This might not be the right thread to ask this question, but I've always wondered... generally speaking, is distortion lower in (i) a ported design or (ii) a sealed design with some bass EQ to bring up the low end response to something closer to the ported version?
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Zaph Audio ZA5.2 DIY Kit Speaker Review
- Thread starter amirm
- Start date
This might not be the right thread to ask this question, but I've always wondered... generally speaking, is distortion lower in (i) a ported design or (ii) a sealed design with some bass EQ to bring up the low end response to something closer to the ported version?
A ported alignment means that (ideally) at resonance the driver is stationary and the output is from the port. This puts some perspective on the advantages of a ported design. At low frequencies, where excursion is largest, the port takes over the output duties. The result is that the driver is, in general, not required to provide as large excursions as a sealed speaker would with the same diameter driver for the same output.
If you go sealed and EQ (or to be more correct, apply a Linkwitz correction) the bass driver must make larger excursions to reach the same output as the ported system - you must move the air somehow. So, you would generally expect that the ported system would have lower driver excursion related distortion for the same deep output. However, sealed systems may benefit by using the air mass as a linearising part of their operation - basically the idea being that air compresses more linearly than a linen spider. So the sealed system may claw back excursion linearity. Even then there are second order effects. An enclosed volume of air is not perfectly linear, and is asymmetric on compression versus rarefaction. But the effect is only really apparent at extremes. Overall, there is no hard and fast answer. But It is always worth emphasising that an eq'ed sealed box inherently requires significant additional woofer excursion over a ported one with the same diameter driver. Some people seem to regard active alignment correction as a magic bullet, and forget that there is no substitute for swept volume.
Significantly higher power levels needed to get the low bass out of sealed box also mean additional voice coil heating and magnetic gap modulation distortion effects.
If you go sealed and EQ (or to be more correct, apply a Linkwitz correction) the bass driver must make larger excursions to reach the same output as the ported system - you must move the air somehow. So, you would generally expect that the ported system would have lower driver excursion related distortion for the same deep output. However, sealed systems may benefit by using the air mass as a linearising part of their operation - basically the idea being that air compresses more linearly than a linen spider. So the sealed system may claw back excursion linearity. Even then there are second order effects. An enclosed volume of air is not perfectly linear, and is asymmetric on compression versus rarefaction. But the effect is only really apparent at extremes. Overall, there is no hard and fast answer. But It is always worth emphasising that an eq'ed sealed box inherently requires significant additional woofer excursion over a ported one with the same diameter driver. Some people seem to regard active alignment correction as a magic bullet, and forget that there is no substitute for swept volume.
Significantly higher power levels needed to get the low bass out of sealed box also mean additional voice coil heating and magnetic gap modulation distortion effects.
Last edited:
Exactly and as long as you don't play below that port tuning frequency there aren't really any downsides that I'm aware of. I like to take advantage of ports by running mains with a 4th order high pass so they benefit from the low distortion port output but are down at least 18db by the port tuning frequency.
Nice analysis. Pity we dont' see more of transmission line designs as IMO they offer some advantages over bass reflex design albeit with increased cost due to more complex and larger box design. What is your opinion on this?
- Joined
- Jan 15, 2020
- Messages
- 6,897
- Likes
- 16,900
In engineering / technical solutions everything is a compromise, to obtain with the same drivers used more level you need a resonator were a transmission line has usually a larger resonance/delay compared to a BR and also side-effects that are measurable in the FR https://www.stereophile.com/content/pmc-db1iii-loudspeaker-measurements
These were nice solutions when passive loudspeakers were the way to go and amps had low efficiency and output, but nowadays with active loudspeakers and class-D amps they are kind of dinosaurs.
These were nice solutions when passive loudspeakers were the way to go and amps had low efficiency and output, but nowadays with active loudspeakers and class-D amps they are kind of dinosaurs.
The main problem with discussions about TL speaker designs is that they are not, and cannot be, true transmission lines*. The handwaving that went on about low speed of sound in the line, and other stuff, was just fantasy. OTOH, the design of a tuned line with carefully placed damping is very much alive. The work described over at Quarter Wave will tell you more than you probably wanted to know. These speakers seems to yield good results and have a solid mathematical footing. Some of the bass speakers designed using Hornresp are very similar, and there seems to be general agreement between both sets of design regimes and their respective mathematical modelling.
These designs can get you some serious deep powerful bass. The group delay might not be to your taste.
A true transmission line is possible for mid-range drivers. But there you just get a long damped line, and one tweaks the stuffing and measures until the line is dead. No open termination. IMHO such an approach is worth investigating (cf the B&W Nautilus). But you can't use it for real bass. Just silly big.
* As generally understood, a true transmission line is something where the energy put into it never comes back. The finite length of any real line makes this a difficult problem, and they will all behave as either quarter or half wave resonators (depending upon termination.) If you don't put any energy into the line near the quarter wavelength, close the end, and properly damp it, you can get close to a black hole for the rear radiation from a mid-range driver. It still isn't exactly a small enclosure. Short open ended lines for mid-ranges (Statements for instance) are another matter again.
These designs can get you some serious deep powerful bass. The group delay might not be to your taste.
A true transmission line is possible for mid-range drivers. But there you just get a long damped line, and one tweaks the stuffing and measures until the line is dead. No open termination. IMHO such an approach is worth investigating (cf the B&W Nautilus). But you can't use it for real bass. Just silly big.
* As generally understood, a true transmission line is something where the energy put into it never comes back. The finite length of any real line makes this a difficult problem, and they will all behave as either quarter or half wave resonators (depending upon termination.) If you don't put any energy into the line near the quarter wavelength, close the end, and properly damp it, you can get close to a black hole for the rear radiation from a mid-range driver. It still isn't exactly a small enclosure. Short open ended lines for mid-ranges (Statements for instance) are another matter again.
This whole data resolution thing made me a bit suspicious so I went and added a detailed resolution chart to Loudspeaker Explorer:
It's only in the low frequencies that the resolution changed. Above 100 Hz or so the resolution has always been 20 points/octave. That actually makes sense if we assume that the data is generated from a smaller FFT length: because of the linear-to-log-spaced conversion, that would leave high frequencies unaffected.
This does mean that with the new parameters @amirm is using, we're only getting 10 points per octave around 40 Hz. Not a big deal, though.
OP
- Thread Starter
- #108
I will be increasing the resolution when a speaker has substantial low frequency output. I did that for subs. But for most of speakers out there, the response down low is non-existent and differences are so small.
I was happy to see that as well. It really helps to put things in perspective at a glance. This is a quite good sounding speaker for a very reasonable price.
Well this one is back with me now. Overall, I was pleased with the outcome.
Despite somewhat lower sensitivity, it measured very favorably. As most of us know, speakers designs are trade-offs. The trade-off made here is likely lower sensitivity in favor of being a much easier amplifier load.
If you look over the designer's website for this speaker, this seems to make sense for another reason - he uses multiples in the larger variants. This allows the woofer be paralleled with another and still be a reasonable amplifier load. Hope you found this a worthwhile effort and look forward to the C-Note review.
Despite somewhat lower sensitivity, it measured very favorably. As most of us know, speakers designs are trade-offs. The trade-off made here is likely lower sensitivity in favor of being a much easier amplifier load.
If you look over the designer's website for this speaker, this seems to make sense for another reason - he uses multiples in the larger variants. This allows the woofer be paralleled with another and still be a reasonable amplifier load. Hope you found this a worthwhile effort and look forward to the C-Note review.
btw, one of the neat aspects of DIY, is reusability. You don't have to take a huge hit every time you want to upgrade...
If you are like Amir and need more bass for a larger room, you can take the drivers from the smaller ZA5.2 and build one of these:
If you are like Amir and need more bass for a larger room, you can take the drivers from the smaller ZA5.2 and build one of these:
Or perhaps you want to try an open baffle and a budget active design...
http://gainphile.blogspot.com/2012/06/s20-z-compact-open-baffle-using-zaph.html
http://gainphile.blogspot.com/2012/06/s20-z-compact-open-baffle-using-zaph.html
ta240
Major Contributor
- Joined
- Nov 7, 2019
- Messages
- 1,427
- Likes
- 2,863
that looks like a fun future project/upgrade to try.
Am contemplating the open baffle one. I have all the parts except the woofers and they are pretty reasonable.
ta240
Major Contributor
- Joined
- Nov 7, 2019
- Messages
- 1,427
- Likes
- 2,863
Am contemplating the open baffle one. I have all the parts except the woofers and they are pretty reasonable.
That looks like a great way to get into open baffle; which has interested me for quite some time.
100%
I suppose coming from DIY guys objectivity might be skewed but there is no doubt in my mind that DIY surpasses at the very least the quality of the build and crossovers compared to commercial speakers and at the most making a speaker worth in the few thousands with just around $400 or $500 worth of components. I understand more DIY speakers have to be measured for us to see how they match up against branded speakers on the measurement scale. I am a little disappointed why it won't be part of the database, for a forum like this I would expect for most to not waste money on overpriced speakers and make them themselves. I do understand that there are other goals at play here including getting the industry to shape up but that doesn't take away what us audio fanatics do and that is in essence DIY with whatever our knowledge base allows us to do. Point being, we want more DIY speakers measured and added to the database or a separate data base for DIY, this is an important part of what shapes speaker design, builds and listening.
So far, ASR has only tested one DIY speaker and it tested better than many other speakers. Despite some initial objection, it is in the database. Like the electronics, it has a DIY prefix (which is pointless IMO).
Aside from Amir's time and a backlog of commercial speakers, there are some notable hurdles. The most notable is lack of DIY speaker submissions. Not sure about the lack of membership loaners, but the other reason is simply part availability for the more popular designs (at least the ones with cabinets). Both the Overnight Sensations and the Samba are backordered into June and August respectively. The S2000 is out of stock too.
Would love to be building something else to test, but am in a holding pattern for now.
I'm hoping to contribute to that a bit as soon as it's safe to leave the house.
didn't you (or someone else, maybe) send in a C-Note? Is that still pending?
Last edited:
Meh, I had PMC Twenty.22's and at very low frequency's they produced some weird port noise. I wasn't impressed by them at all.
Similar threads
