Gladly,
Things change of course but this is where I'm at. All EQ and Xover thru the DCX2496.
Xover & levels:
48db high pass "brick wall" at 20Hz
1.17khz 24db/oct between the woof/tweet
Relative levels: the hp is 3.8 db down from the LP. The LP is ~50wpch and the HP is ~12wpch (the hp is a mini aleph with low gain). The compression driver is enormously efficient as compared to the woofer.
EQ:
Can't do it without a parametric IMHO
+5dB at ~24hz - Q=2.5 boost for BSC (for you Salas)
-3.5dB at ~3k - Q= forget cut for 'horn honk'
+6dB at 20k high Q to mimic Tannoy passive.
Damping is all walls lined with thin 'egg crate' open cell foam. Works good, only bass from the vent as I can hear.
Final performance....what's that? Who's ever done?
- right now.... Very good IMHO. I think you would have to like the 'type of sound' though. It is not like your average 'audiophile' speaker which can sound lifeless to me sometimes regardless of accuracy. The Tannoys are dynamic and detailed. Upper bass is sharp and fast. Female vocals and guitar plucks are right where I like them. The bass now is fairly deep. Good 'nuff for my app.
Thanks all.
By the way GM, your predictions on the vent were real close. I only had to make it 1/2 inch wider than your guess. Thanks.
Things change of course but this is where I'm at. All EQ and Xover thru the DCX2496.
Xover & levels:
48db high pass "brick wall" at 20Hz
1.17khz 24db/oct between the woof/tweet
Relative levels: the hp is 3.8 db down from the LP. The LP is ~50wpch and the HP is ~12wpch (the hp is a mini aleph with low gain). The compression driver is enormously efficient as compared to the woofer.
EQ:
Can't do it without a parametric IMHO
+5dB at ~24hz - Q=2.5 boost for BSC (for you Salas
)-3.5dB at ~3k - Q= forget cut for 'horn honk'
+6dB at 20k high Q to mimic Tannoy passive.
Damping is all walls lined with thin 'egg crate' open cell foam. Works good, only bass from the vent as I can hear.
Final performance....what's that? Who's ever done?
- right now.... Very good IMHO. I think you would have to like the 'type of sound' though. It is not like your average 'audiophile' speaker which can sound lifeless to me sometimes regardless of accuracy. The Tannoys are dynamic and detailed. Upper bass is sharp and fast. Female vocals and guitar plucks are right where I like them. The bass now is fairly deep. Good 'nuff for my app.
Thanks all.
By the way GM, your predictions on the vent were real close. I only had to make it 1/2 inch wider than your guess. Thanks.
Hi Mp,
keep fiddling with DCX until you are reasonably happy.
Then build an analogue crossover to mimic the High Pass section using the slopes frequencies EQs and Q etc.
Finally mimic the Low Pass with the same info extracted from the DCX.
You may need to do a little bit of tweaking after going analoque into the power amps.
BTW. I would omit the bass High Pass (forming a bandpass) and go Low Pass on it's own.
keep fiddling with DCX until you are reasonably happy.
Then build an analogue crossover to mimic the High Pass section using the slopes frequencies EQs and Q etc.
Finally mimic the Low Pass with the same info extracted from the DCX.
You may need to do a little bit of tweaking after going analoque into the power amps.
BTW. I would omit the bass High Pass (forming a bandpass) and go Low Pass on it's own.
Yep, that GM fella is pretty good.
Your gonna keep me working till I'm old and feeble!
Seriously, I hear you, and the deficiencies of the DCX are obvious. (It's not terribly dreadful though)
The flexibility carries a lot of weight with me tho. I think I have to go thru the DCX thread and check in with the yahoo group before I junk the box I think there is hope there, but YOUR RIGHT...it's the weak link right now.
The reason for the HP on the woofer is to make my TT happy for the time being.
Your gonna keep me working till I'm old and feeble!
Seriously, I hear you, and the deficiencies of the DCX are obvious. (It's not terribly dreadful though)
The flexibility carries a lot of weight with me tho. I think I have to go thru the DCX thread and check in with the yahoo group before I junk the box
I think there is hope there, but YOUR RIGHT...it's the weak link right now.The reason for the HP on the woofer is to make my TT happy for the time being.
mpmarino..
ok, so I've done a little work.
the spec quoted on the Tannoy page is
(BL)^2/Re=22
Re=5.8 Ohms, so that implies:---> BL= 11.296
Vas=9,54 ft^3,
Fs=32 Hz
I also input the Qts=0.4
I got a pretty nice "response" from Win ISD, Basically what I would consider good. 3.5 ft^3..4"diametre port, 4.5" long
-3db@47 Hz , +1.5dB@60Hz
I could easily live with this.
ok, so I've done a little work.
the spec quoted on the Tannoy page is
(BL)^2/Re=22
Re=5.8 Ohms, so that implies:---> BL= 11.296
Vas=9,54 ft^3,
Fs=32 Hz
I also input the Qts=0.4
I got a pretty nice "response" from Win ISD, Basically what I would consider good. 3.5 ft^3..4"diametre port, 4.5" long
-3db@47 Hz , +1.5dB@60Hz
I could easily live with this.
Attachments
Changing the BL from 22 to 11.5 does little to the modeling. Have you tried it both ways? No difference in predicted response that I can see.
Yes, your right. It does SEEM that way.
However, I have to be honest. After a couple of weeks since I ditched the numbers and software you are using now (because it is garbage) (for my driver at least). Spending hours trimming the vents through trial and error endlessly, your stating that it seems easy is mildly insulting IHMO.
If you want to discuss the driver spec'd in the data sheet, which has nothing to do with my drivers or my project (my conclusion), that's fine. We can predict that all day long
It really has nothing to do with this though
With all due respect, I have already been there; in the real world. It could be a simple issue that the driver I have is not a K3149, although many Tannerds say that it is.
The drivers I have will NOT tune in a box that size with a 4" vent that long, I have tried 4" diameter vents in 1/2" increments 6" long on down. The box literally chokes. Just look at the vent that ACTUALLY makes bass: 4.5" dia X 1.5" long. That's about equivalent to a 4" X 3/4" vent... a far cry from your 4.5" length. REAL WORLD.
Yes, your right. It does SEEM that way.
However, I have to be honest. After a couple of weeks since I ditched the numbers and software you are using now (because it is garbage) (for my driver at least). Spending hours trimming the vents through trial and error endlessly, your stating that it seems easy is mildly insulting IHMO.
If you want to discuss the driver spec'd in the data sheet, which has nothing to do with my drivers or my project (my conclusion), that's fine. We can predict that all day long
It really has nothing to do with this thoughWith all due respect, I have already been there; in the real world. It could be a simple issue that the driver I have is not a K3149, although many Tannerds say that it is.
The drivers I have will NOT tune in a box that size with a 4" vent that long, I have tried 4" diameter vents in 1/2" increments 6" long on down. The box literally chokes. Just look at the vent that ACTUALLY makes bass: 4.5" dia X 1.5" long. That's about equivalent to a 4" X 3/4" vent... a far cry from your 4.5" length. REAL WORLD.
mpmarino
Q, why is it that I am trying to state the obvious, and somehow I offend you? I am certainly not trying to be disrespectful in any way to you. If it appears that way please accept my sincere apology
As I am very interested in fullrange and dual concentric drivers, I am trying to reconcile the parameters that are on the "Tannoy Gold" site with what you have. Believe me, I have spent a fair bit of time looking at the mathematics and not just WinISD sims.
Have you done any frequency sweeps of the loudspeakers? If so, can you provide the -3dB point?
Also is you enclosure 3 ft^3 net?, (including allowing for the driver , all cross bracing , etc).Again, just to profide clarrification so that I might try to reconcile the differences, that's all. It hardly matters what some simulation may say, ultimately it's your ears, equipment, effort and money. The speakers look good, by the way.
I too would be frustrated with being so disappointed initially given the amount of work, and care you have put into this project.
Q, why is it that I am trying to state the obvious, and somehow I offend you? I am certainly not trying to be disrespectful in any way to you. If it appears that way please accept my sincere apology
As I am very interested in fullrange and dual concentric drivers, I am trying to reconcile the parameters that are on the "Tannoy Gold" site with what you have. Believe me, I have spent a fair bit of time looking at the mathematics and not just WinISD sims.
Have you done any frequency sweeps of the loudspeakers? If so, can you provide the -3dB point?
Also is you enclosure 3 ft^3 net?, (including allowing for the driver , all cross bracing , etc).Again, just to profide clarrification so that I might try to reconcile the differences, that's all. It hardly matters what some simulation may say, ultimately it's your ears, equipment, effort and money. The speakers look good, by the way.
I too would be frustrated with being so disappointed initially given the amount of work, and care you have put into this project.
Hi,
that response hump (post46) indicates that the box is too small. I suggest you make it bigger than 3.5cubft, not smaller.
A Tannoy with a hump like that will sound one note boom boom in a small to medium room even if it is well furnished/damped.
The chosen Qts=0.4 gives the lowest FLAT frequency response.
A Qts that is either higher or lower will raise the F-3 flat response.
If Qts is 0.4, then for a maximally flat response then the box volume (after subtracting all the bracing, vents and driver) will be about equal to the Vas volume.
Most domestic Tannoys have a Qts much lower than 0.4
You really need to measure your Tannoy.
that response hump (post46) indicates that the box is too small. I suggest you make it bigger than 3.5cubft, not smaller.
A Tannoy with a hump like that will sound one note boom boom in a small to medium room even if it is well furnished/damped.
The chosen Qts=0.4 gives the lowest FLAT frequency response.
A Qts that is either higher or lower will raise the F-3 flat response.
If Qts is 0.4, then for a maximally flat response then the box volume (after subtracting all the bracing, vents and driver) will be about equal to the Vas volume.
Most domestic Tannoys have a Qts much lower than 0.4
You really need to measure your Tannoy.
Because I had already been through the obvious, which I thought was obvious
Anyway, it seems my reaction was ill conceived and I misunderstood you. My apologies to you moving on
I did send some sine waves thru via a cd I made at 5Hz increments. Output seems to build considerably with port contribution at somewhere around 35Hz.
Andrew is right. The only way to 'get to the bottom' of this is for me to measure my drivers. Then we will see why things are the way I see them, and the numbers see it a different way. My next 'project' is going to be beating a bit of the learning curve of speaker workshop.
mpmarino said:By the way GM, your predictions on the vent were real close. I only had to make it 1/2 inch wider than your guess. Thanks.
Greets!
You're welcome!
Interesting, considering their in-room placement I figured you would tune them a bit lower than higher.
WRT to a few of things in later posts, my base tuning was 3 ft^3/4" d x 1.5" l = ~47.7 Hz Vs what you arrived at: 3 ft^3/4.5" d x 1.5" l = ~51.5 Hz, suggesting you traded some LF extension for a bit more midbass 'slam' to keep from shelving the HF more; and converting Nanook's alignment to yours: 3.5 ft^3/4" d x 4.5" l = ~34.8 Hz = 3 ft^3/2.51" d x 1.5" l, a much smaller vent that may 'chuff' at high SPLs, though they would sound somewhat different due to the cab's different Vbs. Regardless, both would require what I consider excessive HF shelving if not corner loaded and of course any added series resistance in the signal chain combined with room acoustics can significantly alter any alignment for better or worse.
WRT the questionable published specs, I arrived at a 0.359 Qes and chose a 2.2 Qms based on some other models, so I used a 0.30863 Qts for half space sims. Anyway, calcing BL using this Qes yields a mean ~11.33 N/A since it varies depending on what 'c', 'po' is used, so if the 22 BL^2/Re = 11.296 N/A is reasonably accurate, then the 0.4 Qts is BS.
GM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.