Dismiss Notice
Attention AVForums app / Tapatalk users
Sadly GDPR means that, from 25th, we can no longer offer access to AVForums via the branded app or Tapatalk.
Click here for more information.

Please help me to calibrate my SVS PB12-Ultra

Discussion in 'Subwoofers' started by Gary_W, Aug 18, 2005.

  1. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    Hi

    My Ultra turned up yesterday. Generally very impressed, but I'm having quite a bit of difficulty getting it just right.

    A very rough calibration followed by the Incredibles was very nice. The rocket launch scene caused the glasses in the kitchen cupboards to rattle. They are on the other side of a wall that is 15 feet from the sub. Scary :eek:

    I digress. Room size is 22 feet by 13 feet by 7.5 feet high. Concrete floor with carpet.

    Main speakers are Acoustic Energy EVO 3's, an Evo Centre and a pair of Evo 1's on surround duties. Receiver is a Marantz SR 5200, which was about £300 or so 3 years back. Hardly high end, and I intend to update that next. I am only using the LFE cable at the moment and am trying to get things right for movies.

    TV and front speakers are all along one of the short walls. The main listening couch is 14 feet back from this wall, so we have about 8 feet behind the listening position.

    So far, I have tried the sub in both front corners and also more central to the front wall (actually behind the Evo speakers).

    Calibrated with Avia and an SPL meter, I've get it with a nice flat response everywhere and then am running the sub slightly hot (+4dB).

    I then made myself a CD of the demo tones pointed out by Nimby in the sticky at the top of the forum (thanks to Nimby and to Ian J for putting there permanently).

    I seem to have a very sharp spike at 50Hz, as you can see from the attached Excel file. The blue trace (1st set of readings) is with the PEQ on the Ultra doing its best to kill the spike. And that's as well as it can do it, about a 7dB cut. The manual seems to suggest that it can perform a 12dB cut, but 7 is all I get with the PEQ flat out.

    It's also interesting that I seem to have lots of room gain below 30Hz. The dip in response at 100Hz or so is explainable, as the receiver only does one crossover frequency, which is 100Hz.

    Not sure if I've got the wrong end of the stick here, but Avia's test tone to set up the sub is a noise going from 40 to 80Hz. Just where I've got the worst peak. If this is the case, setting the sub 4dB hot is still leaving it wanting in that trough at around 30-40Hz where lots of good stuff is to be had.

    Any advice on what to do here gratefully received. I can try the sub up the back of the room, but as the receiver insists on a crossover of 100Hz, I'm concerned that it may be possible to localise it. And there is the joy of running cables to the back wall :( So I'll give it a go, but kind of hope it doesn't improve things due to the practicalities...

    Thanks

    Gary
     

    Attached Files:

  2. BH_

    BH_
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Try putting the sub more away from the corners of the room (if you can).
    Have you tried playing with the phase?
     
  3. eviljohn2

    eviljohn2
    Well-known Member

    Joined:
    Jul 8, 2002
    Messages:
    7,545
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Near London.
    Ratings:
    +208
    50Hz noise may also be related to mains noise (although unlikely with a model of this type). Are you using any form of mains filter or even just a good quality mains block?

    Adjusting the phase as mentioned may help or make it worse - the rest of your plot isn't too bad.

    Alternatively you could electronically correct the peaks with additional electronics such as a BFD (dirt cheap at the moment), Onix RDES or the forthcoming Velodyne SMS1. :)
     
  4. Nimby

    Nimby
    Member

    Joined:
    Jan 27, 2004
    Messages:
    9,204
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    The Danish Bacon Factory
    Ratings:
    +670
    Did you place your SPL meter at your usual listening position at sitting head height for the tone tests?

    Is the trace subwoofer only or speakers running as well? (I can't read half of the info label in Open Office)

    Try playing just a 50Hz tone then altering sub phase and checking the result with your SPL meter at your listening position. It may have no effect at all but might be worth a try.

    I wouldn't put the sub at the back of the room because of all the phase effects with your speakers scattering troughs and spikes all over the place.

    Nimby
     
  5. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    Thanks for the replies so far, folks.

    BH_, I have tried repositioning as best I can. Unfortunately, no great change in the 50Hz peak.

    EvilJohn, I am running through a clean multiblock, so I don't think it's that. It's probably no coincidence that there is a peak at 50Hz and then another peak at 25Hz. Points more to room gain in my (very limited knowledge) opinion. A BFD would certainly cheer this up, and I know they're only £65 but not sure if I'd get away with that one at the moment :thumbsdow

    Nimby - Head height at listening position and it should be just the sub playing down at that level. The receiver crossover is set to 100Hz (it's the only choice) and the speakers are all set to 'small'. Unless I'm being very silly, this should mean that phase plays no part down at 50Hz as it's only the sub working.

    I can see how the phase could be responsible for the relatively low output around 80 - 90 Hz; I daresay the 100Hz 'cutoff' isn't exactly stop dead and phase probably isn't doing me any favours up there.

    I've just tried one of the back corners. It actually flattened out the trough between the 50 and 25Hz peaks and made the dynamic range of my curve rather a lot flatter. Downside is that there are some passages you could tell were behind you. I daresay this problem would lessen if I could set my receiver crossover down at, say, 60Hz. Oh, and I'm told that it 'looks even worse there than it does at the front' :eek:

    Too late to play any more tonight. I'll get there sooner or later.

    Having said that, I have to drive past a shop that sells the BFD first thing tomorrow. Tempting, but I'm not sure I want to admit defeat and throw more cash at it yet.

    Any more ideas greatly appreciated.

    Gary
     
  6. eviljohn2

    eviljohn2
    Well-known Member

    Joined:
    Jul 8, 2002
    Messages:
    7,545
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Near London.
    Ratings:
    +208
    Definitely try and get around the problem without external EQ if possible, I was just trying to outline the options rather than make you spend any more money. :laugh:

    Have you got your speakers set to large or small? You ought to try whichever setting you're not using at the moment as I would guess that the 50Hz peak is related to the reinforcement between the sub and your fairly large main speakers. The phase dial may help a lot here too. :)
     
  7. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    EvilJohn,

    They are set to small, crossover in the receiver is 100Hz.

    I can't see how it can be sound reinforcement for this reason.

    I will certainly try setting them to 'large' and see what happens. In fact, that has inspired an idea.

    My mains have a -3dB point of 40Hz. I'll try turning the sub off altogether, setting the mains to 'large' and see what the peak looks like at 50Hz. Forgetting about positioning of the sub in the corners reinforcing things, it should still give a large peak if my room rattles wildly at 50Hz as seems to be the case. It will be interesting to see if it does.

    Thanks

    Gary
     
  8. eviljohn2

    eviljohn2
    Well-known Member

    Joined:
    Jul 8, 2002
    Messages:
    7,545
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Near London.
    Ratings:
    +208
    That's a good idea Gary, you're very good at this. :smashin:

    Have you checked to see if there's anything in particular within your room which could be rattling around at that frequency? The glasses in one of the cabinets here tend to bounce all over the place at times which would obviously skew my results. :blush:
     
  9. Nimby

    Nimby
    Member

    Joined:
    Jan 27, 2004
    Messages:
    9,204
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    The Danish Bacon Factory
    Ratings:
    +670
    The real problem here is that a room of your dimensions shouldn't show a strong peak at 50Hz. :D

    Nimby
     
  10. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    EvilJohn- Thanks :blush:
    I was an engineer before moving into sales. I cannot help tweaking and experimenting. Even if I'm not sure what I'm trying to prove :D

    There is a big TV cabinet and the TV itself is a 32" CRT. The cabinet was a custom made job, wood veneer over very thick MDF. Not ruling it out, but not over likely to rattle IMO. However, I wouldn't mind betting that the TV has a sponsored rattle at certain frequencies, but I can do precious little about it even if it is the culprit. Other than that, there's a couple of leather couches, a CD cabinet (very solid) and a digital piano (very solid, no soundboard so I don't think it's that).

    Nimby - Can you please elaborate? What does theory say *should* be the peaks in my room? It's a long while since I studied physics and all I can remember is that one of the symbols involved was lambda (sic?). The term 'use it or lose it' is entirely true :oops:

    Thanks

    Gary
     
  11. eviljohn2

    eviljohn2
    Well-known Member

    Joined:
    Jul 8, 2002
    Messages:
    7,545
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Near London.
    Ratings:
    +208
    Gary, unless the maths is obvious to you then you can download the helpful program Modecalc:
    http://www.realtraps.com/pmodecalc.htm
    Which provides you with a list of the excitation frequencies of your room are once you've provided the dimensions. :)

    EDIT:
    All that the program does is calculate which waves fit into your room using

    f = c /(L/2)

    where f is the frequency in question, c is the speed of sound in air (~330m/s) and L is the dimension of your room. It's halved as we need to consider half wavelengths.

    Once you've worked out the first modal frequency in each dimension then the 2nd, 3rd and so on are just multiples of the first one. :lesson: :zonked:
     
  12. Nimby

    Nimby
    Member

    Joined:
    Jan 27, 2004
    Messages:
    9,204
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    The Danish Bacon Factory
    Ratings:
    +670
    Or there's:

    http://www.mcsquared.com/modecalc.htm

    This shows a 50Hz peak but only in the tangential modes.

    I wouldn't have thought this would produce such a massive peak compared with the usual parallel boundary effects.

    What about trying the sub on the side wall? Or just pop it in the post and I'll check if there's a 50Hz peak in my room. :D

    Nimby
     
  13. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    Thanks for the links, folks. All interesting stuff :)

    I turned off the sub and set the main speakers to 'large' on the receiver and hit it again with the 50Hz tone. Same huge response on the meter, telling me that this is nothing to do with sub position.

    I also emailed SVS for advice. They have been superb (as usual) and asked for a fax of my room layout.

    In response to this, Tom V recommended moving the main listening position slightly. I would not have believed that this would make such an immense difference in the response. It is startling. Going back 3 feet eliminated the 50Hz spike altogether :eek: Having said that, it causes the trough from 28 - 40Hz to become a mineshaft. In terms of 'thump me' kind of effects, it actually makes matters worse.

    What I can say for definate is that the whole room resonates at 50Hz; you can really hear it 'take off' at that frequency. I opened the door, and instantly lost one rattle. And 2dB off the signal meter :rotfl:
    Perhaps I need to re-nail the plasterboard on....

    Isn't this stuff fascinating?

    Anyway, I've put the seat everywhere possible within the very small limits I have (any further back and we won't be able to see the telly :) )

    So I've emailed Tom back with this info. I have a sneaking feeling that a BFD maybe called for here. It's just such a weird response and I'm not able to move things to ideal positions. Which don't seem to exist in this room :confused:

    Gary
     
  14. Nimby

    Nimby
    Member

    Joined:
    Jan 27, 2004
    Messages:
    9,204
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    The Danish Bacon Factory
    Ratings:
    +670
    Any chance you can listen across your room instead of along it?

    Nimby
     
  15. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    Hi Nimby,

    No chance at all, I'm afraid.

    Windows, patio doors and the door leading to the hallway mean that there is nowhere sensible to put the TV, let alone either of the couches, if we set the room up any other way.

    Gary
     
  16. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    I bought a BFD today.

    I only had 45 mins or so to play around with it tonight, but already I am far closer to a flat response than I would have thought possible. The 50Hz peak has all but gone, the far low end is tamed and generally things don't stray more than 4 or 5dB either side of the 80 that I am calibrating at :D

    Had to do a 21dB cut at 50Hz to get rid of it. Even now, it still gives about 84-85dB. But that's close enough for jazz. Avia's test tone for calibrating the sub includes this 50Hz peak in its wide range. May well explain why, even running 6dB hot according to Avia, I was still lacking thump in the 25-40Hz region that everyone tells me is where most of the movie fun is at. That trough is all but gone too :clap:

    I actually found the BFD very easy to program. Not sure quite what that says about me :eek: I'll post the new graph for those interested once I've played a little more in a day or two.

    Still work to do, but even at very low level listening tonight it was impressive. I will crank the Incredibles rocket launch scene tomorrow. That will tell me what I need to know.

    Gary
     
  17. stevefish69

    stevefish69
    Well-known Member

    Joined:
    Jun 25, 2004
    Messages:
    4,227
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    106
    Ratings:
    +439
    Gary,

    Did you manage to plot the new chart after buying the BFD. I'm interested to see how much it straightened your chart out.

    I've atached mine and am wondering whether it would make "That much" difference. I'm stuck with the position of the Sub where it is. Gotta say that it sounds great, but would it sound better by putting More in the signal path?

    The other thing i'm wondering is that i use a ISO-TECH SLM-1352A Digital sound level meter. Does anyone know if i still need to add compensation when plotting the charts or should i leave them as the raw data comes out.

    Steve
     

    Attached Files:

  18. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    Hi Steve,

    The pink plot is my 'before' and the blue is after. The 25-30Hz range needs work, but it's a night and day improvement.

    I got this on day 2 with the BFD and haven't touched anything since. Apart from the on switch ;) . I will tweak more, just haven't had the time. And it's at a stage where it sounds fab, so I'm in the 'if it works, don't mess about with it' frame of mind :D

    Regarding your chart, I'm not sure if the meter you have needs correction factors or not. What I will say is that they are huge in terms of you need to add lots at the low end (e.g. with the R.S. meter, if you measure 80dB at 20Hz, you add on 8dB so the actual SPL is 88dB). If your readings are 'as is', there is some pretty scary low end going on there above and beyond the volume it should be at (IMO, anyway). If your meter needs dB's added on to compensate (and this is likely) then they are terrifying :eek:

    How are you connected (LFE only or LFE + high level). What are your bass management settings etc.

    Gary
     

    Attached Files:

  19. stevefish69

    stevefish69
    Well-known Member

    Joined:
    Jun 25, 2004
    Messages:
    4,227
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    106
    Ratings:
    +439
    Gary,

    Thanks for the reply. I've got all speakers set to "Small" and the crossover on the amp set to "80Hz"

    I am only using the low level connection, have the SW level on the amp at -5db and the SVS gain at about 11 o'clock. From what i remember the readind were taken with the main volume at around -20db.

    Because the SVS is Not magnetically shielded it rules out 2 of my prefered positions up front, so it currently resides along the L/H wall. This is oposite to the listening position because our viewing is diagonal to the room.

    The missus has said she'll get me a BFD for Crimbo but was wondering whether it'd make a vast improvement. My main worry is the large peak at 45Hz followed by the Grand Canyon at 63Hz.

    Steve
     
  20. Gary_W

    Gary_W
    Active Member

    Joined:
    Jun 6, 2005
    Messages:
    450
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +21
    IMO, the BFD should help out with that a great deal. It has 12 different EQ bands, and they are assignable with very narrow Q factor, huge gains / cuts and with very fine control over the exact frequency.

    What this means is that you can have a very narrow band extreme cut right next to a gentle boost next to a strong boost etc. To get my curve as it now is took about 9 of the EQ bands to be used.

    My method was to take care of the worst peak first and retest. Then add a filter at the next problem frequency and retest. Ad nausium.

    Even using this (fairly unscientific) method, it didn't take long at all. Hour, maybe two.

    Roll on Christmas for you, then :)
     

Share This Page

Loading...
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice