Two cylinder sub-woofers in SVS style. - AVS Forum
Forum Jump: 
 1Likes
Reply
 
Thread Tools
post #1 of 158 Old 04-28-2014, 09:30 AM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Hello everybody,

I'm considering building two sub-woofers in a cylinder shape like the SVS PC13-Ultra and more or less of the same size. The problem is that I don't have any experience as so far I only built a pair of speakers but those were based on an existing project already and anyway it's a completely different story altogether to build a sub-woofer from what I've read so far.

Since I'm somehow limited in terms of space I'm considering a 12" to 15" drive at most (I'm modelling a Scan-Speak 30w-4558t00 right now). I have however some latitude in building a taller sub-woofer, although for aesthetic reasons I'd prefer to keep them under 50" hight or so. Actually I'm also thinking about taking advantage of these subs as if they were bass traps since I will place them in two corners (no options here). Maybe I'm being too optimistic here...

So, I've started some simulations with winISD and the problems began. I'm more inclined for a vented box as these subs are only going to be used for AV and not for stereo and I give most importance to deep frequency response above all. In fact I would like to go as low as 15Hz in my room with about 110dB SPL but here I'm sure if I'm being realistic or not so this is for now just a rough estimate.

I need help in understanding some basic issues. For instance there are 5 different alignments possible such as QB3, BB4/SBB4, C4/SC4, EBS3 and ESB6. for me this is Chinese...
I chose BB4 (as it was the default) and then I get a volume of 93.10l and a Tuning frequency of 17Hz. I also entered 350W.

Questions:

Are these optimum values? From my rough calculations I can probably build a box up to 200l or a little more. What do I lose (or gain) if I increase this 'optimum' volume? And can I play with the frequency as well?

And finally, I noticed on the graph for 'Cone excursion' that bellow 44Hz the value increases above the Xlim and even worse eventually will increase above XMax destroying my driver!!! eek.gif

How do you guys solve this excursion problem? By decreasing the amount of power? But then so decreases the SPL...

Sorry if I'm making a lot of stupid questions but like I wrote I'm newbie in what subs concern.

Thanks!
cochinada is offline  
Sponsored Links
Advertisement
 
post #2 of 158 Old 04-28-2014, 12:48 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Quote:
Originally Posted by cochinada View Post

Hello everybody,

I'm considering building two sub-woofers in a cylinder shape like the SVS PC13-Ultra and more or less of the same size. The problem is that I don't have any experience as so far I only built a pair of speakers but those were based on an existing project already and anyway it's a completely different story altogether to build a sub-woofer from what I've read so far.

Since I'm somehow limited in terms of space I'm considering a 12" to 15" drive at most (I'm modelling a Scan-Speak 30w-4558t00 right now). I have however some latitude in building a taller sub-woofer, although for aesthetic reasons I'd prefer to keep them under 50" hight or so. Actually I'm also thinking about taking advantage of these subs as if they were bass traps since I will place them in two corners (no options here). Maybe I'm being too optimistic here...

So, I've started some simulations with winISD and the problems began. I'm more inclined for a vented box as these subs are only going to be used for AV and not for stereo and I give most importance to deep frequency response above all. In fact I would like to go as low as 15Hz in my room with about 110dB SPL but here I'm sure if I'm being realistic or not so this is for now just a rough estimate.

I need help in understanding some basic issues. For instance there are 5 different alignments possible such as QB3, BB4/SBB4, C4/SC4, EBS3 and ESB6. for me this is Chinese...
I chose BB4 (as it was the default) and then I get a volume of 93.10l and a Tuning frequency of 17Hz. I also entered 350W.

Questions:

Are these optimum values? From my rough calculations I can probably build a box up to 200l or a little more. What do I lose (or gain) if I increase this 'optimum' volume? And can I play with the frequency as well?

And finally, I noticed on the graph for 'Cone excursion' that bellow 44Hz the value increases above the Xlim and even worse eventually will increase above XMax destroying my driver!!! eek.gif

How do you guys solve this excursion problem? By decreasing the amount of power? But then so decreases the SPL...

Sorry if I'm making a lot of stupid questions but like I wrote I'm newbie in what subs concern.

Thanks!

 

I'm no pro, but I have some experience building subs, so I'll try to help where I can. Alignment is basically a specific frequency response, so when you choose one, winSID varies the enclosure volume and tuning to try to match that alignment for your particular driver. So when you choose an alignment, it's not necessarily the 'optimim' values, it's just whatever values it takes to match the frequency response of that alignment. What I would do is ignore that and start playing around with enclosure volume and tuning yourself. With ported subs, bigger usually has a flatter response, so maybe start by putting in the largest your sub could be and lower tuning until you get a decently flat response. 

 

And yes, it's a property of ported subs that driver excursion needs to be tamed below the tuning frequency. This is done a number of ways. The simplest is just by using a plate amp. Most if not all have a high pass filter somewhere around 20hz for exactly this reason. But since you're trying to hit 15hz, this might not be your best bet. What I've done is use a pro amp (Behringer epq2000) along with a miniDSP. The minidsp gives you full digital control over high and low pass, as well as eq, gain and delay settings. It's incredibly powerful, especially for diy subs. You can even model eq with winISD as you're planning your build, and then add it with miniDSP once it's built. I believe you could also do this with pro amps with built-in dsp like the Behringer iNukes, but I don't know for sure how much control those give you over filters.

 

So you should model the high pass while planning as well. Once you've chosen your tuning frequency, add a high pass at or a few hz below your tune and then check your excursion. 

 

These are definitely not stupid questions :) I hope I was able to help.

cochinada likes this.
TheGiantPeach is online now  
post #3 of 158 Old 04-28-2014, 02:47 PM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Many thanks. That helped a lot smile.gif

Here are some curves with the modeling I'm doing for this drive Scan-Speak 30W-4558T00 and the following parameters defined:

Volume: 200,00l (based on an EBS3 alignment)
Fs: 17,00Hz
System input power: 350W

The filters can be seen on the curves.


Here I'm very close to the limit of 28mm.



This convex shape at the middle is awkward...


Same here...




I'm way bellow the target of 15Hz @110 dB but at least the TFM curve seems nicely flat.

Comparing with the specs for the SVS PC13-Ultra bellow, particularly @16 Hz mode, I'm also not up there yet. One thing that I don't know is the SPL achieved on this PC13-Ultra but since they use a 1000W power plate I guess it's not bad...

18-200 Hz ± 3 dB (20 Hz mode)
15-200 Hz ± 3 dB (16 Hz mode)
24-200 Hz ± 3 dB (sealed mode)

I don't think that with this driver and volume limited to 200l I have room for much more improvement. What do you think?
cochinada is offline  
post #4 of 158 Old 04-28-2014, 04:53 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Honestly, 15hz at 110dB is a pretty lofty target. I definitely can't do that on my system. I'm guessing you would need something like at least two 18 inch drivers each in a 10-15 cubic foot box tuned really low to do it.

But I do think you can improve what you do have. You have 8db of boost at just under 10hz. You don't want your eq's center frequency under your high pass because the HP will just eat it up. Try keeping your volume the same but changing the tune to 14hz, your HP to below there, and then putt your eq around 15-16hz and adjust gain and q until your response is flat. How does it look now?

Also, as is, you're not doing too bad. You're +- 3dB at 16hz or so by looking at the graph.

Do you have the winisd file for that driver? Or the T/S values? I could play with it a little
cochinada likes this.
TheGiantPeach is online now  
post #5 of 158 Old 04-28-2014, 05:22 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76

Okay here's what I got. A 200l box tuned to 14hz with these filters and eq will be +-1 dB from 13.85hz to 74hz, including the assumed crossover at 80hz. Here's the bad news. Xlim for this driver is 28mm, but xmax is only 12.5mm. Which means that past 12.5mm, you lose linearity which means you start introducing distortion. How much and if it will be audible depends on a number of things, but really I wouldn't want to push more than 100 watts into it. Because of the eq added, this will require at least a 150 watt amp, as you see from the apparent amp power screen. But this will get you really flat response and 104db to below 15hz. So if you built two of these, you should get 6dB more and there's your 110dB :) Actually that's not a bad plan. It wouldn't take very much power at all. You would need a dsp of some kind though.  

 

 

 

 

cochinada likes this.
TheGiantPeach is online now  
post #6 of 158 Old 04-29-2014, 01:58 AM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Boy, you are a genius! biggrin.gif I'm learning a lot from your experience.

So to see if I'm grasping all this new information you gave me please correct me if I'm wrong but I think it's something like this:

1. One should try a Tuning frequency bellow the target frequency desired, so in my case fs < 15Hz.

2. the system input power should not be greater than causing the cone excursion to go above Xlim. Even better it should not surpass Xmax to avoid distortion.

Now come the filters...

3. the main purpose of the High pass filter is to tame the cone excursion under the Fs frequency chosen, hence it should be applied with fc < fs.

4. the Parametric EQ is for me perhaps the most tricky part. It's goal is to flatten as much as possible the curve response. At the same time the compromise is to keep the cone excursion under control and finally, like you wrote "You don't want your EQ's centre frequency under your high pass because the HP will just eat it up." This is a though one for sure...

5. the Low pass filter must be applied above the EQ's fc and it's goal is to help flatten the curve even further on the higher usable frequency part of the sub-woofer. Here there are not significant impacts on cone excursion.

Now you also wrote something that I heard before but don't quite understand yet: "Because of the eq added, this will require at least a 150 watt amp, as you see from the apparent amp power screen."

How do we interpret this graph Amplifier apparent load power and how do you know I will need at least a 150W power amp?
How come the EQ requires extra watts?

Thanks a lot!
cochinada is offline  
post #7 of 158 Old 04-29-2014, 08:08 AM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76

Oh I'm no genius, I just read these forums a lot lol. 

 

You've got most of it. Let me help clarify.

 

1. You can choose your tuning frequency at or slightly below your frequency desired. But neither can be too far below your driver's resonant frequency, or excursion will get too high. You choose a good driver with a relatively low (17hz) resonant frequency. 

 

2. Exceeding xlim is immediate and possibly fatal damage to your driver. You can go past xmax, but you shouldn't really plan to. Being able to exceed xmax is basically just to protect the driver if you decide to really crank your system one time. But you shouldn't regularly exceed xmax either. When I'm planning, I'll let excursion go a few mm above xmax, because in the real world, it's better to have more than enough power than not enough.

 

3. Yes. The HP goes below the tuning frequency, but where exactly it goes depends on what your excursion looks like. 

 

4. If your eq is below your HP, the filter is just going to filter it out and you won't see any effects of the eq. Does that make more sense?

 

5. No, the low pass I added is the crossover that most people have their receiver applying at 80hz. It just helps see more realistically what the sub will be doing.

 

6. The apparant amp load power screen shows what the 'apparant' power draw from the amp will be for whatever input signal you put in. So if your enclosure is very efficient, requesting a 100 watt signal from the amp might only actually draw 80 watts. In your case, we're using eq, which boosts volume at a certain frequency. It takes power to add that volume, so you need more power where that eq is applied. In my picture of that screen, you can see the power peak right there at 14hz where we added our eq. It's up to 140 watts, so you know you need an amp capable of more than that.

cochinada likes this.
TheGiantPeach is online now  
post #8 of 158 Old 04-29-2014, 09:07 AM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
I'm always mixing Xmax with Xlim but eventually I'll settle this once and for all.

Very interesting that thing about the apparent power and the volume added almost by magic when using equalization. Clever stuff indeed!

I've been researching and searching for alternative drivers and come up with a couple more but can't find anything with Xmax greater than 15.1mm (Dayton Audio DVC310-88).

Here are three more simulations I made with 3 different drivers as bellow:

1. Dayton DVC310-88 (Xmax= +-15.1mm, Xlim = ?)
2. Scan-Speak 30W-4558T00 (your simulation with a minor tweak on the EQ filter) (Xmax = +-12.5mm, Xlim = +- 28mm)
3. Dayton RSS315HF-4 (Xmax= +- 14mm, Xlim = ?)






Projects.zip 3k .zip file

Some interesting points:

Project 1 - 104,117 dB @ 14.97 Hz
Project 2 - 104,435 dB @ 14.97 Hz
Project 3 - 105,170 dB @ 14.97 Hz

Project 1 - 13.07 Hz < f < 72,6 Hz above -3dB
Project 2 - 13.95 Hz < f < 73.68 Hz above -3dB
Project 3 - 12.96 Hz < f < 74.77 Hz above -3dB

Flatter curve -> Project 1
Greater SPL -> Project 3

Unfortunately they are all slightly above maximum linear excursion.

What do you think?
Attached Files
File Type: zip Projects.zip (3.5 KB, 14 views)
cochinada is offline  
post #9 of 158 Old 04-29-2014, 09:30 AM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76

Good work, those look great. I think would all perform pretty similarly. I think either of the Daytons would probably be happier playing loud since they have higher xmax. So I guess it comes down to the price of the drivers and your preference. I would probably go with project 3 because it uses a lot less power than project 1. 

 

If you're interested in high xmax, check out the Dayton Ultimax drivers. I used the 12 incher on my build, but they make a 15 inch, and are releasing an 18 inch this summer. They all have at least 19mm xmax. But they are more expensive than the Daytons you modelled.

cochinada likes this.
TheGiantPeach is online now  
post #10 of 158 Old 04-29-2014, 12:28 PM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Thanks! I will certainly have a look. Strange because I browsed a few shops and they didn't have those Ultimax drivers you mention. But they certainly look promising.
If you know more high excursion drivers please tell as I'm most interested in it.
I wonder what drivers those guys at SVS or JL use. After all the SVS PC13-Ultra can't have more than about 150l volume so I ask myself, how can it play so loud (and low) without distortion? eek.gif
Quote:
Originally Posted by TheGiantPeach View Post

I would probably go with project 3 because it uses a lot less power than project 1. 

Errrr... didn't you mean the other way around? The maximum Amplifier apparent load power is 250W @14Hz for project 3 whereas it peaks 152W @14.92 for the first project.
cochinada is offline  
post #11 of 158 Old 04-29-2014, 02:28 PM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,839
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 1162
a couple of quick items.

1. room gain will kick in when the long dimension of the room is about 1/2 wavelength at about 6db or so per octave. winisd is for 2pi space or outdoors, so you will be higher/flatter it shows on the low end.

2. a sub designed for low tuning has the fundamental problem that the port needs to be very long/or very narrow. check out your port velocity. 17m/s is the neighborhood that you want to be in. a little higher is fine. too high and port chuffing will occur and may be a distraction.

3. the subs will not work as bass traps in the corner. for a wave to be absorbed, it has to either find a resonant source to suck the energy out of the room in a pressure zone or absorb the particle velocity of the air in the velocity zone. a corner is a pressure zone. placing the subs in a corner will maximize their efficiency and also maximally excite room modes. if that works out ok, great. it is difficult to predict which rooms corner loading will work best in.

Listen. It's All Good.
LTD02 is offline  
post #12 of 158 Old 04-29-2014, 02:52 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Quote:
Originally Posted by cochinada View Post

Thanks! I will certainly have a look. Strange because I browsed a few shops and they didn't have those Ultimax drivers you mention. But they certainly look promising.
If you know more high excursion drivers please tell as I'm most interested in it....

Errrr... didn't you mean the other way around? The maximum Amplifier apparent load power is 250W @14Hz for project 3 whereas it peaks 152W @14.92 for the first project.

I know Parts Express sells the Ultimax drivers. Other than that, I don't really know. 

 

Uh, I meant the yellow line, which ever project that one is.

 

Quote:
Originally Posted by LTD02 View Post

a couple of quick items.

1. room gain will kick in when the long dimension of the room is about 1/2 wavelength at about 6db or so per octave. winisd is for 2pi space or outdoors, so you will be higher/flatter it shows on the low end.

2. a sub designed for low tuning has the fundamental problem that the port needs to be very long/or very narrow. check out your port velocity. 17m/s is the neighborhood that you want to be in. a little higher is fine. too high and port chuffing will occur and may be a distraction.

3. the subs will not work as bass traps in the corner. for a wave to be absorbed, it has to either find a resonant source to suck the energy out of the room in a pressure zone or absorb the particle velocity of the air in the velocity zone. a corner is a pressure zone. placing the subs in a corner will maximize their efficiency and also maximally excite room modes. if that works out ok, great. it is difficult to predict which rooms corner loading will work best in.

All very good point. I designed my sub to be flat to 17hz, but in my room I had a pretty big peak at 18hz, so I had to pull that down with eq. And absolutely true about ports. Once you decide on a tune, you need to make your port large enough to get your air velocity down below 17m/s or so and then see if it's practical to fit that port inside your enclosure. Bear in mind that enclosure size doesn't account for the space taken up by the port, so you'll need to compensate for that. 

TheGiantPeach is online now  
post #13 of 158 Old 04-29-2014, 03:15 PM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Thanks a lot!
Quote:
Originally Posted by LTD02 View Post

a couple of quick items.

1. room gain will kick in when the long dimension of the room is about 1/2 wavelength at about 6db or so per octave. winisd is for 2pi space or outdoors, so you will be higher/flatter it shows on the low end.

My room is a rectangle with 5,37m x 3,92m and the sub-woofers will be on the corners of the narrower wall so exactly what am I to expect?

Quote:
Originally Posted by LTD02 View Post

2. a sub designed for low tuning has the fundamental problem that the port needs to be very long/or very narrow. check out your port velocity. 17m/s is the neighborhood that you want to be in. a little higher is fine. too high and port chuffing will occur and may be a distraction.

Ok. That was another thing I was going to focus on. I've played with it trying to lower that speed under 17m/s and you were right. I came up with extraordinary long ports. eek.gif Perhaps they are feasible but it will be very close as I'm estimating that my sub-woofers will be about 1,4m high (not including the feet). So here is one more obstacle on the way...

Quote:
Originally Posted by LTD02 View Post

3. the subs will not work as bass traps in the corner. for a wave to be absorbed, it has to either find a resonant source to suck the energy out of the room in a pressure zone or absorb the particle velocity of the air in the velocity zone. a corner is a pressure zone. placing the subs in a corner will maximize their efficiency and also maximally excite room modes. if that works out ok, great. it is difficult to predict which rooms corner loading will work best in.

Well, there goes my idea out of the window. smile.gif Win some, lose some.
cochinada is offline  
post #14 of 158 Old 04-29-2014, 03:26 PM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Meanwhile I simulated another driver: that famous Dayton UM12-22 and the results were...well I leave it to you to comment. Enough to say that meanwhile I lower the signal power of my projects in order to be as much linear as possible and so, without further due here are the curves of two projects, both using Dayton drivers:

P1 -> Dayton UM 12-22
P2 -> Dayton RSS315HF-4








Projectos.zip 2k .zip file


Some thoughts about these:

Xmax for UM12-22 = 19mm
Xmax for RSS315HF-4 = 14.3mm

so they are both under the limit for linearity response.

P1 is able to use more power and so achieve greater SPL than P2. We are talking about +2.5dB on average.

f (-3dB) for P1 is 12.62Hz and f (-3dB) for P2 is 12.97Hz so they are almost identical.

P1 requires at least 300W and P2 150W (I think).

Care to comment please?
Attached Files
File Type: zip Projectos.zip (2.4 KB, 9 views)
cochinada is offline  
post #15 of 158 Old 04-29-2014, 03:57 PM
AVS Special Member
 
rlj5242's Avatar
 
Join Date: Dec 2000
Location: Southaven, MS
Posts: 3,283
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 18
Quote:
Originally Posted by cochinada View Post

....
I've been researching and searching for alternative drivers and come up with a couple more but can't find anything with Xmax greater than 15.1mm (Dayton Audio DVC310-88).
....
As mentioned earlier, Parts Express has a wide variety of Dayton drivers as well as others from companies like TC Sounds.

There are also lesser known drivers in your 12" and 15" requirements from companies like Sundown Audio and PSI Car Audio. Sundown may custom build you one but PSI definitely will. I have a pair of 12's, a 15 and an 18 all custom built to my specs.
rlj5242 is offline  
post #16 of 158 Old 04-29-2014, 04:02 PM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Quote:
Originally Posted by rlj5242 View Post

As mentioned earlier, Parts Express has a wide variety of Dayton drivers as well as others from companies like TC Sounds.

There are also lesser known drivers in your 12" and 15" requirements from companies like Sundown Audio and PSI Car Audio. Sundown may custom build you one but PSI definitely will. I have a pair of 12's, a 15 and an 18 all custom built to my specs.

I didn't mention but the problem is that I'm in Europe and importing drivers from outside the EU is prohibitive so I'm afraid Parts Express is not an option for me but I will search if I can find those brands here.
Thanks!
cochinada is offline  
post #17 of 158 Old 04-29-2014, 04:25 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Quote:
Originally Posted by cochinada View Post

...


Some thoughts about these:

Xmax for UM12-22 = 19mm
Xmax for RSS315HF-4 = 14.3mm

so they are both under the limit for linearity response.

P1 is able to use more power and so achieve greater SPL than P2. We are talking about +2.5dB on average.

f (-3dB) for P1 is 12.62Hz and f (-3dB) for P2 is 12.97Hz so they are almost identical.

P1 requires at least 300W and P2 150W (I think).

Care to comment please?

 

 

Quote:
Originally Posted by cochinada View Post


I didn't mention but the problem is that I'm in Europe and importing drivers from outside the EU is prohibitive so I'm afraid Parts Express is not an option for me but I will search if I can find those brands here.
Thanks!

Well 2.5dB is about 20-25% "louder sounding", so it would be noticeably more powerful. So two um12's would give you 111 or 112dB at 15hz. That's some serious power. But it's all up to you. Is the extra cost and power requirements worth it? BTW I'm running my um12 from a Behringer pro amp giving me 700watts rms at 4ohms times two channels, so if I ever add a second sub, I'll still have plenty of power. It's working great right now in combination with a miniDSP.

 

But then again, if you can't get them in the EU, that would be a problem lol. 

 

Any particular reason you modelled the untimax 12 instead of the 15? The 15 is only like $20 more and will give you more spl while also requiring more power.

cochinada likes this.
TheGiantPeach is online now  
post #18 of 158 Old 04-29-2014, 04:36 PM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Quote:
Originally Posted by TheGiantPeach View Post


Well 2.5dB is about 20-25% "louder sounding", so it would be noticeably more powerful. So two um12's would give you 111 or 112dB at 15hz. That's some serious power. But it's all up to you. Is the extra cost and power requirements worth it? BTW I'm running my um12 from a Behringer pro amp giving me 700watts rms at 4ohms times two channels, so if I ever add a second sub, I'll still have plenty of power. It's working great right now in combination with a miniDSP.

But then again, if you can't get them in the EU, that would be a problem lol. 

Any particular reason you modelled the untimax 12 instead of the 15? The 15 is only like $20 more and will give you more spl while also requiring more power.

How many dB do you have with your setup @15Hz?

The only reason is that I didn't have time yet and it's already very late here but rest assure that tomorrow is another day tongue.gif

How much is this UM12-22 over there by the way?

I have to read more about this miniDSP. About the power I'm considering only power plates by the way as I don't want more boxes on my rack.
cochinada is offline  
post #19 of 158 Old 04-29-2014, 04:42 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Quote:
Originally Posted by cochinada View Post


How many dB do you have with your setup @15Hz?

The only reason is that I didn't have time yet and it's already very late here but rest assure that tomorrow is another day tongue.gif

How much is this UM12-22 over there by the way?

I have to read more about this miniDSP. About the power I'm considering only power plates by the way as I don't want more boxes on my rack.

My box is tuned to 17hz, so I have very few dB at 15hz :p

 

I paid $180 for it. 

 

If you're planning on getting a plate amp, you're not going to get flat response to 15hz, unless you can find a plate amp with without a built-in high pass around 20hz, but I haven't seen any. I actually originally built mine with a plate amp but switched it out for the Behringer/miniDSP a few weeks later. The HP on my plate amp killed everything below 20hz pretty well.  

cochinada likes this.
TheGiantPeach is online now  
post #20 of 158 Old 04-29-2014, 04:52 PM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,839
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 1162
- some of the plate amps can be modified for high pass, the bash series comes to mind. most others, as noted, tend to have something around a 2nd order highpass at 20hz built in.

- room gain will likely be coming in around 24hz or so, so if the response starts to roll off at that point in the winisd model, it is ok. the room will help boost it back up.

- also, keep an eye on the "first port resonance". if that gets too close to the pass band, you may hear it. doesn't seem to be a real problem for most folks though.
cochinada likes this.

Listen. It's All Good.
LTD02 is offline  
post #21 of 158 Old 04-29-2014, 04:57 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Quote:
Originally Posted by LTD02 View Post

- some of the plate amps can be modified for high pass, the bash series comes to mind. most others, as noted, tend to have something around a 2nd order highpass at 20hz built in.

 

 

I had the Bash amp and AFAIK, you can modify the eq to combat the HP, but the filter itself is fixed. But finding any sort of documentation on high pass filters built into plate amps is near impossible. I specifically asked the Parts Express customer service and all they could say is that most plate amps have some sort of high pass filter around 20hz. 

 

But if I'm mistaken in this, please let me know. I would be really curious, since I still have the Bash sitting around...

TheGiantPeach is online now  
post #22 of 158 Old 04-29-2014, 06:12 PM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,839
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 1162
http://www.parts-express.com/pedocs/specs/300-752-bash-specifications-45313.pdf

the documentation for the bash amps isn't very user friendly. the "fc" in the tables is the high pass filter point, though it is sometimes not wholly intuitive how it works with the higher q's. there is quite a range over which the high pass filter point can be set from 14hz to 35hz. might even be able to go lower by extrapolating values, though i haven't seen anybody try that yet.
TheGiantPeach likes this.

Listen. It's All Good.
LTD02 is offline  
post #23 of 158 Old 04-29-2014, 06:29 PM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76

Super interesting. I've seen that page before, but I always thought Fc was the center frequency that the eq was applied at. But I guess that makes sense since it does go down that low. Hmm.

TheGiantPeach is online now  
post #24 of 158 Old 04-30-2014, 01:36 AM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Hi guys!
Quote:
Originally Posted by TheGiantPeach View Post

If you're planning on getting a plate amp, you're not going to get flat response to 15hz, unless you can find a plate amp with without a built-in high pass around 20hz, but I haven't seen any. I actually originally built mine with a plate amp but switched it out for the Behringer/miniDSP a few weeks later. The HP on my plate amp killed everything below 20hz pretty well.  

I guess you mean without, right? As in my modelling I defined the Highpass with fc around 12Hz , if these plates have an inbuilt filter with fc=20Hz that would ruin everything and prevent me from getting any output bellow 20Hz. Is this correct?

However there seems to be a catch...

Quote:
Originally Posted by LTD02 View Post

- some of the plate amps can be modified for high pass, the bash series comes to mind. most others, as noted, tend to have something around a 2nd order highpass at 20hz built in.

The ideal would be to be able to define the filter exactly as I modelled it but I'm seeing that this is going to be almost impossible... So, how do you guys do it? Simply don't use an amplifier plate and instead use a 'normal' power without any filters nor equalization and then use something like a miniDSP between this and the sub-woofer or what? In this case then we are talking always about passive sub-woofers.

Quote:
Originally Posted by LTD02 View Post

- room gain will likely be coming in around 24hz or so, so if the response starts to roll off at that point in the winisd model, it is ok. the room will help boost it back up.

This is the catch I mentioned. So despite the inbuilt filter perhaps there is a chance of getting a lower response after all but it's not something one can control unfortunately and that I don't like.

Quote:
Originally Posted by LTD02 View Post

- also, keep an eye on the "first port resonance". if that gets too close to the pass band, you may hear it. doesn't seem to be a real problem for most folks though.

Sorry but what is this exactly? I didn't look at it before but I have 145Hz (Dayton UM 12-22) and 125Hz (Dayton RSS315HF-4). Is this all right?

I need to do more modelling and think about all these new inputs because what good it is to be so precise in the modelling if at the end one cannot control such things as the highpass filter?
cochinada is offline  
post #25 of 158 Old 04-30-2014, 01:58 AM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,839
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 1162
"Super interesting. I've seen that page before, but I always thought Fc was the center frequency that the eq was applied at. But I guess that makes sense since it does go down that low. Hmm."

to model the bash filters in winisd, start with any sealed driver and apply a linkwitz transform to 1hz so that you have a flat line to work with. the actual level is completely arbitrary. I used 110db because that was toward the middle of my screen.

the add a second order high pass filter using the "User SOS" filter type and enter the numbers like this: n=2, fc=25hz, q=1.1

n=2 always stays 2. fc and q change based on the values in the chart. so then you can compare the filter against a flat line.

they might refer to this filter as "boost of +2db at 30hz", but that doesn't describe what is going on nearly as good as a picture. they should have included pictures!!!

cochinada and TheGiantPeach like this.

Listen. It's All Good.
LTD02 is offline  
post #26 of 158 Old 04-30-2014, 02:00 AM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,839
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 1162
here is fc=14, q=1.

cochinada and TheGiantPeach like this.

Listen. It's All Good.
LTD02 is offline  
post #27 of 158 Old 04-30-2014, 02:11 AM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,839
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 554 Post(s)
Liked: 1162
"Sorry but what is this exactly?"

if you blow over the port tube like a coke bottle, that is the resonance. not quite right, but kind of. it can create a big peak in the frequency response, which may or may not matter depending on how the subwoofer is used and how close the to passband the peak is.

also, if I may suggest, the eq that you are applying is way too much. don't feel like you have to torture the response into a particular shape. the room will add some gain. the room will also create peaks and dips might mess up the smooth model response anyway. a protective high pass filter is a good idea and then a little boost is required and most ported cabs should need much if any to be pretty flat in room. adding boost below the tuning point is also almost never a good idea.

here is a rough example of a sealed enclosure outdoors (2 pi space) vs. the response that you might get in your room (not including modes that will create peaks and dips)



so you can see that you won't need as much eq as you may think to get a good response. only 6-8db of eq would likely provide for a pretty flat response even with this sealed sub.
cochinada likes this.

Listen. It's All Good.
LTD02 is offline  
post #28 of 158 Old 04-30-2014, 02:55 AM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Quote:
Originally Posted by LTD02 View Post

to model the bash filters in winisd, start with any sealed driver and apply a linkwitz transform to 1hz so that you have a flat line to work with. the actual level is completely arbitrary. I used 110db because that was toward the middle of my screen.

the add a second order high pass filter using the "User SOS" filter type and enter the numbers like this: n=2, fc=25hz, q=1.1

n=2 always stays 2. fc and q change based on the values in the chart. so then you can compare the filter against a flat line.

they might refer to this filter as "boost of +2db at 30hz", but that doesn't describe what is going on nearly as good as a picture. they should have included pictures!!!


Sorry to ask so many basic questions but what exactly is a Linkwitz transform and what is f0 and fp?
Do you mean that those bash filters have implemented a kind of internal Linkwitz transform + Highpass filters?
Are there any 'equalizers' out there that can apply a customised Linkwitz transform according to our specs?
Why the mention to a 'sealed box' together with this Linkwitz transform?
cochinada is offline  
post #29 of 158 Old 04-30-2014, 03:14 AM - Thread Starter
Member
 
cochinada's Avatar
 
Join Date: Apr 2006
Location: Portugal
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
Quote:
Originally Posted by LTD02 View Post

"Sorry but what is this exactly?"
if you blow over the port tube like a coke bottle, that is the resonance. not quite right, but kind of. it can create a big peak in the frequency response, which may or may not matter depending on how the subwoofer is used and how close the to passband the peak is.

I think I got it more or less. This is the mechanical resonance frequency intrinsic to the port size and shape. I guess that the air speed going through the port is the variable that matters here as if it is critical it can cause resonance and be very audible. I just don't see the connection of this with the band pass of the subwoofer as we are talking of mechanical vs. electrical parameters. What am I missing?

Quote:
Originally Posted by LTD02 View Post

also, if I may suggest, the eq that you are applying is way too much. don't feel like you have to torture the response into a particular shape. the room will add some gain. the room will also create peaks and dips might mess up the smooth model response anyway. a protective high pass filter is a good idea and then a little boost is required and most ported cabs should need much if any to be pretty flat in room. adding boost below the tuning point is also almost never a good idea.

here is a rough example of a sealed enclosure outdoors (2 pi space) vs. the response that you might get in your room (not including modes that will create peaks and dips)



so you can see that you won't need as much eq as you may think to get a good response. only 6-8db of eq would likely provide for a pretty flat response even with this sealed sub.

You may of course and are most welcome! I can't get too much advices on this journey. I will be more genlte than and try not to torture so much my poor driver. I'm also going to model some closed boxes as all this mumbo jumbo with the incredible long ports is messing with my nerves.


EDIT: stupid me! Sound is after all frequency, so of course that blowing on those ports at that specific frequency can interfere with the driver's response!
cochinada is offline  
post #30 of 158 Old 04-30-2014, 08:07 AM
Member
 
TheGiantPeach's Avatar
 
Join Date: Mar 2014
Posts: 192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 76
Quote:
Originally Posted by cochinada View Post

Hi guys!
I guess you mean without, right? As in my modelling I defined the Highpass with fc around 12Hz , if these plates have an inbuilt filter with fc=20Hz that would ruin everything and prevent me from getting any output bellow 20Hz. Is this correct?

However there seems to be a catch...
The ideal would be to be able to define the filter exactly as I modelled it but I'm seeing that this is going to be almost impossible... So, how do you guys do it? Simply don't use an amplifier plate and instead use a 'normal' power without any filters nor equalization and then use something like a miniDSP between this and the sub-woofer or what? In this case then we are talking always about passive sub-woofers.
This is the catch I mentioned. So despite the inbuilt filter perhaps there is a chance of getting a lower response after all but it's not something one can control unfortunately and that I don't like.
Sorry but what is this exactly? I didn't look at it before but I have 145Hz (Dayton UM 12-22) and 125Hz (Dayton RSS315HF-4). Is this all right?

I need to do more modelling and think about all these new inputs because what good it is to be so precise in the modelling if at the end one cannot control such things as the highpass filter?

 

That's exactly how I do it. It's a lot more money up front than simply buying a plate amp, but you end up with far greater control over every aspect of the sub's performance with miniDSP, and far more power than I need with the pro amp. 

 

I played with the high pass on the Scan Speak model I was doing, based on LTD02's super helpful illustrations and it looks like you could still get your high pass pretty close to where you want it by using the 1 or 2dB boost from 20-24hz. In order to do this though, you would need to be pretty confident with a soldering iron as you would have to remove two resistors from the board and solder on two new ones. Or pay to have an electronics shop do it. 

cochinada likes this.
TheGiantPeach is online now  
Reply DIY Speakers and Subs

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off