• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Zilch's AK Design Collaborative - Econowave Speaker

Hey Don, do you know how to get the minimum phase frd files, as were used to do the ewave designs? If not, we could make do with Troel's response measurement, although you could find out if your coating manged to knock out the resonant peak from 2-3k at all. Impedance measurement of the woofer is what's needed most of all. edit: It's entirely possible that a good one is posted out there somewhere - I have not searched yet.
 
Last edited:
DT,

Is this what you need? Impedance exported from DATS, measured with WT3, 125A in-box. Also D220Ti on "standard" waveguide with high-efficiency crossover with L-pad at 12:00 position.

Let me know if something doesn't look right. I have very little experience with this gear and I'm learning on the fly.

Check back later for FR sweeps.

EDIT.....

Impedance is not phase...DOH!

Added sweeps from HolmImpulse. Let me know if you can open them and use them. I should re-run these since bass/treble not flat.

Revised sweeps...

All measured at distance of 40" from point between woofer and horn.
 

Attachments

Last edited:
Thanks Don, the impedance looks right at a glance. I will try loading these up when I have a chance.

I have not checked out the response files, but I'll say this now anyway: To make use of the frd response file from Holm, you have to get minimum phase in there, lock the "time zero" for your measurements of both drivers, and gate it to remove reflections. I can't recall exactly how to do all of that because I have been using ARTA, but I did get it to work once. Oh, and you also need a pretty precise physical measurement of the relative vertical position of tweeter-to-mic-to-woofer.

Also, for all measurements including the D220ti, if you're going to include the HF crossover, you'll need to adjust the L-pad to match the woofer prior to taking your final measurements. Reference its average level just above where your gating kicks in, eg. 200-500Hz, not in the crossover range (where it will be louder). After that, you'll also need to take an impedance measurement of it through the crossover with the L-pad still set at that level. Alternatively, you can take all the measurements without any crossover.
 
yeah 26 lbs means they're not heavy duty cabs but a few 2 by 4s would fix that!

Honestly if this is your first build, I would just thro the Parts Express 6 by 6 waveguide in the midrange hole, and leave the tweeters in place. NO Cutting!

If you decide to go the full route later, you are only out about 12 bucks for the 6 by 6 WGs.

I'm very pleased with mine.
 
dumptruck said:
I have not checked out the response files, but I'll say this now anyway: To make use of the frd response file from Holm, you have to get minimum phase in there, lock the "time zero" for your measurements of both drivers, and gate it to remove reflections. I can't recall exactly how to do all of that because I have been using ARTA, but I did get it to work once. Oh, and you also need a pretty precise physical measurement of the relative vertical position of tweeter-to-mic-to-woofer.

Also, for all measurements including the D220ti, if you're going to include the HF crossover, you'll need to adjust the L-pad to match the woofer prior to taking your final measurements. Reference its average level just above where your gating kicks in, eg. 200-500Hz, not in the crossover range (where it will be louder). After that, you'll also need to take an impedance measurement of it through the crossover with the L-pad still set at that level. Alternatively, you can take all the measurements without any crossover.
Click to expand...

Second attempt at measurements uploaded into post above. Fixes included..
* set tone controls flat
* Loaded generic calibration file for ECM8000 microphone (may ship off for proper calibration)
* adjusted L-pad to get LF & HF levels closer
* set gating to about 300Hz

I get some fuzz in the measurements that might be part background noise in the garage. Also I'm using a beater computer and 25 (?) year old Dennon receiver that could be part of the problem. Will swap out amp later to test...

I'm not having any luck with the time lock setting. Automatic (detect time zero) seems to work but clicking the "time zero locked" setting seems to turn everything to garbage as the attached graphic shows.
 

Attachments

  • timelocked example.png
    timelocked example.png
    28.9 KB · Views: 32
the beauty of this build is that the speakers were headed to the dump. I got them for $30 which is a lot less than a pair of jbl woofers.

Another idea is to use the 125A woofer from my L26. The L26 will probably go out of rotation so I may as well use the best woofer for this build. I can put the woofer from the 4800 into the L26 cabinet as it is just a spare or will be donated to one of the kids.

For the cabinet, I have a real robust cabinet from a pair of Optimus 400 that I gutted. One speaker had a bad woofer so I removed the drivers. It has a 12" woofer but I could remove the front panel with a flush mount router and cut a new one to match the woofer + waveguide. The cabinet is slightly larger at 2.55 cubic feet feet vs 2.35 for the L26.

By using the woofer from the L26 and a surplus cabinet, I do not have to permanently destroy a great speaker. I can always put the woofer back or just wait until a pair of 125A's pop up.
 
Last edited:
donprice said:
DT,

Is this what you need? Impedance exported from DATS, measured with WT3, 125A in-box. Also D220Ti on "standard" waveguide with high-efficiency crossover with L-pad at 12:00 position.

Let me know if something doesn't look right. I have very little experience with this gear and I'm learning on the fly.

Check back later for FR sweeps.


EDIT.....

Impedance is not phase...DOH!

Added sweeps from HolmImpulse. Let me know if you can open them and use them. I should re-run these since bass/treble not flat.

Revised sweeps...

All measured at distance of 40" from point between woofer and horn.
Click to expand...

Thanks Don

so it looks like a 125a is a good fit with the D220Ti.

I think this is the way I am going to go as the QSC waveguide may not be available. I took the plunge and ordered a standard crossover kit with high sensitivity woofer c/w the D220Ti from Parts Express. It will be 2 weeks before I get the parts so the project is on hold. I am going to try to read the entire thread while I am waiting.
 
Last edited:
Best practice would be modelling in PCD but I'll take a stab at trial and error solution. As some VP at Dow Chemical once said "Six weeks in the lab will save you an afternoon in the library every time".

First try...standard ewave LF section L2 = 1.5 mH C2 = 12 uF
Doesn't roll off woofer fast enough...xo point ~2480 Hz

Second try...dig up parts L2 = 2.5 mH C2 = 14.7 uF
Not bad. Drop HF a little via the L-pad...xo point a little over 2000 hz

Third graph shows the null from phase reversal.

Notice the sqiggles popping up periodically in the graphs. :scratch2:
 

Attachments

  • 125A D220 L2 1.5 C2 12.png
    125A D220 L2 1.5 C2 12.png
    50.5 KB · Views: 30
  • 125A D220 L2 2.5 C2 14.7.png
    125A D220 L2 2.5 C2 14.7.png
    48.5 KB · Views: 32
  • 125A D220 L2 2.5 C2 14.7 - phase comparison.png
    125A D220 L2 2.5 C2 14.7 - phase comparison.png
    46.9 KB · Views: 35
Sorry don, I haven't had a chance to try to help out yet. That second crossover point looks pretty good, but it's hard to know what's happening with that massive treble roll off. Looks a lot like DigitalChris's troubles from a few pages back. Maybe it's the receiver?
 
dumptruck said:
Sorry don, I haven't had a chance to try to help out yet. That second crossover point looks pretty good, but it's hard to know what's happening with that massive treble roll off. Looks a lot like DigitalChris's troubles from a few pages back. Maybe it's the receiver?
Click to expand...

It doesn't sound like a pillow was stuffed into the horn.

I need to look back at some other measurements made with a different PC and receiver to see if there is an easy answer. I think I'll send the mic of for calibration since the ECM8000 can be all over the place (says the guy selling the calibration service).
 
On all of the graphs, you have a 6db roll down by the time it gets too 200hz. That is as questionable as the high end rolloff. If that trend continues below 200hz, they would sound bass lite.....I think.
 
djnagle said:
On all of the graphs, you have a 6db roll down by the time it gets too 200hz. That is as questionable as the high end rolloff. If that trend continues below 200hz, they would sound bass lite.....I think.
Click to expand...
That trend is false. We don't really know what it's doing below ~800Hz or so in these measurements due to the gating, but Troels' measurements show it approximately flat from 300-1000Hz. However, everything above the bass could be 2-4dB too high on Don's crossover making it weak on bass, you just can't tell one way or the other from these measurements.

P.S. Does the calibration service cost more than the EMM-6 with the Dayton calibration ($48)? I know you can get some sort of "extra-good" calibration from other sources, but... mine seems quite accurate, at least above 30Hz or so..
 
Last edited:
Okay the files seem like they are loading properly and I could give PCD a try if you want, but I can't really do much more than what you did without re-doing the HF part to go with it. For that, I would need a physical measurement of the center-to-center driver spacing, and a new set of measurements where the D220ti is not on a crossover (the drop should be about 10dB from 2k to 20k without crossover). If you want the L-pad included, pull it from the crossover, set it to 1/2 and and wire it up to the D220. Measure that first, so that you can measure at a reasonable volume, and then measure the woofer and the combined with all your levels the same.

In your current config, it does look like 2.0mH would sum better, but phase looks suspiciously linear... I think maybe the whole minimum phase thing didn't work out.
 
Last edited:
The calibration service is ~$55 so I ordered the EMM-6 and another pile of crossover parts. I should have it for the weekend. No point in trying to work with the questionable curves...
 
ECM800 (generic calibration curve) vs EMM-6 with custom calibration curve. Same speaker. More or less same level (1 dB delta?) and measuring point. Smoothed 1/6 octave rather than gated. Looks like 5 dB difference beyond 10 kHz.

Will try to drop in $20 chip amp to check for differences in amp section. Check back later for edits.
 

Attachments

  • ecm8000 vs emm6.png
    ecm8000 vs emm6.png
    15 KB · Views: 38
You must log in or register to reply here.
Back
Top Bottom