First 3.1 w/ DSP

Looking to get some input from some about my first DIY using DSP. Sold my Infinity RS2.5s years back and have wanted to get something with similar sound. Heard a pair of BG z-92s and would like to duplicate something like the Z-92.



For speakers, BG Neo 3 PDR mounted coax in front of the BG Neo 10 as in the Z-92. I will move the Neo3 to the top for a test but I would like the coax because of the size. I was thinking about a pair of Dayton RS150s for bass, any other ideas for the bass drivers? Front ported box with full width rectangle port.


DSP is a MiniDSP 2x8


Amp will be an Emotiva UPA-700. I was going to build my own amps but I don’t think I could build a 6 ch amp to compete with the 700 for $450. Would have loved to get the amp-DSP, DSP4You PWR-DSP3 3 channel plate amp with DSP built in but the cost of $1200 each was too much. Tried to get just the plate amps from pascal-audio, they would not sell them to me; I am not a “loudspeaker
manufacturer”.


Speakers will be tuned and will roll off to a sub somewhere in the 80hz. To start, going to use x-over freq of the z-92, 500hz and 2.3khz. Obviously, with the DSP changes to the x-over freq, slopes etc. will be easy to implement.

Any ideas?
 
I'm surprised nobody has commented yet. This will be one great speaker! I'm excited for you!

Will you be able to get the coax mount for the planar drivers or do you plan to construct one yourself? I think DSP is the way to go, it's much easier to prototype and you can take your time tweaking the driver integration until you get it perfect. You could even experiment with delaying the Neo3 to compensate for the larger planer sitting further behind.

Although a pair of RS150s will cover the low end well I would use bass drivers that are as large as I could get away with. If you want thin speakers you might consider mounting the bass drivers on the side walls. For better integration with the sub I would advise against porting the bass enclosure. You can apply a bass boost with the DSP if needed. You could also get away with running the pair of woofers from a single amplifier channel if you don't plan to play insanely loud. With a pair of subs driven by the DSP this will be a killer system!

I have one last suggestion - consider Open Baffle!
 
Boris, thanks for the reply. I cant find the coax mount anywhere. Didn't think I would. I do have a machine shop here at work so I am going to have to design one on Solidworks and get them to mill a few out for me. Cost me a few 6 packs!.

Going with a larger bass drive I was a little worried about the upper end x-ing over to the neo10 at 350 or so. Do you think an 7" like the rs180 would be a better match? Any other bass to consider? What is the reason of not porting to integrate with the sub?

I didn't consider open baffle. I think Ill give it a test run.

Brad
 
Earl, again thanks for a reply. I have the amp on order already. I looked up many reviews and did not see one that was negative. In fact most were very positive. A search on Absolute Sound sight brings up a lot of info.

" The THD versus signal level is absolutely terrible"

Looking at numbers the amp is rated at 0.01% THD at output with a 870mv input to reach full power. Does this seem that bad? Maybe I didn't understand the statement.
A $6000 McIntosh at 0.005% THD. I am a little concerned about the signal to noise at 97db 1w.

Side question, Have you ever worked with the B&C 8 CTX Coax Driver?

Brad
 
What is the reason of not porting to integrate with the sub?

Phase integration. This topic is a bit over my head but I will explain how I understand it and if somebody objects they can correct me.
The signal leaving the port is not in the same phase as the signal in front of the driver. This causes a shift in the overall system phase. Classic crossovers like the Linkwitz-Riley have symmetrical slopes and expect linear phase from both signals for proper summation. You may be able to get good summation with asymmetrical crossover slopes but that will require precise measurements and tweaking which will not be easy to obtain that low.


I only suggested larger bass drivers because as a general rule sound quality degrades with increased excursion. I use the RS180 and they have a great low-end, they are probably the best woofers for the money. (But make sure you kill the metal resonance with a large notch!) On the other hand I stare at the wide frame and the phase plug and keep thinking how they could have squeezed a bit more surface area. There are a lot of woofers that will do 80Hz without a problem and integrate well with the Neo10.
 
Earl, again thanks for a reply. I have the amp on order already. I looked up many reviews and did not see one that was negative. In fact most were very positive. A search on Absolute Sound sight brings up a lot of info.

" The THD versus signal level is absolutely terrible"

Looking at numbers the amp is rated at 0.01% THD at output with a 870mv input to reach full power. Does this seem that bad? Maybe I didn't understand the statement.
A $6000 McIntosh at 0.005% THD. I am a little concerned about the signal to noise at 97db 1w.

Side question, Have you ever worked with the B&C 8 CTX Coax Driver?

Brad

If the THD rises as the signal level falls then the amp has crossover distortion. Crossover distortion is the most insidious of all forms of distortion because levels as low as .001% can sound terrible. THD as a single number is relatively meaningless. Look at the charts for that amp and you will see a rapid rise of THD as the signal level falls, and it is different for each amp. This is clearly a sign of a bad setting (or design) for the Class AB bias.
 
Gedlee, did a lot of reading last night after getting your post on the amp. I am just starting to learn in depth about this stuff. I read thru most of your web sight and papers also. Nice light reading! Reading thru the Audio Precision sight on audio test standards the speaker they use on there front page sure looks like one of yours with the foam on the high driver. I could not find a lot of amps that publish a THD vs Power graph. The ones I could find all seem have a rising THD as the signal level falls from a minimum THD at about the 80% power level. All graphs I found besides the Emotiva had a minimum power level of 100mW, some stopped at 200mW; the Emotiva graph goes to a minimum of 0.1mW. Which looking at the Emotiva graph the THD goes very bad at 12mW. Heck, McIntosh states their THD specs as "from 250 milliwatts to rated power". Wonder what McIntosh would look like at 10mW? If Emotiva cut the graph at 100mW would it still look "insidious" to you? Maybe just a case of the graph range? I just reread what I wrote and it sounds like Im arguing with you. Please do not think I am. I am just learning and have a long way to go. It is odd that all the amps channel do not track each other with the THD. Bad quality control? Bad parts? Bad Design?.....