AVS › AVS Forum › Audio › DIY Speakers and Subs › DE250 XO designs for 1.0 to 1.2 kHz?
New Posts  All Forums:Forum Nav:

DE250 XO designs for 1.0 to 1.2 kHz?

post #1 of 38
8/7/11 at 6:05am
Thread Starter 
Can someone point me to an example of a XO design where the B&C DE250 is used with an acoustic crossover of 1.0 to 1.2 kHz? I'm familiar with the 3pi/4pi and eWave deluxe designs but all of them seem to fall into the 1.5-1.6 kHz range.

FWIIW, the DE250 (on QSC HPR152i waveguide) is paired up with the kappalite 3015 and I assume the lower XO point would be a good idea with the 15" woofer.

Thanks!

Don
Reply
Reply

AVS Top Picks

post #2 of 38
8/7/11 at 12:13pm
Any xover for those two drivers is ideally going to need to be designed for those two drivers, so there is a good chance no one else has used that combo and xover F, so none exists free. That leaves 3 choices:
1: active
2: measure the drivers and design one yourself using PCD or the like, or
3: maybe adapt the 4Pi xover. Email Wayne and ask and he'll send it to you for free. I haven't looked at this in any detail to see how well it might work though. If you don't already have the drivers, buying what he suggests would be the simplest solution to getting it plug and play.
Reply
Reply
post #3 of 38
8/7/11 at 1:43pm
i think that is the driver used along with the 15tbx100 in geddes' summa speakers, which are crossed at 900hz. you might see if you can find somebody to open up a summa. would be interesting to see what he has going on in there. :-0

"FWIIW, the DE250 (on QSC HPR152i waveguide) is paired up with the kappalite 3015 and I assume the lower XO point would be a good idea with the 15" woofer."

oops. i just saw the second part of your message. that horn loses directivity control around 2khz, so the 4 pi style crossover is probably a pretty good compromise. the geddes horns are about 18 inches on the summas and hold pattern control much lower.
Reply
Reply
post #4 of 38
8/7/11 at 2:58pm
Thread Starter 
I've made a couple of attempts with the 4pi XO but it didn't work as well as I had hoped. Probably because I didn't make the right changes to account for the higher sensitivity of the 3015. I was looking for another example to see what others have done on the with the HF section. Tweaking the LF section is pretty straightforward since it is (or can be) a cap and a coil.

Yes, ultimately I'll give it another try in PCD. I think I've re-read enough to figure out some of the mistakes I made the first time through. I'm slowly figuring out the measurement gear and HOLMImpulse.
Reply
Reply
post #5 of 38
8/7/11 at 4:36pm
this forum isn't too good for such passive advice. have you visited htguide? might get some better help there.
Reply
Reply
post #6 of 38
8/7/11 at 4:48pm
Thread Starter 
Passive advice?
Reply
Reply
post #7 of 38
8/7/11 at 5:01pm
passive crossover design. :-)
Reply
Reply
post #8 of 38
8/7/11 at 5:10pm
This isn't what you're looking for, but maybe it will give you a starting point?

D220ti 900hz crossover for LE14H version of Econowave:

http://www.audiokarma.org/forums/att...1&d=1243017683
Reply
Reply
post #9 of 38
8/8/11 at 6:33am
Quote:
Originally Posted by LTD02 View Post

i think that is the driver used along with the 15tbx100 in geddes' summa speakers, which are crossed at 900hz. you might see if you can find somebody to open up a summa. would be interesting to see what he has going on in there. :-0

"FWIIW, the DE250 (on QSC HPR152i waveguide) is paired up with the kappalite 3015 and I assume the lower XO point would be a good idea with the 15" woofer."

oops. i just saw the second part of your message. that horn loses directivity control around 2khz, so the 4 pi style crossover is probably a pretty good compromise. the geddes horns are about 18 inches on the summas and hold pattern control much lower.



LTD, that is false about the directivity losing control @ 2KHz on the QSC HPR 152i . We had this discussion in the past too and you pointed out you looked at the wrong polars then

You can go here to see all measurements with the DE250/QSC HPR152i
http://sites.google.com/site/driverv...uency-response





To the OP. You can go here for DE250 XOs.

http://techtalk.parts-express.com/sh...ighlight=DE250

I know its not the kappalite 3015 match so you probably have to do some measurements if you want it perfect.

Remember, The 15" is going to beam much earlier then 12" woofers.
Reply
Reply
post #10 of 38
8/8/11 at 6:35am
Quote:
Originally Posted by mobeer4don View Post

Yes, ultimately I'll give it another try in PCD. I think I've re-read enough to figure out some of the mistakes I made the first time through. I'm slowly figuring out the measurement gear and HOLMImpulse.

If you are having HOLM measurement/setup issues.

Maybe this will help??

http://www.avsforum.com/avs-vb/showt...highlight=holm
Reply
Reply
post #11 of 38
8/8/11 at 6:51am
Like these guys have said, it isn't quite as simple as I think you are hoping it will be.

If you don't want to go the route of measuring and modeling (either active or passive), I'd suggest selling the 3015 and buying a driver that was used in one of the Econowave Deluxe designs. Those crossovers are done. Sure, you won't get the benefits of the 15" woofer, and the other woofer prices have gone up, but getting the crossover right trumps all of that.

If you are willing to measure and model, I'd suggest using Passive Crossover Designer (http://audio.claub.net/software/jbabgy/PCD.html). If you have Excel, this is free software that is relatively easy to use. You have to understand crossover basics, but it can model most run of the mill topologies and should adequate for your purposes.

If you want to go down this path, I'm more than willing to help if you can upload your impulse response measurements. Once you see how the modeling works, you'll realize it is not all that tough.
Reply
Reply
post #12 of 38
8/8/11 at 6:57am
Quote:


I'd suggest selling the 3015 and buying a driver that was used in one of the Econowave Deluxe designs. Those crossovers are done.

+1

The QSC HPR 152i/ DE250 and Eminence 3012HO is a perfect match see the line below
http://techtalk.parts-express.com/sh...&postcount=128

Contact EricH on this forum and ask him about the Eminence drivers, I think he has some!!
Reply
Reply
post #13 of 38
8/8/11 at 3:40pm
"LTD, that is false about the directivity losing control @ 2KHz on the QSC HPR 152i."

we disagree about the point where the horn loses pattern control. based on the graphic, it is equally fair to conclude at many different points where this horn loses pattern control.



it is nice horn and now available from p.e. for $35.

http://www.parts-express.com/pe/show...number=245-625
LL
Reply
Reply
post #14 of 38
8/8/11 at 4:07pm
Thread Starter 
All good suggestions. The 12" driver swap is an absolute last resort since the baffle board is mounted from the rear and the speaker (> 80 lb?) is loaded into the entertainment center from the rear. Lots of work moving stuff to do that. Swapping in a JBL 2226 would be easy....oh how I regret not finding the 4pi design before buying those 3015.

So let us get on to some of what I've actually tried...all graphs are 1/6 octave smoothing, mic @ ~42 inches from HF/LF center. These were done over multiple sessions so levels are not the same in each test.

Graph # 1 LE14 eWave

Here's where I'm at with my current set-up using the "LE14" version of the eWave crossover. Designed for different woofer, CD, and waveguide. Looks flat enough on the high end but the acoustic XO is 1840 Hz. I believe the L-pad is set to max or pretty close to it.

Graph # 2 (run 4 - ewave deluxe)
This was what I came up with based on the ewave XO topology used with the 2512 and 3012HO woofers. C1 = 6.2 uF, C2 = 18 uF, L2 = 2.5 mH, R1 = 3 ohms, R2= 7.5 ohms, L1 = 1.0 mH (0.71 ohms). HF is rolling off and I get a peak at 2000 Hz....acoustic XO is 1600 Hz. Trail and error changes in C1, C2, and L2... L-pad at 12:00

Graph # 3 - 4pi
I changed the 4pi to 8 db attenuation (from Wayne's chart). No L-pad.
I don't know if that 6 dB peak is a sign I wired it up wrong or what. Rolls off too early.

Graph # 4 - Z19 XO
I worked up Zilchster's Z19 XO for the Altec Model 19...pretty close anyway, not exact values here or there. Nice camel hump...acoustic XO down to 1400 Hz. Mid/high L-pads at 9:00 and 5:00

So the reason I was looking for a 1.0 to 1.2 kHz XO design for the DE250 was to see if that hump at 2 kHz would go away or if was something I was doing or measuring wrong...and why do all of them roll off too early except for the LE14 version? Is it a problem with HF CD compensation and the high efficiency 3015?

Back to beating on PCD soon enough...
LL
LL
LL
LL
Reply
Reply
post #15 of 38
8/8/11 at 4:38pm
You are fighting an uphill battle trying to make one of the crossovers fit. It just isn't that simple.

Since you have the ability to take measurements, I'd suggest starting with some raw measurements sans crossover. If you can post the impulse response wav files and ZMA impedance files in a zip, I think the group should be able to work something up. It'll be fun. Just make sure the measurements are done from the same spot and that the signal is the same to both woofer and CD.
Reply
Reply
post #16 of 38
8/8/11 at 4:50pm
Quote:
Originally Posted by LTD02 View Post
"LTD, that is false about the directivity losing control @ 2KHz on the QSC HPR 152i."

we disagree about the point where the horn loses pattern control. based on the graphic, it is equally fair to conclude at many different points where this horn loses pattern control.
Its not about agreeing or disagreeing. This is not a subjective topic. Its in the measurements and you are not reading the plot properly if you are concluding it loses directivity @ 2K. I can easily see directivity in that response...remember I was the one that sent them to augerpro for measuring and we have been through this 1000 times. Even in general horns do not lose pattern control at "different points" that just does not make any sense.

HTH, When you are looking at directivity, you are looking for uniform off axis response. That chart shows uniform (directivity control down too 1KHz. Maybe you are looking at the "slope" itself and the curve dropping at 2KHz? That curve has nothing to do with directivity, its just a raw horn response....you should not look at that since it all flattens with even a 1.2uF cap. I mean have you ever seen the raw 15 OS waveguide response (it has a serious dip arond 5KHz and its why Geddes tells everyone to place his speakers at 45 deg angles)??? You would incorrectly make the same conclusion. What is important is how the off axis response curves follow the on axis curve but a few dBs down, that is what directivity is all about.



Besides, I can not believe you do not remember this discussion since we have been through this once with you already. Zilch, Augerpro, Wayne or Pete...More or less anyone with experience using this waveguide will tell you there is directivity down close to 1Khz. You are welcome to ask any expert this, post that response on any board that has waveguide experts and ask them where it loses directivity. Lets bet $1K on what the answer will be
Reply
Reply
post #17 of 38
8/8/11 at 6:15pm
LTD, I have to agree with Penn. Unless you are talking about vert directivity, it holds pattern pretty nicely down to 1khz and my theoretical calculations also support this.

Anyway, to the OP, this WG will work nicely crossed to the 3015 crossed around 1.2khz.

Penn as an aside, I don't think it is completely accurate to say that Geddes suggests the 45deg placement with a cross just in front of the listener solely due to the axial null inherent in his design. That is part of his motivation, but others, Parham of Pi namely, like this orientation as well due to its significant reduction in first reflections off of the near side wall. The main reflections will be the far side wall and those will be very late relative and reduced in strength. Anyway, just a nit. I think it is a good approach with any horn speaker.
Reply
Reply
post #18 of 38
8/8/11 at 7:07pm
" When you are looking at directivity, you are looking for uniform off axis response"

i don't know what to say guys, cd by definition means the same frequency response on and off axis.

the off axis 152i spl begins to collapse around 2.2khz. at that point the off axis power is about 105 to 95 db, a 20 db spread.

at 1khz, the off axis 152i spl is 96 db vs 91 db, a 5 db spread.

a 15db difference in spl vs. frequency is not controlled directivity.
Reply
Reply
post #19 of 38
8/8/11 at 7:25pm
Thread Starter 
Quote:
Originally Posted by coctostan View Post
You are fighting an uphill battle trying to make one of the crossovers fit. It just isn't that simple.

Since you have the ability to take measurements, I'd suggest starting with some raw measurements sans crossover. If you can post the impulse response wav files and ZMA impedance files in a zip, I think the group should be able to work something up. It'll be fun. Just make sure the measurements are done from the same spot and that the signal is the same to both woofer and CD.
I will take you up on your offer. I should be able to run off the wife long enough tomorrow to get the raw driver measurements and give PCD a second try.

My first attempt at PCD was flawed due to some questionable measurements. The XO I built didn't even come close to matching the model. So that's when I went on the kick of building a few XOs that should be close to see where I was going wrong (and hope for success anyway) . I later figured out the CD measurement file was bad, the first couple of XOs were wired wrong, and I'm not sure they were set-up properly in PCD. It wasn't pretty.
Reply
Reply
post #20 of 38
8/8/11 at 7:29pm
Quote:
Originally Posted by mobeer4don View Post
I will take you up on your offer. I should be able to run off the wife long enough tomorrow to get the raw driver measurements and give PCD a second try.

My first attempt at PCD was flawed due to some questionable measurements. The XO I built didn't even come close to matching the model. So that's when I went on the kick of building a few XOs that should be close to see where I was going wrong (and hope for success anyway) . I later figured out the CD measurement file was bad, the first couple of XOs were wired wrong, and I'm not sure they were set-up properly in PCD. It wasn't pretty.
No sweat. I can guarantee you are not the first person to make those mistakes. Penn has screwed all that up before. Kidding! Once you get used to doing this stuff it becomes pretty easy, but there are a bunch of holes to step in at first.
Reply
Reply
post #21 of 38
8/8/11 at 10:33pm
That comment caught my attention as well, and there's yet another important reason for toe-in - widening the sweet spot for multiple listeners via the directivity rendering the nearer speaker more off-axis and the farther one more on-axis, equalizing (or tending to) their SPL.

Quote:
Originally Posted by coctostan View Post

Penn as an aside, I don't think it is completely accurate to say that Geddes suggests the 45deg placement with a cross just in front of the listener solely due to the axial null inherent in his design. That is part of his motivation, but others, Parham of Pi namely, like this orientation as well due to its significant reduction in first reflections off of the near side wall. The main reflections will be the far side wall and those will be very late relative and reduced in strength. Anyway, just a nit. I think it is a good approach with any horn speaker.
Reply
Reply
post #22 of 38
8/9/11 at 5:15am
Quote:
Originally Posted by coctostan View Post

LTD, I have to agree with Penn. Unless you are talking about vert directivity, it holds pattern pretty nicely down to 1khz and my theoretical calculations also support this.

Anyway, to the OP, this WG will work nicely crossed to the 3015 crossed around 1.2khz.

Penn as an aside, I don't think it is completely accurate to say that Geddes suggests the 45deg placement with a cross just in front of the listener solely due to the axial null inherent in his design. That is part of his motivation, but others, Parham of Pi namely, like this orientation as well due to its significant reduction in first reflections off of the near side wall. The main reflections will be the far side wall and those will be very late relative and reduced in strength. Anyway, just a nit. I think it is a good approach with any horn speaker.

Very true, the more important point was the null on axis as a reference point the discussion about what is defined as constant/controlled directivity.
Reply
Reply
post #23 of 38
8/9/11 at 5:19am
Quote:
Originally Posted by LTD02 View Post

" When you are looking at directivity, you are looking for uniform off axis response"

i don't know what to say guys, cd by definition means the same frequency response on and off axis.

the off axis 152i spl begins to collapse around 2.2khz. at that point the off axis power is about 105 to 95 db, a 20 db spread.

at 1khz, the off axis 152i spl is 96 db vs 91 db, a 5 db spread.

a 15db difference in spl vs. frequency is not controlled directivity.

THere is not much to say really, you are not looking at the chart properly and you do not understand what constant/controlled directivity means.

Again, it seems you want to ignore my opinion so I kindly request you create a thread on any site with people you think are experts and ask them about it. This is about the 3rd time you have posted this very misleading information maybe confusing people just learning about horns. Every other thread this opinion has been corrected (not by just myself).
Reply
Reply
post #24 of 38
8/9/11 at 9:57am
if you equalize the response flat, the off axis roll-off will be much higher at 2khz than at 1khz. constant directivity means a flat off axis response. this can easily be seen by looking at the on axis line (the top one) and the light blue line (~45 degrees off axis). they are parallel from 2.2khz up. below 2.2khz they begin to collapse. ergo, this horn holds constant directivity to 2.2khz. it still holds some pattern control down to about 800hz and below that point, it holds zero pattern control.
Reply
Reply
post #25 of 38
8/9/11 at 10:29am
Again, Im sorry but you are wrong. You have never built any of these. I have built 6 different pairs now. Myself, Zilch (RIP), Pete, Augerpro or anyone else that has built speakers with the QSC HPR152i understand a little bit about the directivity. Do you really think Zilch and Pete over on PE would have been so active with their contribution if the QSC was not the BEST waveguide we could buy period?? (Wayne's Pi horn choice do not compare but he didnt have another option) You should search your own discussion on this topic because you already agreed once with me...do I need to find that discussion for you?



You are even ignoring all the formulas about directivity. Ie. Its a 14" wide waveguide which automatically dictates directivity down to around 1KHz.
Reply
Reply
post #26 of 38
8/9/11 at 10:38am
"Again, Im sorry but you are wrong."

what is wrong with what i said in post #24?

"You are even ignoring all the formulas about directivity. Ie. Its a 14" wide waveguide which automatically dictates directivity down to around 1KHz."

what formula specifies that a 14" horn will hold directivity to around 1khz?

perhaps you are referring to the fact that a direct radiator with a 14" diameter will be down -6db at 45 degrees off axis at around 1 khz, but that is a different topic.
Reply
Reply
post #27 of 38
8/9/11 at 10:42am
Here is one post from you....

Quote:
Originally Posted by LTD02 View Post

"With the lower extension of the CP385 we can get a better feel for how low the QSC horn holds pattern control."

that is actually a very surprising set of measures. i thought this horn gave up the ghost at around 2k. you are showing strong 1k performance. not sure if i am going to give up my thinking that the horn should be at least as wide as the horn for directivity matching, but you are testing my beliefs. :-)

your work is very, very much appreciated!


Start here..

Quote:
Originally Posted by augerpro View Post

Polk's Powerport is not the same as these Powerports. I don't believe the Polk's are available to retail.

The foam is the black 30ppi 4" stuff. I got the 13"x13" square.

With the lower extension of the CP385 we can get a better feel for how low the QSC horn holds pattern control. Here is a plot from 0 to 60 degrees horizontally:



Are you seriously saying that does not have directivity below 2KHz???

And read posts after that....Zilch, MBentz and Augerpro have no question whatsoever about its directivity. No one ever has posted " below 2.2khz they begin to collapse", the only person to post that is you. Everyone else owns these drivers and have measured and built designs that have 100% controlled directivity so I asked again how can you be right and everyone else is wrong when everyone else is actually designing speakers with that waveguide?
Let me get more, You should know that my memory is damn good so I pretty much remember most of our discussions on this
Reply
Reply
post #28 of 38
8/9/11 at 11:07am
that looks like about 1khz cd to me. :-)

actually, that is a very nice looking result.

seems like the driver was obfuscating the horn's potential in the other measurement.
Reply
Reply
post #29 of 38
8/9/11 at 11:19am
See, was that so hard.

The problem is that you forgot that horns have a raw response like the original plot above. I guess they need to have XOs maybe for you to see the directivity properly? The only difference between the measurements is really a 1.2uF cap which flattens the response out nicely.

Either way, both plots told me it has 1Khz CD.
Reply
Reply
post #30 of 38
8/9/11 at 11:37am
"The problem is that you forgot that horns have a raw response like the original plot above. I guess they need to have XOs maybe for you to see the directivity properly?"

no. in the plot data that i was working with, the directivity collapses under 2.2khz. the light blue line is not parallel with the on axis response, which means directivity is collapsing (it goes from a 6db difference at 2khz to a 3 db difference at 1khz).

in the beyma data, the blue line is parallel with the on axis response to 1khz (6 db at both 1 khz and 2khz).

"The only difference between the measurements is really a 1.2uF cap which flattens the response out nicely."

how does plot one become plot 2 with a 1.2uF cap? looks like a different driver to me not just a cap.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs

AVS Top Picks

AVS › AVS Forum › Audio › DIY Speakers and Subs › DE250 XO designs for 1.0 to 1.2 kHz?