Zilch's AK Design Collaborative - Econowave Speaker

[Wayne Parham]

Let me make sure I am correct on this: so basically, the < 95 and >95 versions are melded into one, just select the parts for the amount of attenuation that is needed?

That's right, exactly.

There is only one crossover schematic. Select the values of R1, R2 and C1 depending on how much attenuation is required.

If you think about it, that's how it always was. You had two schematics, but the only difference between them was the values of R1, R2 and C1. Now you have a chart of values, so you can pick the right values for any level required from -6dB to -20dB.

 

[Zilch]

An L-pad may be useful after that, I don't know. My problem with L-Pads has always been that they don't lock in the R1/R2 values and they take away some of the range. You really need 10dB of fixed padding in the R1/R2/C1 network to get it to work properly.

Yes, we recognized early on that while an L-pad may not be the "ideal" solution for any particular combination of drivers, we needed to provide sufficient adjustability to accommodate a wide range of woofer sensitivities. Even using this approach, Jack had to design two versions to cover the available spectrum.

As Jack has suggested several times, now, there are many opportunities in the design for E'Wavers to "customize" according to their specific driver choices, system performance objectives, and listening tastes, consistent with their own desires and abilities to do so, including tweaking of the highpass filter, attenuation, and compensation. The lowpass is certainly fair game also, of course, all of which adds up to high-order speakergeek FUN! :thmbsp:

 

[Kegger]

Basically what Wayne has given is "Exact" attenuation, so not using an L-Pad at all here.

If you know how much attenuation you want or need then you don't have to use a L-Pad.

For those first building and want to dial it in the L-Pad can be handy, after you have what
you want with the L-Pad setting then you may want to convert it to fixed attenuation.
(Sonically just using a pair of resistors, not going through a L-Pad should be better as well)

 

[Wayne Parham]

That's right. If you want variable attenuation, I suggest using a switched attenuator rather than an L-Pad. I think it might be useful to have a three-way switch, say at levels -10dB, -12dB and -14dB. Seems like that might be a pretty useful range.

If you prefer the L-Pad, you would probably want to start off with the 8dB or 10dB R1/R2/C1 values even if 12dB padding was what was required for flat response. That way you would have some control above and below.

The problem is going to be that most woofers that are best suited are probably going to need 10dB or 12dB padding, and you'll need all that for top-octave compensation. The EQ works because attenuation is removed at high frequency, so there has to be some attenuation there in the first place or there is not enough left to remove for HF augmentation. When low level (under 8dB) base padding is used, the top-octave is under represented and the speaker sounds dull. You lose the sparkle.

 

[mrTT]

Then again, to me, it is fine to just swap out R1/R2 to make changes

Does it really matter what kind of resistors we use as long as power rating is right? I would prefer non inductive kinds but ordering them is a pain. Is it ok to use non inductive kinds? Sonically speaking, has anyone found significant difference between those two?

 

[skywave-rider]

Does it really matter what kind of resistors we use as long as power rating is right? I would prefer non inductive kinds but ordering them is a pain. Is it ok to use non inductive kinds? Sonically speaking, has anyone found significant difference between those two?

You got to use non inductive resistors, T.

I have no local parts source myself, so it's real expensive for me to do Parts Express orders for all kinds of little things. Alas.

Surprised you are not "agitated."
JK.

 

[Wayne Parham]

Does it really matter what kind of resistors we use as long as power rating is right? I would prefer non inductive kinds but ordering them is a pain. Is it ok to use non inductive kinds? Sonically speaking, has anyone found significant difference between those two?

Honestly, you probably won't hear a difference between wirewound and non-inductive resistors. I would have expected more difference, but don't hear (or measure) any difference. Could be because wirewound resistors have relatively low inductance compared to voice coil inductance.

On the other hand, non-inductive resistors don't cost much. I suggest you get them. They're like 50 cents a piece for 10w Daytons and only about three bucks for 12w Mills.

I always use a block of four 10 watt resistors in series/parallel. This may not be required for SET amps, but then again, the more power handling you have, the less thermal stress. With just one resistor, you'll notice it will get hot enough to melt plastic when the amplifier is turned up. Even at just 28 volts input, which is 100 watts into an 8 ohm load, the level required for about 115dB/M, the resistor will scorch the circuit board, possibly getting hot enough to melt the solder. This is still well below the failure point of the driver. So I suggest using a block of four resistors to form R1 and R2. I also suggest a 100 watt resistor for the Zobel. The 100 watt resistor is about ten bucks.

 

[mrTT]

Surprised you are not "agitated."

You talking to me? R U T A L K I N G TO MEEEEE?:uzi:
I am going through major depression now. The reason is...my head spins with all the options before me. I just want to raise my hf volume. Please somebody....

 

[mrTT]

I always use a block of four 10 watt resistors in series/parallel. This may not be required for SET amps, but then again, the more power handling you have, the less thermal stress. With just one resistor, you'll notice it will get hot enough to melt plastic when the amplifier is turned up. Even at just 28 volts input, which is 100 watts into an 8 ohm load, the level required for about 115dB/M, the resistor will scorch the circuit board, possibly getting hot enough to melt the solder. This is still well below the failure point of the driver. So I suggest using a block of four resistors to form R1 and R2. I also suggest a 100 watt resistor for the Zobel. The 100 watt resistor is about ten bucks.

Well, my amp is a wee little 14wpc eicos. But I don't mind adding more security into my crossover. 100watt? Wow! I've never thought of that one.

By the way. I don't mind paying 50 cent for a resistor or even three dollars. But that $10 shipping for 1 dollar resistor is just a killer.

 

[Zilch]

I just want to raise my hf volume. Please somebody....

Uhmmm, do you have the drivers on the waveguides yet?

BTW, should NOT be playing them without that.... :nono:

 

[Kegger]

An L-pad may be useful after that, I don't know. My problem with L-Pads has always been that they don't lock in the R1/R2 values and they take away some of the range. You really need 10dB of fixed padding in the R1/R2/C1 network to get it to work properly. After that, there's not much wiggle room. If you're using a woofer with less than 90dB/2.83v/M sensitivity, then sure, you could tack on an additional 10dB padding by using an Lpad. But woofers like that aren't well suited for this application, in my opinion. I'd suggest using woofers with greater than 90dB/2.83v/M sensitivity, and then choosing the right R1/R2/C1 values to match.

That's been my thinking when looking at the project as a whole, if I've got a high frequency
driver with high sensitivity I do not want to throw it all away on some low sensitivity woofer.

Yah the Waveguide and CD are what the systems design goals are but I also see this as an
fairly high sensitivity system and want to utilize that feature as well, not just for low output
amplifiers but the great dynamics that come with high efficiency as well.

My goal was to find woofers between say 92 and 98db sens to mate to the wave, I'd like to
upgrade the present HPM woofers for something with higher sens, but it's what I had at the
time that I could use quite easily, it works great but I would certainly prefer higher sens.

 

[Wayne Parham]

my head spins with all the options before me. I just want to raise my hf volume. Please somebody....

You are using an Altec 414, right?

If so, and if your compression driver is an 8 ohm unit, use these values:

R1 = 16 ohms
R2 = 16 ohms
C3 = 0.47uF

If your compression driver is a 16 ohm unit, use these values:

R1 = 50 ohms
R2 = 16 ohms
C3 = 0.33uF

No L-Pad

For 16 ohm resistors R1 and R2, use four Dayton 16 ohm resistors connected in series/parallel. If you need to use a 50 ohm resistor R1 for a 16 ohm compression driver, use four 12.5 ohm resistors in series.

The Parts Express part number for the 16 ohm resistors is 004-16. Order 16 of them if you're building a pair of speakers with 8 ohm compression drivers. The 12.5 ohm resistors are part number 004-12.5. If you're building speakers with 16 ohm compression drivers, get eight 16 ohm resistors and eight 12.5 ohm resistors. Either way, it will cost less than ten bucks and they'll be delivered in just a few days.

While you're there, get a pair of the 100 watt resistors for the Zobels, part number 019-020. They're about ten bucks each.

When you study the circuit and see the power dissipated in the Zobel resistor, you'll understand why I suggest a 100 watt part in that position. See page 66 of my Crossover Document. The Zobel resistor receives about 1/3 the input power, so a 10 watt part will be exceeded when the loudspeaker is presented about 9 volts at its input terminals, 30 watts across an 8 ohm load.

 

[mrTT]

Uhmmm, do you have the drivers on the waveguides yet?

Hey Zilch. Well of course. They arrived today. Sounds good. Really really good. In fact, so much better than 806s. To my ears they sound really clean and very smooth. I worried that it would be harsh but not at all. I am thrilled.

May be I've been hf deprived. On one hand the hf sounds pretty good at l-pad all the way up. On the other hand, I may use a little bit more. Say about 3db?

 

[skywave-rider]

....

I always use a block of four 10 watt resistors in series/parallel. This may not be required for SET amps, but then again, the more power handling you have, the less thermal stress. With just one resistor, you'll notice it will get hot enough to melt plastic when the amplifier is turned up. Even at just 28 volts input, which is 100 watts into an 8 ohm load, the level required for about 115dB/M, the resistor will scorch the circuit board, possibly getting hot enough to melt the solder. This is still well below the failure point of the driver. So I suggest using a block of four resistors to form R1 and R2. I also suggest a 100 watt resistor for the Zobel. The 100 watt resistor is about ten bucks.

I wondered why you did that. Had no idea they'd dissipate that much energy!

 

[skywave-rider]

You talking to me? R U T A L K I N G TO MEEEEE?:uzi:
I am going through major depression now. The reason is...my head spins with all the options before me. I just want to raise my hf volume. Please somebody....

C'mon, ur a psych major.

 

[Russellc]

That's right, exactly.

There is only one crossover schematic. Select the values of R1, R2 and C1 depending on how much attenuation is required.

If you think about it, that's how it always was. You had two schematics, but the only difference between them was the values of R1, R2 and C1. Now you have a chart of values, so you can pick the right values for any level required from -6dB to -20dB.

Thanks Wayne! Now you've brought it down in a simple way that is very clear! It just threw me that the I saw the .47 cap, but not the other value, 1.5, for the other crosover.

Thanks loads again!

Russellc

 

[mrTT]

You are using an Altec 414, right?

Correct.
Thanks Wayne.

 

[Russellc]

Hey Zilch. Well of course. They arrived today. Sounds good. Really really good. In fact, so much better than 806s. To my ears they sound really clean and very smooth. I worried that it would be harsh but not at all. I am thrilled.

May be I've been hf deprived. On one hand the hf sounds pretty good at l-pad all the way up. On the other hand, I may use a little bit more. Say about 3db?

Another falls under the econowave spell:beer:

Russellc

 

[mrTT]

C'mon, ur a psych major.

They are the most psychotic people on earth.
Hey man. Give me a break. We all need a place where we can hang luuse. :nutz:

 

[mrTT]

Another falls under the econowave spell

Count me in. I read the thread with much skepticism. What can a $40 CD and a $7.50 waveguide do? My 811Bs and 806s are full of muscle like MrT is! Well. I am really impressed. That's all I can say. I am extremely satisfied. My pursuit of 3DB may be coming out of my OCD.