AVS Forum banner

Placement Advice

5K views 156 replies 9 participants last post by  Alan P 
#1 ·
Hey all,

I've had my four PSA XS15s for quite awhile now and have tried many, many different placement options...but I'm at my wits end trying to make sense of these graphs. With all 4 subs up front (in 2 stacks of 2) I get great output, sounds great, but my seat-to-seat variance suffers a bit....and my graphs look horrible! I only have to bump the trim +2db in this configuration.

With 2 up front and 2 in back, I get great looking graphs but the bass is thin and weak. I have to bump up the trim at least 6db to get it even close to sounding good. You will notice that I don't get much response below 30hz in this config.

What on earth am I doing wrong? I don't understand why the better looking graphs don't actually sound better....?


All 4 subs up front, Aud off blue, Aud on green:



2 up front/2 in back, Aud off blue, Aud on green:

 
See less See more
2
#2 ·
Phase issues?, have you tried sub pink noise with rew rta to mess with the phase control?
 
#3 ·
I have NOT tried the RTA + phase, good idea! I typically just adjust phase for highest SPL at the MLP. With that method, the back subs are at 180...maybe that's not the correct way to go about it....?

What exactly would I look for in the RTA while adjusting phase?
 
#4 ·
My theory is you like a lot of bass and a flat response does not always mean much if you can not get the output you desire. Placing all 4 subs up front is creating 6db more headroom, but placing them around the room sacrifices output. Like I said back before you purchased the 2 extra XS15's you should of went with dual XS30's. Now for the good news you still can. :)

Simply stack the XS15's up front and purchase a pair of XS30's for the rear. Or save yourself some money and go diy and build some SI 18's.
 
#6 ·
SHUT UP BASSHEAD!! :D

I know, I know....shoulda never went sealed in the first place....shoulda not invested in 2 more sealed subs....but now I know, right??

I've actually been hanging around a lot in the DIY forum....gettin' some good ideas (like quad SI 18s). The only problem I have is I think it would be pretty difficult to sell one PSA sub in my area, let alone four. Maybe I'll just post 'em up on Craigslist and see what happens.

For now, I think I'm gonna just stack up the XS15s again and leave it alone. Then, try to decide - DIY or BUY?? ;)
 
#9 ·
~5000 cu. ft.

No pics, but:

 
#10 ·
Didnt know your room was that large
 
#11 ·
@ Alan P


I see your dilemma as for placement ! only thing I would suggest is pulling the K horns out and away from the corners into the room and placing the XS15se that are up front behind them ! this may boost a room mode and are help with the time domain, it may be the horns cancelling the bass your looking for! Just how far are they in the corners?
 
#16 ·
OK, been doing some pondering and browsing PSAs site....

What would you do if you were me?

1) Upgrade my 4 XS15s to SE + 2 more XS15SEs = $2424 (probably the worst option)
2) Keep the 4 XS15s as-is and add 2 XS30ses = $2349
3) Do I only need 2 more XS15ses to get the job done? = $1499 (I think I would always wonder about option 2 though)

Could take advantage of the Summer of Bass and if the 2 15s weren't enough, well.....XS30s it is! ;)
 
#17 ·
dual XS30's make the most sense if you are buying and not building. two of those would be close to 4 XS15's. Stack the XS15's up front and put the dual XS30 in the rear.


-Reason why-


- adding two more XS15 is only going to yield another 3db headroom...not enough gain imo.


-upgrading all 4 to the SE versions would likely only yield 3db gain but compression performance is greatly improved along with thd.


- adding 2 XS30's will likely add another 6db of headroom which I think is what you need to get satisfying bass in that large room.


that being said I think you could save your self a lot of coin by purchasing a couple SI 18's, a pair of flatpacks from diy sound group, and build two subs for the rear of the room. A Inuke DSP 3000 would power them well. You could complete this build easily for around a grand.
 
#19 ·
Thanks guys!



Gotta agree with bass on this one. You'd be much better served by adding the SI18's. Honestly, I'd add those, then I'd try and sell the XS15s and build more 18s.

My room is about 4300, and my single THT does ok, but I'll be building another in the future.

Good luck Alan!
So you both think that a couple DIY 18s would play well with the XS15s? That's great news!

Could you link me to the particular flat packs you would recommend? Are the SI 18" drivers still available? I can't seem to find 'em anywhere.
 
#18 ·
Gotta agree with bass on this one. You'd be much better served by adding the SI18's. Honestly, I'd add those, then I'd try and sell the XS15s and build more 18s.

My room is about 4300, and my single THT does ok, but I'll be building another in the future.

Good luck Alan!
 
#21 ·
Probably a really basic question that I could find the answer to pretty easily, but what are the advantages of getting the 4 ohm vs. the 2 ohm??
 
#23 ·
Its so you can match amps to subs, some amps are .5 ohm stable and some are only 8 ohm rated
 
#25 ·
2 ohms are less resistive so they are easier to push if your amp is stable at 2 ohms or w/e resistance your load is
 
#24 ·
Gotcha, thanks. I'm just gonna make a new thread in the DIY section before I order anything. :)
 
#27 ·
It all comes down to the amp you are using. Being the SI comes with dual coils you can wire them in different ways to achieve the impendence you are looking for. The D2 wired in parallel would equal a 1ohm load, the D2 wired in series would equal a 4ohm load. The D4 wired in parallel would equal a 2ohm load, the D4 wired in series would equal a 8ohm load.

You could get the SI18 D2's, wire them in 4ohm and purchase a Inuke 6000 to power them. Since you have a mini dsp, you might not even need the dsp series, so you could save some money there.

There also is the Dayton Ultimax 18" if the SI is not available.
 
#28 ·
Hi Alan,

I suspect the need to boost the level more with the front/rear locations comes from the dip in the middle of the front sub response causing Audyssey to set the level a little higher since that's in the middle of the range used to set the level. Bumping the level is just fine, so long as you're not out of headroom and hearing things overload or stop getting louder. The 4 front sub response does also have much more signal in the 20-35Hz range which is where much of the loud, deep sounding bass resides.

I would say you don't have (or haven't posted) enough info yet to know if you're getting the most from the setup. Are these graphs including the main speakers? Do you have all 4 subs at the same volume setting on their dials or set to match some level at the seats? Is this using a single sub out from Audyssey or 2 separate channels?

What we really need to see are measurements from each subwoofer location without EQ and all set to the same output level and main speakers turned off. This would be the same as sliding a subwoofer around the room taking measurements without touching the volume dial. Then you want to test combinations with all of the subs still at the same level to see how they combine. If they happen to be similar in level, the maximum gain for 2 subwoofers is +6dB over the response of one. All 4 can potentially add 12dB over the response of one. How much of this type of gain is what you want to see, and why you leave the level the same when turning on more units.

Take a look at some of the measurements I posted in the thread craig john linked. You can see the efficiency of the different ranges from different locations. There's a possibility that the left subwoofer locations vs the right could be causing some issues as well. Best to know what you have to work with individually as this can also help determine how to make adjustments with the MiniDSP.
 
#29 ·
Thanks for posting Mark!

All the above measurements are the LFE channel only (in REW). No MiniDSP in the chain at that time.

All subs are gain matched, not level matched.

Single sub out, split into both MiniDSP inputs. My AVR doesn't do dual subs.

What we really need to see are measurements from each subwoofer location without EQ and all set to the same output level and main speakers turned off.
You're saying I need to measure each sub individually? I can do that, but what would I be looking for? Wouldn't the response change anyway once I turn on one or more of the other subs??
 
#31 ·
You're saying I need to measure each sub individually? I can do that, but what would I be looking for? Wouldn't the response change anyway once I turn on one or more of the other subs??
Alan,

Take a look at the measurements I posted here: http://www.avsforum.com/forum/90-re...80129-accueq-vs-audyssey-14.html#post26481265 They show individual measurements of each sub and the combined measurements.

You are looking for how each individual sub contributes to the overall response, so you can individually optimize each sub's contribution.

Craig
 
#30 ·
Oh, and this...

Bumping the level is just fine, so long as you're not out of headroom and hearing things overload or stop getting louder.
...is exactly the problem I have with that configuration; running out of headroom when I get to around -10 to -5 MV.
 
#32 ·
Thanks Craig, I'll do some measurements of the individual locations and update here. I'll try some new locations as well.
 
#35 ·
Well, after a PM convo with Tom from PSA, we found (with a simple little test with a 9v battery) that my original 2 base plate version XS15s have voice coils that are wired backwards. :eek: Tom said that this was basically a "one in a billion" chance.

After I swapped the red and black leads to the drivers, everything sounds much better now!

Here's a comparison of what I had before (with 2 subs up front and 2 in back) GREEN TRACE;



And here's what I got now:



Much better extension, and everything just sounds better, tighter, more "cohesive". Kinda pissed I spent all those months struggling with what amounts to crippled subs. :mad:

Anyone think I need to worry about these drivers that were wired basically with reversed polarity for over a year? Should I demand replacements from Tom??
 
#36 ·
Nah I'm sure they are fine. glad you got it figured out, I thought it was phase related
and sure enough your response is there now.
Time to enjoy!
 
#37 ·
Do you have the output you wanted though?
 
#39 ·
Not sure yet....haven't had a chance to really crank the system...wife needs a girls-night-out! ;)

Nah I'm sure they are fine. glad you got it figured out, I thought it was phase related
and sure enough your response is there now.
Time to enjoy!
Thanks, that eases my mind a bit....still irks me about all the time I wasted. :(

Good to hear you sorted that oddity out. Now turn off Audyssey and take the measurements I suggested. Without keeping the drive level to each sub the same, you can't tell what locations are more efficient or less at different frequency ranges. If you have the volume levels matched, you can just mute the others in the MiniDSP and turn on 1 each, and then the various combinations.
Will do Mark!

What do you mean by "Without keeping the drive level to each sub the same, you can't tell what locations are more efficient or less at different frequency ranges."? All 4 are gain matched...or do you mean something else?
 
#43 ·
Mark, et.al.,

OK, I re-gain matched all subs tonight (2 were off by about 3db) and took a bunch of solo measurements. No MiniDSP, no Audyssey, LFE only in REW (ch4).

In my less than expert opinion, the 2 front placements kind of suck....it looks like I would be better off placing all 4 in back, maybe stacked in the rear corners.

Anyway, here is the graphic legend and the link to the MDAT file. Let me know what you think!

BTW, positions 1,2,3 & 4 are the current placements.

https://www.dropbox.com/s/x1id7pi0ca1xv2i/placement options2.mdat

 
#47 ·
After looking at these measurements for awhile this morning, it looks like 5 & 11 give the best response. Maybe I should try all 4 subs in that back right corner...?? Would definitely give the most output, and I'm pretty sure it would be quite a ride at the MLP. ;) The other seats, I'm not so sure.

Maybe 5+11 and 6+12?
I won't have time to look thorugh the MDAT files for a while. If you can post a screen capture or 2 I'd be happy to give some input. Based just on your comments I would suggest also testing the location near location "1" but on the opposite side of the KHorn, between the KHorn & the couch (front right side wall).

Remember you are looking for a complimentary set of measurements, not just the highest levels. In short, when you overlay the graphs by clicking "AllSPL" you should see that where one sub has a significant dip, another picks up the slack. Here is the graph of 2 subs at the front of thebland's room vs. 2 at the right rear corner when driven to the same exact level:


As you can see, one location is very weak below 15Hz, but very efficient in the 20-35Hz range.
 
#44 ·
After looking at these measurements for awhile this morning, it looks like 5 & 11 give the best response. Maybe I should try all 4 subs in that back right corner...?? Would definitely give the most output, and I'm pretty sure it would be quite a ride at the MLP. ;) The other seats, I'm not so sure.

Maybe 5+11 and 6+12?
 
#45 · (Edited)
Whats wrong with keeping the setup that got you the last graph after figuring out the phase issue? That graph looks pretty good to me...I would sit back and enjoy it for awhile. When you constantly make changes you do not give your ears and mind enough time to process if the change was good or bad. Trust me I have been there!! :)

What I would like to see is the same setup as the last graph but keep turning it up 5db and run sweeps until you start to see compression and post the graph with all measurements on one graph.
 
#48 ·
I did all this because Mark Seaton asked me to....when a Sub God like Mark asks to see some measurements, who am I to deny Him?? :)

The reasons I'm still looking at different placement options are:
1) I believe now that the subs are wired correctly, the in-room response has most likely changed
2) I've put the MiniDSP back in the chain, so I now can delay the rear subs (although, not enough it seems)
3) While watching a movie the other day (Mr. Nobody) I hit the limiters on a couple of explosions with the MV at -12. :(

With the subs in positions 1,2,3 & 4 I have to bump up the trim by 8db to get it to my preference. With the subs stacked at 1 & 2, I only have to bump the trim 2db. I'm hoping that I can find a combo with front & back placement (for smoothing) that will provide more output.

With the MiniDSP set to delay the rear subs by 7.5ms (its max setting) I'm getting approximately 6db more output than I would with the delay at 0. Do you think it would be a good idea (since I can't add any more delay) to adjust phase on the rear subs to see if I can get even more output?

I'll take your suggestion and do some compression graphs tonight to see where I stand at 1,2,3,4.
 
#46 ·
It would be nice if a sub was near the mains. 1,2,11, &12 seem like ideal spots. I have a large room like you and I spreaded the subs out as far as I could to the corners, and surprisingly I did not loose spl or tactile response. The corner gain more than made up for the longer distance. Good luck.
 
#49 ·
Yeah, 11 & 12 look better than 3 & 4 (FR), a definite possibility with a nice spread of subwoofage.
 
#52 ·
Off the cuff I would try 1,2 & 5,6 to get the best results with subs front & rear.

To get rid of the delay limitation and get more max output, I would try some combo of 5/6 , 3/4 and 11/12 as all 3 possible comb of these look useful.
 
#54 ·
Off the cuff I would try 1,2 & 5,6 to get the best results with subs front & rear.

To get rid of the delay limitation and get more max output, I would try some combo of 5/6 , 3/4 and 11/12 as all 3 possible comb of these look useful.
Thanks Mark!

Don't you think 11/12 looks slightly better than 5/6 (no giant null at 102hz [pos 5]) for the front/rear setup? Or, are you thinking 5/6 would provide more output and would be worth the slight FR compromise (which is my thinking as well)?
 
#55 · (Edited)
OK guys, great REW session last night!

First of all, the loopback/delay trick with the MiniDSP worked like a charm! Thanks for that Mark.

After getting the rear subs time-aligned, my response - and my distortion (holy cow!) - have improved greatly. But, the best is that it all sounds so much better. Bass sounds like it's coming from my speakers now (instead of from everywhere at once), it's tighter, more defined, (insert your own awesome bass descriptor here)....in short, color me a happy camper.

As for headroom, not sure where I stand on that because it was midnight last night before I could get a listen so I couldn't really crank it up. At low levels (-25 to -15) it sounds excellent. However, see below for my "max output sweeps". ;)

Now, for the measurements.

Just as a reminder, here's my room with possible sub placements. All the measurements below are using positions 1/2/5/6 with the crossover at 80hz, all speakers set to small, DynEQ off.



To start off, here's my FL/R, Audyssey off - pretty sad actually (considering they're HUGE speakers....my room really does suck for these speakers). Notice the null at 60hz.



Here are subs only, Audyssey off, positions 1/2. That null at 63hz is MASSIVE! Pretty much corresponds with the FL/R null - obviously it's my MLP.



Now, things get a little better - here are sub positions 5/6, Audyssey off.



Trying to time-align the rear subs, I measured from 7.5ms delay to 15ms delay. My rear subs are 7' and 8' away from the MLP respectively, but for the sake of time I just time-aligned them as one.



Here are the 2 best candidates (IMO). I ended up going with 12.5ms.



And then all 4 subs combined, LFE only, Audyssey ON. Killed the null completely with the time-aligned rear subs, yeah! :)



Now with the FL/R thrown into the mix. Here's before/after the distance tweak. Audyssey set the subs at 20'. I ended up at 17'.



Here's the final response. Don't know what's going on around 140hz but some bass traps might make quick work of that.



Waterfall looks great (compared to what I'm used to seeing in my room).



Spectrogram.



Distortion. Looking really good...at least I think it does.



And now, just for basshead, here's my max output sweeps. Looks like I might start compressing at -5MV....is that good??

 
  • Like
Reactions: 3ll3d00d
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top