Finally built a prototype, dual FRS8 in pentagon-TL

[AEIOU]

Well, the knuckle test on my enclosures never reveals a low "thud." More like you were knocking on brick or thick tile. My enclosures are usually adequately braced, if not, I damp them with damping materials.

You will never have no resonances. The only thing you can do is try to make it so they are never excited. A LF thud that quickly dissapates is a LF, LowQ resonance and much more likely to be xcited and because of the low Q more audiable.

It has been 30+ years of of questing for the holey grail of no cabinet rsonance, and i'm pretty comfortable with how far i'v gotten.



Exactly. Low diffraction on the outside & a box that has resonances that never get excited is my practical target.



No issues here with plywood.

dave

 

[planet10]

if not, I damp them with damping materials.

In general adding mass without adding stiffness will lower the frequency of resonance and lower its Q, making it more likely to get excited and when it is more audible.

dave

 

[planet10]

Before we take this thrad furthr off-track,,, dig in: http://www.diyaudio.com/forums/cons...rials-build-speakers-out.html?highlight=build

dave

 

[ßart West-VL.]

I finally got around to start a first test on the ugly prototype.
Sounds descent but the FRS8 is known to need a bit of attenuation around 3-4khz iirc.
Since the TL is too long, the tuning is off and results in a bit of a 'boxy' sound.
→ otoh, the impedance-peak measurement shows 150hz! what could cause this?

It's only one pillar, two of them playing stereo would sound fuller, I'm sure.

What to measure next and what tweaks might be needed?

edit:just 'watched' the cone excursion and it moves less at 62Hz then above and below that frequency.
...Time for bed, tomorrow I'll pull out the stuffing...

 

[ßart West-VL.]

If the volt meter reads 2V is this RMS (sine, 50Hz)
On the scoop I see the sine take 4 divisions and a setting of 2V/div.
So this is 8V p/p
Isn't this 2,828V RMS?
What is wrong?

 

[ßart West-VL.]

Did another impedance-peak check and it's at 127Hz now, big difference from 150Hz!
I think it's from the driver break in, played a few hours at about 20Watt.

The visible lowest movement of the cone is now at 60Hz, 2Hz down.

 

[ßart West-VL.]

Thoughts anyone? I really could use some guidance here, this is a first for me to build a TL and actually take measurements.
I'm planning to buy some stuff too, but not before I know what I need and for that I'll have to know what I need to measure.

Later on I'll start measuring a 10" driver for a sub, JBL, unknown type.
First this pentagon!

 

[ßart West-VL.]

I'm a bit amazed that no one reply's to this work in progress.
Help would be much appreciated, otherwise I don't see any point in posting further results as the interest seems to be minimal.

 

[ßart West-VL.]

Hi all,

Did some listening sessions with the fugly prototype and the equalizer seems to be staying at following settings for most of the time:
-2.5dB @ 2kHz
-5.1dB @ 4kHz
-2.5dB @ 8kHz
that's it...
The lower section is not important in such a small driver, but is surprisingly audible!

Now I need to calculate some form of a basic filter for the area around 4kHz.
The problem is that if there is a passive filter, the components of the driver have an influence too and iteract with the new components.
Should I just do a basic calculation and not be worried about driver specs?
Anyone who has a basic method for this?
Ultimately I want a TL that doesn't need any tone-control.

I'm still puzzled to why there is so little interest in this thread, why is that?
Anyway, I hope there are some answers to previous post too.
Thanks for watching.

 

[Bigun]

Now I need to calculate some form of a basic filter for the area around 4kHz.
The problem is that if there is a passive filter, the components of the driver have an influence too and iteract with the new components.
Should I just do a basic calculation and not be worried about driver specs?
Anyone who has a basic method for this?

As I understand it, the T/S parameters of the driver are not as 'fixed' as we might think but vary under load due to heating effects for one thing. Also, depending on your amplifier (e.g. damping factor, gnf or not) there will be an interaction between the driver and the amplifier. What all this means I think, is that you shouldn't try to over design the filter, just build a basic passive version and try it out. It might be even better if you could place it before the amplifier so that the filter only has to work at low current.

 

[ßart West-VL.]

Thanks for your reply, Bigun.
What should I be looking for if I wanted to filter pre-amp?
I'm sure the values would be very different then post-amp, right?

 

[Bigun]

If I understand correctly, you are looking for a notch filter for 4kHz and with little impact below 2kHz or above 8kHz. Such a filter might not have a flat phase response.

I believe you would want a parallel RCL circuit in series with your speaker to make a band-stop filter.

Yes, values will be very different pre vs post amp because the filter after the amp is being fed from low impedance and is looking into a nominal 8 Ohm load. Preamp the impedances are all higher (which is why the components can be smaller valued and cheaper for equivalent quality).

You need more expertise than I can provide - you could start by searching the web for speaker notch filters.

 

[ßart West-VL.]

You have already been very helpful, thank you!
I'll have to look in to that phase issue. My receiver has no equalizer, only bass/treble control. So maybe I'll have no choice but to take the phase problem together with the rest, or listen to screamish boxes...
Does the phase issue exist for pre-amp passive filters too?
Oh, come to think of it, that's not an option, I use my receiver's amp, so no pre-amp correction here.