Onken, anyone?

[zogu]

Even running known good Onken boxes and drivers, the results of the calculator do not equal the real boxes. At least, I have not been able to make them match.

I tested calculator with minionken and I got exactly the same results, n=6,5. Here is T/S for FOCAL 10C01: Fs=26, Qes=0,22, Qms=2,05, Qts=0,20, Vas=288,0 Re=6,1, Sd=343cm2, Xmax=3,5, Pe=60, anyone can test calculator.
And I agree, i got same results with winisd, even port length is same. But onken is not bassreflex? Calculation is probably same, but because of big S of vents and specific design it has a lot better sound than classical designed box.
Does anyone has some formula for calculating box dimension, i want to respect onken approach as much as possible, and I think it is not enough to use L' as only data for designing box. As I sad in previous post, in minionken L is width of speaker, but if I design speaker with this approach, I will get box with height of about 90cm and width~35cm, it does not look like onken at all. GM explained few methods for getting box dimension, but I don't understand this. Maybe is not important? If someone can help, please post.
Thanks, Jovan

 

[sigurd]

Onken

Hello zoqu,
To calculate boxes it is not sometimes that simple.
Maybe look at Google: ' The Loudspeaker Design Cookbook ' by Vance Dickason. It's an interresting book and it is translated in a few languages.
Succes.

 

[sigurd]

onken

Vas of the speaker is important to calculate box dimensions.
Qts of the speaker is best around 0,3 and 0,4 for best sounding bass.

 

[GM]

But onken is not bassreflex? Calculation is probably same, but because of big S of vents and specific design it has a lot better sound than classical designed box.

Greets!

I'm not familiar with this particular 'mini-Onken', is the design posted on-line?

Onkens define bass reflex as it was meant to be! The original 'Onken' is one of Thuras's test boxes in his bass reflex patent, with the other being a single large ducted port: http://www.google.com/patents?vid=USPAT1869178&id=MS1bAAAAEBAJ&dq=1869178

The only difference between the two is that the Onken's vent system adds more resistance through surface friction which increases damping Vs a single large vent and due to its need for longer vents for a given alignment, it has more vent harmonic distortion! IMO this is why Onkens are so beloved since this ~euphonic distortion adds a 'richness'/'fullness' higher up in our acute hearing BW similar to what a tube amp does. If you want accuracy first and foremost though, you'll need to heavily damp the vents, which will roll off output around/at Fb somewhat.

Bottom line, you don't need a ~accurate Onken calculator if you have one of today's accurate vented box designers that allows multiple vent selection, just select a 4th order Butterworth (B4) alignment to get an accurate 'Onken' n = 6.34 and use the number/size of vents that ~ = Sd and are < ~35 cm (13.78") long to keep vent harmonic distortion acceptable. If the vents are too long with an area = Sd, then reduce the vent size and/or number as required, though some folks who have experimented say that below ~75% Sd/six vents, you begin losing some of the Onken's 'sound', so this effectively limits driver selection.

If the goal is to have an 'Onken' that Thuras intended, then measured impedance peaks should be of equal amplitude, ergo Vb, Fb will be ~ = Vas, Fs and if it's under-damped, then it will have to be adjusted with some form of series resistance to reach a B4 alignment. For example, when I load published 414-8C specs into the calculator and adjust 'Rg' to get a ~Vb = Vas, Fb = Fs and plug its specs into LspCad I get impedance peaks that are only a few % off, so the calculator is accurate enough for me and further proof that it's not needed to design one if you already have a decent box designer.

WRT the basic cab shape, it ideally needs to be the 'golden' or an acceptable acoustic ratio with the driver ideally centered within this duct system and between 40-54 cm (~15.75 - 21.25") off the floor, so most Onkens I've seen aren't all that optimum, though the sonic benefits of optimum driver location WRT the HF/listening ear height is by far the more important consideration, so this is one requirement that should only be adhered to when it's appropriate.

GM

 

[zogu]

WRT the basic cab shape, it ideally needs to be the 'golden' or an acceptable acoustic ratio with the driver ideally centered within this duct system and between 40-54 cm (~15.75 - 21.25") off the floor, so most Onkens I've seen aren't all that optimum, though the sonic benefits of optimum driver location WRT the HF/listening ear height is by far the more important consideration, so this is one requirement that should only be adhered to when it's appropriate.
GM

Thank you very much for all your posts, now i have enough information to start with project. It will take a while until I finish it, I will post results and some pictures here.
About minionken, it is published in l'audiophille, I don't have full text, I have only pictures. Box is designed for focal 10c01, but it is suitable for two more drivers, one is lowther pm6c and other i forget. If I get complete text from friend which has it i will post a scan. Noe I have only box dimensions and design, I attached picture.

Thanks for all your help
Regards,
Jovan

 

[zogu]

Better picture of minionken.
Regards

 

[Pano]

Thanks Greg, that clears up a lot of questions. Might motivate me to go back and look at Onken boxes again, now that the mist has lifted. =)

The mini Onken was a great sounding box for its size. I listened to it often over at the Maison de l'Audiophile when I used to hang out there. Might have been the first thing I heard there. It was the same one that was in the published article.

That Focal 10" driver is long gone, alas. IIRC, the system used a Fostex for the mid and a bullet tweeter of some sort. Maybe a Technics?

That was going to be my "next system" but never did get it built.

One of the people behind the design was Gérard Chrétien who wrote and edited for the Audiophile review - as well as the Revue du Son. Gérard is now president of Focal/JM Labs. Nice guy, very knowledgeable.

 

[GM]

Thank you very much for all your posts, now i have enough information to start with project.

About minionken, it is published in l'audiophille........

Greets!

You're welcome! Looking forward to seeing/hearing about the results!

Thanks, don't recall seeing this one, but now I know what drawing Dave began with for his mini-Onken.

GM

 

[GM]

Thanks Greg...........

One of the people.........

Greets!

You're welcome!

I'll bet it does, Dave's has been getting rave reviews with pretty mundane drivers, so a properly done Lowther would be quite the mini-monitor. I love it when DIYers do well in the industry.

GM

 

[planet10]

but now I know what drawing Dave began with for his mini-Onken.

I don't know that i have seen that one before. My miniOnken was inspired by the Solo 103, and aan Onken bass cabinet published in a late 70s Audax publication. The angled bits at the front are all mine,

The 1st one out the gate (below) was a cardboard mockup (BTW, the insides follow the golden ratio).

dave

 

[galac]

i did some calc's w/ the onken spreadsheet to unterstand the "mechanics" but there's one thing (due to my english skills^^) i didn't understand.

What ist the real length of the vent when you build the enclosure in practice - L vent or L' vent?

And, does anybody know the exact SD of an GPA or Iconic 604/704?

Greetings from germany

galac

 

[GM]

I don't know that i have seen that one before.

Greets!

Sorry about that! At a glance it seemed like a drawing to copy/paste and adapt to your design to save some time.

GM

 

[GM]

Greets!

L is the acoustic length and L' is its physical length (shorter of the two).

The published volume displacement (Vd) is 19.2"^3 and the linear excursion (Xmax) is 0.15", so 19.2/0.15 = 128"^2 (825.8048 cm^2.

GM

 

[galac]

@ gm

thx for your fast answer!

 

[CLS]

I'm not sure if this is the right thread to add...

Hello there,

As I can see (via the "ONKEN_CAL.xls" ), this won't be a proper Onken, but....

I'm trying to help a friend designing a bass cabinet for his TAD1601a.

The parameters are:

Fs 28.000 Hz
Re 6.600 ohms
Qms 6.800
Qes 0.320
Qts 0.306
Mms/d 117.000 gr
Sd 8.810E-02 m^2
Vas 300 litter


I put these numbers (with Rg=3 Ohm) in the "ONKEN_CAL.xls", I got:

F-3 26.921 Hz
Fb 25.067 Hz
S vent 328.240 cm^2
L vent 44.126 cm
L¡¦ vent 34.007 cm

Vb 324.691 litres
Vb Total 335.853 litres

So, as you can see, when keeping the L' within the 35cm limit, the vent area is just too small to be an Onken


I also used <Box Plot> to verify the tuning and got this:

An externally hosted image should be here but it was not working when we last tested it.

Ok, the conditions in <Box Plot> are not exactly the same as the above (like Rg), but the Vb and vents are pretty close.


Well, looking good, so I still drew a box with slot type vents on its sides and utilizing the corner:

An externally hosted image should be here but it was not working when we last tested it.

Questions:

1. Is it good to push the vents against the side/back wall? (looks like a big mBVR?)

2. How will this fake Onken perform? I'm not sure about the corner loading.

 

[CLS]

Sigh ~~

Some other Onkens got finished one by one, and I'm still stuck here.

Meantime, I also found some interesting numbers by that <ONKEN_CAL.xls>

Just for fun, I put in the numbers of Eminence Legend125 (a guitar driver).

Fs 94.000 Hz
Re 7.440 ohms
Qms 6.150
Qes 1.180
Qts 0.990
Mms/d 32.000 gr
Sd 5.067E-02 m^2
Rg 0.500 ohms
Vas 32.210 litres
n = 5.700 (best 5.7)

F-3 37.667 Hz
Fb 35.072 Hz
S vent 400.000 cm^2

L vent 44.461 cm
L¡¦ vent 33.290 cm

Vb 200.597 litres
Vb Total 213.913 litres

What a surprise that a driver with 94Hz fs can play down to mid 30's! (with big volume and big ports!)

I made up these numbers in Box Plot and get this:

Yikes!

Bring up the tuning, and it looks much better:

But this does not follow "onken rule" any more.

By this and some other cases I found that higher Q can get "better" results in <ONKEN_CAL.xls>, but would probably get resonance peak

Argh ~~~

 

[zogu]

Hi,
I never used boxplot software, but when i simulated some other drivers in onken_calc and later check them in winisd (software which i use) I always got almost ideal rusults, very flat curve with slow down, what is main characteristic of onken speakers. Can you check your designs with some other software?
Regards, Zogu.

 

[CLS]

Thanks.

I tried winsd just now. For this case (Legend125), the result looks very similar. The contours of SPL plots are almost identical.

 

[zogu]

Now I looked drivers parameters, this driver is not suitable for onken, with fs=94hz and vas=32l. Onken is bass speaker, you can't expect bass from spekaer with Fs=94hz.
For onken designs choose drivers with very low Fs (below 40hz) and big Vas (>100l). You can find much better 12" PE drivers, this eminence driver is very cheap, but it is not suitable for home audio.
My friend built onken speakers with audionirvana 12" drivers, i didn't have time for serious listening, but first impression was very nice, deep bass, excellent middle and highs. Box is 550l, so you need a lot of space in room for such a big enclosure. Driver price is about 200$ for pair.

 

[CLS]

Thanks for reminder.

Yes I know that guitar driver is not for bass or home audio, so I already mentioned "just for fun" at the very begainning

My real "problem" is around the TAD-Onken, which seems also not a good combination.

Just wondering all these numbers and struggling to find a good balance. So I threw in such extreme example to see what happens. And the results are indeed very "interesting"

I once built a pair of big cabs right at the living room's coners. Loud and deep alright, but also the strongest mode I've ever experinced. So I kind of hesitate about the ports against walls....


ps. 550L cab for a 12incher! WOW