Genelec 8351B Teardown (2nd Disassembled)

[Pearljam5000]

Looks like they use the same amplifier and charge substantially more money!

I wonder what amp they're using in the 8361

 

[Descartes]

I wonder what amp they're using in the 8361

May the same!

 

[MAB]

The general consensus is beryllium is a better material. I'm sure you could google all the pros and cons.

Is there really a consensus? Google is about the worst way to investigate pros and cons of something like this, Google struggles to generate answers to simple technical problems, just a mish mash of opinions raked by popularity and monetizability. Google can't even align a woofer in a sealed box, it's not designed to give the answer you suggest.

I'm not sure if that's true, first breakup for aluminium or titanium is around 10khz. Don't forget the speaker is active with correction. Where beryllium it's around 16khz.

Can you show the data, including the magnitude of the breakup? Yes, all other things being equal, the high specific stiffness/weight of Be pushes the dome's resonances and artifacts higher.

Specific modulus - Wikipedia

en.wikipedia.org

But there are consequences to this stiffness.
High stiffness to weight has other tradeoffs, the dotted line on the graph below is the 2908's of-axis response superimposed on the 2904's response. The Be dome in the 2908 beams more than the Textile dome.

Also, Be gets a large breakup mode at 30kHz, that will have odd order harmonics at 1/3 = 10kHz, 1/5 = 6kHz, etc... While I have never measured these drivers, past experience measuring such breakup modes is they will have real harmonic distortion consequences.

I don't understand your comment about not forgetting that the speaker is active with DSP, none of that will deal with a resonance or breakup mode. You could use a series notch filter to increase the impedance at the resonance to control the breakup, or spray the Be dome with aerogel!

All of this is a bit silly, since the best sounding speakers in the world use all sorts of driver materials (mostly not Be) And to this day nobody has shown me how Be solves the problem of integrating a tweeter better than other materials.

Do you think you could use some heat and remove the tweeter from the glue?

Test the tweeter independently.

Find a matching tweeter, something that would work as a good replacement.

3d print a small part if needed.

And put a new tweeter in?

I want beryllium

Uh, are you asking OP to destroy one of their speakers due to your fantasy about Beryllium? I hope you understand how naïve this is.

 

[Pearljam5000]

Interesting comparison of titanium , beryllium and aluminum
Skip to 2:40

 

[voodooless]

Interesting comparison of titanium , beryllium and aluminum
Skip to 2:40

Let’s do the same test with cloth, guess what you’ll want to have after seeing that

 

[dfuller]

I wonder what amp they're using in the 8361

Similar discrete class D, just with more power.

 

[Impossible]

If they used titanium instead of aluminum
That would also be a an upgrade ,no?

Commonly for dynamic range improvement over aluminium and also some times because people preffer the tonality over aluminium.

 

[Impossible]

None of that implies there would be any audible improvement.. There really isn't any evidence that the 8351/8361 tweeter has any kind of significantly audible issues.

I personally think Be tweeters are more about bling and "uniqueness" than the material properties. They don't seem to make much difference in measurements usually and having heard Be Revels and non-Be I didn't think there was a huge treble difference. I mean does this look like a speaker that has significantly better treble than this? Sure doesn't to me.

Personally if I was wanting to upgrade the Ones I'd be looking to improve the bass, it could use some form of cardioid or other directivity improvement without the huge/expensive W371A. But that would necessitate redesigning the entire cabinet so it would end up an entirely new product in terms of development and manufacturing.

I think Amirm has his revels not just because they play louder then the 8361a but also because of the tonality that beryllium brings.

Even in the m2 vs revel salon ultimate blind test the revel came out.

Maybe (I'm no expert) is it you don't see the measured difference because of smoothing especially clouding anything over 10khz where the difference is mostly expected. I would really like to know. Just a guess.

DrBA Test Bench: Eighteen Sound ND4015Be | Sphericalhorns

See picture, same. Driver only difference is the diaphragm material. Blue is beryllium.

 

[test1223]

Has anybody an explanation why they use a relatively low amount of damping material for the midrange driver? From my experience you get an audible cleaner midrange if you use more / denser material. Maybe they haven't experimented with it since short time reflections can't be displayed easily in measurements? Or Genelec knows more about it like a potential cause of none linear distortions? Less heat transfer which becomes an issue at very high volume?

I think a modification of the 8135B would improve the sound at least at not very high volume.

It relates to the question I did asked a while ago here.

Does a lot of damping material have any negative effects in a closed box?

Hello, from my experience a lot of midrange and bass enclosures sound much better with a lot of damping material in it. I often use very dense material for the walls and stuff a lot of usual material in the box so that it is very compressed with the exception directly behind the driver. The...

audiosciencereview.com

 

[holdingpants01]

Has anybody an explanation why they use a relatively low amount of damping material for the midrange driver?

What do you mean? The mid + HF combo is completely filled with damping material and closed from the back, the photo is in the original post.
I really like this thread, people who never heard them trying to fix imaginary problems, suggesting that genelec engineers don't know how to test something. I wait for someone to demand "upgrade" of resistors for "audiophile grade" ones. No wonder Genelec doesn't really aim at hifi market other than a few small models

 

[test1223]

What do you mean? The mid + HF combo is completely filled with damping material and closed from the back, the photo is in the original post.
I really like this thread, people who never heard them trying to fix imaginary problems, suggesting that genelec engineers don't know how to test something. I just wait for someone to want to change the resistors for "audiophile grade" ones. No wonder Genelec doesn't really aim at hifi market other than a few small models

I have built some loudspeaker and did some experiments .

I am sure that the Genelec engineers are good engineers. But you have to optimize the whole speaker for a production. E.g. It is much harder to stuff a cabinet consistently with material. So the midrange might improve but in the end you get more deviations from speaker to speaker. So there is much more to it.

The amount of damping material is relatively low and you need very dense material or stuffed material to effect any waves below about 1kHz significantly.

 

[holdingpants01]

I have built some loudspeaker and did some experiments .

I am sure that the Genelec engineers are good engineers. But you have to optimize the whole speaker for a production. E.g. It is much harder to stuff a cabinet consistently with material. So the midrange might improve but in the end you get more deviations from speaker to speaker. So there is much more to it.

The amount of damping material is relatively low and you need very dense material or stuffed material to effect any waves below about 1kHz significantly.

As you can (or maybe can't) see the midrange+tweeter enclosure is filled with material to the edge. I used them for the last 1,5 years for my work, about 8 hours every day of critical listening, this gives me more or less 3000 hours of experience, so I'm pretty familiar with them and I would say there's zero problems with their midrange or high frequencies, they're as good and neutral as it gets, bar none

 

[dj94030]

yes but i just learned about genelec and i must fix it !

 

[dfuller]

I have built some loudspeaker and did some experiments .

I am sure that the Genelec engineers are good engineers. But you have to optimize the whole speaker for a production. E.g. It is much harder to stuff a cabinet consistently with material. So the midrange might improve but in the end you get more deviations from speaker to speaker. So there is much more to it.

The amount of damping material is relatively low and you need very dense material or stuffed material to effect any waves below about 1kHz significantly.

The coax is a sealed unit, it doesn't care about the cabinet's stuffing. If it weren't they'd have to have a separate sub-chamber.

 

[MAB]

Has anybody an explanation why they use a relatively low amount of damping material for the midrange driver?

You keep asserting this. Look at the picture, the mid enclosure is the small cylinder and it is 100% filled. It's not 'packed', but that is not typical.

Hard to get more damping in the mid enclosure without packing it tight.

Damping material isn't a game changer, so trying to stuff more damping material into the enclosure isn't of much utility and risks contacting the cone.

 

[test1223]

The coax is a sealed unit, it doesn't care about the cabinet's stuffing. If it weren't they'd have to have a separate sub-chamber.

The waves from inside the midrange cabinet will be transmitted outside through the cone of the midrange driver, which is very thin compared to the cabinet.

You keep asserting this. Look at the picture, the mid enclosure is the small cylinder and it is 100% filled. It's not 'packed', but that is not typical.

Hard to get more damping in the mid enclosure without packing it tight.

Damping material isn't a game changer, so trying to stuff more damping material into the enclosure isn't of much utility and risks contacting the cone.

Yes you have to be sure that the material doesn't get contact with the cone. Such polyester wool has almost no effect in the lower midrange. You either have to stuff it or use much denser material. It is effective in the higher mids though.

 

[Sancus]

I think Amirm has his revels not just because they play louder then the 8361a but also because of the tonality that beryllium brings.

There are a lot of differences between these speakers other than the tweeter, like SO MANY differences. The Salon 2s are not coaxial and they have much wider dispersion for starters. Also, one person's sighted comparison in non-ideal conditions doesn't even consistently translate to any other person in the first place.

Not sure how you can just randomly pick "tweeter material" out of those things.

Maybe (I'm no expert) is it you don't see the measured difference because of smoothing especially clouding anything over 10khz where the difference is mostly expected. I would really like to know. Just a guess.

Your claimed differences are all over 10khz too. Frequencies over 10khz are generally down by -20 to -30dB in music relative to the SPL at around 1khz. And there isn't that much content there either. Is it audible? Yeah, of course, if you *remove it from the audio completely* you can tell a difference.

But hearing extremely narrow artifacts and slightly higher distortion? And not only hearing it, but deciding that tonality change is the biggest factor out of all the other major differences?

Nah, no way.

 

[Pearljam5000]

There's no 8361 teardown anywhere

 

[holdingpants01]

There's no 8361 teardown anywhere

mid+tweeter combo is the same as 8351B, the rest is just bigger and more powerful

 

[YSC]

The waves from inside the midrange cabinet will be transmitted outside through the cone of the midrange driver, which is very thin compared to the cabinet.


Yes you have to be sure that the material doesn't get contact with the cone. Such polyester wool has almost no effect in the lower midrange. You either have to stuff it or use much denser material. It is effective in the higher mids though.

So what the…. Are you trying to “improve” suggesting other stuffings should be used? When distortion or linearity isn’t a problem at the midrange anyway?????