Jump to content

(4) Sealed 21": Funk Audio UH-21v1


SME

Recommended Posts

  • Replies 167
  • Created
  • Last Reply

Argh!  Well, my plans to get these installed fell through, and my friend is not available tomorrow because of the holiday.  Actually it's just been one of those weeks, ya know.  Me and my friend both suffered catastrophic cell phone failures.  I'm still without my phone because I had to order the one I wanted online.

 

I'll try to get him here on a weekday evening, but worst case, it'll be next weekend.

 

Sounds amazing, looking forward to seeing pictures, measurements, and thoughts.  And can you throw in a picture of the trim ring as I'm interested in what you came up with there.

 

Tracking says my new phone will be here on Monday, so I'll try out the new camera on the trim rings.  I like them a lot, even though installing them is a bit awkward.  The driver has to be loaded vertically.  Then the interlocking trim pieces ring are laid onto the driver frame with washers placed between them and the frame under each mounting bolt hole.  The mounting bolts then thread all the way through the rings, the washers, and the frame, into the cabinet.  I did a dry run using a naked driver and some spare threaded inserts, and the end result looks surprisingly tight.

 

Do you want pictures of the pieces or just the finished product?

Link to comment
Share on other sites

I finally got the subs up tonight!  It took hours to prep and do the switch, but it's all done now.  (sigh)  Apart from some mysterious software issue in which the bypass mode for my custom DSP wasn't passing audio, the whole thing went smoothly.  The trim rings fit beautifully and came out almost dead on flush with the cabinet face.

 

I checked each driver individually for leaks using a 6 Hz briefly pushed high enough to heroic levels (rough guesstimate: 79V less the roll-off at 6 Hz for the SP2-12k).  With each driver, I heard just a little bit of noise that sounded like the surround going "squish", but no whistling or other offensive sounds.  The suspension noise was quiet and would almost certainly be masked with any content.  The only other thing I heard was ... my window glass creaking.  I guess I'll need all four woofers firing to "hear" 6 Hz, but I'm really impressed that I *didn't* hear 18 Hz 3rd harmonic or any other harmonic distortion.  Wow!  I'll have to do some more formal measurements later, but my first impression is *very* positive.

 

After getting them hooked up, I only had time to listen to a couple things before my wife went to bed.  My wife and friend were with me.  I did a very rudimentary calibration and attempted to match the response of my old subs, very roughly.  I played some 16 Hz organ music.  Then I popped in SW:TFA, and watched the intro, the first interrogation scene, and the scene with Rei in the forest.

 

My first impressions?  Very very clean, but it was a bit too low in level to get a good feel for everything.  The forest scene did have some nice ULF wobbling going on, and it seemed to be a lot more complex of a motion than it did before.  I'll do more tuning tomorrow.

 

And now for some proper data:

 

post-1549-0-71104600-1492755864_thumb.png

 

My impedance measurements indicate Fb @ 39.1 Hz and a Qtc of 0.33.  These numbers came in lower than simulations, especially the Q, which modeled to be closer to 0.5.  I ran the standard Fs/Qts test in Woofer tester and took the resulting Qts value as the Qtc.  I *think* this is valid, but I do wonder why the Q is so low.  My cabinet stuffing is pretty sparse too, so I doubt that had anything to do with it.  All said, I have *zero* complaints about these numbers.  The subs appear to be absurdly efficient.  The DCR is 4.75ish ohm, and I measured 8.5 ohm at 10 Hz and 14 ohm at 20 Hz.  The impedance is very high from 20-60 Hz where I get a lot of room gain.

 

Take a look at my in-room sensitivity, measured at the MLP and normalized for 2.83V 2V:

 

post-1549-0-14534600-1492759798_thumb.png
 
Note that my mic is only calibrated to 5 Hz, so what's going on below there is anyone's guess.  I may have more output than it looks.
 
If I'm reading this right, I'll be able to do up to 120-125 dB from 5-20 Hz and 135 dB+ from 20-70 Hz.  If I want to play something real loud, I can hit 145 dB a around 60 Hz, where I have a nasty room mode.  :o  And if that wasn't enough, I could throw a second SP2-12k on these babies for up to another 6 dB over most of the range.  My guess is that I won't need be needing that second amp.  Because the Q came in so low, I suspect 'll need to keep an eye on the excursion with the ULF and may not be able to realize  ULF numbers quite as high as indicated above.
 
I am fairly certain that I now have enough sub to destroy my house if I allowed it to happen.
 
Edit: Just as a reminder, each box is 800 mm x 700 mm x 600 mm (31.5" x 27.5" x 23.5") with about 240L (8.5 cuft) internal volume.  I'll get some pictures of these babies hopefully tomorrow, but needless to say, this bass system is remarkably compact.
 
Edit: Corrected post to reflect that the in-room sensitivity measurement was normalized to 2V and not 2.83V.
Link to comment
Share on other sites

I'm, way too busy today to do much listening today, but I've tried a few things.  I did some signal shaping last night that should put me flat to 5 Hz, but I haven't measured yet to verify.

 

The 16 Hz organ pedal seems to be reproduced with better temporal accuracy and is maybe a bit cleaner without port noise.  I listened to the canon blasts in a 2 channel down-mix of the "1812 Overture".  The single digit content is undoubtedly present, as evidenced by the motion of the woofers, but it's not loud enough to appreciably hear or feel.  Bummer.  I also tried the Space Shuttle recording, and at "-3 dB" from its natural level, I get a decent amount of feeling of lateral motion, but it's not a whole lot more than I noticed with my last subs, which went down to 15 Hz.  So far, I'm thinking that single digit bass may be a dud in here, despite my suspended floor.  It does rattle the glass in the living room window quite a bit though.

 

Of course, this is all preliminary.  I need to do a lot more tuning and optimization before I render any final judgments.

Link to comment
Share on other sites

I'm, way too busy today to do much listening today, but I've tried a few things.  I did some signal shaping last night that should put me flat to 5 Hz, but I haven't measured yet to verify.

 

The 16 Hz organ pedal seems to be reproduced with better temporal accuracy and is maybe a bit cleaner without port noise.  I listened to the canon blasts in a 2 channel down-mix of the "1812 Overture".  The single digit content is undoubtedly present, as evidenced by the motion of the woofers, but it's not loud enough to appreciably hear or feel.  Bummer.  I also tried the Space Shuttle recording, and at "-3 dB" from its natural level, I get a decent amount of feeling of lateral motion, but it's not a whole lot more than I noticed with my last subs, which went down to 15 Hz.  So far, I'm thinking that single digit bass may be a dud in here, despite my suspended floor.  It does rattle the glass in the living room window quite a bit though.

 

Of course, this is all preliminary.  I need to do a lot more tuning and optimization before I render any final judgments.

Turn it up ????
Link to comment
Share on other sites

Turn it up

 

I'm working on that.  :)  Today I reversed the polarity of the signal going to them and adjusted the delay to fix a big hole around 40-50 Hz where they are crossed to the MBMs.  The response looks good at the MLP now but still isn't integrating well with the MBMs at the left seats where I often sit.  I may have to re-do the whole EQ optimization to get the smooth, multi-listener in-room response I had before.

 

The Space Shuttle Launch is a very challenging audio recording.  I wasn't sure if I would be able to play it at full level.  Today I tried it, and those extra 3 dB along with the polarity fix helped a lot.  In terms of amp power, I believe I had another 3-6 dB to spare not counting what I'll gain by upgrading to 240V power, but the excursion was looking pretty wild at times.

 

The biggest problem I have right now is that my living room window is rattling a lot with all kinds of content.  This was a problem before, but it is much more severe now.  Pretty much anything 60 Hz and below seems to aggravate it.  Other than that, I feel like I have less tactile sensation with 20-40 Hz content.  Hopefully I can find a way to improve on this, but if some of that was from mechanical vibration transmission, it's possible that I gave it up by doing a D.O configuration.  We'll see.

 

Other than that, I tested some sine waves to see how the house reacts.  It seems to hold up quite well against single digits bass, except for the stupid window.  My floor structure does have a resonance at around 5 Hz, so that's a good thing.  OTOH, the resonance is quite high Q, and the rest of the single digits do pretty much nothing.  Above 10 Hz, I'm able to feel the floor shaking as well as hear and feel the sound more directly.  The stuff above 10 Hz also shakes the house a lot.

 

Tonight, I watched "Star Trek 2009".  I noticed a lot of ULF including a lot of detail that I never noticed before, but it was almost all subtle.  I also struggled to find a good "house curve" that provided enough mid bass while not muddying the dialog.  And without enough mid bass, the deep bass sounds dry and distant.   It may be a theatrical feature, which always seem to have weird issues with bass balance.  I ended up with the bottom end about a few dB hot, and wonder if it overwhelmed the mid bass too much.  Next time, I'll try a more recent home mix.  Headroom wise, "Star Trek" was easy peasy.  I guess none of the ULF is that strong.  Either way, the track barely moved the peak indicators, even during the warp effects.

 

About the sound.  They sound incredibly clean.  Maybe that's a lot of my problem.  It may take me a while to get accustomed to deep bass that's so clean.  Or I'll find I just need MOAR.  The challenge is to keep the sound reasonably balanced.  I have plenty of work to do including try to fix that window pane rattle.

Link to comment
Share on other sites

What range do you run the MBM's for, have you tried running just the UH's up to the mains?

 

Also I think something is off with your sims if you got a Q ~0.5, two drivers in 240L should be closer to 0.35, which is a lot closer to what you measured.

 

The MBMs run 50-110 Hz right now.  The old subs ran up to 50 Hz, but one of them was aggressively tweaked around the XO point to improve coverage in the high 50s where there is some weird modal behavior.

 

For the short term, I was hoping to be able to just drop in the EQ configs for the old subs and make coarser adjustments to account for the difference in native response shape.  It appears that their location in the room has shifted enough for that to not work as well as I hoped.  The old subs dropped off above about 45 Hz or so, probably due to side-wall reflections.  It looks like the new subs drop above 65 Hz or so instead, being that their centers are closer to the side-walls.  Note from my in-room sensitivity measurements that their response at the MLP is quite terrible around 70-100 Hz.  It's not fault of the subs.  Here is a 6" close mic measurement:

 

post-1549-0-18392900-1492982369_thumb.png

 

Ceiling reflections are likely also causing trouble here. If I were to try to boost the output of the subs to be flat at the MLP, it would likely sound quite terrible due to a huge build-up of energy in the room.  And my neighbors would hate me.  If I were to try to match them directly to the mains, I'd want to cross at 60 Hz, and would probably have a lot of trouble getting the blend right there too because of the crazy phase shift.  MBMs are essential for optimum sound in this room configuration.

 

For the long run, my plan was to run the subs up to where the mains start and possibly beyond and use the MBMs to fine tune response at the seats, but it looks like that won't work for the MLP.  Perhaps the subs will do better at the off-center seats, and the MBMs will be able to fill in at the MLP.  I'll need to go through the whole optimization process before I know what can be done.  I expect it will be very worthwhile.

 

As for my simulations, I unfortunately cannot review them because the computer they were on died, literally the morning after my subs got installed.  I haven't had time to troubleshoot the computer, being that I'm focused on the subs.  :)  IIRC, my original plan was for 280L boxes, which delivered closer to Q 033 using the specs I was originally quoted for the driver.  Then you sent me specs taken from the actual driver in which BL/Re and BL^2/Re came in a bit lower and Mms came in a bit higher, and I got a Q around 0.4 when I used these and adjusted the simulation for boxes closer to 250 L.  Then, when I designed the actual boxes, I believe I estimated that the volume would come in closer to 220 L, not 250 L like I said above.  Using this smaller box size and the T/S specs you pulled from one of the drivers, the sim came in with Q closer to 0.5.  Instead, it appears my Q came in closer to my original projection despite the smaller box.  As I said, I'm not complaining.  As it stands, I may have more amp power than I'll ever need and don't know if these drivers will ever see more than 1 kW or so each.

 

Anyway, I'll do some more listening and testing today.

Link to comment
Share on other sites

Just tried "Bass I Love You".  I had to boost the sub about 15-20 dB from usual to get more bass while maintaining a comfortable listening level for the highs.  If I push the level up to within 3 dB or so of the amp limits, I'm able to feel a very subtle pulse sensation on the lowest, single digit notes.  At that point, the excursion is pretty crazy.  I need to walk up to the sub to get a better look, but the wife in the kitchen complained about ear pressure.  :)  I doubt I can go much higher without encountering a lot of distortion.  OTOH, the 18 Hz notes are very clean and powerful.  I feel them in my body more than I feel it in the floor or couch.  They also need a fraction of the amp power to reproduce.

 

So aside from the floor structure resonance at 5 Hz, it looks like anything below 10 Hz or so just doesn't do much except suck power from the amp.  I guess that's not far out of line with what many others here report.  Maybe my opinion will change after I get more time to experiment with other content, but I'm thinking that single digits content may require Crowsons to do justice to.

Link to comment
Share on other sites

Maybe my opinion will change after I get more time to experiment with other content, but I'm thinking that single digits content may require Crowsons to do justice to.

Or a wood or suspended floor. I did some tests and found that even a pair of 5/8" OSB sheets over 1/4" rubber gives quite a bit of tactile feel, much closer to a true suspended floor than concrete.

Link to comment
Share on other sites

Or a wood or suspended floor. I did some tests and found that even a pair of 5/8" OSB sheets over 1/4" rubber gives quite a bit of tactile feel, much closer to a true suspended floor than concrete.

 

 I *am* on a suspended wood floor.  As I mentioned above, I have a floor resonance at around 5 Hz, but that's it for the single digits.  The resonance is quite narrow and takes a few seconds to reach full strength with a 5.5 Hz sine wave above 100 dB.  I find it unlikely that it will be appreciably excited by real film content.  I'm not convinced that the single digits won't contribute anything at all, but what it contributes will probably be slight.  I may end up rolling them off, and aim for a goal 10 Hz extension with minimal phase shift.

 

Tonight, I watched "Despicable Me", which isn't measured on here but definitely should be.  It's has a mix by "Randy Thom" with a nice full-bandwidth bottom end.  I did have to fight with the mix again, like I did with "Star Trek".  This time I was able to strike a better bass balance that was only slightly heavy in the deep bass and below.  Nice!  The bass was overwhelming and overwhelmingly tactile, even though it also shook the hell out of my house.  There is a rocket launch scene, wouldn't ya know?  I had forgotten until it started happening and it felt like my body was being pummeled by a fire hose of bass from all directions.  That scene was always powerful and impressive, but this was on a whole new level.  I did have to boost the bottom a bit more than my measurements suggested though.

 

In any case, the subs are doing a great job.   I'm quite convinced all my challenges are to do with getting them properly integrated with the room and the rest of the speakers.  I vaguely recall that the house shaking settled down a lot after I completed my original multi-seat sub EQ configuration, so revisiting this may help address two different problems at the same time.

 

Anyway, I often regret giving subjective judgments of sound, especially before I mangled the frequency response to my liking, but the bass from these subs is very special.  It is clean, pure, and physical.  It seems to hit from nowhere.  I'm looking forward to experiencing what these are capable of once they are properly playing the room instead of just playing in it.  I just hope I can reduce some of the shaking, because it really does distract from their great sound.

Link to comment
Share on other sites

Close-mic Response

 

This was measured approximately 6" from the cone, and a signal level normalized to 2V.  It should not be taken as quantitatively accurate but reveals the approximate native response shape:

 

attachicon.gifUH-21v1-close-mic-response.png

 

 

In-room Sensitivity

 

Here are in-room sensitivity measurements, normalized for 2V (more precisely than above) and taken at the main listening position:

 

attachicon.gifUH-21v1-sealed-in-room-sensitivity.png

 

 

 

 

Was your close mic performed close to walls or a corner? Next time try putting the mic on the same plane as the baffle or top of the surround.

 

 

Your raw response shape is a bit rough to be honest. You have a 10-20dB dip above 70Hz and a 15dB dip near 10Hz. Is there a low pass filter involved with these sensitivity measurements? I'm kind of surprised by the dip near 10Hz. Looks to be covering a fairly wide bandwidth and have two nulls. I've seen regions of diminished gain below 20Hz before but they are usually fairly narrow band and a bit higher in frequency. Have you noticed that your suspended floor resonates easily down in that range?

 

Listening position placement inside the space? Do you have more room and placement details? Perhaps a floor plan?

Link to comment
Share on other sites

This thread needs more pictures. 

 

Also Excursion Video

 

I'll try to get to these soon.  I'm going to probably have to hook up some external photo hosting because I'm running out of storage on the forum here.

 

Was your close mic performed close to walls or a corner? Next time try putting the mic on the same plane as the baffle or top of the surround.

 

As I've been saying for years, this is a weird room.  Perhaps that's why I'm a big fan of bringing heavy DSP horsepower to bear.  No low pass filter is involved in either of these measurements.  The extreme dip above 70 Hz is likely due to interference from the side-wall reflections (~8.5' path length difference), but also the room contribution ramps up substantially as you go from 40 Hz or so down toward 20 Hz or so.  It drops off abruptly below 20 Hz and again below 15 Hz.  Above 40 Hz, the rising native response of the subs takes over, and there is a strong room mode a little above 60 Hz.  It may be better to view my MLP measurement as having a big hump over the 20-70 Hz range rather than dips above 70 Hz and around 10 Hz.  Like I said in another thread, my room contribution behaves a bit like a horn. 

 

I don't have a floor plan drawn up.  I might need to draw one as it is difficult to describe the layout.  The living room, dining room, and kitchen comprise a space about 24' x 19' by 8'.  A pair of partial walls of about 9' length with a stairwell to the basement between them divides the living room and kitchen.  The living room itself is about 19' x 11' but is wide open to the dining room, which is wide open to the kitchen.  There is a smaller opening that leads from the kitchen back to the living room off of which is the hall to the bedrooms and bathroom.  Most of the seats are along the partial (stairwell) wall in the living room, and the MLP is centered along the 19' span of the room.  The two subs are placed on the wall with the rest of the front stage with centers a little more than 5' from each sidewall.

 

Does that help at all?  The deep dip around 9-12 Hz is a big bummer.  IIRC, the entire house length is about 48 feet, so I'd expect to see a length mode at around 12 Hz.  However paradoxically the "Right Sub", the one *farther* away from this end of the house actually has less more output over this range.  What's with that?  Yes, the floor moves a little bit at around 11-12 Hz (1st harmonic of 5.5 Hz mode?), but not very much.  OTOH, the dining room hanging lamp bounces like crazy at 11.5 Hz or so.  I may have to notch that frequency out.

 

Edit: I mis-typed that the sub farther from the end of the house has less output in the 10-12 Hz range.  I meant that it has more, not less output.  The seats on that side of the room see more output too.

Link to comment
Share on other sites

I posted new pictures to the start of the thread.  For now, at least, It looks like I can link directly to pictures stored on Google Photos, and I have unlimited storage because I bought a new Google Pixel phone.  The phone also has a much better camera than what I was using.  Woohoo!

 

I got a few more measurements in today and did some EQ tweaks to the system as a whole.  I think I got my deep bass under control finally, but not before I toned down some of the upper bass on my mains.  I later re-did the lower mids from 1k on down.  The speakers are set on top of the subs, and the change of acoustic environment necessitated the adjustments.  These issues probably explain why I was fighting so much with the EQ on the movies.  I also still had some lingering issues on my center channel, since re-EQing the treble.  I'll definitely have to re-visit "Star Trek" when I have things better optimized again.

 

The fact that I run a tightly hand-optimized system means that things almost always sound worse whenever I make any changes.  In this particular case, these vastly better subs sounded unimpressive when I first got them plugged in.  The funny thing is that I expected as much.  No matter how much I try to plan my equipment migrations, something weird goes wrong or I end up missing something seemingly simple but essential.  I remember when, a few days after I got my MBMs, my Mom was over and I gave her a demo.  She seemed impressed.  I didn't share her enthusiasm, and only after she left did I figure out they weren't turned on.

 

After my latest tweaks, the bass is sounding pretty good again with music.  We'll see where things go with movies.  The house noise is the biggest issue, one I know is hard to fix.

 

Now that it's sounding better, I can readily tell that the quality of the bass has definitely changed, and this is apparent at low as well as high playback levels.  Is it lower distortion?  More linear transient response (via well-controlled inductance)?  Maybe it's the lack of vibrations being transmitted directly to the floor due to the D.O. design.  The tactile sensations do seem a bit different with more feeling directly in my body and less in my feet, legs, and butt (the parts touching other vibrating things).  I think it's a bit more realistic to feel more from the air than from the furniture and floor.  Maybe it's all of the above.  I do notice that pure tones seem just a bit drier than they did before.  With transients, I get the feeling that they get out of the way of the MBMs and mains.

 

Tomorrow, I will work on prep for doing a big set of phase accurate measurements at a bunch of locations.  Then I can work on overhauling the whole room EQ including all the crossovers.  I want to send more mid bass to these things and maybe more to my mains, which seem to have the best handle on the 70-95 Hz region of any speaker right now.  They really do sound fantastic, and I need to hear more of them.  It was unfair of me to cut off the response graphs at 250 Hz, because these extend way higher in frequency... just in case that matters to anyone.  :)

Link to comment
Share on other sites

For some reason, I can't see the pics....

 

JSS

 

Same here. I can see some of the pics on the first page but there are a few missing.

 

Bummer.  I thought it might be too good to be true.  It looks like Google closed that loophole or something.  I'll probably have to set up a proper image hosting service account after all.

Link to comment
Share on other sites

  • 3 weeks later...

Sorry for the lack of update.  I'm knee deep in developing the next gen tools for speaker and room EQ.  These tools will allow me to take full advantage of my custom DSP solution including the ability to do sophisticated crossovers in which each mains channel gets rendered to each of the subs differently.  This is taking longer than I expected, for various reasons, but I'm getting close.  I'm real eager to get this done so I can hear these subs at closer to their full potential.

 

My preliminary work suggests I'll be able to get a very tight, near zero decay response across all 7 seats of my sectional from 20-70 Hz or so using the front subs alone.  I'm very happy about that.  I hope that by steering more sound into the middle of the room, I get a higher ratio of sound to room shake, at least for frequencies above 30 Hz or so.  The reduction of decay time may also help.

 

I am surprised by how uneven my response is in the octave between 9-18 Hz.  Seat-to-seat variation is quite high for some frequencies in that octave if I use identical EQ for the two subs down there.  Being that I don't have nearly as much headroom to work with down there as I do above, I have to make some compromises.

 

I also noticed a significant narrow dip at around 11 Hz with both subs in all locations measured.  I now suspect this dip is caused by my ceiling resonating and transmitting (leaking) sound through it.  The evidence?  A sine wave of around 105 dB at the seats at around that frequency makes my hanging dining room lamp hop on its chain and start swinging.  I'm a bit concerned about what more SPL will do there if sustained, so I think I will leave the notch there or even deepen it on purpose.  Indeed, one key to minimizing house shaking may be to find and notch out the worst offending resonances that often leak too much sound to be worth "fighting for" in the first place.

 

Above 70-80 Hz, the subs in-room get a lot messier.  I'm still working on the bits of tools I need to EQ higher than that, particularly being that the crossovers are important.  I think I'm going to still want the near-field MBMs.  The near-field locations seem best for these troublesome frequencies.

 

I need to post pictures of all this stuff.  I apologize for being so lame about pictures, but they will come once I get through this current work.  Then I need to seriously start watching some movies.  :)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...