Econowave Speakers

[richidoo]

Been messing around with the PCD crossover simulator, it is fun! Learning a lot about what parts do what. Very informative exercise. 

I discovered that there is a null at the crossover freq with the Econowave Delite crossovers I am using. Must be different polarity conventions with the different brand drivers. I have them wired in reverse according to the colors on the driver terminals, but measurements don't lie, one of them is wrong. I'll figure that out later.

But in viewing the measurement made with HolmImpulse, versus the simulation of the drivers and crossover made with Jeff Bagby's Passive Crossover Designer there is great correlation!

Note the reversed measurement and sim are very alike, the null is the same shape at the same freq.

The hopeful thing in all this is how nice the simulated speaker with the correct polarity is - nice and flat. I will use a battery to determine the reverse polarity of woofer, and the correct for the tweeters tonight and post a measurement tomorrow. I have two baseball games to attend tonite. My oldest is pitching!  

No wonder the sound has been bugging the hell out of me. I wonder if the Eminence drivers were this way also. Time will tell

[richidoo]

I used a D cell battery to determine that the AE red post is positive. Same for the Eminence drivers. So I wired the AE in reverse per LR2 requirement. However, the Radian compression driver is in phase with the woofer when it is wired opposite of what the colors would indicate. Black post is positive.

I have read about this being typical of CDs, done to make field replacement repairs less confusion to the techs. The mfgs know that most applications are LR2, and most will reverse the tweeter so they can keep the positive thump from correct polarity in the bass. So they help the techs avoid crossover null by anticipating their ignorance and coloring the posts opposite of the reality.

So, now with the polarity fixed, the measurement shows well behaved crossover region,(red) compared to reversed polarity (blue.) This recorded with sine sweep from 1m at height between driver centers.  I have been listening to this crossover null since I built these things. I forgot about the child-psychology polarity fake out thing.

I'll tune the tweeter level flat with the measurement equipment, then fix and tune the other one, and try some listening! Just in time for the weekend.

[richidoo]

Wow, what a huge improvement fixing the polarity has made, as expected from looking at the change to the FR. Now I understand the excitement over the econowave concept. Honestly I was starting to scratch my head for a while, and got frustrated that I couldn't get any traction with this project, but it was my own mistake. This explains my comments previously about midrange energy lacking. Now it is engaging and easy to drift into the music, while bringing lots of delicious detail in the lower mids and bass. 

Still to come are replacing the adjustable tweeter LPad with fixed resistors to allow contouring the treble response, building an adjustable line level baffle step correction box to add some bass boost, removing more FG from the box to loosen up the woofer a little more, and applying damping material to the outer surface of the plastic horn.

[richidoo]

The bump at 450Hz is too much. I can't figure out what is causing it. Floor bounce dip should be ~470, but it is peaking there instead.    And the bassiness is audible even when walking around. So the floor bounce is probably helping some.

Could be the woofer is rolling off too early with the Zilch crossover, making a bump that should stay flat at that level. I'll have to do some sweeps with driver only and adding crossover parts to figure it out.

[Face]

Is it still there if you measure it nearfield? 

[richidoo]

I'm gonna check that, and also at listening position 3m, and maybe outside too. I'll post.

The box is 120l, way more than the AE driver needs for Qtc .5, so it is overdamped. Can that cause a bump like that?

I think the bump is too low freq to be caused by internal reflections?

[richidoo]

Looks like room reflections.  

Measured with no xo filter, amp to driver.
Red is 1" from phase plug, blue is 1meter, green is listening position 3m.  In the 1st pic all are gated 100Hz/3.5m   In the 2nd pic, each is gated to it's actual distance from the driver.

I didn't suspect room, because it sounds similar standing in different places in the room. I'll try moving it around, and outdoor measurement.

[richidoo]

Here is the sweep at 3m listening spot, gated to 3m. Green is raw driver, red is with xo filter engaged.

[richidoo]

Here's a shot of the whole speaker, in the same location. With the tweeter turned up flat with the woofer's flat zone there is a big dip at the crossover, caused by too sooon rolloff of the woofer. Tuning the level of the tweeter by ear to match that dip would explain the woofer bump. This is the basic problem I had before when I suspected the Eminence woofer inductance to be the cause of the rolloff, but it turned out to be in spec. The AE is lowest inductance of any woofer, but still rolling off too soon.

I will verify the inductance of the Erse super Q 3.3 woofer filter coil that I am using now. Geddes uses 2mH in Abbey, this is similar design.

[richidoo]

Replaced the 3mH coil with 1mH aircore, no more dip. But there is too much overlap @1500-2k. The 1mH coil is too small, allowing the woofer to play too high, combining with tweeter output to be too loud. I guess 2mH is worth a try.

[richidoo]

This is gonna piss off 2two, er I mean Evan who hates the same thread hogging up the whole forum... hehe

Blue is 2mH on the woofer, two 1mH in series. Red is 1mH. 2mH  the 1.5-2k peak at xo. 1/6 octave smoothing.

Now for the peak at 650 and dip at 1200. Hopefully just acoustic. although I'm not so sure about the 1200.

[richidoo]

Face, how do you recommend to measure drivers for importing into PCD? Should I do it in room where the speakers will play, or should I do it outside up high to eliminate as many reflections as possible?   Tom Danley says measure without reflections, then apply filters. What do you do?

[richidoo]

1st pic is Moving the speakers around on floor. Blue is original spot. Red is 1 foot forward. Green is 2 feet backward. Doesn't make it go away.

2nd pic is height. Green is on floor. Red is lifting it up 3.75" Makes it worse, but could be on track if it's a ceiling mode.

Ceiling is 9 or 10 feet. Fundamental of 9 feet is 63Hz. No way that big of a higher order harmonic mode at 650.

What do you think is causing the 650 bump?  It seems to be narrower with the 2mH coil than with the original 3mH coil.

[2two]

This is gonna piss off 2two, er I mean Evan who hates the same thread hogging up the whole forum... hehe

Blue is 2mH on the woofer, two 1mH in series. Red is 1mH. 2mH  the 1.5-2k peak at xo. 1/6 octave smoothing.

Now for the peak at 650 and dip at 1200. Hopefully just acoustic. although I'm not so sure about the 1200.

Not that I know Evan but after reading about what went on I think he was more upset about the worthless posts . IMO

[richidoo]

Yes, (and thanks)