Hello all,
First time posting, so I thought I’d take the opportunity to thank the community for all the info I’ve come across on this forum. I’m completely new to this and you guys have been a huge help up to now, but the more research I do, the more questions I have, so I’ve decided to jump down the rabbit hole and register as a member.
A friend of mine wants me to design a cabinet for a audio Nirvana Super 15 inch ferrite full range driver, and the enclosures that Audio Nirvana suggests are a one size fits all configuration of very basic bass reflex boxes, which makes me think there’s room for improvement.
I found out that the cabinet I’d sketched out was called an Onken type enclosure, and that those had been developed by “mad scientists” (not my words) who wrote down a rudimentary chart for proportions, but no actual formula.
I’ve found a calculator online : mh-audio.nl/Calculators/Onken.html which lets you put in the parameters of your driver and the size and number of vents and that gives you the total volume of the box and the length of the vents. Perfect!
I’d read you could simulate these in Winisd as well, so I figured it would give me additional information and check the maths. And this is where my issues start.
I’ve modeled the data I got from the online Onken calculator and found out that the box was tuned to 44 Hz.
As I mentioned above, I am new to this, and there is a lot of information to digest, so I have a bunch of questions, most of which will seem very basic. I am a fast learner though so bear with me please, and thank you in advance for your patience.
- The driver is rated for a free air frequency of 31,7 Hz, which is the default value Winisd gave me when I started the project. I am guessing I should aim to tune the box by tweaking the vents to be as close to that value as possible, correct? What is the consequence of the 44 Hz tuning that the online calculator suggested?
- The driver has a test bench frequency response from 31 Hz to 16000 Hz. Does that mean that I should set the minimal frequency on my model at 31 Hz and assume anything that happens on my model below 31 Hz will have no impact because there wouldn’t be a signal?
- Should I aim to have F3 as close to 31 Hz as I can?
I understand these parameters are determined through box volume and vent volume, and I can fiddle with those until I find the best curve. This will also determine wind velocity which should not exceed 18 m/s.
Then I started looking at cone excursion. And this is when the real issues started. The driver is rated for a 50 watt RMS which is what I set my signal to. And at that rate, the curve only drops beneath the drivers Xmax (1 mm) after 140 Hz or so. I understand you can filter out the signal up to a determined frequency on your source, but 140 Hz seems ridiculous. I can also get the curve to drop if I lower the input power, but I have to drop it to 3 W to get it below the 1 mm line, and that’s ridiculous as well. I’ve noticed the dimensions of the box have an influence on the cone excursion, and the smaller the box, the lower the curve, but if I want to drop below the 1 mm line, the box gets so small that I’m better off not bothering with a box at all.
I’d like to avoid filtering the signal as much as possible. There must be a way of designing the box so that it doesn’t blow the driver up.
So what am I missing? Up to cone excursion, I could (almost) get my head around the various parameters. But at this point, my model makes it look like the driver is unusable, which I’m sure is not the case, so what’s going on?
Many thanks for your insights, I’ve only just started and I am already very confused and frustrated!
Max
First time posting, so I thought I’d take the opportunity to thank the community for all the info I’ve come across on this forum. I’m completely new to this and you guys have been a huge help up to now, but the more research I do, the more questions I have, so I’ve decided to jump down the rabbit hole and register as a member.
A friend of mine wants me to design a cabinet for a audio Nirvana Super 15 inch ferrite full range driver, and the enclosures that Audio Nirvana suggests are a one size fits all configuration of very basic bass reflex boxes, which makes me think there’s room for improvement.
I found out that the cabinet I’d sketched out was called an Onken type enclosure, and that those had been developed by “mad scientists” (not my words) who wrote down a rudimentary chart for proportions, but no actual formula.
I’ve found a calculator online : mh-audio.nl/Calculators/Onken.html which lets you put in the parameters of your driver and the size and number of vents and that gives you the total volume of the box and the length of the vents. Perfect!
I’d read you could simulate these in Winisd as well, so I figured it would give me additional information and check the maths. And this is where my issues start.
I’ve modeled the data I got from the online Onken calculator and found out that the box was tuned to 44 Hz.
As I mentioned above, I am new to this, and there is a lot of information to digest, so I have a bunch of questions, most of which will seem very basic. I am a fast learner though so bear with me please, and thank you in advance for your patience.
- The driver is rated for a free air frequency of 31,7 Hz, which is the default value Winisd gave me when I started the project. I am guessing I should aim to tune the box by tweaking the vents to be as close to that value as possible, correct? What is the consequence of the 44 Hz tuning that the online calculator suggested?
- The driver has a test bench frequency response from 31 Hz to 16000 Hz. Does that mean that I should set the minimal frequency on my model at 31 Hz and assume anything that happens on my model below 31 Hz will have no impact because there wouldn’t be a signal?
- Should I aim to have F3 as close to 31 Hz as I can?
I understand these parameters are determined through box volume and vent volume, and I can fiddle with those until I find the best curve. This will also determine wind velocity which should not exceed 18 m/s.
Then I started looking at cone excursion. And this is when the real issues started. The driver is rated for a 50 watt RMS which is what I set my signal to. And at that rate, the curve only drops beneath the drivers Xmax (1 mm) after 140 Hz or so. I understand you can filter out the signal up to a determined frequency on your source, but 140 Hz seems ridiculous. I can also get the curve to drop if I lower the input power, but I have to drop it to 3 W to get it below the 1 mm line, and that’s ridiculous as well. I’ve noticed the dimensions of the box have an influence on the cone excursion, and the smaller the box, the lower the curve, but if I want to drop below the 1 mm line, the box gets so small that I’m better off not bothering with a box at all.
I’d like to avoid filtering the signal as much as possible. There must be a way of designing the box so that it doesn’t blow the driver up.
So what am I missing? Up to cone excursion, I could (almost) get my head around the various parameters. But at this point, my model makes it look like the driver is unusable, which I’m sure is not the case, so what’s going on?
Many thanks for your insights, I’ve only just started and I am already very confused and frustrated!
Max
That's why i don't like thos (partly). They are stated 50w but don't do it untill low. With that excursion at +3.5W they run out of steam. Idem with the frequency response below, they say they go to 31Hz (FS), but the lowest i would tune it is 35Hz or you get a very uneven bass response. And the topend is very ragged, even for a fullrange driver. Basicly they are crap in my book...
Btw, the best tuning i could get out of it is a 250L cabinet with 2 round vents of 7cm diameter and 2.4cm long. As long as you don't use it with an amp higher than 3-4 watt (basicly a 2A3 or EL84 SE amp) it will work to an F3 of 35Hz. That's what i designed for a guy down here last summer. He did use a BFC type of filter altough at 500Hz (to take 5dB of the top) and added a tweeter above 10kHz as this driver is not fullrange or flat responding at all. I'm building a similar cabinet now for a much cheaper driver, the Fane Sovereighn 12-250TC that will do much better (F3 of 30Hz in 174L ported and 20w before running out of xmax) and a much less ragged response. So i don't know how this guy keeps selling his crappy drivers...
Btw, the best tuning i could get out of it is a 250L cabinet with 2 round vents of 7cm diameter and 2.4cm long. As long as you don't use it with an amp higher than 3-4 watt (basicly a 2A3 or EL84 SE amp) it will work to an F3 of 35Hz. That's what i designed for a guy down here last summer. He did use a BFC type of filter altough at 500Hz (to take 5dB of the top) and added a tweeter above 10kHz as this driver is not fullrange or flat responding at all. I'm building a similar cabinet now for a much cheaper driver, the Fane Sovereighn 12-250TC that will do much better (F3 of 30Hz in 174L ported and 20w before running out of xmax) and a much less ragged response. So i don't know how this guy keeps selling his crappy drivers...
Last edited:
Member
Joined 2009
Paid Member
This is a superb driver, one of the best. Whether it is worth trying to get an Onken alignment I’m not sure because I think this driver does well in a ‘simple box’. Albeit a large box. I suspect there’s little to be gained in a reflex/onken alignment as this driver has a very large Vas.
A Big'un - the Audio Nirvana Super 15
A Big'un - the Audio Nirvana Super 15
I only have a moment, but quickly trying to cover a few general points:
Firstly, the (mathematical) result will depend slightly on variations in the data input. I assume in this case we're talking about the current spec. AN unit https://www.commonsenseaudio.com/an15cfspecs.jpg rather than its significantly different predecessors which shared the name).
-An Onken 'proper' is a specific vented box alignment descending from the Jensen Ultraflex and its predecessors in turn. Despite appearances, that's simply what it is: a vented box. It has a massive symmetric vertical vent system, usually with a CSA of about 80% - 100% of driver Sd for high acoustic efficiency, albeit at the price of potential 1/2 wave resonant modes if ducts get excessively long.
-Yes, you can model them in WinISD -as noted, it's 'just' a vented box so there is little issue in assessing a basic alignment in that sense -from a modelling perspective, there's nothing difficult in a macro-sense (some detail variations, but for broad alignment trends, this holds). Compared to most others, it simply has a large number of rectangular vents, so you can model them in almost any piece of lumped element design software that allows different numbers of vents, or even those that don't if you simply use the sum area of all the vents, and add a minor correction for potential frictional losses.
-The Onken calculator is simply giving you an acoustically efficient volume and tuning, which is one of the things it was set up for; you will likely get a relatively significant peak at its suggested 44Hz tuning. Unlikely to be catastrophic, but I'd expect a handful of dB (no time to check at this moment).
-The default tuning WinISD is giving you is simply a different alignment, probably SBB4 which in this case is a nicely damped alignment under anechoic conditions (i.e. suspended in free air rather than being in a room), so may have less of a peak at tuning in practice than the unadjusted / undamped Onken.
-Yes and no, re tuning to Fs; it depends what alignment you're aiming for and what sort of power-handling you're after as it affects the loading.
-As far as power-handling goes, you're not really missing much. Rated power input is just how much the voice coil can take, thermally speaking, without failing. This is a high[ish] sensitivity driver with very little nominally rated linear suspension travel available to it (1mm). It has more mechanical travel than that, but as you exceed rated linear travel, the bass frequencies are likely to compress more as distortion rises. SBB4 is one of the better default T/S alignments in that sense for this unit (assuming the published data is correct) but it's not a subwoofer so if you want to be playing bass-heavy material at high uncompressed SPLs, it's not going to do that I'm afraid.
Firstly, the (mathematical) result will depend slightly on variations in the data input. I assume in this case we're talking about the current spec. AN unit https://www.commonsenseaudio.com/an15cfspecs.jpg rather than its significantly different predecessors which shared the name).
-An Onken 'proper' is a specific vented box alignment descending from the Jensen Ultraflex and its predecessors in turn. Despite appearances, that's simply what it is: a vented box. It has a massive symmetric vertical vent system, usually with a CSA of about 80% - 100% of driver Sd for high acoustic efficiency, albeit at the price of potential 1/2 wave resonant modes if ducts get excessively long.
-Yes, you can model them in WinISD -as noted, it's 'just' a vented box so there is little issue in assessing a basic alignment in that sense -from a modelling perspective, there's nothing difficult in a macro-sense (some detail variations, but for broad alignment trends, this holds). Compared to most others, it simply has a large number of rectangular vents, so you can model them in almost any piece of lumped element design software that allows different numbers of vents, or even those that don't if you simply use the sum area of all the vents, and add a minor correction for potential frictional losses.
-The Onken calculator is simply giving you an acoustically efficient volume and tuning, which is one of the things it was set up for; you will likely get a relatively significant peak at its suggested 44Hz tuning. Unlikely to be catastrophic, but I'd expect a handful of dB (no time to check at this moment).
-The default tuning WinISD is giving you is simply a different alignment, probably SBB4 which in this case is a nicely damped alignment under anechoic conditions (i.e. suspended in free air rather than being in a room), so may have less of a peak at tuning in practice than the unadjusted / undamped Onken.
-Yes and no, re tuning to Fs; it depends what alignment you're aiming for and what sort of power-handling you're after as it affects the loading.
-As far as power-handling goes, you're not really missing much. Rated power input is just how much the voice coil can take, thermally speaking, without failing. This is a high[ish] sensitivity driver with very little nominally rated linear suspension travel available to it (1mm). It has more mechanical travel than that, but as you exceed rated linear travel, the bass frequencies are likely to compress more as distortion rises. SBB4 is one of the better default T/S alignments in that sense for this unit (assuming the published data is correct) but it's not a subwoofer so if you want to be playing bass-heavy material at high uncompressed SPLs, it's not going to do that I'm afraid.
Last edited:
I haven’t read what has come before yet, but i started a miniOnken for these, but abandoned them when the size prved too large. These drivers do not need any extra work to get bass out of.
The big Boffle-like enclosure i recommended and BigUn built seems to be the more practical “box” for these drovers.
dave
The big Boffle-like enclosure i recommended and BigUn built seems to be the more practical “box” for these drovers.
dave
They are really bad.
No.
Fs is only a very weal indicator of how low a box can reasonably go. The Qt is also very important.
Note that F(-3) is a useless number AFA human hearing is concerned (ref Toole). Look at F6 & F10. Keep rook gain in mind.
Probably not something you need to worry about. And the same with power handling which is a prety useless number.
dave
Might have to look at the driver again in that case. They should give the driver a new name if it is different.
dave
Published specs on the original were a mile from actual - but I absolutely love my pair.
Stuck them into some huge cheap PA cabs that dad had (maybe 200litres or so, with a couple of 3" ports) and they really rock even with low power tube amps. As mentioned above, you don't need to do anything fancy to extract bass.
Stuck them into some huge cheap PA cabs that dad had (maybe 200litres or so, with a couple of 3" ports) and they really rock even with low power tube amps. As mentioned above, you don't need to do anything fancy to extract bass.
Good to know and based on published specs that includes a Qts > than a BR's ~0.403 upper limit that further increases its net volume [Vb], i.e. it ideally requires a ~1401.23 L cab tuned ~a 1/3 octave below Fs [~21.5 Hz] for a max flat alignment, so power handling down low is non-existent!!!
Ergo the only advantages to the Onken are a smaller, unique looking cab tuned higher with a large vent system's organ pipe harmonics comb filtering with the driver's ~200-600 Hz BW where 'warmth', 'fullness' ['mudiness' if too much] and last, but not least, 'honk', resides with the attendant under-damped response, which ideally will require significant added damping to smooth/'tighten' it up, negating the main reason folks prefer it ['colored' upper mid bass, lower mids]!
Yeah, I figured as much and was going to address this all to common issue next.
OK, mind sharing them for prosperity?
Member
Joined 2009
Paid Member
The thing is, when you just shove them in a big box they sound wonderful, the spec which might help in designing a resonant enclosure don’t enter into it.
But why they sound so good I don’t know, perhaps the whizzer is optimal (Bud Purvine thought so) or perhaps big cones with traditional cloth surrounds in rigid chassis mean the usual parasitic issues are relatively small. Nevertheless, this particular driver from David is a potential ‘final destination’ driver.
But why they sound so good I don’t know, perhaps the whizzer is optimal (Bud Purvine thought so) or perhaps big cones with traditional cloth surrounds in rigid chassis mean the usual parasitic issues are relatively small. Nevertheless, this particular driver from David is a potential ‘final destination’ driver.
Thanks for the replies guys! Having no knowledge of acoustics whatsoever, a lot of the things you mention come across as complete gibberish until I get a chance to look them up. This is very much like reading a book in a foreign language with an open dictionary next to me. If there is any comprehensive “must have” reading material you can recommend, I’d love a few references!
Opinions seem to vary regarding the driver in itself, but that may be a result of enclosure configurations.
@Scottmoose : I clearly need to read up on alignments, I will do than ASAP!
@Planet10 : you make my lack of knowledge painfully obvious hahaha I always welcome an opportunity for education and you make it sound like this isn’t a completely lost cause so I will soldier on!
The basic idea behind this project is to build a pair of stand alone speakers that are as user friendly as possible, hence the full range driver (I don’t really feel brave enough to start messing with crossovers, one thing at a time) .
Size isn’t really an issue, but the sketches I've come up with are around 200 L (NOT 1400 L!!).
I’m a furniture designer by trade so both function and form are important. I am not set on the specific driver, and I do like the look of vented boxes, so if there are other options that are better suited to the Onken type configuration, I will gladly look into those!
Opinions seem to vary regarding the driver in itself, but that may be a result of enclosure configurations.
@Waxx : I haven’t done the research on the drivers, but I’ll look into those, thanks for the tip
@Bigun : I had a quick look at your thread, and i will do some more reading tonight, but the first three pages make it quite clear this driver doesn’t work well with vents... only 50 more pages to go!
@Scottmoose : I clearly need to read up on alignments, I will do than ASAP!
@Planet10 : you make my lack of knowledge painfully obvious hahaha I always welcome an opportunity for education and you make it sound like this isn’t a completely lost cause so I will soldier on!
The basic idea behind this project is to build a pair of stand alone speakers that are as user friendly as possible, hence the full range driver (I don’t really feel brave enough to start messing with crossovers, one thing at a time) .
Size isn’t really an issue, but the sketches I've come up with are around 200 L (NOT 1400 L!!).
I’m a furniture designer by trade so both function and form are important. I am not set on the specific driver, and I do like the look of vented boxes, so if there are other options that are better suited to the Onken type configuration, I will gladly look into those!
Member
Joined 2009
Paid Member
You can likely get away with a smaller box than mine, you might be happy with 18” wide (mine is 24”), 40” tall (mine are 49”) and 18” deep (mine are 24’).
You're welcome!
Understood, sympathize. Doubt there's any single book regardless of one author calling his a 'Cookbook', though at a glance this on-line 'Bible' I wasn't aware of may suffice: The Speaker Building Bible -
Techtalk Speaker Building, Audio, Video Discussion Forum
Otherwise, need you to ask specific questions about what particular 'gibberish' you're referring to. 😉
OK, but now wondering which driver you have, Bigun's 'cast frame' with its super large ideal cab or the newer 'Ferrite' model, which 'only' requires 201.47 L for a max flat alignment tuned to 37 Hz based on published specs, though without measuring them, AN's 'one size fits all' cab may suffice as quite a few folks have been fine with them over the years with other AN drivers.
Seen some bodacious Art Deco floor models in my time, ditto Danish Modern............
Yes. Completely different. If the specs are on, it fits into a 135-150 litre miniOnken. Could be squeezed into a 100 litre one and surprisingly a 50 litre (this might be impractical), but 88 litres sealed (unstuffed, say 76 litre with damping) does better than the little one, and reaches just as low as the 100 litre mO, and has a nicer shape roll-off if the room has some gain.
If the FR is to be believed, it is mor elike 93dB and you need a substantial broadband notch to bring down th ebig bump above 800 Hz… at a guess just into the driver’s chaotic response zone.
dave