or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › Multi-sub: adding DIY-SG flatpack 18"+15" sub to 4x15" IB line array
New Posts  All Forums:Forum Nav:

Multi-sub: adding DIY-SG flatpack 18"+15" sub to 4x15" IB line array - Page 2

post #31 of 89
Thread Starter 
Easier for me to answer into your original post, thx for time.
Quote:
Originally Posted by Stereodude View Post

Quote:
Originally Posted by mtbdudex View Post

I have the new Denon 4520CI with true independent subwoofer pre-outs.
So you can set different distances for each subwoofer?
I've not set up the 12" box sub into the new 4520CI as a 2nd sub, it's borrowed from upstairs and I did not want to mess up my prior Audyssey calibration.....though if I do save that calibration that would be a way to do that.

Looks like I will be doing this:
Run both thru the AVR manually input the distances, turn off Audyssey but it will apply distances to each sub, then using front input for REW input, turn off my XPA-3 and then both subs have distance corrected
Rinse and repeat for 3rd sub location, then I should be able to use REW with A +B, save that result as A',then add the 3rd location to that as A' + B (B being 3rd location) for simulation of 3 subs....

Quote:
Your suggestion - find peaks/valleys in opposite location....YES, but the $ question is where is that?
Well, IMHO, it's not likely to be in the adjacent front corner. When I was setting my system up I had a finite area where I could place the "fill" subs so I moved them around within that envelope, got things the best I could, and fixed the rest with EQ and delay.
Understood, I'll do same approach, pick real world placement zones and focus on those
I'm trying for both 1st and 2nd row, 8 seats......this will take time, I wonder how many have gone thru the whole multi-sub thing....and lived to tell about it wink.gif
Quote:
My room really is non-symmetric from LF/ULF viewpoint:
My great room ends up being very non-symmetric also. It's sort of a rectangular room by the footprint. However, the long interior wall on the side is missing opening the great room to the whole house. When I wanted to integrate my DIY subs with my previous UFW BBK subs I started with them in the back corner pointed forward for complementary fill. That didn't give me very good results. I tried messing with the direction they were pointed but that only made it worse. I started sliding them along the wall pulling them toward the front of the room and found things got better. I pushed them as far forward as I could until I started to encroach on the side surround (tower speaker). By comparing the raw frequency and phase responses I manipulated the net response as flat as possible with a delay on the UFW BBK subs. Finally, I EQ'd all the subs together.

Here's roughly what I ended up with:


Looks like we both have OCD for achieving flat bass as much as possible, that's what these forums do to some people...
Thx for educating me and sharing your experience, it gives hope that with tenacity and help from others it can be done


One step I omitted above was determining an appropriate output level. It's just an important as placement since if you change the output level of the two subs relative to each other the overall frequency response changes. You don't want one sub to run out of steam before the others or go into distortion while the other(s) are still playing nice and clean. Since the two UFW BBK subs each have a 15" driver and a 1000W amp and each of the two DIY AV15-H sonosubs has 2 15" drivers and each enclosure gets 1200W I decided to set the combined two UFW BBK subs equal to one of the sonotubes. This way each set of 2 15" drivers was being called on for roughly the same SPL. Before any EQ'ing or messing with phase I used pink noise and found the two sonotubes were almost 6db apart in output (due to location) even though they were getting the same signal and power. So, I set the combined two UFW BBKs subs equal to the average of the two sonotubes and went from there. I should also point out that the two AV15-H sonotubes are treated as a single sub electronically. I plan to reevaluate that over the holidays and see if any improvement can be gained (specifically a reduction in the amount of EQ needed to flatten things out) by treating it as a 3 sub system instead of a 2 sub system.

Well that complicates things, as I like to "show off" the IB sub at times and crank up the gain on the EP2500 for events, so "small" 10" or even 12" helper subs won't keep up with the IB, which keeps going and going....hence I'm back to at least 15" driver in 3cu ft box powered by pro amp not plate amp, those flat packs by EricH may be bought after all if this analysis comes together and leads into that direction
post #32 of 89
Thread Starter 
Had some time this morning before heading out to another family visit....
Double checked my set-up, cal file via loopback, etc, all checked.

Set each sub independent to 75bd, and ran REW to test the A+B summation.....something is not working like expected, below is my graph.
Purple is IB sub alone, Dark green is 12" box sub alone, RED is with both running, while light green is using REW A+B.
(seat#6 only)


Now I'd expect the REW A+B to truly match the real world combo measurement, but it's not.
The dips at 27hz and 36hz in the RED simply don't make sense at all....

I've tried to attach the mdat file to share here...I'm getting upload file error
(I made it a zip file, it's 7.6MB in size)



I'll post this to JohnM in the REW forum over at HTS.
Edited by mtbdudex - 3/17/13 at 4:14am
post #33 of 89
Quote:
Originally Posted by mtbdudex View Post

Had some time this morning before heading out to another family visit....
Double checked my set-up, cal file via loopback, etc, all checked.
Set each sub independent to 75bd, and ran REW to test the A+B summation.....something is not working like expected, below is my graph.
Purple is IB sub alone, Dark green is 12" box sub alone, RED is with both running, while light green is using REW A+B.
(seat#6 only)

Now I'd expect the REW A+B to truly match the real world combo measurement, but it's not.
The dips at 27hz and 36hz in the RED simply don't make sense at all....
I've tried to attach the mdat file to share here...I'm getting upload file error
(I made it a zip file, it's 7.6MB in size)
I'll post this to JohnM in the REW forum over at HTS.

I suspect IB does not act as point source through full passband.
You are in fact summing 5 sources (4 + 1) with distinct acoustic centers and phase characteristics instead of two.
is there an easy way to check A+B summation with only 1 of the 4 IB subs active?
post #34 of 89
Thread Starter 
Quote:
Originally Posted by zheka View Post


I suspect IB does not act as point source through full passband.
You are in fact summing 5 sources (4 + 1) with distinct acoustic centers and phase characteristics instead of two.
is there an easy way to check A+B summation with only 1 of the 4 IB subs active?

Hmmm, but the mic should pick up that at the seat location, so the REW "math" should equal the real world measurements??

That would be a PIA, since I'm running the 4 x 15" line array drivers parallel/series, and the EP2500 is bridged mono....I'd hate to re-wire it, this is getting more complex and overtaxing my 50 year old brain eek.gif, I do appreciate support and learning as I go.

I've posted this in the HTS REW forum
http://www.hometheatershack.com/forums/rew-forum/64240-why-b-not-working-my-2nd-subwoofer-placement-analysis.html#post575698

I'd like to get this REW A+B summation figured out before proceeding with many measurements.

Here is the mdat file via HTS upload to thread there
http://www.hometheatershack.com/forums/attachments/rew-forum/39548d1356534727-why-b-not-working-my-2nd-subwoofer-placement-analysis-dec-26-2012-ib-study-2nd-row.mdat
Edited by mtbdudex - 12/26/12 at 7:26am
post #35 of 89
Quote:
Originally Posted by mtbdudex View Post


http://www.hometheatershack.com/forums/rew-forum/64240-why-b-not-working-my-2nd-subwoofer-placement-analysis.html#post575698
I'd like to get this REW A+B summation figured out before proceeding with many measurements.

i am very curious what John would have to say about it. subscribed.
post #36 of 89
Thread Starter 
Quote:
Originally Posted by zheka View Post

Quote:
Originally Posted by mtbdudex View Post


http://www.hometheatershack.com/forums/rew-forum/64240-why-b-not-working-my-2nd-subwoofer-placement-analysis.html#post575698
I'd like to get this REW A+B summation figured out before proceeding with many measurements.

i am very curious what John would have to say about it. subscribed.

Me too!! JohnM is very helpful.
post #37 of 89
Are you using the 2nd channel loopback as a timing reference? link
post #38 of 89
Thread Starter 
No, when doing ETC charts I do, but for this clicked off and cable not plugged in, should I?
post #39 of 89
Quote:
Originally Posted by mtbdudex View Post

No, when doing ETC charts I do, but for this clicked off and cable not plugged in, should I?
Maybe. I'm not sure how REW determines phase without the loopback. If it isn't able to accurately determine phase the A+B will never work correctly.
post #40 of 89
Thread Starter 
Per JohnM I need to have the ref loopback connected and checked in the analysis tab.
Glad the answer is known, I'll redo measurements tomorrow.
Small steps for success.
post #41 of 89
Thread Starter 
After much reading here via search in "multiple suboofers", I'm going to NOT do a 2nd corner sub, rather follow member coctostan's advice:
-1 corner sub front wall; My current IB sub will have the ULF duties
-2nd sub front wall near the other corner, I'll build a sealed box sub with the Dayton Audio RSS460HO-4 18" driver , EricH flatpack most likely will be used
-3rd helper sub along side wall, I'm looking at 12" sealed box with Dayton Audio TIT320C-4 12" Titanic Mk III, again EricH flatpack most likely will be used
-I'll use a NU3000DSP to drive both of them and set phase/delay

>>Any thoughts on the above 2nd and 3rd subs I'm considering?

I could burn hours taking measurements with my 16 year old 12" boomy box sub, when I can move forward with a tried and true multi-sub approach.
I liked the idea of the 2nd sub in the corner, but many people kept saying don't put the 2nd sub in corner if doing the Geddes approach, a box sub gives me more options for optimal placement on the front wall vs being forced into the corner.
Quote:
Originally Posted by coctostan View Post

I like the Geddes approach. Here is a nice walkthrough: http://mehlau.net/audio/multisub_geddes/


This is for 3 subs, but you can do the same with 4. Just treat every sub beyond your first two like the 3rd. The key is that the subs in the rear of the room produce only a fraction of the total output.


I don't know what drivers you presently have, but you might be best served to keep the four you have up front and simply add 1-2 smaller cheaper subs in the back. Using larger drivers designed for significant low frequency output is a waste on those rear subs.


I don't think that using a single Bassis would be advisable. At a minimum you need level and phase control for each sub location (not per woofer, but per woofer group that is colocated). This is part of the "blending" process that is described in the link above. Using something like the DCX or MiniDSP is ideal IMO, but the phase and level control on some plate amps would be fine.


There might be other approaches, but I think this is a good practical and feasible route.
Quote:
Originally Posted by coctostan View Post

Although you are asking DS-21, we tend to agree on this stuff (except I like content below 20hz and DS doesn't much care for it) so I will respond.


The reason for the individual phase control is not for dialing in the system for a single position but to "alter" how each bass source interacts with the other bass sources. Delay or phase is not as simple as adjusting for a specific distance, especially at these frequencies. You are simply trying to excite a combination of modes that gives you the most even response across multiple positions. Small phase adjustments is just an additional tool.


Like I said before, you are best served using your "heavy lifters", the 4 RSS390HFs up front in two separate positions. These front bass sources can be set up to provide your full SPL and frequency tilt that you desire. I would put them in 5 cu ft each wired in series for an 8 ohm stereo load on the NU3000DSP. That is 315w RMS/channel which happens to be about how much power it will take to reach Xmax. Place one of the pairs in a front corner (Sub A) and place the other pair (Sub B) a few feet from the other corner along the front wall.


Start bringing up the level of Sub A (the corner sub) until its average level is about 2-3db below your final target (probably 75 or 85db). Once you have that set, adjust the phase/delay until you get the nicest response you can get (this should be done with your mains running the same test signal).


Next, while Sub A is still running, bring up the level of Sub B (front wall non-corner sub) until you have reached your desired level match. Adjust phase/delay to get the best response you can.


Finally, add in the smaller 10-12" sealed sub with the plate amp. The ideal location for this is elevated above the halfway point of the wall, but that is not necessary. Now, while everything else is running, bring up the level of this subwoofer until you see some smoothing effect. You shouldn't see any significant increase in level. Then adjust phase until you get the smoothest response across multiple positions. This sub will put out very little relative to the fronts.


You should see smoothing from the 3rd sub and more smoothing as your tweak the phase. Peaks should come down and nulls should fill.


Finally, adjust eq to flatten the response at your primary position. It should be pretty flat already, but some tweaking might be needed. Test in other listening positions to make sure you aren't throwing things way off.


The only catch to this is that you will only have EQ on your front subs and not the third sub. This shouldn't be much of an issue IMO, but you do lose some flexibility. You could go with a MiniDSP instead. I'd probably suggest that route anyway so you can do a proper Linkwitz Transform on your subs.


Which brings up another point. You should setup your Linkwitz Transform prior to any of this. How you do this will be based on how your sub reacts to your room (aka room gain) and how much headroom you have down low. If you are using the NU3000DSP you will apply a shelf filter. With the MiniDSP you will apply a proper LT. I wouldn't apply any more than 4-5db of boost with your setup. (that would target an LT with a tuning around 25hz with your subs) How deep that will get you depends on your room.


Feel free to ask questions. I'm just writing this off the top of my head and is the process I use. Some of it may not make as much sense to other people as it does in my head.
post #42 of 89
Quote:
Originally Posted by mtbdudex View Post

After much reading here via search in "multiple suboofers", I'm going to NOT do a 2nd corner sub, rather follow member coctostan's advice:
-1 corner sub front wall; My current IB sub will have the ULF duties
-2nd sub front wall near the other corner, I'll build a sealed box sub with the Dayton Audio RSS460HO-4 18" driver , EricH flatpack most likely will be used
-3rd helper sub along side wall, I'm looking at 12" sealed box with Dayton Audio TIT320C-4 12" Titanic Mk III, again EricH flatpack most likely will be used
-I'll use a NU3000DSP to drive both of them and set phase/delay
>>Any thoughts on the above 2nd and 3rd subs I'm considering?
I could burn hours taking measurements with my 16 year old 12" boomy box sub, when I can move forward with a tried and true multi-sub approach.
I liked the idea of the 2nd sub in the corner, but many people kept saying don't put the 2nd sub in corner if doing the Geddes approach, a box sub gives me more options for optimal placement on the front wall vs being forced into the corner.

I think coctostan recommended the second sub a few feet away from the corner, which maybe enough to decorrelate from the other front sub. you will see when you try. in general, using symmetrical positions is against the basic idea behind this approach.

Three subs is definitely the way to go, but you can get a significant improvement with just two.

This is how it worked with a pair of Epik Legends in my room



Gold - midpoint front and back wall, per Welti.
Blue - Geddes approach. First sub is in the front left corner, second - 1/3 from the left back corner. The LPF on the second sub is set at approx @ 50Hz and volume is trimmed down compared to the main one.
Purple - after audyssey XT

I am using 3 subs now - one fairly capable dual 18" sealed (CHT SS-18.2) upfront and the pair of Legends as supplemental subs along side and back walls. The Epiks are probably an overkill for what I am using them for.
I am contemplating second powerful sub up front to boost the output. But I do not expect it to help with the FR. As far as the multisub, both front subs will form single "first" unit and the Epiks will act as the "second" and the "third" ones.
post #43 of 89
Thread Starter 
zheka - Looks like Audyssey XT did a decent job for you!
Neat to see how the different multi-sub approaches did in your room.

I've got XT32 in my AVR, it will be neat to see how it handles my 2-3 subs on its own following the FAQ for audyssey multi-subs, I do plan on using the NU3000DSP for the integration also.
Plus I got the Pro key, I'll borrow a pro mic system from a friend when these subs are done for integration ability.
A nice winter project.

Now, for the sub cabinets, I'll have to see where EricH is in his 12" and 18" flat packs.......
post #44 of 89
Quote:
Originally Posted by mtbdudex View Post

zheka - Looks like Audyssey XT did a decent job for you!
Neat to see how the different multi-sub approaches did in your room.
I've got XT32 in my AVR, it will be neat to see how it handles my 2-3 subs on its own following the FAQ for audyssey multi-subs, I do plan on using the NU3000DSP for the integration also.
Plus I got the Pro key, I'll borrow a pro mic system from a friend when these subs are done for integration ability.
A nice winter project.
Now, for the sub cabinets, I'll have to see where EricH is in his 12" and 18" flat packs.......

I am running XT32 now (4311ci). One thing I learned is that using dual sub capabilities of Sub EQ HT along with the multisub may not work well. I ended up using single output so that all subs were EQ'd by Audyssey as if it was a single unit.

good luck!
post #45 of 89
Thread Starter 
I just got a call from Parts Express, they offered me the Dayton Audio TIT320C-4 12" Titanic Mk III a re-coned one for $125 vs the $167 I paid for a new one.......I was leery on getting it as who knows how else if was abused, so I kept with the new 12" driver.

Looks like both my 18" and 12" drivers will be shipping possibly Wed Jan 3rd, small back order on both but they are getting stock.

Also, the loop back worked for adding A+B in REW, so glad that is cleared up for me as well.
Red IB alone, Purple 12" box sub alone, Dark green is both via real time measurement, light green is via REW A+B function.
I'll leave the loopback cable on always from now on.




Last thing.....it sure is a lotta wires and cables doing these acoustic measurements!
Plus, I had a slight 60hz ground loop hum, until I remembered to put the 12" box sub and the iMac onto the same 20a supply as the IB sub - that orange cord is plugged into the front of my 20a Furhman delay, and not a completely different circuit (my wall plugs).

Which also pointed out to me no way can I ever use plate amps here....even though I wired for 2nd/3rd sub on LH wall front/rear, that is RG6U coax, I should have also ran dedicated circuits there with same tie into circuit breaker as my IB sub is running off.

Edited by mtbdudex - 3/17/13 at 4:15am
post #46 of 89
Thread Starter 
Re-cap for those new to this thread:
I am using a 12" box sub to simulate different potions for multi-sub arrangement, my RH IB sub is fixed as 1st sub.

With REW now better understood by me, I did measurements yesterday, here are some of the comparisons organized hopefully so easy to follow:
(yea, after the fact I need to override REW and force same colors on different measurements when making charts for visual comparison, I started doing that somewhat.....I also tried to make the font larger so can post small charts readable, that did not work for me)

1st row analysis, baseline IB sub only:
1st%2520row%2520seat%25201-4%2520ib%2520only%2520overlay.jpg

Baseline IB + 12" box sub in LH corner.
I actually like what this did except the 3rd seat null that appeared, the yello dip at 27hz.
Honestly this graph is making me re-think going to the corner location again, I'm hoping a 3rd 12" helper sub would further flatten these peaks/valleys out.

LH corner 12" box sub
1st%2520row%2520seats%25201-4%2520ib%2520and%2520sub%2520in%2520lh%2520corner%2520overlay.jpg

I then did (4) 12" box sub locations along the front wall:
LH corner, and from LH wall 3ft, 4ft, 5 ft. Closer than 3f is physically constrained by the LH main, and further than 5' is constrained by the center channel shelf.
Note:I have no way to change the phase/distance easily on these - my NU3000DSP will not be here till next week, I'm hoping since along the front wall phase is not an issue....?

Seat 1 and (4) 12" box sub locations along the front wall
dec-28-2012%2520seat%2520%25231%2520combo%2520overlay%25204%2520locations.jpg

Seat 2 and (4) 12" box sub locations along the front wall
dec-28-2012%2520seat%2520%25232%2520combo%2520overlay%25204%2520locations.jpg

Seat 3 and (4) 12" box sub locations along the front wall
dec-28-2012%2520seat%2520%25233%2520combo%2520overlay%25204%2520locations.jpg

Seat 4 and (4) 12" box sub locations along the front wall
dec-28-2012%2520seat%2520%25234%2520combo%2520overlay%25204%2520locations.jpg

My thinking after going thru the above:
Re-consider the corner location again, I'll do plots of the 4 seat locations overlaid with each sub in 3', 4'. and 5' location for comparison for visual confirmation.

I did take measurements in 2nd row also, to confirm the multi-sub locations would not really mess that up...will post those also.
Edited by mtbdudex - 3/17/13 at 4:16am
post #47 of 89
Thread Starter 
Here are charts, same colors, for 1st row seats with sub in each of 4 positions

Corner
1st%2520row%2520seats%25201-4%2520ib%2520and%2520sub%2520in%2520lh%2520corner%2520overlay.jpg

3' from LH wall
1st%2520row%2520seats%25201-4%2520ib%2520and%2520sub%25203ft%2520from%2520lh%2520corner%2520overlay.jpg

4' from LH wall
1st%2520row%2520seats%25201-4%2520ib%2520and%2520sub%25204ft%2520from%2520lh%2520corner%2520overlay.jpg

5' from LH wall
(ignore the label, I forgot to change it ti 5 ft in there)
1st%2520row%2520seats%25201-4%2520ib%2520and%2520sub%25205ft%2520from%2520lh%2520corner%2520overlay.jpg

That LH corner location for 2nd sub is looking pretty decent considering the above charts.

For my triangle subwoofer concept, is there any basic design reason not to do it?
Like a critical minimum distance between the speaker cone and back wall for internal reflections?
Edited by mtbdudex - 3/17/13 at 4:17am
post #48 of 89
Thread Starter 
Here are more charts from the data I took 2 days ago, for 2nd row seats with 12" box sub in corner and 3 feet from LH wall position
(I did not take 2nd row measurements at the 4 and 5 foot position from LH wall like I did for 1st row)

Baseline 2nd row seats w/o 2nd sub, nearfield is top line
2nd%2520row%2520seat%25205-8%2520ib%2520only%2520%252B%2520nearfield%2520overlay.jpg

2nd row seats with 2nd sub in corner
2nd%2520row%2520seat%25205-8%2520ib%2520%252B%252012%2522%2520corner%2520sub%2520overlay.jpg

2nd row seats with 2nd sub 3feet from LH wall
2nd%2520row%2520seat%25205-8%2520ib%2520%252B%252012%2522%2520sub%25203ft%2520from%2520lh%2520corner%2520overlay.jpg

For subs we look at +- 5db as the target for flatness, not +-3db correct?

What I'm actually looking at doing is for sure building the new Dayton 12" driver sealed box, and using that as the 2nd sub for simulation purposes along front wall, and then using the "old" 12" box sub as 3rd filler sub during that simulation if desired.
>>I keep thinking that 16 year old 12" box sub is not giving me good response and changing these reading from where they could be, and possibly skewing my analysis....

I can then determine if building the triangle corner sub with 18" driver is acceptable, or should I build a cube box for the 18" Dayton driver.

Here is an image of the layout I'm working with, I also added to post#1:
HT%2520Plan%25202x4%2520Seats-Riser-acoustics%2520multi-sub.jpg

There are lots of articles out there on multiple subs, this one is helpful for me, http://www.hifizine.com/2012/06/bass-integration-guide-part-3/
Going thru the 150 pages here is daunting to say the least, I have skimmed it http://www.diyaudio.com/forums/subwoofers/134568-multiple-small-subs-geddes-approach.html
Edited by mtbdudex - 3/17/13 at 4:17am
post #49 of 89
Thread Starter 
Using WinISD here are overlays of my 4 x 15IB line array and the 2 boxes I will be building, forget the yellow plot as I will not be going with the Dayton 12" Ultimax.
12%2522%252018%2522%2520IB%25%2020sub%2520model.jpg

Am I mis-understanding the multi-sub approach of Geedes?
The 18" sub will be able to fill in and help low/nulls and achieve full reference 115db, but the 12" sub will not.......
I've got room for a 15" driver/sub - I'd go for the Titanic mkIII, since my 12" Titanic mkIIIdriver is on back-order I could swap for the 15" driver as 3rd sub.

Just saying out loud/typing, with the whole multi-sub approach based on the incremental adds of the next sub filling nulls/lows, so that's why each individual sub alone does not need full 115db ref capability, correct?

Just those occasions when I want to truly show off the IB sub and run it "hot", then the helper subs may run outta headroom, that's what I'm trying to avoid.
post #50 of 89
Quote:
Originally Posted by mtbdudex View Post

Am I mis-understanding the multi-sub approach of Geedes?
The 18" sub will be able to fill in and help low/nulls and achieve full reference 115db, but the 12" sub will not.......
I've got room for a 15" driver/sub - I'd go for the Titanic mkIII, since my 12" Titanic mkIIIdriver is on back-order I could swap for the 15" driver as 3rd sub.
Just saying out loud/typing, with the whole multi-sub approach based on the incremental adds of the next sub filling nulls/lows, so that's why each individual sub alone does not need full 115db ref capability, correct?
Just those occasions when I want to truly show off the IB sub and run it "hot", then the helper subs may run outta headroom, that's what I'm trying to avoid.

my understanding the helper subs do not need to be ref. capable. the nulls do not get "filled" but rather the modes that cause the nulls get broken by "selective cancellations"
post #51 of 89
Thread Starter 
I cancelled the 18" Dayton and ordered the new Stereo Integrity HT 18" Subwoofer instead
This driver just looks bad a$& eek.gif It will go in a 6^3 sealed triangle box, that modeled nicely in Win ISD.


The 12" Titanic mkIII driver is shipping today, I'm waiting on EricH flatpack going to baltic birch for the 12" cabinet.
post #52 of 89
Quote:
Originally Posted by mtbdudex View Post

It will go in a 6^3 sealed triangle box, that modeled nicely in Win ISD.

How much power are you going to be applying? At 6cf you'll want to keep a wary eye on excursion.
post #53 of 89
Thread Starter 
Quote:
Originally Posted by bass addict View Post

Quote:
Originally Posted by mtbdudex View Post

It will go in a 6^3 sealed triangle box, that modeled nicely in Win ISD.

How much power are you going to be applying? At 6cf you'll want to keep a wary eye on excursion.

I ordered the Dual 2 Ohm, figuring running 4 ohm load on my NU3000DSP , which says 2 x 880 Watts into 4 Ohms, but I won't have the gain dialed to full of course.
The SI website states their recommeded box sizexs, I plugged in 600 watts and the cone excursion appeared ok:
I'd like a slightly smaller box, but was trying to have the box q close to 0.7, I'm open to advice/suggestions
Quote:
600 watts RMS power handling
Recommended enclosures: 18"
Sealed = 6 ft^3
Ported = 7 ft^3 tuned to 20 Hz
23.5mm Xmax (one-way linear)
36.2mm Xmech (one-way)
post #54 of 89
FWIW, a Q lower than .7 isn't a bad thing.
post #55 of 89
Thread Starter 
Here's my WinISD models and charts, the cone excursion is a non issue for both boxes @ 600 RMS:

6 cu ft, power 600 watt, cone excursion and spl
stereo%2520integrity%252018%25206%2520cu%2520ft%2520cone%2520excursion.jpg . stereo%2520integrity%252018%25204.75%2520cu%2520ft%2520spl.jpg

Actually for my corner location 4.75 cu ft is physically better - so I show it here also, same power 600 watt, cone excursion and spl
stereo%2520integrity%252018%25204.75%2520cu%2520ft%2520cone%2520excursion.jpg . stereo%2520integrity%252018%25206%2520cu%2520ft%2520spl.jpg

So a Q of .718 for 6 cu ft versus .775 for 4.75 cu ft, what does that mean to me in real world performance difference?
post #56 of 89
Thread Starter 
Came home from work, look what I was greeted with smile.gif


Unboxing porn shot, the Kevlar 12" cone looks sweet, while the 18" is simply BIG.


Both were well packed/protected, however the SI 18" D2 had 2 small magnet voids/crack as shown.
I'll send Nick picture directly.

Edited by mtbdudex - 1/11/13 at 2:06am
post #57 of 89
I see you haven't updated this in a while, where are you on the project? I hadn't been around much either, but had a look in a few days ago to check in on the DIY sub arena and noted you had some stuff going.

I haven't read the thread but I don't think you've posted some pertinent info, so I'll throw it in. I believe the newest articles Earl has written about his multi-sub setup methods are actually on his site under http://www.geddes-audio.com/forum/forumdisplay.php?5-Subs ..there is also a video of a presentation he gave a local audio group (I was there), do a google search for "smwtms geddes".

About the arithmetic discussion, I think it's likely saying the same thing as the "adding point sources" point, but I suspect the responses won't add at frequencies where either one of them is not minimum phase. So if you still see differences, check the excess group delay for each of the measurements and see if it correlates to the areas that don't add properly. I'm interested to know this because I'm interested in understanding the limits of the software Earl uses to figure gains and filter sets for each sub.

I've been studying up on multisub because I'm planning to go multisub too. For my room I'm planning to build 4x 12" or 15" sealed subs, driven with 2x nu1000 and probably run through miniDSP. I'm plotting and planning whether I can do Geddes-style overlaps with the mains, because I think it makes sense to use the subs to adjust response up through the transition frequency if they're capable.

I haven't given up on horn subs, and I did study up on them enough to answer all the questions I had before--but I decided it wasn't right for my goals. I'm going to build a THTLP for my girlfriend's great room, maybe two if it needs moar though. No turning back on that one as I already have wood, driver, amplifier.
post #58 of 89
Thread Starter 
Hi John - are you going to the HEMI meet in April? I'll be there.
Actually today I'm building my sub box for my 15" SI D2 driver, flat pack from EricH.
Glued all the pieces but the front baffle early this morning, now I'm actually drilling the holes for the mtg and sinking the 1/4-20 thd inserts into the second baffle backside as soon as I hit "post" here.

So busy at work, 9 engineers staff report to me and 3 more hiring effective 4/1, plus I'm cub scout leader for my 9 year old, soccer coach for my 7 year old, and of course supporting my 11 year old daughter in her activities.
Plus, we have a 10th grade foreign exchange student Aug-2012 thru June 2013, so I'm a instant teen parent also!
(wow - have I learned lots, he's a good kid, with hormones kicking in full speed)

The "arithmetic discussion", REW A + B worked perfect once I did loopback as a timing reference.

Once I build this 15" box/sub, then I'll move the 24" iMac and REW gear, along with the 12" sub I used prior, and do more real world testing.
As that's my wife's basement computer I need to plan for 2-3 days of total REW/Multi-Sub geek-fest and put it back, or I'm in the doghouse.

Part of me desires to build the triangle sub with the 18" driver in the corner for a clean front sound stage look.
Yet part of me thinks future wise when I go AT screen and baffle wall with 3 speakers there, then the constraint of floor stander nearby will be gone so a 4 cu ft box for now may be best long term solution, I'll re-test that.....
Measurements will dictate what to do...

Love this hobby, sorta busy with "distractions" like all of us are.
post #59 of 89
Thread Starter 
3 cu ft flat pack build:
Dry fitting, Took me 10 minutes to un-pack and get my 3 cu ft sub box dry fit and tape together.:
photo%25201.jpg . photo%25202.jpg
Double top baffle plates removed to show nice fit inside
photo%25203.jpg . photo.jpg

I posted the below in EricH flat pack thread , tweaked the below since then:

Build viewpoint is described this way:
"front" = where driver is mounted, "back" is the opposite piece with the CNC cuts for the braces.
Outside front has the roundover and driver recess cut, inner front has no roundover
"top" and "bottom" piece are the long piece with step cut for more surface contact and less seems visible
"side" pieces are shorter and no step cut

Build is with the back piece laying on the work surface, CNC cuts facing upward.
I used Titebond II for glue, and clear caulk to seal all inside seams.

Text here, below this text next post I'll add pictures (when I remembered to take)
step 1: back, glue where braces go
step 2: install braces to back
step 3: glue on braces edge to top panel edge, glue on back to top edge, install
step 4a, 4b, 4c, repeat step 3 in this order side, then side, lastly bottom
Tape, them add clamp's, progressively slowly tighten and wipe glue with slightly damp rage off MDF board, increase clamp force till pretty snug
Let dry 24hrs.

Next day, seal all seams with caulk then:
step 5a: Dry fit front baffle inner and outside piece, put 15" driver in place, use 3 thin items to position driver centered
step 5b: Pre-drill front baffles for 1/4-20 screws and thd fasteners, using box as a holding jig and the 15" driver holes as drill template - be careful when doing this
step 5c: carefully remove 15" driver - do this by pulling up outside front piece with one hand and then remove driver
step 5d: remove both outside and inner front baffles,
step 5e: drill 3/8 hole just 1/2 deep in the "inner" side of the inner front baffle, this is where the 1/4-20 thd fastener will go next step
step 5f: install thd fasteners into inner front piece, use 1/4-20 screws with washer to pull the thd fastener into the inner baffle from outside baffle side, put small amount of glue into each hole before doing this
>>This will ensure the 1/4-20 thd fasteners are perpendicular to the front baffle assy and also exact alignment to driver
step 6: glue along inner front piece & top/bottom/both sides contact, install inner front piece (slight tap may be needed), clamp, wipe glue with slightly damp rage off MDF let dry 2 hrs
step 7: apply glue to Outside front, temporary lay onto inner front and push down for glue contact both pieces
step 8: Remove Outside front, see where glue not contact, add glue and spread
step 9: Final assy time! put Outside front back on, gently tighten the 8 1/4-20 screws as clamp for driver area, wipe glue with slightly damp rage off MDF the circle cutout
step 10: Apply 4 clamps to 4 corners approx 1" in both, wipe glue with slightly damp rage off MDF
step 11: add more clamps on outside and inside perimeter of front baffle, progressive tighten and wipe glue
(here is where I am right now, Sunday Mar-17-2013 1pm)
Next up:
step 12: let dry 24 hours
step 13: seal inside seams of inner front baffle with caulk to all pieces
step 14: Let Box fully dry 24 hrs,
prep for finish method (paint in my case) and binding posts install.
Can't forget for the bottom spikes vs feet discussion...and of course the ceremony of the driver wiring and install.....

[edit 3-23-2013] EricH posted these graphic assy instructions, I'll modify the above to match



Edited by mtbdudex - 11/25/13 at 2:35am
post #60 of 89
Thread Starter 
What's a build w/o some actual pictures?

Here is the 15" D2 Drwg:


step 5a: Dry fit front baffle inner and outside piece, put 15" driver in place, use 3 thin items to position driver centered


step 5b: Pre-drill front baffles for 1/4-20 screws and thd fasteners, using box as a holding jig and the 15" driver holes as drill template - be careful when doing this, here you see some sawdust from drilling

Outside front

and the inner front all done same time and line up nicely


step 5c: carefully remove 15" driver - do this by pulling up outside front piece with one hand and then remove driver
step 5d: remove both outside and inner front baffles,
Lesson learned: Watch where your braces are...I had my screw holes lined up at 12/3/6/9 o'clock....and they each went into the brace end just slightly.
I had to open them up just a tad.


step 5e: drill 3/8 hole just 1/2 deep in the "inner" side of the inner front baffle, this is where the 1/4-20 thd fastener will go next step
I used these from Lowes, The Hillman Group 4-Count 1/4"-20 Zinc Plated Standard (SAE) Wood Insert Lock Nuts http://www.lowes.com/pd_148251-37672-881674_0__?productId=3013891&Ntt=the+hillman+group+1%2F4%22-20+wood+insert&pl=1&currentURL=%3FNtt%3Dthe%2Bhillman%2Bgroup%2B1%252F4%2522-20%2Bwood%2Binsert&facetInfo=


step 5f: install thd fasteners into inner front piece, use 1/4-20 screws with washer to pull the thd fastener into the inner baffle from outside baffle side, put small amount of glue into each hole before doing this
>>This will ensure the 1/4-20 thd fasteners are perpendicular to the front baffle assy and also exact alignment to driver

All 8 done


Edited by mtbdudex - 3/17/13 at 6:24pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › Multi-sub: adding DIY-SG flatpack 18"+15" sub to 4x15" IB line array