Directiva r2 project: market requirements gathering

[Rick Sykora]

True, with that 750Hz being mainly set by baffle width. A wider baffle would send you down the Grimm route. My point being if you hold to a normal box shape then following the 8C route is inevitable, so might as well just say so.

If we were doing r2 as a commercial product, agree that we might have to make a decision earlier. But it is not. We can have multiple good variants and flex the cost if need be. If one variant is going to have ultimate directivity, then seems reasonable that it could be more costly to make.

 

[gy-k]

True, with that 750Hz being mainly set by baffle width. A wider baffle would send you down the Grimm route. My point being if you hold to a normal box shape then following the 8C route is inevitable, so might as well just say so.

I'm just stating the obvious here, the 8C has a fairly large waveguide and a crossover point of 1250Hz so down to that point the directivity control is from the waveguide.

 

[Rick Sykora]

Do we need some MoSCoW prioritization for the requirements? That might add some weight to some, and make it more clear that others are less important.

Have done before too, but in wanting to make this a fun effort, do not want to go overboard in repeating the commercial world. Always thought the W part was forced anyway…

So for now will keep this simple, am not willing to tradeoff features (in the basic version) and want quality to be high. This leaves time and cost. Time has to flex in a volunteer effort but we still need to get it done. For me, that mean cost has to flex most.

This all matters most to the design team and, as I have asked, I want this thread to focus more on requirements now. As part of cleaning it up, I may take some of the sidebars and design-oriented comments to another thread. So, if anyone is worried about finding their post here in the future, please avoid further design or lengthy discovery content not related to market requirements. Note I have added market to the thread title just to emphasize that these are NOT design requirements.

 

[voodooless]

Have done before too, but in wanting to make this a fun effort, do not want to go overboard in repeating the commercial world. Always thought the W part was forced anyway…

Well, yes, there is some valid criticism for doing it this way for sure, especially for commercial use Quite often because the parts that bring the biggest wow factor to products are declared as Co or W, and therefore degrade the total value or appeal of a product if they are not implemented at first. The trick here is to define them in terms of user value and not (only) technical merit.

So for now will keep this simple, am not willing to tradeoff features (in the basic version) and want quality to be high.

I think that will be unavoidable. But the goal is good to have for sure.

Might be good that a topic about the technical brainstorms on the R2 is on your name as well? I think it's clear that there is a need for a place to put that stuff.

 

[Rick Sykora]

IME, a W basically was seen by the engineering team as not really needed and so would get tossed to make schedule. So, ends up a waste of time for all involved.

I found that simple numeric ranking was a good compromise and allowing multiple features to have the same rank/priority allowed the lower ones to stay around as stretch goals. If the design team feels this is needed, am open to it...

One key diff now is that am also contributing to implementation too. So, if a task is dragging along, I can either help or quickly compromise to get it moving again.

 

[JanRSmit]

IIRC it had to do with the mechanism by which cancellation was achieved, and I suspect accurate modeling might be useful when "designing" the box, although it may be moot now that another line of attack is being deployed. I couldn't get to Small's original dissertation, but the pertinent page is attached wherein he describes the system as 4'th order, but also notes that the alignment peak and poor transient response is relieved by allowing the box to be leaky, and mentions flow velocity in this context while observing that f3 extension is sacrificed whereupon he wonders what the whole point is. I suspect he assumed ad lib choices of driver and cabinet, not appreciating that there may be occasions when life isn't fair, and one is forced to accept such compromises. I have no dog in this fight, and it remains that notecard in the dark recesses of the brain when it might be useful--such as a sealed alignment where space is at an absolute premium. Cheers.
Edit: BTW, I am curious how pressure relief is provided w/o leakage of some sort, except by storing the energy in a flexible balloon (with significant hysterisis) like membrane. Which will obviously represent a very temporary solution and unknown effect on damping--I suppose flexible wooden cabinets are a case of such where the pressure is converted to heat, but maybe not so much as to e useful. We are sealing with a resistive loss which implies flow, no?

One could try with very clean wool in the enclosure. When seen under a microscope it looks like a long pine-cone. The 'leafs' can open and close with pressure variations, thus absorb energy.
In closed box it is my choice, but can be overdone quite easy. Should be short fiber very well cleaned wool, typically for shipping packed densely pressed. Which is great, als you can pull it into small pieces of ramdomly varying size, thus also very effective in killing cavity resonances of the box.

 

[Rick Sykora]

Looks as though the forum requirement comments have slowed, so am going to close the review tonight and do a final pass with the design team before I finalize them.

Thanks for your interest!

Rick

 

[Morpheus]

Looks as though the forum requirement comments have slowed, so am going to close the review tonight and do a final pass with the design team before I finalize them.

Rick

Sorry, was going to wait till last minute to pitch in.
Excellent as the DXT is, especially for the money, I don't think we should make it the "de facto" only choice by not allowing more money into it for better performance. Paying more but getting something with even better ability ,especially better IMD, cleaner decay and ability to play loud and undistorted in its low operating range, and/or enabling lower crossover points frees us of the need of a smallish sized, but very large signal capable and extended upper operation woofer like the Purifi, thereby freeing a lot of resources...Plus, if we are building the bass modules, it may not be needed for the bass capability either. Both the the Kii ( with active cancellation, really cheap Peerless basss drivers for the bass) and the DD do without it. Also, tellingly, Grimm, chose to evolve its awesome LS1 at just by upping the tweeter, leaving the midwoofer well alone ( and also adding digital motional feedback to the sub, truth be told..), so that there sort of tells us where they found the glass ceiling....

 

[Rick Sykora]

Sorry, was going to wait till last minute to pitch in.
Excellent as the DXT is, especially for the money, I don't think we should make it the "de facto" only choice by not allowing more money into it for better performance. Paying more but getting something with even better ability ,especially better IMD, cleaner decay and ability to play loud and undistorted in its low operating range, and/or enabling lower crossover points frees us of the need of a smallish sized, but very large signal capable and extended upper operation woofer like the Purifi, thereby freeing a lot of resources...Plus, if we are building the bass modules, it may not be needed for the bass capability either. Both the the Kii ( with active cancellation, really cheap Peerless basss drivers for the bass) and the DD do without it. Also, tellingly, Grimm, chose to evolve its awesome LS1 at just by upping the tweeter, leaving the midwoofer well alone ( and also adding digital motional feedback to the sub, truth be told..), so that there sort of tells us where they found the glass ceiling....

No worries. The design team is not under any major driver selection constraints other than cost…

Based on experience, the Purifi woofer will not be easy to replace and expect that the new driver may consume more of the budget. That said, want the design to have great midrange performance. So, as we already know tweeter choices are limited, but do not know until other choices are made.

 

[gy-k]

Excellent as the DXT is, especially for the money, I don't think we should make it the "de facto" only choice by not allowing more money into it for better performance. Paying more but getting something with even better ability ,especially better IMD, cleaner decay and ability to play loud and undistorted in its low operating range, and/or enabling lower crossover points frees us of the need of a smallish sized, but very large signal capable and extended upper operation woofer like the Purifi, thereby freeing a lot of resources...

I'm not sure there's anything else really? Crossing a HF driver lower would need a waveguide, ideally, if the goal is to extend directivity control lower. There are a few dome tweeters with a larger sized waveguide vs. DXT, but I'm thinking right now that difference wouldn't be very substantial. The D&D uses a ~8" waveguide. If that approach is out, the Kii Three uses the DXT, but crossed to a 4" mid driver, supposedly at 2.5Khz, so it must be the woofers on the side providing directivity control below that, which would be more complicated.

 

[voodooless]

There are a few dome tweeters with a larger sized waveguide

You don’t need a total product. There are enough separate waveguides and tweeters that would make for great combinations.

 

[abdo123]

I feel like it's pointless to talk about distortion and output with a directivity focused design. something has to give in if there is no custom baffles in plan, and it's likely going to be the driver choice / matching.

If you want this:

Then the distortion will probably look like this:

 

[voodooless]

Then the distortion will probably look like this:

I really don’t see how that follows? Sure there are trade-offs to be made, but it doesn’t need to be this bad.

 

[abdo123]

I really don’t see how that follows? Sure there are trade-offs to be made, but it doesn’t need to be this bad.

it's barely even reaching 1% so it's not bad, yet a long 'this bad'.

It's difficult to match a 6.5 inch woofer to a tweeter, so for a directivity focused design something has to give, it won't be DIY work or price, so what other compromises are we left with?

 

[voodooless]

it's barely even reaching 1% so it's not bad, yet a long 'this bad'.

It's difficult to match a 6.5 inch woofer to a tweeter, so for a directivity focused design something has to give, it won't be DIY work or price, so what other compromises are we left with?

Find a woofer without a massive resonance at 1.8 kHz?There should be plenty. Or use a waveguide so you can cross much lower.. which I would prefer anyway.

 

[abdo123]

Find a woofer without a massive resonance at 1.8 kHz?

and you think they didn't think of that when they made the speaker?

 

[TimVG]

One of the R2 requirements is that the system should be relatively easy to build - that will include routing. This means elliptical waveguides and the likes will be hard if not impossible to implement in a flush mounted configuration.

Another requirement is that the parts need to be readily available and usable without further modifications. I know a lot of waveguides require further adapter plates and the likes to be used with otherwise readily available tweeters.

Also I believe 6,5" for the woofer is not set in stone, since this will be a 3-way system - it's a possibility, but not a certainty.

 

[voodooless]

and you think they didn't think of that when they made the speaker?

Probably they did, but didn’t care or made other trade-offs

 

[voodooless]

Another requirement is that the parts need to be readily available and usable without further modifications. I know a lot of tweeters require further adapter plates and the likes to be used with otherwise readily available tweeters.

Is a dome tweeter a requirement? Otherwise that would open additional options.

As for a generic waveguide: turns out they fit quite af few tweeters without modifications. Check out this article: https://www.soundimports.eu/en/blogs/blog/wave-guiding-your-favorite-tweeter/

 

[Tangband]

No worries. The design team is not under any major driver selection constraints other than cost…

Based on experience, the Purifi woofer will not be easy to replace and expect that the new driver may consume more of the budget. That said, want the design to have great midrange performance. So, as we already know tweeter choices are limited, but do not know until other choices are made.

Isnt the GRIMM style option maybe the one that makes most sence ?
If you think about it :

1. A wide baffle down to about schroeder frequency is a good thing. That means about 300 Hz.

2. Instead of the Seas excel unit used in GRIMM, why not use two SB 17cac-08 in 2,5 way figuration ? The distortion is lower than purifi.
One SB 17cac plays from 100 Hz ( crossed active ) up to the tweeter crossover at 2 kHz ( crossed active ) and the other from the baffle step frequency, ie from 300 Hz down to 100 Hz . A simple 6 dB aircoil passive at 300 Hz for this unit. No need for baffle step correction then.

3.The tweeter could be seas DXT crossed at about 2 kHz.

4.The bass could be GRIMM style closed box with LT with SEAS r0y crossed at 100 Hz active and the digital 3-way crossover a dbx driverack pa2.

whatˋs your opinion about that ?