Still learning ......
I was just reading the technical explanation of how two drivers in push-pull cancel out a lot of harmonic (2nd order) distortion. Evidently, the driver movement is not the same going forward as it is going backwards. Yes?)
I kind of, sort of understand it. Just enough to ask a potentially dumb question: (Or I could actually be a genius, you never know...... OK, probably not)
I am currently using Line 6 dual 12" subs. They are standard BR, with no divider inside. They sound great at low volume, but turn to absolute mush when pushed. (I realize that some of this might be port chuffing, but that can't be the main problem. I also realize that I might be hearing mostly 3rd & 4th order harmonics, but there's nothing I can do about that short of building small horns.)
Since each driver has it's own amp channel, I'm wondering if I couldn't turn these into a push-pull configuration, and maybe gain some cleanliness.
----------
SO:
What would happen if I mounted one of the 12" drivers facing inward, then reversed the wiring phase from its amp?
Crazy, or brilliant?
I realize that to do this properly I'd have to lose a very small amount of internal volume, hence the tuning would be slightly off, but if it lower distortion significantly, that might be worth it. (Plus, they are fairly shallow drivers, and / or I could easily extend the grill forward an inch or so.)
I was just reading the technical explanation of how two drivers in push-pull cancel out a lot of harmonic (2nd order) distortion. Evidently, the driver movement is not the same going forward as it is going backwards. Yes?)
I kind of, sort of understand it. Just enough to ask a potentially dumb question: (Or I could actually be a genius, you never know...... OK, probably not)
I am currently using Line 6 dual 12" subs. They are standard BR, with no divider inside. They sound great at low volume, but turn to absolute mush when pushed. (I realize that some of this might be port chuffing, but that can't be the main problem. I also realize that I might be hearing mostly 3rd & 4th order harmonics, but there's nothing I can do about that short of building small horns.)
Since each driver has it's own amp channel, I'm wondering if I couldn't turn these into a push-pull configuration, and maybe gain some cleanliness.
----------
SO:
What would happen if I mounted one of the 12" drivers facing inward, then reversed the wiring phase from its amp?
Crazy, or brilliant?
I realize that to do this properly I'd have to lose a very small amount of internal volume, hence the tuning would be slightly off, but if it lower distortion significantly, that might be worth it. (Plus, they are fairly shallow drivers, and / or I could easily extend the grill forward an inch or so.)
Last edited:
Well, maybe. That's why I'm still considering building small horns, instead.
Still, I want to see how far I can take what I have now, and hearing 3rd order ore clearly doesn't mean the overall sound is worse.
Anyway, it's an easy experiment to reverse. Sounds like you guys mostly think it's worth a try. But it raises 2 more questions:
-----------------------------------------------------
1: I realize how long the bass frequencies are, even the harmonics of 100 Hz, but still I wonder if alignment is critical. Where exactly, on the driver, would be the "acoustical center?" The center of the coil? Halfway-deep on the cone?
Obviously, I can't align the coils, as one speaker would end up crazy-deep into the cabinet, messing the tuning too much. (I assume.) Let's say I mount the reverse-driver only about halfway in. That would put the cone-centers maybe 5" to 6" away from each other. Should this be enough? Could this actually cause more mud that it cancels? (I'm probably overthinking it, I know)
2: I've read that push-pull works better (for reasons I don't understand, if both drivers share the same box, as opposed to having a divider inside.
Is this true? If so, why?
I ask, because I've also read that two drivers in one box, when mounted standard BR, should ideally HAVE a divider as this sounds better.
A pair of 12's isn't going to go very loud in a large space.
https://line6.com/data/6/0a06434d10...e6 L3s Specifications - English ( Rev A ).pdf
One thing you could also try is highpassing them as high as 50-60Hz. If that works, you just need more woofers to get as loud as you want to be..
https://line6.com/data/6/0a06434d10...e6 L3s Specifications - English ( Rev A ).pdf
One thing you could also try is highpassing them as high as 50-60Hz. If that works, you just need more woofers to get as loud as you want to be..
Alignment isn't critical. Give it a try, but I really don't think that the distortion cancellation on offer is going to save these boxes.
Where are you in the world?
Add your location and you might find there's someone nearby that'll let you listen to what they have, so you've got an idea of what's out there.
Chris
Where are you in the world?
Add your location and you might find there's someone nearby that'll let you listen to what they have, so you've got an idea of what's out there.
Chris
Cableaddict,
A) Horn loading loads drivers more heavily- they do more work (produce more SPL) for a given voltage input. More work comes with more cone stress and more distortion. A bit less distortion at the same SPL, but more distortion at full power, which you obviously will use, since you are looking for more SPL than small boxes are capable of.
B) Even order distortion produces octaves of the fundamental, which musically changes nothing. Odd order distortion adds notes which were not in the original musical composition. "Worse" is a subjective term which would imply those notes don't sound as good as what the original composer chose. I'd probably use "funny", or "different" or "odd", depending on the type of music - OK, some music simply sounds "wrong" when odd order distortion predominates.
1) The higher the order, the more critical the alignment becomes. At 100 Hz, anything within 1/4 wavelength (33.9 inches) is "close enough". At 200 Hz (second harmonic), 16.95", 400 Hz (fourth harmonic), 8.47".
1A) The acoustical center would be dependent on material composition, voice coil and cone depth.
1B) Usually neither.
1C) Might be very close, depending on the drivers, but probably not required.
1D) Some guys would argue the "mud" comes from odd order distortion ;^).
2) Who wrote that?
2A) I can't think of any reason why sharing the same box would work "better" for push pull.
2B)There are several valid reasons explaining why separate chambers "sound better"- stiffer box, less intermodulation between drivers, less chance of destruction of the both drivers if one fails, etc.
Art
Wow, I had not read this before.
Thanks, Art, that's a hugely important piece of information.
Lot's more thinking to do now ....
And they would be wrong! (So wrong, it's kind of funny.)
- On this one point, I am 100% positive. (I'm a life-long pro recording engineer.) -----------------------------------------------
I've read it many times. Luckily, I found the last reference in my browser history:
PiSpeakers Forum - Push-pull versus shorting rings - Wayne Parham, June 21, 2006 at 16:18:33
(A very good read, overall.)
Valid point, in many cases. Understand that (if I can get the sound a little better) I'll be using THREE of these 2 X 12" subs, and maybe a fourth outdoors.
Also, with the right design, there are actually arguments to the contrary:
Note, this info comes from many sources, including Jeff at JTR, with whom I recently discussed his decision to retire his small horn-loaded "Growler" sub, and replace it with a 2 X 12" BR design.
Two 12" gives the same cone area as one 18", so theoretically the 2 designs should be capable of the same output, if the drivers have the same Xmax. (And no, the 18" being deeper does not matter, as some folks have suggested elsewhere.)
So it mainly comes down to efficiency. The Line 6, the Yorkville, and the JTR 2 X 12" designs all use EXTREMELY efficient drivers. Jeff claims 103 dB efficiency! He says it's the most efficient sub he's ever made. The Line 6 and Yorkie aren't quite that efficient, but pretty close.
Now factor in that a 12" driver needs less box volume for the same bass extension, and this design starts to look very promising, for guys like me that need portability.
The downside is that my Line 6 subs, while they sound GREAT at low volume (really, really great) completely fall apart when pushed hard. This could be due to several factors, of course. Harmonic distortion is probably the least of them, but it's the main subject of this particular thread.
This does help a little (I tried it today) but it defeats the whole point of these subs, which is good LF extension in a small package. If I didn't mind the sound of a 50 Hz rolloff, I'd already be using small horns.
Sadly, I love that extended LF, even for rock music. I don't need 20Hz, but I DO need 40 Hz at a solid amplitude.
Yeah, I realize that everything is a trade-off ! I will probably build something else this year, but even then, I want to experiment with these L3S subs, to see what can & can't be improved. The next step is retrofitting flared ports.)
Thanks for your thoughts.
Last edited:
If you want peak output from two 12s, consider a series-tuned push-pull bandpass enclosure instead.
1. The push-pull arrangement will reduce even-order distortion.
2. The bandpass arrangement (external vent) will reduce odd-order distortion as the vent acts as a filter
3. The series-tuned 6th order arrangement reduces excursion at frequencies at the lower end of the passband, also reducing distortion. And any out of band noise that makes it through the internal vent should be filtered a bit by the second vent.
The reduction in distortion is real and can be measured. Here's the measured distortion for a clamshell 4th order BP design that I built (which covers steps one and two above). Distortion only gets significant below the usable passband of this system.
Disadvantage? You'll definitely need something to cover the kickbass region.
BTW, peak SPL at low frequencies is related to cone displacement, not cone area.
1. The push-pull arrangement will reduce even-order distortion.
2. The bandpass arrangement (external vent) will reduce odd-order distortion as the vent acts as a filter
3. The series-tuned 6th order arrangement reduces excursion at frequencies at the lower end of the passband, also reducing distortion. And any out of band noise that makes it through the internal vent should be filtered a bit by the second vent.
The reduction in distortion is real and can be measured. Here's the measured distortion for a clamshell 4th order BP design that I built (which covers steps one and two above). Distortion only gets significant below the usable passband of this system.
Disadvantage? You'll definitely need something to cover the kickbass region.
BTW, peak SPL at low frequencies is related to cone displacement, not cone area.
Attachments
Brian, that's a fascinating idea. - but if it limits the LF, then it's just not going to work for me.
Sure but displacement is cone area times excursion. So again, if the Xmax is the same, peak SPL should be the same. I was briefly referring to the false idea, often read, that an 18" moves more air than two 12" because it's deeper.
Which is obviously false if you think about it.
Sure but displacement is cone area times excursion. So again, if the Xmax is the same, peak SPL should be the same. I was briefly referring to the false idea, often read, that an 18" moves more air than two 12" because it's deeper.
Which is obviously false if you think about it.
It's not going to limit the LF, at least when compared to a vented alignment. The bandpass arrangement will limit the output at midbass frequencies, the amount depending on how wide a passband the design covers. A narrow passband will reduce distortion and increase power handling. A wide passband will not do this. If I was going to design one, I'd probably aim for a 40Hz~100 Hz passband, no wider.
The motor is usually the most expensive part of a driver, so two 12s with the same excursion as one 18 could end up costing more, all else being equal. There are valid reasons for going for two 12s instead though, e.g. form factor. A box containing a 12" driver will have a minimum dimension that's smaller than one containing an 18" driver.
Thanks for the clarification. So you meant I might need to add a MID-bass box, not a low-bass box.
But since I don't need anything below about 40 Hz, that's not an issue. Your last suggestion is exactly where I want it to be.
Actually, 40 - 80 Hz, with a slight rolloff at 100Hz.
As for cost, that is not an issue. This is how I make my living these days. I need small, I need (relatively) loud, I need as much efficiency as possible. - But I don't need cheap, luckily.
But since I don't need anything below about 40 Hz, that's not an issue. Your last suggestion is exactly where I want it to be.
Actually, 40 - 80 Hz, with a slight rolloff at 100Hz.
As for cost, that is not an issue. This is how I make my living these days. I need small, I need (relatively) loud, I need as much efficiency as possible. - But I don't need cheap, luckily.
Nitpick here: power of 2 order distortion (2nd, 4th, 8th...) produces octaves. Other even orders - 6th, 10th, 12th, etc. are not octaves.
I also wonder whether even the power of 2, octave relationship higher order distortion products can produce more dissonant difference tones and beating with other non-power of 2 distortion products. Do we know?
I personally have a couple of PPSL sub cabinets and think they sound great even with cheap drivers (but I may be biased
)
1) You are correct, I had not done the math past 4th order. Past 4th or 5th order, a good speaker has less distortion than I can detect by ear when run at reasonable levels.1)Nitpick here: power of 2 order distortion (2nd, 4th, 8th...) produces octaves. Other even orders - 6th, 10th, 12th, etc. are not octaves.
2)I also wonder whether even the power of 2, octave relationship higher order distortion products can produce more dissonant difference tones and beating with other non-power of 2 distortion products. Do we know?
3)I personally have a couple of PPSL sub cabinets and think they sound great even with cheap drivers (but I may be biased
2) Most "low" acoustic instruments have 2nd order harmonics louder than the fundamental note, they don't cause dissonant difference tones and beating, though acoustic piano tuning is usually "tempered" (or "stretched") to reduce dissonance in certain chords.
3)Cheap drivers tend to benefit from PP because making suspensions and magnet structures that work equally well in both directions ("pushing or pulling") tends to cost more.
A single linear (expensive..) driver may have less distortion below Xmax than a pair of "cheap" PP of equal displacement.
Last edited:
That's a good question.
There is a phenomenon called "tartini tones", which is harmonics from two very close frequencies modulating each other, creating a very audible distortion.
I've dealt with this quite a few times when mixing songs, & you wouldn't believe how bad it can be.
- But I have no idea if that's cause by 2nd order or not. I suspect it's more odd order, and the resulting distortion tends to be thin & bright. But again, I really don't know.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.