First audio project, driver selection

[BrechtVD]

Hi all,

A few months ago I also post something about my first audio project, then I discoverd that it is harder then I first thought it would be...

So now I want to start with a simple PC fullrange speaker (start small, end big ). Because it would be my first project I want to get the parts in local stores. If this project turn out well I would also use internetshops etc. but just for trying I want to stay with that local store.

They have a copple of visaton drivers and not so good X-over parts, but I think it's fine to start with .

I would take a driver around 3 inch, the drivers I then have left are:
-FR 8 - (4 and 8 Ohm)
*http://www.visaton.com/en/chassis_zu...and/fr8_4.html
-FRS 8- (4 and 8 Ohm)
*http://www.visaton.com/en/chassis_zu...nd/frs8_4.html
*http://www.visaton.com/en/chassis_zu...nd/frs8_8.html

Because the FR8 drivers do strange things at high frequency I would take the FRS 8. But there is also a different between the 4 and 8 ohm, and that is the point where I don't know what to choose ... (the low frequency is the same but at high frequency they have both a few peaks )

So wich speakers would you chose if you had the choise of these drivers??

Thank you in advance!

Brecht V.D.

[5th element]

With full range drivers you are almost always going to get some frequency response issues at the very high frequencies. You are right in saying that the FR8 isn't so nice, that peak/dip thing at 3kHz is a big no. The FRS8 is a different matter and both the 4 and 8 ohm versions have a nice flat frequency response, either would make for a decent small loudspeaker.

The main difference between the two though is their T/S parameters. The 8 ohm version has a Qts of 1.04 and the 4 ohm version 0.69. Both are intended to be used without a cabinet - that is in an open backed application - such as in a TV. Having a high Qts makes using these drivers very difficult when you want to put them in a box, a sealed box is about the only option you've got and you most certainly will want to use the 4 ohm version.

When you place a driver in a sealed box you effectively increase the overall Q factor. The usual aim is to end up with a system Q of between 0.5 and 0.7. If you're starting with a driver that has a Q of 1.04 you're already way about this. The driver with 0.69 isn't ideal, but if you place it in a reasonably large box, relative to it's VAS parameter, you will end up with an end system Q of around 0.8, which is perfectly acceptable.

I don't know what other drivers you have available, but out of the above choices, the 4 ohm version of the FRS8 would be the way to go.

[BrechtVD]

Thanks for your reply!

But you say you don't want to have the total Q of your system above 0.7 (and ... 0.5) but what are the effects if I do??
Just so hard for a beginner to understand the T/S parameters. If you look at the frequency respons it's a lot easier to say what would be the best driver...

[5th element]

Usually the frequency response of a driver is measured with the driver in free air, but mounted on a very large baffle. This shows you how a driver behaves without any box.

Now the only area that the box size, or type of box (ported, sealed, etc) affects is the bass. Usually one ignores the manufactures frequency response of the bass of the driver and uses the drivers T/S parameters to simulate how it will perform/behave, in the box type, or size, that you have in mind for your application.

If you google about sealed boxes I am sure you will find some websites that will detail the differences between the various box Q alignments. As always with these things it's about a trade off between what it is you want to achieve.

[JonMarsh]

There is the driver Q, and there is the combined Q with the driver and an enclosure. Many programs are available to model the combination. In general, If your starting Q is fairly high, you need to have a fairly large box to avoid increasing it much more.

To clarify a little further, a Q of 0.707 gives a maximally flat alignment for Fb (resonant frequency in the box), response is -3dB at Fb, and extension before the roll off at 12 dB/octave. There is a small amount of overhang in the transient response.

A Q of 0.8 would give a small bump in the response at the Fb, and a bit more hangover in the transient response (it doesn't start and stop quite as fast as the signal).

A Q of 0.5 is critically damped, no hangover, but the response is down 6 dB at Fb, and due to the larger box size to help get that Q, Fb is likely to be at a lower frequency. Often one can use room placement to tune the boundary loading so that boost in the response compensates for the box roll off to a degree.

[BrechtVD]

I already did some search on the internet like 5th element said (just needed a push ) and I also find the things you said, but now I realize it's not a easy speaker -.-

But thanks for the help and al the information!!

[Face]

I hope you're measuring as Visaton's drivers sometimes don't match up to their spec sheets very well.

[BrechtVD]

No i would not measure the drivers, it's just my first project for simple PC-speakers so...
But I think I'll use other drivers (and buy them on the net, not that local store) , for example the fountek fe85. Looks me a nice little drivers for not so much money, nice to start with

[BrechtVD]

I already did some simulations with unibox (with the foutnek fe85), but bouth (vented and sealed) boxes looks nice, no weird volumes or ports... and no bumps...
Here are some pictures:
Specs Fountek FE85


Sealed box



Vented box



Anyone comment or someone who sees a big mistake of mine
Thanks in advance

[5th element]

Nothing wrong from where I am sitting, those look okay. The Fountek, with it's lower Qts, will work reasonably well in both a sealed and ported box.

[BrechtVD]

Thanks again
I did some reading on the forum and the things you guys make are really insane!! And I'm asking question about a simple fullrange speaker .., so thank you verry much that you guys take the time to answer such simple questions!! (But you have to start somewhere )

Just one more question, if I now have to know something about for example T/S-par. i'm looking on the internet but sometimes you find contradictory or other things about the same subject...
And now the question :
Is there a good book for the basics of speaker(building)?? Would be nice

Did some "research", The loudspeaker design cookbook (good to start with??)

[5th element]

The loudspeaker design cookbook is decent, but it's one of those books that makes far more sense reading, once you know what it's talking about. It's okay, but it's not amazing, imo, at teaching you the basics. What it will do however is introduce you to lots of useful topics that you will need to know about, but you can perhaps research more about elsewhere. I suppose this is subjective, some might find it excellent, but I didn't find it great with the way I learn and usually design things.

[JonMarsh]

I already did some simulations with unibox (with the foutnek fe85), but bouth (vented and sealed) boxes looks nice, no weird volumes or ports... and no bumps...
Here are some pictures:
Specs Fountek FE85
[ATTACH=CONFIG]23657[/ATTACH]

Sealed box
[ATTACH=CONFIG]23662[/ATTACH]
[ATTACH=CONFIG]23658[/ATTACH]

Vented box
[ATTACH=CONFIG]23660[/ATTACH]
[ATTACH=CONFIG]23661[/ATTACH]

Anyone comment or someone who sees a big mistake of mine
Thanks in advance

Note, with this input power, you are exceeding the Xmax versus frequency for SPL- the red line. this highlights the performance limitations of small full range drivers. The port helps, but the driver excursion vs frequency needs to be under the red line.

[BrechtVD]

Thanks for this helpfull comment!!
Again something I learned