that I sometimes just shake my head and laugh about.
The tweeter and mid are crossed at 2.2kh 12db LR. Both in phase. The mid and woofers are crossed at 350 12 db LR but the woofers are wired out of phase. I am getting way more energy from the tweeter mounted to this baffle than the mid so it is generously attenuated down some -12 db. DSP makes this easy without using a great deal of costly resistors.
It measures pretty good with these settings but the mid from time to time exhibits some resonances and colorations I feel that are due to the mid enclosure not having enough mass. It was a small sealed compartment at around .2 cubic feet that I honestly felt would be plenty inert at building with .75 inch plywood. However, it needs more mass so out will eventually come the foam lining and in will go some non drying modeling clay that I use for adding mass to the walls of this sort of thing.
On my tower speakers I used the same .75 inch plywood but knew more mass would be needed to combat resonances so I lined the insides of those cabinets with .5 inch mdf and that has worked nicely. MDF has mass whereas plywood has rigidity with low mass.
Thanks again for posting!
If this means an acoustical 12dB/octave LR response, then the tweeter and mid should be connected in opposite polarity to each other to get flat response and best phase relationship.
If you get flat response and good phase tracking with the drivers have the same polarity, then it is not an acoustical 12dB/oct. LR (called LR2) roll-off but a 24dB/oct. LR (called LR4).
Or if this is really an acoustical LR2 and having a flat freq response, then your phase tracking is bad between the mid and tweet and your xo point is not at -6dB.
I write this because these drivers deserve the best.
Last edited:
Second order is 12 db my friend whether its butterworth or Linkwitz. Why do you think the mid and tweet have to be wired out of phase? So my mtm towers that also use a second order LR that both measure flat +/-( 1-2 db) across the range shouldnt be wired also in phase? Let me ask you another question. Why do I get a distinct dip at the crossover frequency anytime I reverse the polarity of either the mid or the tweeter and then when wired back to normal polarity the response flattens? This is a common place practice amongst any builder whether they are using a DSP or measuring for the modeling of a passive network which reveals the best phase relationship. I think unfortunately you are confused and need to do some more research....
As i wrote, if you have flat frequency response with LR2 and the drivers have the same polarity then your phase tracking is not good. This is all math. LR2 or BW2 is not the same, having huge difference between them.
True LR2 at it's best, have a big null at xo point, if you use the same polarity for the drivers.
True LR2 at it's best, have a big null at xo point, if you use the same polarity for the drivers.
Here is a response graph of an mtm I built. Nevermind below 150 HZ where the room modes start to show themselves but this is a 2nd order LR (12 DB) at 2kh and maybe Im just a fool but that looks to me like a pretty good "phase relationship" between the mid and tweeter which happen to have regular polarity.
Attachments
Where are your measurements? Where are your builds? Which speaker design firm do you work for? I very much want you to go around emailing some of the best speaker manufacturers in the world and tell them how wrong they are also for using LR2 designs with the same polarity and then tell them how crumby their designs are.....(shakes head) I swear internet armchair experts that want to tell everybody how wrong they are when they can't get anything right themselves. Please go troll somewhere else.
This graph reveal nothing about the phase relationship between the drivers. Just flip the tweeter polarity and measure for yourself. LR filters have a big null with wrong polarity at xo point if the phase is good.
Last edited:
Sorry, I didn't want to attack you, I just wanted to clarify a few things about your loudspeaker.
This guy posted on here and is also using an LR2 with his mid and tweeter both in phase. You need to go on his thread "creating the best three way speaker in the world" and tell him what an ill-educated fool he is and how wrong his speaker is designed and constructed and how you just know that it must sound absolutely terrible....So I expect you to go over there also ok.....Deal?....Go on right ahead....
Attachments
Sorry, there is no evidence that this is a real LR2 design. No single measurement shows that it was or was executed accurately.
If you think i am ruining your thread, you are wrong. You have the opportunity the defend yourself, just show your LR2 measurements and you are good and i lift my hat.
Maybe there is, but in loudspeaker design, we always talks about the acoustical response, right?
Yes, I know that not everyone is talking about the acoustic response, but ... you know that it doesn't mean anything in this form.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.