AVS Forum banner

1 - 19 of 19 Posts

·
Registered
Joined
·
45 Posts
Discussion Starter #1
Wow, this is driving me nuts. Any thoughts on these multiple subs results. Getting tired of running all over and messing with phases. This is an odd pack of mixed subs I'm trying to get in sync.

3078230
 

·
Registered
Joined
·
1,527 Posts
Um, it would be helpful if you labeled what graph was what as a start...
 

·
Registered
Joined
·
45 Posts
Discussion Starter #3 (Edited)
Um, it would be helpful if you labeled what graph was what as a start...
I have them labeled, but I'm not sure what my phases and sub abreviations would help you with.

I'm trying to determine the most idealistic curve. I may go back and start with the SVS PB2000 and revolve around that (as it's my strongest ATM)
 

·
Registered
Joined
·
1,527 Posts
What I'm getting at is you have a graph with 3 lines. No explanation whatsoever as to what each one is... It's rather difficult to read minds over the internet...

Something helpful would be like the following as an example:
The teal line is sub xyz, green is sub abc, and purple is both.

That would at least give us something to start working with. As is, All I can say about your graph is there's 3 responses plotted. Without knowing what they represent, how can I possibly offer any advice or suggestions?

Also, please use the little camera icon that says "capture" below it to screenshot the graphs... It's built-in to REW and makes it easier for the rest of us...

See the below examples from my own REW use... They're labelled, which means one can draw conclusions based on just looking at the graphs. Contrast that with yours, 3 graphs, no info at all...

3078252
3078253
3078254
 

·
Registered
Joined
·
5,329 Posts
What are you using for eq? Are these all sealed, ported, or mixed?
 

·
Registered
Joined
·
45 Posts
Discussion Starter #6
What I'm getting at is you have a graph with 3 lines. No explanation whatsoever as to what each one is... It's rather difficult to read minds over the internet...

Something helpful would be like the following as an example:
The teal line is sub xyz, green is sub abc, and purple is both.

That would at least give us something to start working with. As is, All I can say about your graph is there's 3 responses plotted. Without knowing what they represent, how can I possibly offer any advice or suggestions?

Also, please use the little camera icon that says "capture" below it to screenshot the graphs... It's built-in to REW and makes it easier for the rest of us...

See the below examples from my own REW use... They're labelled, which means one can draw conclusions based on just looking at the graphs. Contrast that with yours, 3 graphs, no info at all...

View attachment 3078252 View attachment 3078253 View attachment 3078254
I will provide then. But all 3 lines are 4 subs at various phase settings. 1 SVS PB2000, 1 Klipsch Sub 12 and two 12 BIC V1220's. Each sub was calibrated to 80db prior to test. As they were tested individually first. After hitting 3 subs the line started getting dips. I was only testing from 20-120hz. I'm going to get at least one dip in the curve. It doesn't seem like any combo without testing numerous positions will alleviate that.
 

·
Registered
Joined
·
1,527 Posts
Try running plots with each sub by itself, no extra delay, then with all 4 subs at once. The 4 individual graphs should all be in the 80dB ballpark, the combined (if you have them time aligned properly) should be around 90dB.
Like this (I think I shot for 83dB when I level matched my subs), but the idea is similar...

3078265


Now, it obviously won't look like my response, because I only plotted the combined and time-aligned response aside from the 4 individuals.

It sounds like you're wanting to get them in phase (time aligned). The way I do that is to run a sweep on each sub with an acoustic timing reference, then use the impulse tab to see the start of the response. You can calculate the delay needed to get the subwoofers to positively sum together this way. You will also see if they're in or out of phase... If you're depending on the notoriously inaccurate phase dials on subs... good luck. I use a MiniDSP 2x4HD because I can input delay in 10ths of a millisecond. I randomly chose one of my 4 subs to be sub1 (by that I mean the one with the largest delay or farthest sub from the mic). I then ran a sweep on sub 2, checked impulse, calculated delay and added it to sub 2 and then played both at once. They summed as expected, nearly 6dB gain across the board. Then I played sub 3 with acoustic timing reference and calculated delay needed to match the sub1+sub2 value; after input, I got 8-9dB gain as expected over a single sub. I did the same thing for sub 4 relative to sub1+sub2+sub3. In my case, I had to invert phase on 2 of the subs, and only had to actually delay one of the subs to get positive summation everywhere. But, I was running two pairs of subs, and each pair was of same type (ported), and had port tunings within 2Hz of each other, so it was easy.

If you've got ported & sealed, you can run into issues with phase at near the port tune, where things will diverge. That's why most of us arm-chair "experts" say to avoid mixing ported and sealed. Similar problems can arise if you're trying to align ported subs with very different port tunes...

But, you've got to give us details if you want some help... There's no magic button unless you run MSO and use REW to feed MSO the input. That'll spit out delay and gain adjustments you input into a minidsp and give great results with only a bunch more learning curve of MSO.
 

·
Registered
Joined
·
45 Posts
Discussion Starter #8
Try running plots with each sub by itself, no extra delay, then with all 4 subs at once. The 4 individual graphs should all be in the 80dB ballpark, the combined (if you have them time aligned properly) should be around 90dB.
Like this (I think I shot for 83dB when I level matched my subs), but the idea is similar...

View attachment 3078265

Now, it obviously won't look like my response, because I only plotted the combined and time-aligned response aside from the 4 individuals.

It sounds like you're wanting to get them in phase (time aligned). The way I do that is to run a sweep on each sub with an acoustic timing reference, then use the impulse tab to see the start of the response. You can calculate the delay needed to get the subwoofers to positively sum together this way. You will also see if they're in or out of phase... If you're depending on the notoriously inaccurate phase dials on subs... good luck. I use a MiniDSP 2x4HD because I can input delay in 10ths of a millisecond. I randomly chose one of my 4 subs to be sub1 (by that I mean the one with the largest delay or farthest sub from the mic). I then ran a sweep on sub 2, checked impulse, calculated delay and added it to sub 2 and then played both at once. They summed as expected, nearly 6dB gain across the board. Then I played sub 3 with acoustic timing reference and calculated delay needed to match the sub1+sub2 value; after input, I got 8-9dB gain as expected over a single sub. I did the same thing for sub 4 relative to sub1+sub2+sub3. In my case, I had to invert phase on 2 of the subs, and only had to actually delay one of the subs to get positive summation everywhere. But, I was running two pairs of subs, and each pair was of same type (ported), and had port tunings within 2Hz of each other, so it was easy.

If you've got ported & sealed, you can run into issues with phase at near the port tune, where things will diverge. That's why most of us arm-chair "experts" say to avoid mixing ported and sealed. Similar problems can arise if you're trying to align ported subs with very different port tunes...

But, you've got to give us details if you want some help... There's no magic button unless you run MSO and use REW to feed MSO the input. That'll spit out delay and gain adjustments you input into a minidsp and give great results with only a bunch more learning curve of MSO.
truly excellence stuff.. I'll give it another go today. I think your directions help alot as I was randomly turning dials assuming there was some way to eliminate peaks and dips. Should I be running the sweeps from 0-120?
 

·
Registered
Joined
·
1,527 Posts
truly excellence stuff.. I'll give it another go today. I think your directions help alot as I was randomly turning dials assuming there was some way to eliminate peaks and dips. Should I be running the sweeps from 0-120?
If you're referring to frequency, I like 10-200Hz, others run 20-200Hz or 20-300Hz depending on what they're looking for. If your subs don't do much below 20Hz, then no need to stress them by running sweeps starting at 10Hz, IMHO.

You can minimize peaks and dips, it just takes methodical stepwise actions, and moving subs and measuring the change, changing delays on one sub at a time and measuring the change, and then finally some EQ after you've got them as good as you can get with placement and delays and verifying the change via measurement. Odds of eliminating peaks and dips via delay and placement alone are not good unless you've somehow got the magic room that cooperates; most of us get as good as we can with placement and then whack it with EQ to flatten it out.

Divide and conquer is the motto. Change one thing at a time and verify the change before trying something else.

So, it's possible to go from this:
3078553

To this:
3078554

But, you've got to make single changes, and verify changes after each step, and then throw a bunch of EQ at it.
 

·
Registered
Joined
·
1,527 Posts
3078614


Use this for graphs... ;)

I'd like to see a single graph with all 4 subs (no phase adjustments) individually, and playing at the same time on one graph if you'd be so kind. Mostly so I can see if the placement of the subs is complementary, or if they're interfering with each other...

So it would end up with 5 traces on the graph... sub1, sub2, sub3, sub4, and combined as an example. Do you also have a rough picture of the room layout (picture of room layout hand drawn on paper or ms-paint type stuff is fine). Just trying to get an idea of the room, where the subs are located, and whatnot. A picture is worth a thousand words...

What are you using to drive the subs? For instance, I've got the sub output from my receiver going to a MiniDSP 2x4HD, and then a single output from MiniDSP goes to each subwoofer (I'm using 1 input to minidsp and all 4 outputs). This way I have exact control of the delays to each sub, and to the "combined" subwoofer (and loads of EQ to tame the response). Trying to integrate 4 subs w/o something like a MiniDSP is akin to attempting to herd cats.
 

·
Registered
Joined
·
45 Posts
Discussion Starter #12 (Edited)
View attachment 3078614

Use this for graphs... ;)

I'd like to see a single graph with all 4 subs (no phase adjustments) individually, and playing at the same time on one graph if you'd be so kind. Mostly so I can see if the placement of the subs is complementary, or if they're interfering with each other...

So it would end up with 5 traces on the graph... sub1, sub2, sub3, sub4, and combined as an example. Do you also have a rough picture of the room layout (picture of room layout hand drawn on paper or ms-paint type stuff is fine). Just trying to get an idea of the room, where the subs are located, and whatnot. A picture is worth a thousand words...

What are you using to drive the subs? For instance, I've got the sub output from my receiver going to a MiniDSP 2x4HD, and then a single output from MiniDSP goes to each subwoofer (I'm using 1 input to minidsp and all 4 outputs). This way I have exact control of the delays to each sub, and to the "combined" subwoofer (and loads of EQ to tame the response). Trying to integrate 4 subs w/o something like a MiniDSP is akin to attempting to herd cats.
attached. I'm using ASIO and flipping the subs on and off. I don't have a rough picture because I can't seem to find a need simple application that allows me to do so. The room is large finished basement, drop ceiling 7'4" height. I have some specs written down.

SVS PB2000 - Front Left
Klipsch Sub12 w/ Bash 300s - Front Right
BIC v1220-1 - Mid Left
BIC v1220-2 - Cubbie behind Seats (close to mid Right)

The setup is the best I could achieve. So I'm using the REW software w/ Dayton Mic. I'm swapping between subs by flipping power and using ASIO4All.

3078620


The more I look at this graph, the more I realize I need 4 SVS's. Or maybe just 2 SVS Ultras or equivalent.
 

·
Registered
Joined
·
1,527 Posts
attached. I'm using ASIO and flipping the subs on and off. I don't have a rough picture because I can't seem to find a need simple application that allows me to do so. The room is large finished basement, drop ceiling 7'4" height. I have some specs written down.

SVS PB2000 - Front Left
Klipsch Sub12 w/ Bash 300s - Front Right
BIC v1220-1 - Mid Left
BIC v1220-2 - Cubbie behind Seats (close to mid Right)

The setup is the best I could achieve. So I'm using the REW software w/ Dayton Mic. I'm swapping between subs by flipping power and using ASIO4All.

View attachment 3078620
Ok. This I can work with... First off, that response looks darn good now. Considering you're breaking all the rules; by mixing lower tuned ported with high-tuned ported and sealed, you've got positive summation almost everywhere, and you're not running a MiniDSP to help... I'm impressed.

The BIC subs, as you can see, fall off a cliff below 30Hz, the Klipsh rolls off below 30 rather gently, and the SVS falls as expected below 20Hz.

Overall the combined response looks great considering all the rule breaking you've done. ;)

Constructive criticism inbound: With a MiniDSP, you could EQ your response to just about ruler-flat very easily.

The BIC's aren't really doing anything other than bloating up your >40Hz response; they're not even playing the same sport as your SVS, and hardly the same sport as the Klipsch... Honestly, I'd try just the SVS and Klipsch and see if optimizing just those 2 doesn't flatten the overall response. I'd imagine the current 4 sub response is actually pretty good sounding despite being quite mid-bass heavy, though.

Again, what you've achieved is incredible considering the hurdles.
 

·
Registered
Joined
·
9,766 Posts
Agree with the above. The bloat should be removed. Just use the SVS and Klipsch. Then see what you get.
 

·
Registered
Joined
·
45 Posts
Discussion Starter #15
Ok. This I can work with... First off, that response looks darn good now. Considering you're breaking all the rules; by mixing lower tuned ported with high-tuned ported and sealed, you've got positive summation almost everywhere, and you're not running a MiniDSP to help... I'm impressed.

The BIC subs, as you can see, fall off a cliff below 30Hz, the Klipsh rolls off below 30 rather gently, and the SVS falls as expected below 20Hz.

Overall the combined response looks great considering all the rule breaking you've done. ;)

Constructive criticism inbound: With a MiniDSP, you could EQ your response to just about ruler-flat very easily.

The BIC's aren't really doing anything other than bloating up your >40Hz response; they're not even playing the same sport as your SVS, and hardly the same sport as the Klipsch... Honestly, I'd try just the SVS and Klipsch and see if optimizing just those 2 doesn't flatten the overall response. I'd imagine the current 4 sub response is actually pretty good sounding despite being quite mid-bass heavy, though.

Again, what you've achieved is incredible considering the hurdles.
Much appreciated. I've messed with little bits but nothing significant. What you stated was my thinking too. I was going to try and get the Klipsch and the SVS to work together a little nicer and pick up some in the <35Hz range.

Truth being, I was in the process of selling the BIC's and thought.. eh.. what the heck.. 4 subs, why not give it a try.

Really I'm looking to add more SVS's, but I'm unsure if able to mix various models (say PB2000 + PB4000) or whether I should step up to something like an SVS 16 Ultra. Or to be honest.. I have the room.. I can go pretty big to get the sound I'm looking for.

For awareness, I'm using 2 Emotiva T1+'s, Emotiva C2+, 2 Polk 60 Monitor Series 2 for Surround, 2 older Bose for rear surround and 2 RSL c34e's for Ceiling.

The plan is to use Emotiva's for F/C/Surround, RSL c34e's for Height and SVS for Subs. 90% of use is Home Theatre
 

·
Registered
Joined
·
1,527 Posts
Much appreciated. I've messed with little bits but nothing significant. What you stated was my thinking too. I was going to try and get the Klipsch and the SVS to work together a little nicer and pick up some in the <35Hz range.

Truth being, I was in the process of selling the BIC's and thought.. eh.. what the heck.. 4 subs, why not give it a try.

Really I'm looking to add more SVS's, but I'm unsure if able to mix various models (say PB2000 + PB4000) or whether I should step up to something like an SVS 16 Ultra. Or to be honest.. I have the room.. I can go pretty big to get the sound I'm looking for.

For awareness, I'm using 2 Emotiva T1+'s, Emotiva C2+, 2 Polk 60 Monitor Series 2 for Surround, 2 older Bose for rear surround and 2 RSL c34e's for Ceiling.

The plan is to use Emotiva's for F/C/Surround, RSL c34e's for Height and SVS for Subs. 90% of use is Home Theatre
How large is the room in cubic feet (LxWxH), and what's your budget?

These are the best subs you can get for the money, IMHO...

At $600 (shipped): HSU VTF-2 mk5.
At $900 (shipped): HSU VTF-3 mk5.
At $1100 (shipped): HSU VTF-15H mk2.
At $1500 (shipped): Rythmik FV15HP.
At $1800 (shipped): Rythmik FV18.
 

·
Registered
Joined
·
45 Posts
Discussion Starter #17
How large is the room in cubic feet (LxWxH), and what's your budget?

These are the best subs you can get for the money, IMHO...

At $600 (shipped): HSU VTF-2 mk5.
At $900 (shipped): HSU VTF-3 mk5.
At $1100 (shipped): HSU VTF-15H mk2.
At $1500 (shipped): Rythmik FV15HP.
At $1800 (shipped): Rythmik FV18.
Are you aware of a free application that allows me to recreate my room as a blueprint w/ dimensions? I'd really like to have that to better explain the room.

I'm dealing w/ a finished basement, with drop ceiling. The drop ceiling is approx 7'4" in height. The width of the room is around 17ft, The length of the room is around 30ft+.

Except I'm only utilizing a certain portion of this room. The room is awkwardly broken apart by a stairway. Essentially 2/3 of the room is for Home Theatre, while the other 1/3 I will most likely use a Zone 2 w/ bar, some seating etc..

The area I'm focusing on is more like 17'x17'

I've heard alot of good things about the HSU's. Are they vastly superior to say SVS? Many appear to just love the SVS returns, customer services etc. The upgrade option is a really great way to keep people with their company.

As far as budget? hm.. lol I haven't really set one. I'd like to piece things together iteratively rather than in one big swoop.. I just dropped an $5k into the room over the past month. Maybe another $2k on subs? I do still have to buy another 2 RSL c34e's to complete my 4 height setup.
 

·
Registered
Joined
·
45 Posts
Discussion Starter #18
hm.. So in this graph.. I'm unsure if those 2 BIC's are just causing me issues..

After messing a little with timings. The overall graph of the SVS w/ the Klipsch (purple) doesn't have that 35hz jump. Setting the SVS to 180 degrees w/ all other at 0 made it a tiny bit smoother. I did a number of other tests that aren't shown here. These "appear to be the best"

Any opinion? All 4 or just the SVS and Klipsch. I might try moving them around a little and repositioning.

3079002
 

·
Registered
Joined
·
1,527 Posts
Are you aware of a free application that allows me to recreate my room as a blueprint w/ dimensions? I'd really like to have that to better explain the room.

I'm dealing w/ a finished basement, with drop ceiling. The drop ceiling is approx 7'4" in height. The width of the room is around 17ft, The length of the room is around 30ft+.

Except I'm only utilizing a certain portion of this room. The room is awkwardly broken apart by a stairway. Essentially 2/3 of the room is for Home Theatre, while the other 1/3 I will most likely use a Zone 2 w/ bar, some seating etc..

The area I'm focusing on is more like 17'x17'

I've heard alot of good things about the HSU's. Are they vastly superior to say SVS? Many appear to just love the SVS returns, customer services etc. The upgrade option is a really great way to keep people with their company.

As far as budget? hm.. lol I haven't really set one. I'd like to piece things together iteratively rather than in one big swoop.. I just dropped an $5k into the room over the past month. Maybe another $2k on subs? I do still have to buy another 2 RSL c34e's to complete my 4 height setup.
Just draw it freehand and take a pic... I'm sure there is software, I've just never felt the need to go that fancy.

The entire open area is of concern, because unlike regular speakers, the subs are going to react to the entire volume of space open to them. Bass is pretty much all about room, and how they excite room modes for the entire volume of space. So a ballpark figure for space is in the 3800-4000 cu ft ballpark. That means that the PB2000 is a bit undersized.

The "free" returns are built-in to the upfront price on SVS subs. They're competent performers, but lose bigtime on the value proposition. You get more output for the money with HSU or Rythmik at every price point except the anemic PB1000, and at that level, you should really just save another hundred bucks and get a VTF-2 mk5.

What is the end result you're shooting for?

As to the graphs, maybe mess with the range a bit to get to 5dB increments on the SPL axis; it's easier to compare that way. I'd go with your current SVS & Klipsch (purple) graph because it's not so bloated in the mid-bass. You could EQ your 4 sub response ruler flat with a minDSP... And if the output level is sufficient for you, costs a whopping $200... But being this is AVS, I'll also say that quad PB2000 would be worlds better, quad VTF-3 mk5 an enormous leap better yet, and quad FV18 would be truly end-game. ;)
 
1 - 19 of 19 Posts
Top