I'm using pre-built car subs with good results. The enclosures are really flimsy though. I'm going to replace the 16mm particle board with something the same internal size but made out of 25mm MDF.
The plexiglass front on that JBL looks a bit suss.
col.
The plexiglass front on that JBL looks a bit suss.
col.
TRADERXFAN said:
I think those would really make a better ULF than broad band. I'm not 100% sure which GTO 12" driver is in there, but using JBL's published specs on either of the variations I could find, the rough bandpass I came up with from their own specs suggests a pretty low tuning, and even retuning it and adjusting dimensions, I could not come up with a design that would fit Dr. Geddes defenition of a broad band subwoofer.
JBL GTO Sub sheet
more than likely this is the driver, and the bandpass they spec is the one that you got. It fits, and you can clearly see that the bandwidth, if you use the -6db defenition, is not into the range of a true broadband. This is a side issue, but JBL's sensitivity figures in that spec sheet seem a bit optimistic. They are usually quite good with those sheets, but I don't see 95db's. I would estimate the top -6db point without a filter would be around 88db's and the low end -6db point would be in the low 40 hz range.
By comparison I think the Geddes subwoofer has a -6db point more around 120 to 130hz and a low end -6db at around 55-60hz or so. Also, the resonant frequency is closer to 90-95hz I believe. For the JBL it looks to be around 50hz.
It's my impression that a BB sub needs to not roll off until closer to 120-150hz to be best for this application. As I mentioned area, it's been my impression that the region in the bass range of greatest modal density is the area between say 60hz and 200hz (varying some with room, right?). It was also my impression that this is the range to target with the multiple subs, not the region from say 60hz on down.
Sorry for the less than ideal looking image. Snipe isn't the greatest, and that was the best I could do.
Ok the red curve is the best approximation I could come up with for the Geddes bandpass subwoofer based on his claimed sensitivity, bandpass, etc. The lower line is actually blue and is the JBL subwoofer in a bandpass based on the enclosure given in the spec sheet for the driver. The yellow line, which looks white, is the Abbey speakers response of the 12TBX100 in a sealed box with a box Q of around .5ish. The thick blue lines were my attempt at outlining the 0 and -6db points.
Look at the range between 60hz and 200hz, the area of greatest modal density, and the optimal tuning for the multiple subs seems, at least to me, more clear.
Well, you probably are right on with your projections... that would help explain why it was impossible to get it to "fill" in my room around 100hz! Which is what was causing me to continue experimenting with it. 
Wish I had your skills with modeling these things.
We should probably be discussing this subject over in the "multiple sub" thread though...
-Tony
Wish I had your skills with modeling these things.
We should probably be discussing this subject over in the "multiple sub" thread though...
-Tony
Matt
You are pretty close on your guesses. My subs do go up higher than what you show for the JBL and your guess is about right but a little high. I'm getting the -6 dB points to be at 43 and 100 Hz, with a peak at 60 Hz. And my measurements show the abbey to be -6 dB at about 80 Hz. The JBL is not a bad choice for one of the subs, or maybe two, but not all of them. It's certainly cheap enough!! Its hard to compete with that price point! I can't even get the driver for that price.
You are pretty close on your guesses. My subs do go up higher than what you show for the JBL and your guess is about right but a little high. I'm getting the -6 dB points to be at 43 and 100 Hz, with a peak at 60 Hz. And my measurements show the abbey to be -6 dB at about 80 Hz. The JBL is not a bad choice for one of the subs, or maybe two, but not all of them. It's certainly cheap enough!! Its hard to compete with that price point! I can't even get the driver for that price.
my own models never seem to perfectly match your or others claims, but I can just rack it up to different software and different driver parameters. My B&C 12TBX100 was based off one of the Abbey's drivers I measured with a woofertester. My numbers were a bit off from what the manufacturer had, nothing huge, probably within tolerance, but different. The biggest difference seemed to be an fs that was around 5-6hz higher than spec. That might account for my difference in fs from yours.
My guess on your sub, as I said, was based on what you had indicated in some talks, but also what was speced for the AI sub. I know you had said it was a bit different, but since saying that, I don't recall if we ever got an update on what the specs were exactly like. The last problem I have with accurately modeling your subs is that I'm guessing you are using your foam to dampen the ports. If that is the case, I suspect that slightly lowers the fs and tuning, and slightly lowers the tuning Q as well. None of the softwares I have accurately model the foam, best that I can tell. I took my ported sub box, measured it, compared it to the model, and found it close. Then I stuffed the port with the foam, and added that parameter to the boxes port, and they no longer matched. The measured impedance resonance lowered by about 1hz, but this was not reflected in the model. The output of the port was only reduced around 1.5db's, but the model lowered it by closer to 3db's, or if I got the other parts looking right, it didn't lower it at all.
If you aren't using the foam in the port, can I request it for mine? With my own sub, I was able to reduce distortion from the port tube by a significant margin. I also reduced the higher frequency noises by a significant margin. My assumption was that the foam had properties that could both slow and straighten the air. This property would help to reduce port turbulence, and thus reduce distortions, or extend the onset of audible turbulence distortions. I even looked into some of the literature on the properties of reticulated open cell foams on sound, air movement, water movement, you name it. I was trying to find some math I could use to model the full effect of it, but felt that the math was either beyond me, or I was unable to account for the same effects I could measure. for instance, based on the best model I ever came up with, port turbulence still should have been popping up at audible levels only a small bit above where it had been at, yet no matter what level I turned it up to, I did not get the onset of port turbulence. Even my measurements seemed to confirm that distortion levels remained fairly stable and low, most of the rising distortion did measure was coming from the driver and amp.
My guess on your sub, as I said, was based on what you had indicated in some talks, but also what was speced for the AI sub. I know you had said it was a bit different, but since saying that, I don't recall if we ever got an update on what the specs were exactly like. The last problem I have with accurately modeling your subs is that I'm guessing you are using your foam to dampen the ports. If that is the case, I suspect that slightly lowers the fs and tuning, and slightly lowers the tuning Q as well. None of the softwares I have accurately model the foam, best that I can tell. I took my ported sub box, measured it, compared it to the model, and found it close. Then I stuffed the port with the foam, and added that parameter to the boxes port, and they no longer matched. The measured impedance resonance lowered by about 1hz, but this was not reflected in the model. The output of the port was only reduced around 1.5db's, but the model lowered it by closer to 3db's, or if I got the other parts looking right, it didn't lower it at all.
If you aren't using the foam in the port, can I request it for mine? With my own sub, I was able to reduce distortion from the port tube by a significant margin. I also reduced the higher frequency noises by a significant margin. My assumption was that the foam had properties that could both slow and straighten the air. This property would help to reduce port turbulence, and thus reduce distortions, or extend the onset of audible turbulence distortions. I even looked into some of the literature on the properties of reticulated open cell foams on sound, air movement, water movement, you name it. I was trying to find some math I could use to model the full effect of it, but felt that the math was either beyond me, or I was unable to account for the same effects I could measure. for instance, based on the best model I ever came up with, port turbulence still should have been popping up at audible levels only a small bit above where it had been at, yet no matter what level I turned it up to, I did not get the onset of port turbulence. Even my measurements seemed to confirm that distortion levels remained fairly stable and low, most of the rising distortion did measure was coming from the driver and amp.
Matt
What I did that was new was to use four small ports instead of one big one. This dramatically lowered the noise from the ports, but also raised the flow losses. As we speak I am testing the use of foam in the ports since I knew all the benifits that you have already noted. I will use it if I can. But sometimes the foam is just too much loss and complete detuning occurs. I've seen this several times. If that happens then I won;t use it as the subs work well without it.
What I did that was new was to use four small ports instead of one big one. This dramatically lowered the noise from the ports, but also raised the flow losses. As we speak I am testing the use of foam in the ports since I knew all the benifits that you have already noted. I will use it if I can. But sometimes the foam is just too much loss and complete detuning occurs. I've seen this several times. If that happens then I won;t use it as the subs work well without it.
Matt,
To oversimplify somewhat, at the lowest freq's efficiency is determined by the box volume because almost all of the energy is being used to compress the air.
What we need is regenerative braking.
Add a VC that draws off the air spring energy, full wave rectify it, and use it to charge capacitors in the amp's PS.
A sensing/control circuit attached to this VC or perhaps a separate sensing coil would PWM the energy drawn off so as to keep the cone following the input signal.
BTW you can't increase mass while holding all else equal.
To oversimplify somewhat, at the lowest freq's efficiency is determined by the box volume because almost all of the energy is being used to compress the air.
What we need is regenerative braking.
Add a VC that draws off the air spring energy, full wave rectify it, and use it to charge capacitors in the amp's PS.
A sensing/control circuit attached to this VC or perhaps a separate sensing coil would PWM the energy drawn off so as to keep the cone following the input signal.
BTW you can't increase mass while holding all else equal.
I don't think so; servo control doesn't mean it recaptures the energy put into the air spring.
it's true.
My Prius does a better job than my speakers at this particular task. It just doesn't sound as good with a 1w input though.
Chris
My Prius does a better job than my speakers at this particular task. It just doesn't sound as good with a 1w input though.
Chris
I've been using some really cheap amplifiers with my Summas, literally under $200. Couldn't hear a difference between any of them, so I just stuck with the one that didn't have too many extra features, which was an old Kenwood from the 80s.
(I know, I know, this is an absurdly cheap amp...)
I was going to pick up an Adcom from a local pawn shop for $250, but decided to hoop up my headphone amp(!) to the Summas.
WOW
What a difference. The noise floor is blacker than night now, and the resolution of this amp is just jaw-dropping.
So...
Did you ever make a tube-friendly crossover for the Summa? I thought I read somewhere that you had. The output impedance on a tube amp is very high, so a crossover mod may be in order. I know that compression drivers require complex crossovers, due to their falling response and multiple impedance peaks.
Also, any solid state amps you might recommend? I know you're not a fan of tubes. I considered some of the ICEPower units from Pioneer, but I don't need seven channels of amplification. Which leaves "audiophile" solid state amps, which seem to be basic designs wrapped in an overpriced package. (IE, you're paying for a flashy case, not whats inside.)
Or I could just stick with Plan A, and buy that Adcom GFA 555 from the pawn shop. But it's way more power than I need, and this tube amp transforms the treble on the Summas.
Hi Patrick,
You might want to try one of the First Watt "F" amps. These are very high quality, low power. The diy F5 in my system is more that quiet, it is silent. A DIY version would not be too expensive.
Tom
You might want to try one of the First Watt "F" amps. These are very high quality, low power. The diy F5 in my system is more that quiet, it is silent. A DIY version would not be too expensive.
Tom
Question for Earl. What do you think of the idea of soffit mounting your kits like in a recording studio? Obviously some BSC effects will come into play although that can easily be compensated for with digital EQ that I use already. However, I'm interested in psychoacoustic effects, room integration, imaging and soundstage issues. I'd be curious to hear your thoughts on the idea.
Patrick,
I think this is an interesting observation you just made. When I was back at Earl's chateau in May 2008, auditioning the Summas, I too was very surprised by the amazing sonics that he obtained from a JVC cd player/HTPC as a source and Pioneer receiver. It was way ahead of everybody else.
What this really means to me is not that different amplifiers sound the same, they definitely sound different, and the Summas illustrate this more than any other speaker. What it really comes down to is value. What is it worth to you? Is a $1000 amplifier really five times better than a $200 amplifier? Maybe not to Earl, and I can appreciate his candor on the subject. But to many of us, his speaker expresses these differences even more than the other speaker competition. Its an interesting observation because his simple setup, in a well treated room, sounds like a $50K system at a show!
With the Summas, Abbey, Nathan or Harpers, really, the sky is the limit. Earl's waveguide technology just rocks. It's the most revolutionary concept I have ever heard.
Anand.
P.S. I know quite a few people who are using tube amps with the Abbeys without a problem. I don't know about the Summas though. You will never know until you try. Hook up a pair of well modified Dyna ST70's and see for yourself.
I really lost interest in reading Stereophile years ago because they perpetuated the myth that certain speakers could reveal differences in components upstream. And I just never heard it. So I lost interest in upgrading amplifiers, and focused my resources on the speakers. But I gotta admit, that tube amp sounded remarkable with the Summas. After a few days I put the tube amp back on the shelf, since it was ridiculously under-powered. I wound up putting the cheap Kenwood in the garage, and repaired a Legacy amp that broke down a few years ago.
Sure enough, the Legacy does indeed sound better than the Kenwood. (I know some readers are chuckling, since most would expect this result. Keep in mind, I've never noticed a difference between amps until I bought the Summas.)
Anyways, the Legacy sounds better than the Kenwood, but not as sweet as the Antique Sound Lab amp. My next project will be a Aleph Mini, based on the suggestions from a post a few days ago...
Long story short, if you've never noticed a difference between amps, buy a set of Summas, they're very revealing...
Could it also just be that the low-powered amp is clipping, giving you more distortion up high (especially if the Summa has some impedance minima up there) and thus a more traditional "hifi" sound?
I personally wouldn't recommend the 555, unless you can try it and take it back. Your Summas are more efficient than my Tannoys. While I've not used the 555 I do have a 535II and a 2535. I've found that both are a little bit too noisy for use with even ~96dB/w/m speakers. (The amps in my old and new Denon receivers - 4306, 4308ci, respectively are quieter.) Actually, the Denon 2809ci or 989 would be a good choice for you. Reasonable in cost, plenty of power, and Audyssey Dynamic EQ is the only "loudness" control I've yet heard that really, really works.
Adam is using one with his Nathan10's, I believe.
Another possible option if you want seaparates I've found quiet enough to use with the 12 DMT II's that's not too expensive considering are NHT's A1 monoblocs. Outlaw sells a rebadged version of that amp, I think.
- Status
- Not open for further replies.