I'm still not comfortable with the distortion values you state. I think your stated value for the vented box is too low, and the value you quote for my 12Pi is too high.
At the very least, I think you have "rounded-down" the 30Hz distortion value of your reflex box, and "rounded-up" the distortion value of my 12Pi hornsub. When I look at the 10-meter 800 watt chart of the 12Pi, I see the THD value is at a peak level, 85dB. The fundamental is falling fast, of course, since it is heading into cutoff. It's somewhere between 98dB-100dB, making it 13dB to 15dB above the harmonics. That puts total harmonic distortion in the 15% to 20% range, not 28%.
To expound a little further, one reason I think it is optimistic to state ~10% distortion at 30Hz from the 36Hz reflex alignment is it has exceeded xmax by that point. All of its operating parameters are on the edge - it is maxxed out on thermal capacity, excursion and magnetic flux modulation. All these things conspire to cause distortion to be rapidly climbing at 30Hz and below.
The attached chart (below) shows a LAB12 in a 3.5ft3 box tuned to 36Hz. At 400 watts, the 13mm excursion limit is exceeded at 30Hz, so I would expect distortion to be rising rapidly, in the double digits at least.
This alignment gives a 6dB bump at 36Hz, and reduces excursion in this region. So I think that's a pretty decent alignment for a light-duty prosound "club-sub" box, because it prevents 400 watt excursion from being exceeded in the passband. But it does run out of steam fast at 30Hz, and can't keep excursion (and distortion) from skyrocketing. That's OK though, I mean, everything has limits. Use it with high-pass at 30Hz, no problem.
But you can't really compare a direct radiating dual-LAB12 with the 12Pi hornsub. It just isn't a fair fight, not at 30Hz, certainly not anywhere higher, nowhere.
If we compare a direct radiating dual-LAB12 tuned this way with the 12Pi hornsub, you can do it where SPL matches or where power matches. If you compare the hornsub with the vented box where SPL matches, say at 120dB, then the 12Pi hornsub is barely breaking a sweat. It's there at 100 watts, just sitting there idling. No measurable distortion, it'd down below the noise floor, a fraction of a percent. The direct radiating dual-LAB12 is at full tilt, both drivers at full power. So it's in the mid to high single digits at its best points, and down low, it's getting into the double digits. At 30Hz, it's ramping way up there, as excursion is passing xmax.
If we compare at equal power levels, then what we see is a little different. At 800 watts, the dual LAB12 is still straining, not sounding too bad but definitely at the edge of xmax, so measuably distorting. It's doing about 120dB. The 12Pi is using that same 800 watts to generate SPL fully 10dB louder, around 130dB. Distortion is averaging 2%. Distortion under cutoff rises into the double digits, but just barely - it's about 15% at 30Hz.
So I'm just still scratching my head on the direct radiating comparisons. I mean, even if we're talking equal power levels, the distortion of the 12Pi is much lower than the direct radiator. No question. And if we're talking about equal SPL levels, it's just no contest. The 12Pi hornsub has an order of magnitude lower distortion.
Attachments
Last edited:
That is IMHO the only comparison that matters, is equal SPL levels in these days of very cheap volts and power.
I would assume that people without a distro or operating in venues without them ocassionally would be running a system that takes these things in to account. What that has to do with distortion measurements I don't know. If you want to compare different cabs performance relative to each other you would look at them at the same spl not the same voltage.
Wayne,
Looking closely, I see a >12 dB level difference at 30 Hz for the 12 Pi, a 12 dB difference is 25.12%, 11 dB difference is 28.18% distortion.
So lets say the 12Pi distortion at 30 Hz is 25 % at 30 Hz instead of 28%.
The vented alignment you show does hit Xmax at 30 Hz, but my dual ported dual Lab 12 does not exceed Xmax at 800 watts at 30 Hz, and does not have as much distortion as the 12Pi at 30 Hz. It happens to do well at low frequencies, where the design was optimized for.
I did not use programs to design it, years ago I empirically tested dozens of real cabinets with real speakers to arrive at the alignment. The Bass box Pro simulation does correlate well with my real world tests regarding excursion.
At 30 Hz, according to the simulation, my ported cabinet can do about 118 dB (one meter) with 49 V input, the 12Pi is almost identical there (98 +20 to reduce 10 meters to one =118 dB).
My tests don’t agree with the simulation exactly, I’d say the the Bass box Pro sim may be about 4 dB high at 30 Hz, so I won’t be betting on my ported cabinet to win, place or show in a one to one comparison to the 12Pi, other than having less distortion at 30 Hz.
Having designed, built and tested plenty of designs that exceed 100% distortion at full power below Fb or Fc, I know my dual ported Lab 12 happens to be an exception to what we normally hear or see.
I have agreed with you that the 12Pi has lower overall distortion than my ported cabinet in the range it is designed to reproduce, but you still insist on disregarding the fact that the 12Pi happens to have a greater distortion percentage at 30 Hz.
Disregarding facts never increases our knowledge .
Art Welter
Attachments
Yeah, I still don't buy it. This is where we have a fundamental disagreement.
Not only am I somewhat uncomfortable with your coursely-spaced measurements and possible calibration/accuracy issues, but like I said in my last post, you tend to inflate the distortion when you read the 12Pi charts but state looser figures for your own systems that are overly flattering.
For example, the 12Pi distortion peak occurs at 32Hz, where THD is approximately 85dB in the 800 watt chart. The fundamental is 100dB at that frequency, -15dB lower. That's 15%. Not 28%, not 25%.
Of course, amplitude is falling fast as it enters cutoff. It's falling at the rate of about 1dB/1Hz, so by 30Hz, it's lost 2-3dB. So you wiggle stuff around a little bit and whittle this -15dB difference down to a -11dB difference, but that's not the truth. It's spin, man. I'm back to calling BS on you, bud.
Let's use this kind of approach on your screen shots. I look at your RTA harmonics, for example, and I see the 30Hz fundamental at 108dB, 109dB maybe, The second harmonic looks like 89-90dB to me, about 18dB difference. I also see a third harmonic that you haven't included in the distortion figures you quote. Nor are any of the other harmonics, just the second. You've been running around all this time just using the second, and when called on it, you justify it saying "the difference is insignificant". But, man, how can you be so loosey goosey with your figures? Especially when you're pushing mine down 3-4dB. Round yours one way, round mine the other. Not cool.
Truthfully, the whole 30Hz discussion is minutia, an attempt to find one tiny place on the curve where you might say a 36Hz vented direct radiating dual-LAB12 can be said to outperform the 12Pi hornsub. It's way off base, in my opinion. Move the frequency up or down a tiny bit in either direction and the difference becomes obvious, but in this one place, you seem to think you can find a single metric that might be somewhere close to scale.
I guess I'm back to thinking we've come to a point where we're just not going to agree on the measurements. I think you want to feel that there is something that your 36Hz direct radiating subs do better. Perhaps it might comfort you to know I agree that your subs do have one trait that is much better - they're smaller.
Not only am I somewhat uncomfortable with your coursely-spaced measurements and possible calibration/accuracy issues, but like I said in my last post, you tend to inflate the distortion when you read the 12Pi charts but state looser figures for your own systems that are overly flattering.
For example, the 12Pi distortion peak occurs at 32Hz, where THD is approximately 85dB in the 800 watt chart. The fundamental is 100dB at that frequency, -15dB lower. That's 15%. Not 28%, not 25%.
Of course, amplitude is falling fast as it enters cutoff. It's falling at the rate of about 1dB/1Hz, so by 30Hz, it's lost 2-3dB. So you wiggle stuff around a little bit and whittle this -15dB difference down to a -11dB difference, but that's not the truth. It's spin, man. I'm back to calling BS on you, bud.
Let's use this kind of approach on your screen shots. I look at your RTA harmonics, for example, and I see the 30Hz fundamental at 108dB, 109dB maybe, The second harmonic looks like 89-90dB to me, about 18dB difference. I also see a third harmonic that you haven't included in the distortion figures you quote. Nor are any of the other harmonics, just the second. You've been running around all this time just using the second, and when called on it, you justify it saying "the difference is insignificant". But, man, how can you be so loosey goosey with your figures? Especially when you're pushing mine down 3-4dB. Round yours one way, round mine the other. Not cool.
Truthfully, the whole 30Hz discussion is minutia, an attempt to find one tiny place on the curve where you might say a 36Hz vented direct radiating dual-LAB12 can be said to outperform the 12Pi hornsub. It's way off base, in my opinion. Move the frequency up or down a tiny bit in either direction and the difference becomes obvious, but in this one place, you seem to think you can find a single metric that might be somewhere close to scale.
I guess I'm back to thinking we've come to a point where we're just not going to agree on the measurements. I think you want to feel that there is something that your 36Hz direct radiating subs do better. Perhaps it might comfort you to know I agree that your subs do have one trait that is much better - they're smaller.
Last edited:
Agreed. Reliable measurements are few and far between. A good comparison would involve measurements that are taken on the same equipment in a consistent manner at least. This example here with Wayne using his measurements and weltersys using a completely different setup and method is a clear example.
Anyway the basic point was simply that if you are going to compare the distortion of 2 different cabs it makes most sense to have a matched output level. It makes no sense to compare a cab that is putting out much less spl to begin with.
I agree as well. I think the approach of matching SPL for distortion tests makes the most sense. This is especially true when comparing speakers with significantly different sensitivity.
I also think that measurements, in general, are a whole science in themselves. It is important to use the right tool for the job. The signal type used makes a difference, as does the method and the environment. For example, a swept sine is perfectly fine for an anechoic (outdoor) measurement of amplitude response. But I'd prefer a different measurement when looking at phase, like when designing crossovers. Indoors, the only method I would use is pseudo-anechoic (unless I wanted to meaure the room, and not the speaker).
However, I think some comparisons can be made using data coming from different systems. As an example, amplitude response of a swept sine outdoors can be compared with pseudo-anechoic response measured indoors, provided the reflections are gated out. That usually means the charts can only be compared from ~300Hz up. I think it is important that the measurements be at similar resolutions, and that smoothing be the same too. Otherwise, obviously the low-resolution and/or smoothed chart will appear to be better.
I think also that some measurement gear is just too inaccurate to be useful. As an example, in the 1970s, all I had was an oscilloscope, a signal generator and a fairly good recording microphone. I had the manufacturer's response chart of the microphone, but no way to verify that or calibrate it. So the best I could do was assume the chart was right, and plot data with its conjugate to "normalize" it. I could use that to make some crude measurements, but the truth is most of the computer models made today would be way more accurate than the best measurements I could have made then. I think good measurements trump mathematical models any day - Measurements should be used to verify models and perhaps refine them. They're the final word. But that requires the measurement system be accurate. If you don't have good calibrated gear, there's a good chance the models you can make with software will be more accurate than your measurements. Probably best to keep that in mind.
I also think that measurements, in general, are a whole science in themselves. It is important to use the right tool for the job. The signal type used makes a difference, as does the method and the environment. For example, a swept sine is perfectly fine for an anechoic (outdoor) measurement of amplitude response. But I'd prefer a different measurement when looking at phase, like when designing crossovers. Indoors, the only method I would use is pseudo-anechoic (unless I wanted to meaure the room, and not the speaker).
However, I think some comparisons can be made using data coming from different systems. As an example, amplitude response of a swept sine outdoors can be compared with pseudo-anechoic response measured indoors, provided the reflections are gated out. That usually means the charts can only be compared from ~300Hz up. I think it is important that the measurements be at similar resolutions, and that smoothing be the same too. Otherwise, obviously the low-resolution and/or smoothed chart will appear to be better.
I think also that some measurement gear is just too inaccurate to be useful. As an example, in the 1970s, all I had was an oscilloscope, a signal generator and a fairly good recording microphone. I had the manufacturer's response chart of the microphone, but no way to verify that or calibrate it. So the best I could do was assume the chart was right, and plot data with its conjugate to "normalize" it. I could use that to make some crude measurements, but the truth is most of the computer models made today would be way more accurate than the best measurements I could have made then. I think good measurements trump mathematical models any day - Measurements should be used to verify models and perhaps refine them. They're the final word. But that requires the measurement system be accurate. If you don't have good calibrated gear, there's a good chance the models you can make with software will be more accurate than your measurements. Probably best to keep that in mind.
Indeed I would agree also IF same equipment and consistent measuring methods are used it would make more sense to look to the relation SPL and THD, especially for comparison of two different types of loading.
But the reality is I don’t see much accurate equipment in general. Most SPL meters used are <300$ which means +/- 1.5dB at 1 KHz for the meter. That means already an error rate of max 3dB at 1Khz between two different SPL meters. If you are going to use those meters below 100HZ you will end with an even higher error rate.
And then there are THD measurements. I see careful noise floor measurements but without the amplifier chain. I don’t want to accuse anyone but how accurate are these THD figures when their amps need to provide huge amounts of sine-wave power below 50Hz? The reality is that many amplifiers are not very happy with pure sine-wave signals below 50Hz.
To me that is piling up of errors. If you are planning to compare THD within resolutions of percents and SPL in resolutions of 1dB there will always be enough room for legitimate doubts from both sides.
But the reality is I don’t see much accurate equipment in general. Most SPL meters used are <300$ which means +/- 1.5dB at 1 KHz for the meter. That means already an error rate of max 3dB at 1Khz between two different SPL meters. If you are going to use those meters below 100HZ you will end with an even higher error rate.
And then there are THD measurements. I see careful noise floor measurements but without the amplifier chain. I don’t want to accuse anyone but how accurate are these THD figures when their amps need to provide huge amounts of sine-wave power below 50Hz? The reality is that many amplifiers are not very happy with pure sine-wave signals below 50Hz.
To me that is piling up of errors. If you are planning to compare THD within resolutions of percents and SPL in resolutions of 1dB there will always be enough room for legitimate doubts from both sides.
To be perfectly positive of results to three decimal points, one needs to check speakers under the exact same conditions with the exact same test instruments.
That said, useful data can be extracted from different test procedures and instruments.
When comparing distortion, there are several metrics that all "make sense".
Compare at a matched output level:
Good idea, but at what frequency ? Your Gjallerhorn will kill the 12Pi for output at low distortion below 35 Hz, but the 12Pi will kill the Gjallerhorn at 100 Hz.
Compare at a matched drive level:
This is useful to determine if the cabinet is capable of low distortion and "adequate" SPL at rated power. It will show what a given cabinet can do with a given amount of power, whether a cabinet can put out 3 dB more level using 6 dB more power is immaterial if that power level is out of reach due to lack of amplifier power, or circuit brown out.
My home power is (a lot) better than most, but none of the high powered amps I tested could put out their full rated power due to brown out (low voltage at high peak or average demand). Some amps do far better than others with less than ideal power.
Compare matched output level or drive level with cabinets of a similar size:
Obviously, a cabinet like the 12Pi or the Gjallerhorn is physically too large and has more output than needed (for most, LOL) for a small room, and a Carver Sunfire is not going to impress anyone at an outdoor dance party for 4000 people.
Art Welter
I don’t use just the second harmonic, as I wrote before (and you ignored again) I use whatever the highest harmonic is.
As I have also previously posted (and you ignored again), to figure THD, you sum the squares of the distortion, then take the square root of the sum. If you bother to do all the math, you find the addition of the lesser volume harmonics has little effect on the THD compared to the value of the highest single harmonic.
Just to please you, I went through the math (again, posted previously) on my 35 Hz distortion figures, in the case of the 35 Hz distortion, my “loosy goosy” reported distortion was 6% instead of 6.25%.
And in the spirit of full disclosure, here is the math for the 30 Hz distortion figure you seem so upset about:
30 Hz 109.6 dB
60 Hz 89.1 dB -20.5 dB 9.5% 9.5 x9.5 +90.25
90 Hz 77.6 dB -32 dB 3% 3x3=9
120 Hz 65.4 dB -44.2 .6% .6x.6=.36
150 Hz 60.1 dB -49.5 .4% .4x.4=.16
Sum of distortion squares to the fourth harmonic, 99.77, square root =9.99% distortion.
I have also previously posted that my distortion figures are accurate to within about 1%, the math says they are.
The THD 30 Hz amount is 9.99% instead of 9% as I reported, better call the distortion police...
By the way, a -15 dB difference between the fundamental and a harmonic is 17.78% distortion, not 15%, if you want to get further into math minutia.
Considering distortion less than 10 % up to 100 Hz or so is pretty hard to detect by itself, much less when HF content is present, if you want to think that a difference of 1% is significant, that is your prerogative, and we can agree to disagree on that point.
I did not “spin” my or your information, I simply compared the data points I have to yours, at the frequencies I tested. There was no reason for me to compare my 30 Hz distortion figure to your 32 Hz figure. Notice I also compared the 35 Hz figures, the 12Pi was half the distortion of my Lab 12s there, no spin, no BS, just fact.
If you look at the comparisons, they don’t “flatter” my cabinets, they give an objective look at what they can do under a given, rather extreme, test condition.
So give me a break, look at the WS “C horn” distortion specs (tested the same way as all the other cabinets I tested for distortion) below compared to the 12Pi, I did not put any lipstick on my pigs
.Art Welter
Attachments
For onlookers, first, please forgive the minutia. It can get old sometimes when you're listening in on two people in a debate, once they get into the hair-splitting phase.
Now on to the hair-splitting.
Art, you made my point, one of them anyway, when you admitted that even using your data, you knew the distortion figure to be 9.99% yet you claimed it to be 9%. You rounded down. When you examine my charts, you quote a distortion value that is much higher (28%) than what the data shows (15%). You rounded up.
I also question your microphone, calibration, etc. There is ambiguity in your data that isn't there in mine. I'm using a calibrated microphone on a calibrated system that is known to provide reliable consistent results in this kind of measurement.
The distortion of my 12Pi is maxxed out as it enters cutoff, at 800 watts it's 15% at 32Hz. Below that, the slope of the rolloff of the fundamental and the slope of the rolloff of the harmonics is the same. They both fall at the same rate. So full tilt distortion at cutoff hits 15% and stays there all the way down into the stop band, where there is very little sound. It's pretty darn quiet below cutoff, no distortion, no sound at all. In the passband - where it counts - distortion is only 2-3%, even lower towards the upper end of the passband.
What other kind of high-power system you know of does that? What else out there can do 135dB/M at 2% THD and only 15% distortion even below cutoff? Sure, the passband in the 2-3% range is what's most important, but the fact that distortion is only 15% at cutoff and below really tells you something. It shows how effective push-pull drive is. Horn loading reduces distortion, and the push-pull drive takes what's left and cuts it in half.
That's my point, and it is why I split hairs with you.
Someone asked about the 12Pi hornsub back in April, and you launched into promoting your products. You unfairly and inaccurately characterized your speaker as performing better. But when I discovered this, I called you to task. That's what this is about.
I do want to say this though too, because I think it's important. I like your design. I don't think it compares with the 12Pi hornsub, but it's a different class of speaker. I would recommend your cabinet to anyone that needed a medium duty sub, something I would call a "club sub". It provides adequate output, good performance, and is more than enough for small to medium venues. It is convenient and portable too, easy to transport without a large truck.
It's like a good hotrod Camaro with a well built small block Chevy, the right gears and sticky street tires. Or a small block Mustang, if you prefer. Good stuff, runs great. Good weekend warrior car. But my box is more like a blown big block with wrinkle walls. You just ain't gonna catch it. No contest, stop talking about it.
Now on to the hair-splitting.
Art, you made my point, one of them anyway, when you admitted that even using your data, you knew the distortion figure to be 9.99% yet you claimed it to be 9%. You rounded down. When you examine my charts, you quote a distortion value that is much higher (28%) than what the data shows (15%). You rounded up.
I also question your microphone, calibration, etc. There is ambiguity in your data that isn't there in mine. I'm using a calibrated microphone on a calibrated system that is known to provide reliable consistent results in this kind of measurement.
The distortion of my 12Pi is maxxed out as it enters cutoff, at 800 watts it's 15% at 32Hz. Below that, the slope of the rolloff of the fundamental and the slope of the rolloff of the harmonics is the same. They both fall at the same rate. So full tilt distortion at cutoff hits 15% and stays there all the way down into the stop band, where there is very little sound. It's pretty darn quiet below cutoff, no distortion, no sound at all. In the passband - where it counts - distortion is only 2-3%, even lower towards the upper end of the passband.
What other kind of high-power system you know of does that? What else out there can do 135dB/M at 2% THD and only 15% distortion even below cutoff? Sure, the passband in the 2-3% range is what's most important, but the fact that distortion is only 15% at cutoff and below really tells you something. It shows how effective push-pull drive is. Horn loading reduces distortion, and the push-pull drive takes what's left and cuts it in half.
That's my point, and it is why I split hairs with you.
Someone asked about the 12Pi hornsub back in April, and you launched into promoting your products. You unfairly and inaccurately characterized your speaker as performing better. But when I discovered this, I called you to task. That's what this is about.
I do want to say this though too, because I think it's important. I like your design. I don't think it compares with the 12Pi hornsub, but it's a different class of speaker. I would recommend your cabinet to anyone that needed a medium duty sub, something I would call a "club sub". It provides adequate output, good performance, and is more than enough for small to medium venues. It is convenient and portable too, easy to transport without a large truck.
It's like a good hotrod Camaro with a well built small block Chevy, the right gears and sticky street tires. Or a small block Mustang, if you prefer. Good stuff, runs great. Good weekend warrior car. But my box is more like a blown big block with wrinkle walls. You just ain't gonna catch it. No contest, stop talking about it.
Art,
Sure you can look at power levels, or Spl, or even a distortion threshold being met. Let's say that the tiny sunfire sub generates 11% distortion with a 30v input into the driver and an spl of 100db. Lets assume that giant horn sub Z under the exact same test conditions generates 13% thd at 30v but an spl of 120db. Sub Z has higher distortion but it is also putting out the spl of 10 of the other cabs. Does sub A have bragging rights to lower distortion because the percentage is lower with the same input voltage? Nevermind that the impedance profile and magnitude might be completely different resulting in greatly varying amounts of power being used for each. It stands to reason that the more powerful sub with 20db lower output demands would have much less distortion. The little Sunfire can barely match the output of sub Z with 10x the voltage and power and has much less headroom to begin with. We use speakers to produce sound. I just don't get why you would compare 2 speakers distortion while one is putting out much more sound. When doing an informal listening comparison you wouldnt play one set at a level much higher and use that to form an opinion would you? Voltage sensitivity and efficiency are looked at with other tests. Looking at distortion with a specified input voltage helps you gauge that particular subs performance ,power handling and when output limits are being reached.
Sure you can look at power levels, or Spl, or even a distortion threshold being met. Let's say that the tiny sunfire sub generates 11% distortion with a 30v input into the driver and an spl of 100db. Lets assume that giant horn sub Z under the exact same test conditions generates 13% thd at 30v but an spl of 120db. Sub Z has higher distortion but it is also putting out the spl of 10 of the other cabs. Does sub A have bragging rights to lower distortion because the percentage is lower with the same input voltage? Nevermind that the impedance profile and magnitude might be completely different resulting in greatly varying amounts of power being used for each. It stands to reason that the more powerful sub with 20db lower output demands would have much less distortion. The little Sunfire can barely match the output of sub Z with 10x the voltage and power and has much less headroom to begin with. We use speakers to produce sound. I just don't get why you would compare 2 speakers distortion while one is putting out much more sound. When doing an informal listening comparison you wouldnt play one set at a level much higher and use that to form an opinion would you? Voltage sensitivity and efficiency are looked at with other tests. Looking at distortion with a specified input voltage helps you gauge that particular subs performance ,power handling and when output limits are being reached.
Doug,
I built cabinets for the local bands I mixed from 1970-76, built cabinets and was a front of house technician with Eclipse Concert Systems out of Saint Paul in the late 1970s, started my own sound and light company Southern Thunder Sound Inc., In 1979, providing production for regional events and national tours with artists as diverse as Andy Williams, Jean Luc Ponty, and the Beastie Boys, as many as 800 shows a year.
All the Southern Thunder Sound cabinets (with a few exceptions made by R&R and EV) were designed by me, using the Welter Systems name .
I sold STS in 1992, Welter Systems has continued to be my sole proprietorship since. I moved from the Minneapolis area to New Mexico in 1996.
Although I have continued to design cabinets primarily for my own company’s use, my income comes from mixing and providing sound for live events and some studio work.
I have compared dozens of my own designs to dozens of commercial and proprietary systems over the course of my entire career, and like to weigh the positive and negative characteristics of any specifications or cabinets I see or hear.
Wayne Parham’s 12Pi cabinets have many positive characteristics, the only negatives being large size and weight, not a problem for arena tours.
Art Welter
Like many cabinets now, the Sunfire uses built in amplification and processing, making a side by side comparison to a sub without built in processing and amplification more difficult.
If it were 20dB lower in output than a giant horn sub, it would take upwards of 16 units (not 10) to equal the larger cabinet’s output.
If 16 Sunfires equaled or bettered the output and sound quality (distortion, phase, and other subjective stuff) of the giant sub, weight, size and cost then would be a determining factor.
Here in DYI land, some are only concerned with cost, while others will pay astounding amounts for “improvements” I could never hear .
If a single Sunfire had adequate output for the intended application (say a control room monitor sub) the 20 dB headroom the large sub provided would be a moot point, as the difference between 11 % distortion and 1 or 2 % distortion is not audible to me even when I concentrate on that metric alone.
So to specifically answer your two questions, sub A does have bragging rights to lower distortion if the distortion percentage is lower with the same input voltage than sub B, though that brag is of little consequence my (or Wayne's) world.
When doing an informal listening comparison, I wouldn’t play one set of speakers at a level much higher than another, my ears are old and threshold shift happens too quick for that to be of any use .
Art
Wayne,
Though your 12Pi has admirably low distortion below Fc for a horn sub,you still have not provided the evidence that distortion below Fc “is cut in half” by push pull, and you continue to claim my test results are not valid.
As long as you persist in that (or it becomes too boring), I’ll continue defending my claims.
Many folks reading don’t bother to look at the size of cabinets before considering them, in post #5 I mentioned cabinet sizes and output level, and wrote:
“a brief look at the 12 Pi distortion compared to the Lab 12s in ported or TH cabinets shows that the 12 Pi has perhaps a bit more distortion than the normally mounted ported 12”, and a bit less than the normally mounted 12” in the TH.”
I have since taken a closer look and posted many times that that was an error (with my apology), that the 12 Pi has lower distortion than my designs run at full power, and indeed most other designs I am aware of.
The JBL SRX 728 I mentioned is not mine, and all my designs mentioned are available for free, and I have no intention of building any for sale, and I answered some other questions about other designs brought up in this thread, I have no sales agenda.
That said, I have some used cabinets for sale that perform well for their size ;^).
In post #29, on June 21, 20011, you wrote:
“The fact that my hornsub is only 28% below cutoff is excellent evidence that the push-pull drive reduces distortion a significant amount.”
Now you seem to want to take that back, but whether it is 28%, or 25% at 30 Hz is no big deal to me, but it is not 15%, which requires harmonics to be about 16.5 dB below the fundamental.
No way can you stretch my Dual lab 12 distortion to have it come up to even 15% at 30 Hz.
I gave you the numbers, test my math, if you want I can give you all the harmonic figures until they fall below the noise threshold.
The microphone I used for testing my ported Lab 12s is a B&K 4004 using a 130 volt phantom power supply, rated for <1% THD at 148 dB.
Of course, any microphone distortion would be added to the speaker under test’s distortion with my set up.
As far as “calibration” I never claimed the output of my ported dual Lab 12 was higher than the 12Pi, and a calibrated SPL is not needed to figure distortion percentages.
Art Welter
Attachments
SPL normalization seems a much more useful method than power input normalization, to me.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.