Home Theater Forum and Systems banner

1 - 18 of 18 Posts

·
Registered
Joined
·
591 Posts
Discussion Starter #1 (Edited)
Hi guys, I'm at the stage with my build to start trying to locate and EQ my sub, and plan for broadband treatments. First up: sub location.

My room (see sig) is roughly 17' wide, 23' deep (21' to false wall), 9' high. I have floor to ceiling Roxul AFB superchunks in the front, about half high (will be full height more or less after final inspection - covers outlets) superchunks in the rear, and soffits all the way around are loosely filled with pink fluffy and topped off with FRK OC 703 (FRK out), fabric covered.

The sub is a single THT LP - below are a few locations I'm trying out, to see what the best starting point is to EQ from. All are behind the false wall.

In order:
left firing horizontal
vertical between L and C
vertical shifted by L
vertical far L
center firing horizontal


Thoughts? Seems like "vertical far left" or "center firing horizontal" might be my best bets.

Edit: or maybe left firing horizontal (the first) - more peaks, but the biggest isn't as big.

Edit: using RS digital meter (with cal) and Behringer UCA202 (with cal); sat the RS meter on top of the front middle (money) seat, facing forward.
 

Attachments

·
REW Author
Joined
·
6,711 Posts
Some pretty impressive resonances there, concrete walls?

"vertical far L" looks about the best starting point to me. You can show all the traces on one plot with the "All SPL" graph, when capturing the graph tick the box to include the legend to show the trace names.
 

·
Registered
Joined
·
591 Posts
Discussion Starter #3
Some pretty impressive resonances there, concrete walls?

"vertical far L" looks about the best starting point to me. You can show all the traces on one plot with the "All SPL" graph, when capturing the graph tick the box to include the legend to show the trace names.
No, no concrete - well, the front wall is framed in front of an exterior wall that is hollow block on the lower half, framed on top.

I'll try a few more locations this evening, see if there are any better options on the right side.
 

·
Registered
Joined
·
591 Posts
Discussion Starter #4
You can show all the traces on one plot with the "All SPL" graph, when capturing the graph tick the box to include the legend to show the trace names.
Here's the "top 3" together - dropped the two that have the big null around 65-70

Edit: forgot to include the legend
 

Attachments

·
Super Moderator
Joined
·
9,186 Posts

The sub is a single THT LP - below are a few locations I'm trying out, to see what the best starting point is to EQ from. All are behind the false wall.

In order:
left firing horizontal
vertical between L and C
vertical shifted by L
vertical far L
center firing horizontal
My first reaction was, “Those are pretty bizarre placement descriptions.” We’re used to seeing things like “front right corner,” “two feet to the right of the left speaker,” etc. But then I saw your AVS build thread, and it made sense. :D

You have a 30 dB spread between the worst peak and depression, which is virtually unequalizable, IMO. It looks like you have a capable sub and amp, but trying to EQ a spread that expansive the headroom penalty is going to be really severe. I guess it never hurts to try, but you should be aware of what you’re getting into going in.

Ideally the spread needs to be cut in half, or close to it. I think the problem with the vertical placement is that the driver (assuming it is located at one end of the cabinet) and as such the output ends up being dead-center between boundaries, which is often a bad location in a room with shoe-box dimensions. I’d try horizontal placement with the driver firing into a corner. Move the cabinet out from the corner in say, 1 ft. increments, taking measurements of each.

Regards,
Wayne
 

·
Registered
Joined
·
591 Posts
Discussion Starter #6
Thanks Wayne, I'll try more locations later today, and see if I can come with something a little easier to work with (fingers crossed).

Are there guidelines for how much boost is advisable on a given PEQ? I guess if some are due to nulls, a PEQ boost may not in practice be effective, but wondering how much I should be willing to try during trial and error time.
 

·
Registered
Joined
·
591 Posts
Discussion Starter #8
I’d try horizontal placement with the driver firing into a corner. Move the cabinet out from the corner in say, 1 ft. increments, taking measurements of each.
Made some more measurements - not very encouraging - all seem to have the same nasty spikes and dips:

:scratch:

I tried applying some EQ (those that REW suggested based on Target, plus a couple more for areas that it didn't seem to want to touch) - didn't see to help at all, maybe made things worse.

Any ideas where to go from here / what to try next?
 

Attachments

·
REW Author
Joined
·
6,711 Posts
If EQ had no effect it either wasn't really in the path (e.g. unintentionally bypassed) or the EQ filter frequencies were not what you thought (an easy one to get wrong on some EQ's is kHz instead of Hz). What EQ are you using and how is it connected?

The persistence of the 47Hz-ish resonance is odd, to be so consistent in all the measurements suggests it is either an inherent characteristic of the sub or something else in the room is being picked up - but at those levels it would be hard to miss if it were a background issue!
 

·
Registered
Joined
·
591 Posts
Discussion Starter #10
If EQ had no effect it either wasn't really in the path (e.g. unintentionally bypassed) or the EQ filter frequencies were not what you thought (an easy one to get wrong on some EQ's is kHz instead of Hz). What EQ are you using and how is it connected?

The persistence of the 47Hz-ish resonance is odd, to be so consistent in all the measurements suggests it is either an inherent characteristic of the sub or something else in the room is being picked up - but at those levels it would be hard to miss if it were a background issue!
EQ did have an effect - but not very effective - i.e., shuffled the deck a bit, but didn't look much like the predicted response. I guess a lot more trial and error is in order.



Going to run some manual sweeps, and see if I can localize/pinpoint any resonances at that frequency that may be at fault.

I have an FBQ2496 in the chain, between AV sub out and QSC RMX850 for amp. While running the tests, all the PEQs are flat.
 

·
Registered
Joined
·
591 Posts
Discussion Starter #11
I had a go at applying EQ manually, and thought I was doing pretty good:



Bolstered with confidence, I threw in Dark Knight for a test listen at -10 - when it got to "whatever doesn't kill me makes me stranger - *whoomp*" - the whoomp seemed to make some mechanical noise - not huge, but I turned it down a little, and tried again, same thing - so pulled back on some of the EQ, thinking I may have gone overboard on boost (looking now at the graph above, wouldn't be surprised that there was too much cumulative EQ), and broke something (exceeded xmax??) - then pushed bypass on the EQ, and remeasured:



What do you think? Maybe hurt the driver? Sprung a leak while moving it around trying locations? Maybe nothing, and I'm just getting paranoid?

Not sure what to think about bigger drop at 20, and then jump up below 20?? Maybe RS meter is just plain noise below 20?
 

·
Registered
Joined
·
591 Posts
Discussion Starter #12
Also noticed that it seems like every time I fire things up, even though I have all the same settings as far as I can tell, my gain seems to be different every time - sometimes too low, and I have to boost, sometimes too high. I don't get it.
 

·
Super Moderator
Joined
·
9,186 Posts

Frankly I can’t say I’ve ever seen mode peaks that sharp (narrow) and severe like you’re having at 50 Hz. It’s also curious that it never moves, diminishes or changes is “size” (i.e. width) no matter where you locate the sub. Same thing with the broad depression at 30 Hz. I have to wonder if there are internal resonances that are causing some or all of this. Seems to me something like that could be an issue in a really huge sub cabinet, like it becomes its own “acoustical space” with its own modal issues? Disclaimer: I’m no expert in sub-building theory, but an outdoor measurement of your sub and/or in-room measurements of a different sub might be tell us something. Also you might try running your room’s dimensions through a modal calculator and see how predicted modal issues correspond with what you’re seeing.

One more thing: Make sure the RMX850's high pass filters are turned off.

Regards,
Wayne
 

·
Super Moderator
Joined
·
9,186 Posts

Also noticed that it seems like every time I fire things up, even though I have all the same settings as far as I can tell, my gain seems to be different every time - sometimes too low, and I have to boost, sometimes too high. I don't get it.
Settings on what? What gain? You have to boost what? We have no idea what piece(s) of equipment in your signal chain that you're referrring to.

Regards,
Wayne
 

·
Registered
Joined
·
591 Posts
Discussion Starter #15
One more thing: Make sure the RMX850's high pass filters are turned off.
Thanks Wayne - I had them set off, but forgot to check when I got the amp back from being serviced - sure enough, the QSC guy reset them to their "defaults" I guess, which included turning the HPF back on. I reset the dip switches, and that may have helped the cause - can't hurt.

Settings on what? What gain? You have to boost what? We have no idea what piece(s) of equipment in your signal chain that you're referrring to.
Sorry, that was pretty cryptic - what I meant was that I didn't think any gain anywhere in the chain had changed (AVR volume, sub volume, laptop volume, REW input setting), but somehow when I remeasured, the level was much lower, so I had to bump it up to get a similar reading (the laptop was at max, so I adjusted the sub power amp up). Anyway, that seems to have gone away for now, so not too concerned.

I removed the chamber cover to see if there was anything obvious going on that could explain the mechanical sound I heard on Dark Knight (driver damage, leak, loose driver bolts), but couldn't see anything, so I buttoned it back up and tried again.

Here's what I came up with, before and after applying REW suggested EQ params (I think I must have inadvertently changed the REW measurement level, because it all shows at a lower level than the before):



After re-running MCACC, I then tried a full range measurement (using RS SPL meter as the mic, so not sure how far up in FR I can actually rely on) - MCACC always sets the crossover to 100, so I manually set it back to 80, but that leaves a hole in the FR - perhaps the TCA speakers just don't handle 80-100 very well, so I'll need to keep 100 Hz crossover; or manually adjust what MCACC came up with in that range to boost it.



waterfall:



Any thoughts / suggestions?
 

·
Registered
Joined
·
591 Posts
Discussion Starter #17
Disabled EQ for now, going back to trying to understand why I have such big issues to address, so hopefully I can come up with something that doesn't require so much EQ.

an outdoor measurement of your sub and/or in-room measurements of a different sub might be tell us something.
Would a close mic measurement help at all? I measured about 4" from the mouth, but still in room. Had to reduce the REW input level substantially of course - so not sure if I did this right, if there's any standard way to do close mic measurement.



If not, maybe I can drag the beast outside and measure it - I think I have enough speaker wire to run that far. I also have my old Velodyne sub, that I'll hook up and measure.

Also you might try running your room’s dimensions through a modal calculator and see how predicted modal issues correspond with what you’re seeing.
The room dimensions are roughly 23' deep, 17' wide, 9' high. I ran it through Bob Gold's calc, some of the "yellow" modes do correspond roughly to my problem areas. Why they seem to have such an extreme impact in my case, I'm struggling to understand.

I made a few more measurements from different locations (with the sub in the same location as before) in case they provide some insight. Does the fact that the 30 Hz trough is much shallower when mic'ed from the seat bottom (vs head height) maybe indicate that the back wall is the suspect (since its somewhat blocked when mic'ed from the seat bottom)?

 

·
Registered
Joined
·
591 Posts
Discussion Starter #18
Bolstered with confidence, I threw in Dark Knight for a test listen at -10 - when it got to "whatever doesn't kill me makes me stranger - *whoomp*" - the whoomp seemed to make some mechanical noise - not huge, but I turned it down a little, and tried again, same thing - so pulled back on some of the EQ, thinking I may have gone overboard on boost (looking now at the graph above, wouldn't be surprised that there was too much cumulative EQ), and broke something (exceeded xmax??)
I finally came up with a reproducible test using REW, and was able to locate the source of my problem: gain structure. It rang a bell after digging around and around for ideas, and running across Wayne's article on the topic. I had set my sub amp gain about halfway up; when I ran MCACC full auto, it set the sub trim to +3 dB, which on peaky HT material, resulted in the BFD input clipping. When EQ was disabled (bypass), hitting the clip light on the BFD didn't show up audibly; with EQ engaged (and boosting a bit around 28 Hz), it resulted in loud popping / white noise / crunch sounds.

My fix (so far) has been to raise the power amp gain to max, and drop the AVR sub trim to -5 dB (did it a little at a time, then validated by rerunning MCACC full auto, which left the sub trim about the same level). So far so good, I was at least able to get through the WOTW pod emergence scene at levels that were giving lots of problem before. Need to test out more material, and see how far I can push it (with voltmeter in tow) - crossing my fingers!

Just thought I'd capture this in case anyone else runs into something similar, since I banged my head on the wall off and on for weeks before I made it this far.

Here's the response close mic'ed (1' from mouth - black), at the LP without EQ (orange), and at the LP with REW suggested EQ (green). Certainly not perfect (boosted 28 Hz some, but don't want to get too carried away), but a lot better:

 
1 - 18 of 18 Posts
Top