'LGT' Construction Diary

[ScottG]

I know from experience that my room has a bump between 30 and 40hz so this will likely be better. That's my reason for not porting - an 8" will almost certainly be tuned in that 30-40hz range. Most ported designs don't work well in my room, the B&C sub did because it was tuned to 19hz and didn't have to sit where the loudspeakers normally do.

That is just an spl-thing. If the thing is purposely tuned for a lower spl in this passband then it is equivalent (..or if filtered). Additionally the substantial physical separation will almost certainly alter this.

I like the mid with mid reinforcement - I've used it before to good effect. I've also done something similar with tweeters to reach a lower cut-off freq. while maintaining fairly low non-linear distortion.

Even an 8" max thing is OK (..like you an many others, I also like speakers that are slim). What I think will have to change is volume (particularly if sealed) which means either additional cabinets in the future or, like a porn star going for the money shot, "go deep".
Perhaps even including a compound configuration.

 

[Salas]

The sub driven alone, here. The PHL receives a 6db roll off under 200Hz and its in sealed box too.

Underside, with mean level line, the total response with PHL in. One division = 1dB.

I knew that from the start, because that room gain is my room gain. And I don't predict tarantism I just say things are brighter than initially looking and a dipole monster sub with dsp is only one evening woodwork for our thread host.

 

[ShinOBIWAN]

Originally posted by tinitus Nice pictures, Shin - something that nice will sound good, I am sure

Ah you see! You don't have to worry about whether they'll sound good, just enjoy the pretties.

If they don't sound good I'll come back, let you know and see if it can't be fixed.

Some time ago, we laughed at a designer who proudly claimed that he could hear 20hz from a small speaker, when sitting under a table....ups, sorry Salas I was not thinking of you

I've heard 20hz on headphones. Actually it might have been a mix of 40, 60 and 80hz. I just can't tell

Funny, sometimes when sitting at the computer, I hear very deep well defined tones coming from the other end of room opposite the speakers, but sitting in listening position, nothing of that

Move you listening position to the where the computer is

 

[ShinOBIWAN]

Shin, I think your measurement only show a ressonance below 20hz, nothing else and certainly not the woofer playing music

Not my measurement but taken from the Swedish sub tests done by Ilkka.

Its an FR sweep and shows what it shows. Model the XLS10 in an 18ltr sealed enclosure. Do you still think its a resonance giving wholesale output over what the model predicated? I have a hard time believing room gain is responsible given that you can say best case is +3dB/oct from 100hz down. So I'd like to see a CSD to support your resonance theory but at 16hz I don't think we'll be getting an accurate one anytime soon.

 

[ScottG]

The sub driven alone, here. The PHL receives a 6db roll off under 200Hz and its in sealed box too.

Underside, with mean level line, the total response with PHL in. One division = 1dB.

I knew that from the start, because that room gain is my room gain. And I don't predict tarantism I just say things are brighter than initially looking and a dipole monster sub with dsp is only one evening woodwork for our thread host.

Small room with a monsterous modal gain between 30-40 Hz? In fact it COULD be similar to Shin's - don't know.

What likely won't be the same is the driver compliance - remember he said flat to 130 Hz (which implies a reduction in spl below that). I'm guessing he spec.ed to Per more extension, as a result he has a lot less spl-level in the midbass then you have.

 

[ShinOBIWAN]

That is just an spl-thing. If the thing is purposely tuned for a lower spl in this passband then it is equivalent (..or if filtered). Additionally the substantial physical separation will almost certainly alter this.

You've still got worse time based behavior with the ported alignment, add this to a room node and you can see why ported might not be the best choice if that tuning frequency and room node coincide. If your trying to get around that then I can't see the point of an EBS alignment on an 8", especially one that's half decent regarding efficiency. It just doesn't have the displacement/SPL at those low frequencies to be worthwhile to begin with.

I like the mid with mid reinforcement - I've used it before to good effect. I've also done something similar with tweeters to reach a lower cut-off freq. while maintaining fairly low non-linear distortion.

I just wish there was a sanely priced wide range tweeter that you could cross from 800hz and up but that's just more audio mythology.

Even an 8" max thing is OK (..like you an many others, I also like speakers that are slim). What I think will have to change is volume (particularly if sealed) which means either additional cabinets in the future or, like a porn star going for the money shot, "go deep".
Perhaps even including a compound configuration.

Scott are you trying to bankrupt me? Compound configuration... 8 drivers...

I think there's also a point where you have to draw a line at the relative cabinet depth in relation to baffle width. Otherwise you end up not being able to stretch your legs out

 

[Salas]

Scott:

Why you keep saying that I compare my little 26 litre in total, cheap woofers box, to Shin's 2x35ltr super expensive custom stuff. Its another speaker and room. I say that it can do better with some luck. And I am not a supporter of low bass in small rooms. It sounds awful to me. As for what is being played by a woofer and what is a room mode, its not relevant to the final result quantitywise. Pressure is pressure and we hear it and feel it. If it was only what played by the woofers to be counting then room modes would never present a problem in acoustics. The measurements are honest and you saw I got +8dB what more?

 

[ShinOBIWAN]

The sub driven alone, here. The PHL receives a 6db roll off under 200Hz and its in sealed box too.

I had a similar result for the 9" ATC in a 50ltr enclosure.

In 4pi space I was getting a hump of nearly 15dB between 30 and 40hz. After EQ'ing flat I had a response with a virtually flat(no rolloff) response between 30 and 45hz. The 30hz figure was -6dB. Still fairly low output though despite initially looking impressive for what it was (sealed 50ltr, Q=0.5).

 

[tinitus]

I know I really ought to try a sub once, on the other hand I dont believe in it, and the added complications
I strongly believe that with the right drivers and setup a simple EQ on the main system will do the trick
Greg(SKA) makes such a slick cheap little thing, but Im not sure if its the right alignment, but it might work and cost next to nothing - a "Linkwitz Gainer" should be right

Shin, its a bit premature I know, but what you need is a passive LCR curcuit on bass drivers ressonanse measured in box - it might also have helped on Perceives instead of dealing with it digitally

BTW, have you ordered bass drivers - I dont understand Per recommending Qts=0.6....I found that about 0.4 would suit 35liter
calculations indicated that higher Qts would need much more than 35liter to work ...
but what do I know, Per is the pro ... even so, Qts=0.6 is close to what the dipole guys use
But its your show, and money ... I dare not take any responsibility

 

[ScottG]

You've still got worse time based behavior with the ported alignment, add this to a room node and you can see why ported might not be the best choice if that tuning frequency and room node coincide. If your trying to get around that then I can't see the point of an EBS alignment on an 8", especially one that's half decent regarding efficiency. It just doesn't have the displacement/SPL at those low frequencies to be worthwhile to begin with.



I just wish there was a sanely priced wide range tweeter that you could cross from 800hz and up but that's just more audio mythology.



Scott are you trying to bankrupt me? Compound configuration... 8 drivers...

I think there's also a point where you have to draw a line at the relative cabinet depth in relation to baffle width. Otherwise you end up not being able to stretch your legs out

I question worrying about time behaviour at lower freq.s ..at least from a non-aligned vented system - I've been down that "road". With an eff. driver - not a problem, however lower eff. drivers - yes. In other words even up to 50 milisec.s delay below most fundamentals added onto something the room generates doesn't mean much. Now the linear decay behaviour at vent resonance AND the room mode may cause problems, but because its a very narrow passband for this resonant/linear problem - it isn't something I'd expect (or have found) to be a problem with most recordings (..as long as the resonance was slightly below 40 Hz).

As for excursion.. such a vented system would actually display less excursion in many instances (depending on the recording). Hold that - less DRIVER excursion. In fact, that was the problem with the B&C sub. Vent diameter needs to be large for a given freq. to not run into excursion problems (which effects linear decay and power effects). My bad..

Long story short: IMO its a non-issue.

Don't get me wrong though, this is not an effort to change your mind on the subject.. rather this is just something to keep in the background and perhaps investigate on your own. (..definitely not a suggestion for this project.)

There is such a tweeter! (..otherwise know as "fullrange my @ss")

http://www.ejjordan.co.uk/

It just needs better eff. (especially for a 4 ohm driver), AND a better horizontal polar response. You can even use a first order high pass with it.

8 drivers.. well not AT - thats for sure!

Thats true about the depth thing.. - go too far with the speaker depth and its called a wall.

 

[ScottG]

Scott:

Why you keep saying that I compare my little 26 litre in total, cheap woofers box, to Shin's 2x35ltr super expensive custom stuff. Its another speaker and room. I say that it can do better with some luck.

Umm, because you made the comparison? (either express or implied.) And it did seem to be more than saying it can be done with luck.

 

[ShinOBIWAN]

I question worrying about time behaviour at lower freq.s ..at least from a non-aligned vented system - I've been down that "road". With an eff. driver - not a problem, however lower eff. drivers - yes. In other words even up to 50 milisec.s delay below most fundamentals added onto something the room generates doesn't mean much. Now the linear decay behaviour at vent resonance AND the room mode may cause problems, but because its a very narrow passband for this resonant/linear problem - it isn't something I'd expect (or have found) to be a problem with most recordings (..as long as the resonance was slightly below 40 Hz).

Its relative though and what I mean by that is the combined decay rate of tuning and room node in respect to that of neighbouring frequencies. Flattening the FR just isn't enough when your dealing with over 15dB of room related gain plus port resonance and the cumulative time smear. It isn't noticeable on all music but when there's something heavy such as electronic music then it sounds less than what it could do, not boomy(hate that word but it works) as such just blurred and less detailed. Sealed works better for my room overall though still not perfect of course.

You've got to go with what works best given the constraints right? And that includes leveling the speakers up with the room.

I can understand your enthusiasm for vented, you've obviously had some good success with them. I too have heard some that rate as amongst the best... but not in my room. The B&C was good though but the alignment it used isn't relevant to the 8's were talking about here.

As for excursion.. such a vented system would actually display less excursion in many instances (depending on the recording). Hold that - less DRIVER excursion. In fact, that was the problem with the B&C sub. Vent diameter needs to be large for a given freq. to not run into excursion problems (which effects linear decay and power effects). My bad..

Long story short: IMO its a non-issue.

I know excursion drops at tuning but lets face it, how much output can you get from an 8 below 30hz because that's where you'd have it tune to.

Other than that you'd have to be real gentle and go with a modest QB3 alignment to get good GD and phase plus still maintain the benefits your talking about.

Don't get me wrong though, this is not an effort to change your mind on the subject.. rather this is just something to keep in the background and perhaps investigate on your own. (..definitely not a suggestion for this project.)

Its definitely worth looking into but maybe not whilst I'm stuck with this room and even better would be to get over my phobia of wide baffles holding 15's. I think that would be when what your talking about really comes to life.

There is such a tweeter! (..otherwise know as "fullrange my @ss")

http://www.ejjordan.co.uk/

It just needs better eff. (especially for a 4 ohm driver), AND a better horizontal polar response. You can even use a first order high pass with it.

I should have said excluding fullrangers

 

[ScottG]

That driver really is a tweeter.. And its reputed to sound more than a little like a good ribbon driver. Still, any driver with a 2 inch diameter is going to have "issues" with polar radiation.

This might seem strange - but all else equal I too would prefer a sealed system over a vented one. Its just limitations with drivers/extension/volume that cause me to suggest it.

And "boomy" seems an entirely appropriate word. The thing is, what I was suggesting does not fit that descriptor (just the opposite in fact) - and I don't know why. Use an average or even slightly better than average driver eff., and then you often get that "boomy" sound that has serious "hang time" when coupled with a mode (..not unlike 10% of the cars in the US). I know.. weak. You'll just have to experience it yourself, or perhaps have it confirmed by others. Hmm, I'll be interested to hear your report on the sound of the RAAL in relation to other tweeters (then you might find yourself in a similar position). Then latter on you can re-define that when you use a high output impeadance amplifier on the ribbon!

 

[Salas]

Umm, because you made the comparison? (either express or implied.) And it did seem to be more than saying it can be done with luck.

My intention was to show that we can squeeze spl lower with a good plan, counting in room, help from other drivers, impedance and source impedance. Not necessarily expensive or big. It was a walk on the bright side. And Shin will short it out eventually.

P.S. That heavily modified Korato KVA-20 (Belgrade - Serbia, can be seen in the picture, right side of the speaker brown one, the golden one is my DIY pre) has a 0.5R
source impedance after global feedback (you asked earlier about source Z, I forgot to reply, sorry), but the measurements are done with an L.H. Hood mosfet 160W @ 4 Ohm solid state amp. Not even the crossover coil used in only sub measurement. The sound of the total system is very dynamic (with no boom - nada) with the EL-34 or 6L6GC and the Audionote transformers feeding the 94dB speakers. I wish Shin to have good karma with Serbian RAAL too today.

 

[wigginjs]

And Shin will short it out eventually.

Geez, I hope that was meant to be 'sort'. I really would hate to see Shin short out some of this nice equipment .

 

[ShinOBIWAN]

Hmm, I'll be interested to hear your report on the sound of the RAAL in relation to other tweeters (then you might find yourself in a similar position). Then latter on you can re-define that when you use a high output impeadance amplifier on the ribbon!

I wish Shin to have good karma with Serbian RAAL too today.

Afraid not fella's. The RAAL's arrived this morning and after getting them up and running I quickly realised all wasn't well with one of them.

Its buzzing a fair bit and looks like the ribbon isn't sat in the gap properly and its flexing around badly which suggests an issue with the tension also. If you keep the volume down then it behaves OK but still sounds a little off (maybe not quite as extended?) in comparison to the good one which shows non of what I'm talking about here.

I've had a quick listen but doing so at low SPL's because of the problem. I've done the usual 'board of MDF and open baffle slap together' to get a stereo setup working. I used just a single 5" AT with the RAAL mounted above. Very crude stuff and I haven't even bothered to flush mount anything just hastily cut holes for the drivers in the baffle. The two rather large panels of MDF that form the baffles aren't even the same size But close enough and I couldn't wait so what the hell.

I was listening to Kenny G's "You're beautiful" which I'd class as smooth jazz. There's a sax playing throughout the track and, before today, in places it always sounded a tad artificial, particularly on the powerful sustained high notes which tended towards shrill, thin and unnatural. Well the RAAL and AT nailed it and that's the first time I've heard it sound right. I kept thinking back to the Scan R2904 and the idea that its supposed to be one of the better ones, let me tell you, it isn't or at least not for this track.

I could go on about the sense of space in the sound - its massive. Or maybe the natural tone and overall smoothness that makes music seem that bit more involving. There's not much to be gleamed from this ongong experiment because its a world apart from the planned final setup. All I've done here is use a crude baffle, one faulty tweeter, level match and stick a 4thorder XO in at 1.5Khz, hardly optimised stuff.Still interesting, rewarding and entertaining though.

 

[ShinOBIWAN]

Geez, I hope that was meant to be 'sort'. I really would hate to see Shin short out some of this nice equipment .

Hey remember back when I sent a 20hz test tone, intended for the bass driver, through an R2904 and blew it up? That was an expensive active XO routing error

I never could get a reasonable quote for repair so have the thing on display as a reminder to be careful and not get complacent.

 

[Salas]

Geez, I hope that was meant to be 'sort'. I really would hate to see Shin short out some of this nice equipment .

Typo devil! Cheers mate! (sorry, meant thanks pal)

 

[Salas]

Afraid not fella's. The RAAL's arrived this morning and after getting them up and running I quickly realised all wasn't well with one of them.

Can it be a package handling issue? Alex is personally assembling them and its a bit odd that he sent em out without a test.

 

[ShinOBIWAN]

Alex gave me a ring and we fixed the RAAL over the phone

Very simple stuff and I was impressed with just how he dealt/planned for things like that.

At each end of the ribbon is a section of heatshrink about 5mm in length. I spotted these when I examined them on arrival and thought they they were most likely used to terminate the ribbon ends to stop vibrations traveling back along the ribbon. Well not only do help with that but they also serve to tension the ribbon after overload and shipping.

The box the drivers arrived in looks pretty battered and with the flight over here the combined effects of the pressure/temperature changes plus the ribbons being sat horizontally in a box thats been thrown around has led to the ribbon sagging slightly.

A very easy fix indeed, all we did was get a hot air gun heat the ribbon up for about 10 seconds, left to cool and then briefly heated again. After this the heatshrink did its job and the ribbon was very taught and back within spec.
This technique can be used to fix sag after overloading them, although I certainly don't plan on doing that!

Gone is the buzzing and all I have now is the most gobsmacking treble I've ever heard.

With both ribbons on top form things sound even better. I can finally wick the volume and I'm amazed at the dynamics and SPL these things have, the HF extension seems to extend forever. Even at 1.5Khz and big SPL's they're just clean, clean and oh so clean. Most natural sound I've heard full stop. Domes sound flat, boring and constipated in comparison, they're absolute shite for me now. Seriously, watch me, I will never and I mean never use a dome in any serious project again. I'm so serious about that I'd bet my life on it.

It seems like I've cheated somehow because for such a cruddy implementation the integration is already impressive. Both the AT and the RAAL are just one. I've no doubt its because both drivers are so flat in their operating ranges. The little 5" AT's also extend nicely on those ~80cm wide x ~130cm high MDF baffles.

Its not often you'll see me gush like this but I couldn't care less if I look a little over enthusiastic, these are ******* amazing. Might not be getting much done over the next week or so. I need to get my head around these somemore.