Proposed build (i.e. X-over gurus in the house)

[jinjuku]

I am chatting via PM with Guiria about the Zaph ZA5 builds since I did a center channel.

If you have read my few comments You know that Mr. Krutke has designed a stellar driver almost regardless of price. The fact that it is $40 is just a lot of icing on the cake.

With that said almost any 5.5" driver is not going to be a mid or lower bass output, rock n' roll, monster. His driver is no exception. He speaks to this in his detailed notes section:

I was not after creating a 5" bass monster. For woofers of this size, these days it seems like the world is full of low sensitivity woofers with high Xmax and low Fs. I went in a direction that I consider opposite, creating a woofer that is as much of a dedicated midrange driver as it is a midwoofer. While 87 dB may not seem that high, there are far too many 5" woofers around the 85 dB mark or lower. 87 dB isn't so high that serious compromises or expensive design choices had to be made in other areas, but not so low that it couldn't be used as a midrange in a 3-way system.

I was thinking that it would be killer to do an MTM with the ZA5 but integrate a 10" Dayton Audio RSS265HF-4 with an x-over point of 450'is and let the ZA14W08 handle the 450 on up.

So I guess what I am after is a speaker with some rock n' roll capabilities that are still going to pull off the nuance of other genre's.

The RSS265HF-4 just seems like an incredibly smooth driver given what Parts Express shows as a model. I'm not sure what other drivers to look at for that range that is going to have some displacement for that mid and bass smack.

What modeling tools are going to help me get to a starting x-over design/redesign point that I can voice from there?

I wouldn't mind getting everything worked out with my DCX first and then go from there.

Thoughts, corrections, opinions are always welcome. This would be my next project after my Statements are finished.

Thx all.

 

[Swerd]

I was thinking that it would be killer to do an MTM with the ZA5 but integrate a 10" Dayton Audio RSS265HF-4 with an x-over point of 450'is and let the ZA14W08 handle the 450 on up.

The RSS265HF-4 just seems like an incredibly smooth driver given what Parts Express shows as a model. I'm not sure what other drivers to look at for that range that is going to have some displacement for that mid and bass smack.

What modeling tools are going to help me get to a starting x-over design/redesign point that I can voice from there?

Thoughts, corrections, opinions are always welcome.

The only thought I have is using the RSS-265HF subwoofer with a low-pass filter at 450 Hz may be trouble. It's FR curve shows breakup starts above 500 Hz. Even if you use a 4th order filter, you'll hear that noise. Consider using instead the Dayton RS-270 or the Peerless SLS 830668. Both are smooth out to about 1 kHz. A rough rule-of-thumb has it to use a driver that is smooth for another octave beyond the crossover point. As an example, Dennis Murphy uses the Peerless 10" woofer in a 3-way crossed 2nd order at 450 Hz.

 

[GranteedEV]

I also think 450hz is a tad bit high.... I would go with around 300hz give or take.

The upcoming 8 incher from exodus looks perfect for your application. While you may be looking for low fs low extension drivers you may also want to take a look at some JBL, Eminence, B&C, PHL, Acoustic Elegance, and Sellenium woofers.

 

[jinjuku]

The only thought I have is using the RSS-265HF subwoofer with a low-pass filter at 450 Hz may be trouble. It's FR curve shows breakup starts above 500 Hz. Even if you use a 4th order filter, you'll hear that noise. Consider using instead the Dayton RS-270 or the Peerless SLS 830668. Both are smooth out to about 1 kHz. A rough rule-of-thumb has it to use a driver that is smooth for another octave beyond the crossover point. As an example, Dennis Murphy uses the Peerless 10" woofer in a 3-way crossed 2nd order at 450 Hz.

Looks like the Peerless will work for the application and do it for less. I looked at the RS 270 and it's repsonse curve just looked more ragged than the RSS-265H.

I'm assuming partially by picking drivers that have a clean response curve and operating them easily inside their pass-band the the x-over work becomes a little easier.


The Peerless looks clean which I assume means less X-Over work. My reason for doing this is to get my feet wet in actuall design/voiceing/tweek and final implementation.

 

[jinjuku]

I also think 450hz is a tad bit high.... I would go with around 300hz give or take.

The upcoming 8 incher from exodus looks perfect for your application. While you may be looking for low fs low extension drivers you may also want to take a look at some JBL, Eminence, B&C, PHL, Acoustic Elegance, and Sellenium woofers.

I would be looking for a driver that would start its F3 at around 27-32hz and get me cleanly up around 4/500 Hz.

When you say Exodus are you referring to Kevn Haskins? I was looking to see if the Ex-Anarchy would work.

 

[GranteedEV]

I would be looking for a driver that would start its F3 at around 27-32hz and get me cleanly up around 4/500 Hz.

When you say Exodus are you referring to Kevn Haskins? I was looking to see if the Ex-Anarchy would work.

The current anarchy extends into about the 30s and is a 6.5 inch driver. its breakup mode is around 4khz so it'll work nicely in a 3 way but does lack sensitivity.

the upcoming anarchy drivers are an 8 incher as well as a 5 incher. kevin haskins posted some prototype pics. The 8 has an inverted dustcap instead of a phase plug.

 

[jinjuku]

The current anarchy extends into about the 30s and is a 6.5 inch driver. its breakup mode is around 4khz so it'll work nicely in a 3 way but does lack sensitivity.

the upcoming anarchy drivers are an 8 incher as well as a 5 incher. kevin haskins posted some prototype pics. The 8 has an inverted dustcap instead of a phase plug.

Did he post this at AVS? Link? Any preliminary #'s? NM, it's on his site.

 

[Swerd]

Looks like the Peerless will work for the application and do it for less.

Less money is always good . What kind of cabinet are you thinking of? Will you build the MTM as a sealed cabinet and sit it on top of a separate bass cabinet? You could always copy the separate woofer cabinet that Dennis Murphy used for that woofer.

The other thing I forgot to mention is woofer sensitivity. The Peerless is 89 dB (at 2.83 volts 1 meter away), the RS 270 90.5 dB, and the RSS 265 is 84 dB. I don't know how sensitive Zaph's MTM is, and how much resistance you'll have to add to get it balanced with your woofer, but the RSS 265 will clearly require the most. The less padding you have to add the better. I think it really should only be used as a powered subwoofer.

I looked at the RS 270 and it's repsonse curve just looked more ragged than the RSS-265H.

I'm not sure I understand what you mean here. Between 500 and 1000 Hz, the RS 270 is much smoother than the RSS 265.

I'm assuming partially by picking drivers that have a clean response curve and operating them easily inside their pass-band the the x-over work becomes a little easier. The Peerless looks clean which I assume means less X-Over work. My reason for doing this is to get my feet wet in actuall design/voiceing/tweek and final implementation.

Yes.

I haven't read Zaph's long description of his design, but another thing you might think about is baffle step compensation. As a separate MTM speaker, it needs more if it will be far from walls, less if closer. But as a 3-way like you plan, it probably won't need any. Does Zaph have alternative crossovers for the MTM design that have little if any baffle step compensation?

 

[GranteedEV]

Less money is always good . What kind of cabinet are you thinking of? Will you build the MTM as a sealed cabinet and sit it on top of a separate bass cabinet? You could always copy the separate woofer cabinet that Dennis Murphy used for that woofer.

The other thing I forgot to mention is woofer sensitivity. The Peerless is 89 dB (at 2.83 volts 1 meter away), the RS 270 90.5 dB, and the RSS 265 is 84 dB. I don't know how sensitive Zaph's MTM is, and how much resistance you'll have to add to get it balanced with your woofer, but the RSS 265 will clearly require the most. The less padding you have to add the better. I think it really should only be used a powered subwoofer.

I'm not sure I understand what you mean here. Between 500 and 1000 Hz, the RS 270 is much smoother than the RSS 265.
Yes.

I haven't read Zaph's long description of his design, but another thing you might think about is baffle step compensation. As a separate MTM speaker, it needs more if it will be far from walls, less if closer. But as a 3-way like you plan, it probably won't need any. Does Zaph have alternative crossovers for the MTM design that have little if any baffle step compensation?

That particular MTM only has around 3-4db of BSC.

His MTM center does offer an alternative but it's got a very different design concept with a forward midrange for improved speech.

Additionally, he has an MMTMM 2.5 way version of this with full BSC. I would love to use that 2.5 way in a digitally crossed 3.5 way.

 

[GranteedEV]

The Peerless looks clean which I assume means less X-Over work. My reason for doing this is to get my feet wet in actuall design/voiceing/tweek and final implementation.

With an analog 3-way, you also have to keep cascading slopes and phase shift in mind a lot.

 

[jinjuku]

Less money is always good . What kind of cabinet are you thinking of? Will you build the MTM as a sealed cabinet and sit it on top of a separate bass cabinet? You could always copy the separate woofer cabinet that Dennis Murphy used for that woofer.

Dual cabinets. Vented for the sub woofer, sealed for the MTM.

The other thing I forgot to mention is woofer sensitivity. The Peerless is 89 dB (at 2.83 volts 1 meter away), the RS 270 90.5 dB, and the RSS 265 is 84 dB. I don't know how sensitive Zaph's MTM is, and how much resistance you'll have to add to get it balanced with your woofer, but the RSS 265 will clearly require the most. The less padding you have to add the better. I think it really should only be used as a powered subwoofer.

The ZAW14W08 is 87dB (I thought I read somewhere +/- 2dB) so looks like that peerless will pretty much line up.

I'm not sure I understand what you mean here. Between 500 and 1000 Hz, the RS 270 is much smoother than the RSS 265.

I thought I was refering to the 10" HF driver being smoother.

I haven't read Zaph's long description of his design, but another thing you might think about is baffle step compensation. As a separate MTM speaker, it needs more if it will be far from walls, less if closer. But as a 3-way like you plan, it probably won't need any. Does Zaph have alternative crossovers for the MTM design that have little if any baffle step compensation?

I was hoping to have a wider baffle to reduce / eliminate the BSC circuit.

I am trying to use some already proven designs that I can modify as a starting point for learning. I still plan to get this all worked out with a 6 channel amp (used NAD maybe) and my DCX. Then when that is worked out move on to te passive x-over phase.

I have to get a bafflle board going. Already have REW and a EMC8000. Need to setup a treated space to do measurements in.

 

[jinjuku]

With an analog 3-way, you also have to keep cascading slopes and phase shift in mind a lot.

I have seen with a ton of designs where the mids are wired inverse to the rest. I have a rudimentary understanding of frequency vs phase.

 

[GranteedEV]

As a general rule, in a 3 or four way, you want the crossover points to be a rough decade apart for things to sound cohesive.
So if your Bass-mid XO point is 300hz, you need the mid-tweeter crossover at 3khz or higher.
With that said, some speakers, like the four-way revel salon, totally ignore that "rule" and still get great results. I guess crossover order has significance as well, as that speaker uses all fourth-order acoustic slopes.
I don't know how digital FIR slopes come into this picture. They're a lot simpler in my mind however.

Either way, if going analogue IIR, I would try my hand at going 2-way first, rather than trying a 3-way using someone else's 2-way. That's just me... I learn by experience and then expand upon it.

 

[jinjuku]

Looks like the ZAW can be easily crossed at 2.5K so that means a 250hz xover point which puts the the Peerles 83068 and Dayton SVC into the mix for not a lot of cash and a usable range of 30-250 Hz easy.

The sensitivity is 88.7 on both so I would would only have to cover a 1 to 1.5 dB gap. I guess a BSC circuit pushing the output on the MTM by 1.5 dB is doable.

 

[GranteedEV]

If you don't mind spending, I highly recommend a pair of these:

http://aespeakers.com/shop/catalog/product_info.php?products_id=61&osCsid=9c6871f44afddf973754e4bb126d8b4f

To do the bass.

 

[TLS Guy]

Looks like the Peerless will work for the application and do it for less. I looked at the RS 270 and it's repsonse curve just looked more ragged than the RSS-265H.

I'm assuming partially by picking drivers that have a clean response curve and operating them easily inside their pass-band the the x-over work becomes a little easier.


The Peerless looks clean which I assume means less X-Over work. My reason for doing this is to get my feet wet in actuall design/voiceing/tweek and final implementation.

I vote for the Peerless, it is a good driver.

350 to 400 Hz is a good point for a passive three way. You need to select drivers that can operate up to about an octave above the and below the crossover points.

 

[jinjuku]

I vote for the Peerless, it is a good driver.

350 to 400 Hz is a good point for a passive three way. You need to select drivers that can operate up to about an octave above the and below the crossover points.

The ZAW is good for 3K but Zaph recommends getting closer to 2K. So a 2.5K X-over point looks doable. I would like to let the peerless 10" operate in the fattest part of it's operating band. It's really temping to let it x-over at 400/450. I just don't know what that means for setting the next xover point at 2.5K.

So does it mean that a X-over point of 450 for the Peerless is ok since it is clean up to 1K?

Not sure the where the rule of thumb of a 'decade' above comes from. I believe the thought is get a driver that is not going to break up mechanically in the fundamental above it's x-over point (the next order of distortion). Maybe a detailed explanation would help me about the rule of decade (X10).

Thx.

 

[jinjuku]

If you don't mind spending, I highly recommend a pair of these:

http://aespeakers.com/shop/catalog/product_info.php?products_id=61&osCsid=9c6871f44afddf973754e4bb126d8b4f

To do the bass.

Trying to keep a value perspective on this. If I can do a 3 way pair for $400 in parts I would be stoked.

 

[GranteedEV]

The ZAW is good for 3K but Zaph recommends getting closer to 2K. So a 2.5K X-over point looks doable.

The driver itself would probably be nice crossed as high as 4khz with a higher (4th acoustic) order slope and notch filter..but i guess it depends what kind of directivity control you are looking for more than anything. What I would personally like to do is cross it to the seas DXT tweeter in a tmmw.

Not sure the where the rule of thumb of a 'decade' above comes from.

Im not too knowledgeable but I believe it is mostly about keeping XO points away because of how the midrange highpass may not as well affect the tweeter slope if close enough...and the midrange lowpass might not affect the woofer slope as well... something along those lines? as in, you can't look at any one crossover in isolation.

There are a lot of 'rules' out there, and I'd be one to challenge and put each and every one to the test both in measurement and controlled listening test.

 

[TLS Guy]

The ZAW is good for 3K but Zaph recommends getting closer to 2K. So a 2.5K X-over point looks doable. I would like to let the peerless 10" operate in the fattest part of it's operating band. It's really temping to let it x-over at 400/450. I just don't know what that means for setting the next xover point at 2.5K.

So does it mean that a X-over point of 450 for the Peerless is ok since it is clean up to 1K?

Not sure the where the rule of thumb of a 'decade' above comes from. I believe the thought is get a driver that is not going to break up mechanically in the fundamental above it's x-over point (the next order of distortion). Maybe a detailed explanation would help me about the rule of decade (X10).

Thx.

You have to space the crossover points of the band pass (mid) driver at least three octaves and preferably more, 3.5 or more octaves is best. You just cant control the band pass gain well otherwise and they just sound bad.

That means a mid range driver needs a bandwidth of 200 Hz to 8 kHz. There are very few contenders.

The other option is to lower the crossover and have the bass/mid crossover active. However it is optimal to have the mid range driver cover the whole of the speech discrimination band.

The next issue is that at least for classical music, the band pass driver has to carry the lions share of the power. So you really need a driver that will handle 100 watts plus and have a frequency response out to 8 kHz.

There really is no longer a driver available to home constructors that really fits that bill.

So I was forced to use a slight of hand and use an acoustic crossover high pass and a first order transitioning to second order low pass active, at the same time providing step loss compensation in the low pass signal. The crossover are 45 Hz and 2.8 kHz. The latter because of the fierce break up mode of the SEAS excel drivers at 4 kHz. The LFE signal being shelved in to both 10" drivers below 60 Hz.

So at your price point in my view you will be far better off with a two way or 2.5 way. You will end up with a flatter mid band and much greater power handling where it counts.

I wish I had the resources to return to the Jordan Watts Mk 3, for which I was largely responsible.

This very light 4" foil aluminum cone weighing only 6 GM has no nasty break up modes and flexes such that the radiating cone area decrease predictably as frequency increases. It starts a first order roll off at 6 kHz. It needs no low pass filter, and a single unit sounds well balanced full range.

The problem is limited power handling, only 15 watts and a slight suspension resonance at 325 Hz with a second order mode at reduced amplitude at 750 Hz.

I think with modern materials, the cone is light enough, the rear suspension could be dispensed with. I think ferro fluid would increase the power handling considerably.

I'm convinced that driver could be refined into a world class mid. I think you could get superb power handling from 350 to 400 Hz on up, and you would only need a first order high pass filter to the tweeter starting at 6 kHz.

I just wish I had the resources to pull it off.