Announcement

Collapse

Midwest Audio Fest

It’s that time audio enthusiasts! Registration for the 2019 Speaker Design Competition is now open! Visit midwestaudiofest.com for details and to list your speaker project. We are excited to see all returning participants, and look forward to meeting some new designers this year, as well! Be sure your plans include a visit to the Parts Express Tent Sale for the lowest prices of the year, and the Audio Swap Meet where you can buy and trade with other audio fans. We hope to see you this summer! Vivian and Jill
See more
See less

High powered sealed sub + room eq.

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • camplo
    started a topic High powered sealed sub + room eq.

    High powered sealed sub + room eq.

    I come asking for some wisdom, in the area of sub-woofer implication. I want to upgrade my subwoofers a pair of ported 12" Kevlar woofers... The bass is respectable from current setup, but with best practice in mind, I would like to invest in improving performance in this area. The subs complete a system with 2-ways that take over after 200hz or so. Think three way with low xover point to the woofer and high xover point to tweeter.
    I've heard of people using overpowered sealed subs to create really accurate systems. Sealed advantages for subwoofer performance; better impulse response, less delay, etc, and then the short comings; less bass extension as opposed to ported systems, so with an over powered sealed sub, we use room correction eq to extend the range of the subwoofer, which is only possible with a luxury of amp headroom and sub xmax.....
    This is for studio monitoring, so application is of concern, just wondering is this a good idea for what I'm looking to do. I'm not sure how powerful of sub system I would want, 500rms8ohms per sub (2 subs) is a good starting point I think. I know of xmax calculators to help determine what type of driver specs to look for...Not sure which starting f3 is too high to try and extend with room eq...I'm hoping for some insight, correction if needed. I could always just build some nicer ported subs....I did like the novelty of the over powered room corrected seal sub....what do you guys think?
    I am also open to product suggestions if you know of one that would suite my need.
    Last edited by camplo; 02-06-2019, 01:13 AM.

  • JRT
    replied
    Originally posted by camplo View Post
    ... The average playback level in a club is 116db. Though I know and entertain the philosophy, there are other philosophies that suggest something like 86-87db to be the best mix volume. ...
    Note that dBFS, deci-Bell Full Scale, is commonly used in reference to signal level in the digital domain. Per SMPTE RP155 standard, 0_dBFS in the digital domain corresponds to +24_dBu signal level in the analog domain and +20_dB on a VU meter, and -20_dBFS is an alignment level in the digital domain which corresponds to +4_dBu analog signal level which corresponds to 0_dB on the VU meter. For easier reading in this post below, I take the liberty of converting everything to dBFS. (for Europeans reading this, the EBU R86 has analog signal levels 6_dB lower than SMPTE levels for same digital signal levels, uses -18_dBFS digital alignment level to correspond to 0_dBu analog signal level, and with 0_dBFS corresponding to +18_dBu).

    Monitor level used during mastering...

    In the early 1970s, optimizing for high quality relative to that era, using all analog signal chain and using multitrack magnetic tape as the recording medium and using vinyl LP record pressings as the primary retail distribution medium, the subject matter expert Ioan Allen of Dolby Labs recommended a mastering monitor level of 86_dB at -20_dBFS summed with uncorrelated phase, 10*log(x), at the listening position as propagated from a stereo pair of monitors, which is 83_dB at -20_dBFS at the listening position from each of the two monitors measured individually. That included consideration of s/n ratio of the recording and playback signal chain on high quality gear, the dynamic range and saturation characteristics of magnetic tape, and dynamic range limitations of vinyl LP playback, etc.

    Note that with spherical propagation from a point source there is -6_dB change in SPL with each doubling of distance from the point source, -20*log(x), so for example the SPL at 4 meters distance is -12_dB relative to a reference level at 1 meter distance.

    In the mid-2000s Dolby Labs issued a white paper on mastering music on 5.1 multichannel digital recordings. That whitepaper also included recommendations for conventional 2.0 channel recordings. In that, regardless the number of monitors, Dolby Labs recommended an uncorrelated sum of 85_dB at -20_dBFS at the listening position, which is 82_dB at -20_dBFS at the listening position from each of two monitors measured individually. They further recommended that the loudspeakers should be capable of providing linear behavior to at least +3_dB headroom above 0_dBFS, and the amplifiers and upstream analog signal chain including the analog portion of the DA converters should be capable of providing linear behavior to at least +6_dB headroom above 0_dBFS.

    Note that in theory a peak centered between 0_dBFS samples at a frequency that is onehalf of the sample rate would slightly exceed +4_dB above 0_dBFS, however for good quality results it is necessary to low pass filter the audio to place coherent audio well below the noisefloor at a frequency that is half of the sample rate, and practicable low pass filters have a corner further below that, so a maximum peak would not likely exceed +3_dB above 0_dBFS. Only the tweeter would experience that higher peak at the upper end of its passband on very brief crests. Any electronics upstream of the tweeter would need to swing that voltage unclipped for best quality.

    For mastering high quality music, Bob Katz echos Ioan Allen's 1970s recommendations mentioned above, and that is "K20" on Katz's scale, 86_dB at -20_dBFS at the listening position, summed uncorrelated. For pop-music and film soundtracks, Katz recommends "K14", 86_dB at -14_dBFS. K14 requires more dynamic range compression than K20, as maximum crests in K14 are 6_dB lower than the maximum crests in K20.
    Last edited by JRT; 03-03-2019, 09:21 AM.

    Leave a comment:


  • camplo
    replied
    Thanks for the information guys!

    Leave a comment:


  • billfitzmaurice
    replied
    Originally posted by camplo View Post
    There are more than a few speakers with very similar specs to the Ultimax. Some of the speakers, had the same...exact...diecast spider frame.
    That's because Dayton doesn't make drivers. For that matter very few companies do. Even those who assemble their own drivers do so using components that are mostly Asian sourced. Some companies that are large enough, like Eminence, have proprietary components produced exclusively for them, but even Eminence uses some generic components, the LAB series being one example.

    Leave a comment:


  • JRT
    replied
    Originally posted by camplo View Post
    ...the required box size for a reasonably flat response will be absolutely huge...
    Not specific to this woofer, but generalized consideration of fundamentals, consideration of Hoffman's iron law of loudspeaker enclosures... All else equal, a smaller enclosure requires more power to produce the same SPL over the same bandwidth. All else equal, more power into the box on crests in the program material leads to larger changes in temperature in the voicecoil which result in larger changes in series resistance within that voicecoil, increasing dynamic compression and the nonliear distortion associated with dynamic compression, and increasing the more audible linear distortion associated with varying Q-factor. Without getting into silliness of extremes, the power conversion efficiency of a moderately large box can be an element of improved performance, and going smaller can degrade performance.

    Leave a comment:


  • JRT
    replied
    Originally posted by JRT View Post
    Camplo... If you do not already have Unibox, then download it from Charlie Loab's website. Use that to model one Dayton RSS390HF-4 subwoofer driver and one RSS460-PR passive radiator with PR moving mass increased to a total of 1100 grams in 150 liters net enclosed compliant air volume, without leaks and without stuffing, and with 400 Watts applied.
    Here are the associated links.

    Unibox
    http://audio.claub.net/software/kougaard/ubmodel.html

    Dayton Audio RSS390HF-4
    http://www.daytonaudio.com/specs/spe...p?prod=295-468

    Dayton Audio RSS460-PR
    http://www.daytonaudio.com/specs/spe...p?prod=295-506



    Leave a comment:


  • camplo
    replied
    Some one made this for me =(
    Click image for larger version  Name:	 Views:	1 Size:	599.3 KB ID:	1405148Guess its not a huge downer, aside from the lower xmax, the performance went up the different specs.
    Last edited by camplo; 03-01-2019, 07:02 PM.

    Leave a comment:


  • camplo
    replied
    While searching through the loudspeaker database I noticed a trend. There are more than a few speakers with very similar specs to the Ultimax. Some of the speakers, had the same...exact...diecast spider frame. So I imagine that for those who know where to go, a particular frame and voice coil, and woofer diaphragm, is a best bang for buck, at a large volume purchase....so a buncha speaker manufactures jump on it. Just my guess.

    Leave a comment:


  • wogg
    replied
    Originally posted by camplo View Post

    I gotta be careful with winISd, the files from loud speaker database can be missing a spec or to, and it doesn't always have the omh spec in the right place, but I did plug those woofers into the winISD and the CW vega out performed the Dayton ultimax, but once again I may have a spec out of place

    "Based on the ginormous Vas and fairly high Qts of that 15" CV, the required box size for a reasonably flat response will be absolutely huge. It's designed for a car, where you can get away with a high Q sealed box and rely on cabin gain and EQ to flatten it out."

    Well sir, you know more a bout this topic that I do! I am listening, all ears, over here. The high damping factor is a good thing for a tight driver, is it not? I use room eq as a standard so unless the response characteristic isn''t smooth, is that not enough? In winISD the driver as barely an audiable bump towards roll off?
    Click image for larger version

Name:	
Views:	1
Size:	291.3 KB
ID:	1405121 This modeling is for a 3.5 sealed enclosure.
    You may be quite right, those curves look awful similar... if I get a minute I'll plug them in BassBox and see if it agrees. The Q of a sealed box is quite different than damping / tight driver. Usually Qtc > 0.7 creates a response peak above roll off that has to be dealt with and has lots of talk about "boominess" or other completely unverifiable stuff people say about their audio systems.

    My guestimate is based on lots of modeling, where I've observe that for a sealed box, a driver with a Qts of ~0.39 typically hits a sealed box volume about equal Vas for a Qtc of about 0.7. Honestly though, some of the other guys have more accurate in head modeling calculations. Chris Roemer comes to mind, that guy seems to be able to model enclosures and whole crossovers in his head.

    Leave a comment:


  • camplo
    replied
    Originally posted by wogg View Post

    Xmax and BL won't tell you how a woofer will behave in a box. Based on the ginormous Vas and fairly high Qts of that 15" CV, the required box size for a reasonably flat response will be absolutely huge. It's designed for a car, where you can get away with a high Q sealed box and rely on cabin gain and EQ to flatten it out. I haven't modeled it myself, but I suspect if you pop the parameters in WinISD you'll find the Dayton much more appealing when compared to the CV.
    I gotta be careful with winISd, the files from loud speaker database can be missing a spec or to, and it doesn't always have the omh spec in the right place, but I did plug those woofers into the winISD and the CW vega out performed the Dayton ultimax, but once again I may have a spec out of place

    "Based on the ginormous Vas and fairly high Qts of that 15" CV, the required box size for a reasonably flat response will be absolutely huge. It's designed for a car, where you can get away with a high Q sealed box and rely on cabin gain and EQ to flatten it out."

    Well sir, you know more a bout this topic that I do! I am listening, all ears, over here. The high damping factor is a good thing for a tight driver, is it not? I use room eq as a standard so unless the response characteristic isn''t smooth, is that not enough? In winISD the driver as barely an audiable bump towards roll off?
    Click image for larger version

Name:	
Views:	1
Size:	291.3 KB
ID:	1405121 This modeling is for a 3.5 sealed enclosure.

    Leave a comment:


  • djg
    replied
    Look at these.

    Theorized and tested: "Overdrive 10"... - Techtalk Speaker Building, Audio, Video Discussion Forum

    I theorize: "Vesuvius" (no, I will not be building this one) - Techtalk Speaker Building, Audio, Video Discussion Forum

    Leave a comment:


  • wogg
    replied
    Originally posted by camplo View Post
    A guy at Diyaudio was spouting about how good push pull config is, but he never elaborated. He did make it a point to use multiple "small" drive. Those Cerwin Vegas are cheaper and have better specs (higher xmax and BL and sens), you should take a look at the one I'm fancying "VPRO154D". Otherwise, I completely agree about the Daytons.

    Gebbes Multi sub approach aye? I just found an article, time to read. Thank you very much for the information. If You anything about this "push pull" theory I'm all ears too.

    edit: Oooooooh yeah I know about running multiple subwoofers, I just can't. Its too much gear, though its tempting. For some reason I was envisioning multiple subwoofers in a 3 way fashion. This has to do with placing multiple subwoofers within the room. Its best practice, but for me, this will be in the future.
    Xmax and BL won't tell you how a woofer will behave in a box. Based on the ginormous Vas and fairly high Qts of that 15" CV, the required box size for a reasonably flat response will be absolutely huge. It's designed for a car, where you can get away with a high Q sealed box and rely on cabin gain and EQ to flatten it out. I haven't modeled it myself, but I suspect if you pop the parameters in WinISD you'll find the Dayton much more appealing when compared to the CV.

    Leave a comment:


  • camplo
    replied
    I've come to understand that sealed is less efficient than ported. A person who's opinion is respected, has said, in this thread I believe, that he could not tell the difference in spectral decay of a sealed vs a ported sub, generally speaking, by ear. I'm ok with all that but show me a sub that is as good as the subs discussed, will play a f3 of 30hz or better, in a 3.5ft3 ported box. I want to try sealed anyway but I am not set in stone. The box size is though, as well as range expectations.

    "You need accurate playback at the SPL your customers would use" The average playback level in a club is 116db. Though I know and entertain the philosophy, there are other philosophies that suggest something like 86-87db to be the best mix volume.

    Normal studio equipment has limiters to prevent damage. In my case, the subs will be ridiculously over powered that I'll never actually turn it up, loud enough to damage anything. The amp has a clip indicator, maybe a limiter, and the rms of the speakers I've surveyed are more than double the power I'll need for playback.

    Also, my comment about frequency graphs needs clarification. On a proper spectrum graph, the octaves are spaced equally, so though there is some truth to the sensitivities of the ear, where the graph is wider, the sensitivity is to pitch discernment not exactly spl...still I'd think those areas are wise to avoid with an crossover, yet because crossover curves are derived with roll offs per octave they are not disturbing the balance......yet, the area between 1khz and 5khz is most sensitive to spl, and in a perfect world I'd think you'd avoid a crossover there. Just my theory. I think its easier to space the high pass and low pass on the mid, wide as possible, leaving the largest pass band possible to be afflicted by crossover.

    Leave a comment:


  • rpb
    replied
    I'm surprised that you might listen that loud. The tune that I tested was on my Steely Dan Two against nature disc. To me it was plenty loud at 80dB. Maybe the low battery voltage is affecting the scale on my spl meter. I used to listen at 90dB routinely. If the music you mix is going to be played loud, I see your point. You need accurate playback at the SPL your customers would use. It would also seem that bass is more prominent in what you mix, than it is in most rock that I listen to. What EDM type music is available on you-tube that I can try on my system? I want to hear something that will tax my setup.. (Maybe not as loud, but I do like to push the system on occasion.) Have you ever heard "Spies" on Telarc? Uncle Festive? Ed Mann "Get Up"? Yellow Jackets "Time Squared"? These are some of the discs that I play louder than usual. The Spies disc has a warning about keeping the volume reasonable until you know if the system can handle it without breaking something.

    The JeffB modeling software I use shows how much power the speaker actually uses at different frequencies. If you compare a sealed and ported speaker, you will see differences in the power consumed because of the impedance. You can compare sealed with boost, to ported without boost.

    The bass in my room is uneven, but deep. I sit against the wall. My sub has eq, and is very flat below 40hz in my small room. I mostly cut 45hz., and above 80hz.

    Leave a comment:


  • camplo
    replied
    I mix/master music, I am frequently staring at a frequency analyzer and while doing the work, adjusting those frequencies into "place". The work alone, makes you very familiar with sound/audio engineering. its like I have pieces of the puzzle from sound engineering and the rest of the mysteries are within loudspeaker theory and design plus room acoustics. Its funny because the majority of studio engineers I've come across, are clueless when it comes to speaker design, the ONE TOOL, they NEED to do the job *facepalm**facepalm**facepalm*
    I definitely know if and what music occupies what areas of the frequency range, and EDM and Rap have information in the 30's. That's why I aim for flat to 30. Flat to 20 would be preferable for any sound engineer, maybe I want to mix/master movie sountracks, like you said, HT....well someone has to tailor those sounds before it his the big screen, and if your system isn't flat in the area of critique, you are going to have issues of your results translating to other systems, though there are some other methods/theory, it all starts with having a super smooth frequency response.
    I'll often high pass a mix at 30-35hz... its kinda a golden rule. So music can be less demanding but from the engineering side, it is, or at least can be. When ever I set up room eq, I get a snap shot of the whole frequency curve, with my current 12"s my roll off is somewhere the 30's

    I tried your test at full volume and I hit 110 with or without the mains on, leaving the subs on. The problem is I'm also clipping the amp.....which was the first reason to upgrade my subs. The plate amp is the complimentary amp of the subs, if I turn the limiter off I'm probably just about to the point of clipping the xmax. The subs aren't powerful enough, long story short. I don't mix at these volumes, but when you add room correction into the equation, I think I am hitting the limiter at lesser volumes,but not so obvious. This system also doubles as a HT for movies and video games.
    (my house curve target was flat with a 2db bass tilt, I can't remember how close I actually got it with room eq, but it was close)

    So, I have something to share, I think I'll start a new thread to discuss it, but did you guys know that the way the lines are organized on a frequency chart is not without coincidence!? Where the vertical lines are spaced father away, that is the area where the ears are least sensitive.....or more most sensitive....I can't remember which....but with the correct information, it would lend tell to where to place a crossover....potentially

    Leave a comment:

Working...
X