Looks like the 8" square PT waveguide will fit nicely.... :thmbsp:
Zilch's AK Design Collaborative - Econowave Speaker
- Thread starter Zilch
- Start date
skywave-rider
Slumming in Akihabara
Isolated
OK, I re-measured using 8 Ohm non inductive resistors as dummy loads in place of drivers.
These are taken with DE250 on QSC on top of PE KD Trap box loaded with JBL 123A-1. Crossovers are standard EWave (30 Ohm comp) and 4Pi with additional fixed attenuation plus L-Pad and a notch filter to take the top down some.
Zobel is 8 Ohm/20uF, what Wayne usually uses.
EWave orange: LF; green: HF; red: sum
EWave blue: LF w/Zobel; orange: LF w/o Zobel; purple: sum w/Zobel
4Pi yellow: LF; red: HF; blue: sum [standard 4Pi Zobel in line]
4Pi orange: w/Zobel; red: w-o/Zobel
Seems like the Zobel has to work more at about 1.8k for this woofer.
OK, I re-measured using 8 Ohm non inductive resistors as dummy loads in place of drivers.
These are taken with DE250 on QSC on top of PE KD Trap box loaded with JBL 123A-1. Crossovers are standard EWave (30 Ohm comp) and 4Pi with additional fixed attenuation plus L-Pad and a notch filter to take the top down some.
Zobel is 8 Ohm/20uF, what Wayne usually uses.
EWave orange: LF; green: HF; red: sum
EWave blue: LF w/Zobel; orange: LF w/o Zobel; purple: sum w/Zobel
4Pi yellow: LF; red: HF; blue: sum [standard 4Pi Zobel in line]
4Pi orange: w/Zobel; red: w-o/Zobel
Seems like the Zobel has to work more at about 1.8k for this woofer.
Attachments
Last edited:
skywave-rider
Slumming in Akihabara
Distortion
Don't know if this is meaningful, but here are distortion measurements of DE250 on QSC. HF shown with the lower crossover point and the higher crossover point. Listening, I think it sounds OK crossed at the lower frequency. I have to listen more because I just set up a pair last night.
Red is 4pi, blue is EWave.
Don't know if this is meaningful, but here are distortion measurements of DE250 on QSC. HF shown with the lower crossover point and the higher crossover point. Listening, I think it sounds OK crossed at the lower frequency. I have to listen more because I just set up a pair last night.
Red is 4pi, blue is EWave.
Attachments
Wayne Parham
Active Member
Just to be clear, I think the solid curves shown below are the fundamental sweep of the tweeter circuit by itself, the dotted curves (next down) are 2nd harmonic and the dashed lines (third down) are 3rd harmonic. Can you verify this please?
P.S. haven't had time to check in much in the past few weeks. Looks like you guys have been having some fun!
P.S. haven't had time to check in much in the past few weeks. Looks like you guys have been having some fun!
Don't know if this is meaningful, but here are distortion measurements of DE250 on QSC. HF shown with the lower crossover point and the higher crossover point. Listening, I think it sounds OK crossed at the lower frequency. I have to listen more because I just set up a pair last night.
Red is 4pi, blue is EWave.
lico
Super Member
Nice setup. How they sound?
You have identified the culprit and how to deal with it. :thmbsp:
Looks like Wayne will have some comments; I'll have some, too, once I study it a bit more.
Very nice!
Do you consider DE250/QSC a significant upgrade with respect to the performance?
Last edited:
ROSSO
Active Member
Well, they certainly sound good to my ear. I'm not using anything that would be confused as a serious reference system to drive them, almost embarassingly low tech, really, so I don't really know how good they may ultimately be. I've got my three other ewave varients spread out on the second & third floors, so a direct A-B comparison is not possible. I bought the B&C drivers to use on my Altec H-811B horns, so my standard here was that the QSC waveguides would be an upgrade from that, and I believe they are.
The electronics side of this particular setup is dictated by continued domestic tranquility, and as such, what can be hidden in the 12 inch deep, 1925, built-in cabinets at the opposite end of the room. So what fits the bill, and the cabinet, is an old Bose Lifestyle AM/FM/CD head unit used as a preamp/tuner/CD player that feeds a Carver M400 cube. The Bose has a RF remote, so I can run the thing from the veranda, etc.
skywave-rider
Slumming in Akihabara
Yes, Wayne, you are correct all around.
Since you may not have been following, to be clear, when I say "4Pi" it is not referring to your speaker, but the crossover which I am using as a standard reference on the DE250/QSC waveguide; the woofer in the last post was the JBL 123A-1. I had to use the 20 dB of attenuation derived from your chart, plus an L-pad. (Have to learn Spice.) I know the HF is not optimized. The Econowave crossover is the standard version.
http://www.audiokarma.org/forums/attachment.php?attachmentid=151281&d=1242968714
Last edited:
skywave-rider
Slumming in Akihabara
Group Delay
I have read a post by EarlK that group delay is useful to see where a horn or waveguide's lower useful limit is.
http://www.audiokarma.org/forums/showpost.php?p=3083841&postcount=21
Here is the group delay of the DE250 on QSC and 4Pi x-o:
The big transition is obviously at about 1k. Any interpreters? Earl?
Am I zoomed-in far enough?
I have read a post by EarlK that group delay is useful to see where a horn or waveguide's lower useful limit is.
http://www.audiokarma.org/forums/showpost.php?p=3083841&postcount=21
Here is the group delay of the DE250 on QSC and 4Pi x-o:
The big transition is obviously at about 1k. Any interpreters? Earl?
Am I zoomed-in far enough?
Attachments
Hi Sykwave,
- It's nice to see you putting your software through it's paces . :thmbsp:
- If you can, see if you can subtract the propagation delay from your pic before displaying it one more time ( it seems to be @ 1.5 ms ) .
- I do feel that Group Delay figures into the lower limits that a horn/driver combo can be used to . But, I also feel that a drivers' Fs plays into this topic / though I've never seen its' contribution studied outright ( the 288 has a very low Fs ) .
- I haven't seen any conclusive work in this area / just a lot of anecdotal claims ( just like my post that you referenced above
) .
- Fortunately, like most things in audio , nothing's written in stone .
- FWIW, I'm interested in this waveguide ( for SR work ) / so I do appreciate all the postings made, concerning it .
>< cheers
- It's nice to see you putting your software through it's paces . :thmbsp:
- If you can, see if you can subtract the propagation delay from your pic before displaying it one more time ( it seems to be @ 1.5 ms ) .
- I do feel that Group Delay figures into the lower limits that a horn/driver combo can be used to . But, I also feel that a drivers' Fs plays into this topic / though I've never seen its' contribution studied outright ( the 288 has a very low Fs ) .
- I haven't seen any conclusive work in this area / just a lot of anecdotal claims ( just like my post that you referenced above
) .- Fortunately, like most things in audio , nothing's written in stone .
- FWIW, I'm interested in this waveguide ( for SR work ) / so I do appreciate all the postings made, concerning it .
>< cheers
Last edited:
skywave-rider
Slumming in Akihabara
Wayne Parham
Active Member
Horns become more resistive in their passbands, but down low they become increasingly reactive. This can be described in many ways, as group delay, acoustic phase, etc. A horn's ability to be resistive can be seen as a measure of quality, because horns that provide better acoustic loading are more purely resistive. The impedance curve also shows how smooth the termination is.
All drivers have some reactance, whether direct radiator or horn. Horns are usually resistive over a wider range than most direct radiators, but they also usually have more peaks and dips in their impedance curve. Each device is resistive at certain frequencies, and some are more purely resistive than others. This is reflected in the electrical impedance as well, and that's why speaker drivers aren't purely resistive loads. It is usually important to take account the real impedance when designing a crossover, because the load reactance is a part of the filter, and modifies the transfer function compared to a purely resistive load.
While not specific to midrange or tweeter horns, the post below shows the impedance of a bass horn, including the real (resistive) and imaginary (reactive) parts. These two are 90° apart, so total impedance is a vector. This is not limited to basshorns, but applies to all horns.
All drivers have some reactance, whether direct radiator or horn. Horns are usually resistive over a wider range than most direct radiators, but they also usually have more peaks and dips in their impedance curve. Each device is resistive at certain frequencies, and some are more purely resistive than others. This is reflected in the electrical impedance as well, and that's why speaker drivers aren't purely resistive loads. It is usually important to take account the real impedance when designing a crossover, because the load reactance is a part of the filter, and modifies the transfer function compared to a purely resistive load.
While not specific to midrange or tweeter horns, the post below shows the impedance of a bass horn, including the real (resistive) and imaginary (reactive) parts. These two are 90° apart, so total impedance is a vector. This is not limited to basshorns, but applies to all horns.
Your measurements agree with my own and Jack's with respect to the DE250/QSC on the eWave highpass -- the green line. It is slightly overcompensated and the core frequency is too high, the latter being the primary reason the crossover has moved up to 2.2 kHz. Is this with R1 = 30 or 16 Ohms? Crank the high SPL up 4-5 dB and you'll be the same place as me, ~1.8 kHz.
See Geddes's horizontal polar for how we'd like it to be looking, his green line:
We're very close; we just have to move it lower by adjusting the core filter.
The 4-Pi highpass, the red line, is more overcompensated and underdamped. The culprit at 1.8 kHz is 123A-1. Wayne's lowpass knocks that down, orange, but not optimally, perhaps; the apparent flat response through the crossover region is coming from the underdamped HF driver making up the difference, an effective 4 dB of "boost" in a region below the recommended minimum crossover frequency for the DE250. Once the comp is tweaked, we'll not want that to remain. The fact that red is above blue there indicates the flat response is the result of a phase cancellation:
It'd be nice to get it playing well down there at 1.2 kHz (that's where Geddes uses it in Abbey, I believe), but the QSC is losing pattern control that low according to Brandon's polars, both in the horizontal:
And the vertical:
http://sites.google.com/site/drivervault/driver-measurements/horns-and-waveguides/qsc-pl-000446-gp
Much as we'd like the crossover to be lower, I'd prefer to stay with our 1.6 kHz target 'til we see where the vertical nodes come in; once that is known, we may be able to fine-tune a compromise, but the C/C distance is against us in accomplishing that with this waveguide.... :yes:
Attachments
Last edited:
skywave-rider
Slumming in Akihabara
That's 30 Ohms, Zilch.
I see, cranking the HF brings the acoustical x-o down...follow the lines.
Everything you said makes sense.
Since there are so many potential woofers out there to build with, I assume the generic 1.2kHz EWave low pass will remain.
I'll try to hit the mark on the 123A-1 Zobel and take it from there.
Unfortunately, I took the DC300s to school today for the dumpster bin EWaves.
I see, cranking the HF brings the acoustical x-o down...follow the lines.
Everything you said makes sense.
Since there are so many potential woofers out there to build with, I assume the generic 1.2kHz EWave low pass will remain.
I'll try to hit the mark on the 123A-1 Zobel and take it from there.
Unfortunately, I took the DC300s to school today for the dumpster bin EWaves.
Assuming we keep the highpass in the original place, yes; it's the high frequency we've got to "lock." It's the universal module having a defined response characteristic with which an array of suitable woofers may be adaptively mated.... :yes:
Russellc
Addicted Member
I have read a post by EarlK that group delay is useful to see where a horn or waveguide's lower useful limit is.
http://www.audiokarma.org/forums/showpost.php?p=3083841&postcount=21
Here is the group delay of the DE250 on QSC and 4Pi x-o:
The big transition is obviously at about 1k. Any interpreters? Earl?
Am I zoomed-in far enough?
Somehow Geddes gets it down to 900 hz. Only x-over pic I've seen had at least three notch filters in it.
russellc
Russellc
Addicted Member
How are you adjusting the core filter? you mean the crossover point?
May just be the way it is with this waveguide.....
Russellc
GOYA
Super Member
Greetings gentlemen,
I may very well be joining the ranks of the E-wavers in the near future but I have some questions about an unusual way to go about it. I have no cabs (although I've been on the lookout for some) that will allow me accomplish the normal woofer/horn arrangement. My speakers are DCM TF-600s. My hopes are that I can acheive acceptable results by affixing the horn to the top of the DCM and using an active crossover to biamp. The DCMs would not be altered in any way but the LF would be sent to them. Would this present any showstoppers as far as you fellows can see?
Also, I have yet to acquire an active crossover but I have my sights on a Crown VFX2A. Any opinions on this unit?
I may very well be joining the ranks of the E-wavers in the near future but I have some questions about an unusual way to go about it. I have no cabs (although I've been on the lookout for some) that will allow me accomplish the normal woofer/horn arrangement. My speakers are DCM TF-600s. My hopes are that I can acheive acceptable results by affixing the horn to the top of the DCM and using an active crossover to biamp. The DCMs would not be altered in any way but the LF would be sent to them. Would this present any showstoppers as far as you fellows can see?
Also, I have yet to acquire an active crossover but I have my sights on a Crown VFX2A. Any opinions on this unit?