I didn't add the same level of detail as Davor, instead choosing to focus on the baffle and the compensation.
A: A resistor/inductor baffle step circuit will work with one driver or two in parallel.
B: A single inductor will work on a 2.5 way, which is what I simmed. It uses two drivers in parallel with one cut at the baffle step.
C: A full crossover for a WAW.
If I'm not mistaken as I'm discussing this in another thread as well, there are three options.
A: A resistor/inductor baffle step circuit will work with one driver or two in parallel.
B: A single inductor will work on a 2.5 way, which is what I simmed. It uses two drivers in parallel with one cut at the baffle step.
C: A full crossover for a WAW.
Do you know how much gain the second driver will add from 50-100dB? I am measuring in-room response to be about 15dB down @ 50-100Hz with a single W4-616 and other 4" as well..
A second driver adds 6dB at any lower frequency. It is 6dB at higher frequencies as well except that it gets more directive.
I take it you are thinking of starting your compensation down low so you can create more bass extension..
I take it you are thinking of starting your compensation down low so you can create more bass extension..
If wired in parallel. You get 3dB from the 2nd driver, and 3dB gain from the doubling of power into half the impedance for a toal of 6dB.
If wired in series. You get 3dB from the 2nd driver, and 3dB loss from the jalving of power into double the impedance.
dave
Same thing is shown in my sims, first graph in post #31.
Blue ang green are individual drivers and you can see 50-100Hz area is around 15dB down.
Black is summed response, which shows you the gain. When both drivers are working equally gain is 6dB.
My sims are made with calculated response from W4-616SA ts-parameters and assuming pistonic operation of the cone in free-space. Its fairly correct for point source conditions when cone diameter is less than half of wavelenght. Up to 2kHz for 4in driver.
Allan sims are apparently made with ideal drivers, having perfectly flat response all the way and it seems like that made a bit confusion.
Regarding drivers you asked me in post #34
if you wish to avoid active bass boost you can use vented box with max. flat response. If F3 is 42Hz or lower you won't need additional bass boost, just BSC:
Vb=20*Vas*Qts^3.3
Fb=Fs*0.42*Qts^-0.96
F3=Fs*0.28*Qts^-1.4
For simplicity I would like to start with just one driver and correct the FR.
- Which is better, bass boost with BSC or notch filter to reduce midrange 6-9 dB? I always thought is better to attenuate than boost but Davor says both will increase the bass load. If that is true, i would rather go with the simpler circuit.
- So the 5.62mH and 100mOhm in your post #38 are used in parallel to the + terminal with one driver for BSC?
I have two drivers I want to try this on:
SEAS Excel W12CY-003 (E0044) Nextel Cone 4.5" Woofer
and
Tang-Band W4 616
How do I calculate BSC myself? I have tried some online calculators but don't get he same results as you.
In your equations,* is multiply and ^ means to the 3.3 power / -.96 power / -1.4 power? I didn't think it can be negative number
Yes, I would like to make simplest correction possible so just BSC and tien the cabinet to 42Hz. My current speaker is .25ft^3. Using Dickason's Vent calculator, 42Hz would be 1" dia x 4" long. 50hZ would be 1" dia x 2.65" long.
.
But I was told the box is too big for either driver. In that case, I could use Tang-Band W5-2143 but it's Xmax is only (2.5 x 2) = 5mm.
As you can see, the Seas and Tang-Band have similar FR but the Seas has X max of 9mm vs Tang-Band's (3.2x 2) = 6.4mm
Last edited:
1. The 100mOhm is the resistance of the inductor.
2. Don't use that circuit for one driver.
For one driver you would use this circuit. (You can also use this circuit for two drivers because there are different ways to do BSC.)
The resistor sets the level of the midrange and the inductor sets the frequency of the step.
Yes, * is multiply and ^ is power (or exponent). Power can be negative, its not mistake.
Max linear BR alignment starts from driver specs. Then, if calculated F3 suit your needs you make the box with calculated Vb and Fb.
If existing box volume Vb=0,25cu.ft. don't match the calculated value, tuning itself won't give proper results.
Thanks Planet 10 and allenB. This circuit will reduce midrange? Might as well try this on my current speaker since already have it.
Can you tell me how to find for Alpair 10P
Ok, example for Alpair 10P:
Fs=42,4Hz
Vas=30Liters
Qts=0,33
Assuming parasitic resistance in series with driver (amp,cables,connectors) is 0,5 Ohm, than Qts will rise to 0,353.
Using the formulas for max flat alignment you get:
F3=51Hz
Fb=48,4Hz
Vb=19,3Liters
Note that you need much bigger box than your 0,25cu.ft/7Liter and you still won't reach desired F3.
Edit: you can use the same calculations for any other driver to check its suitability to your needs.
But I doubt any FR driver smaller than 6,5in will reach F3=42Hz without some sort of bass boost.
Fs=42,4Hz
Vas=30Liters
Qts=0,33
Assuming parasitic resistance in series with driver (amp,cables,connectors) is 0,5 Ohm, than Qts will rise to 0,353.
Using the formulas for max flat alignment you get:
F3=51Hz
Fb=48,4Hz
Vb=19,3Liters
Note that you need much bigger box than your 0,25cu.ft/7Liter and you still won't reach desired F3.
Edit: you can use the same calculations for any other driver to check its suitability to your needs.
But I doubt any FR driver smaller than 6,5in will reach F3=42Hz without some sort of bass boost.
Last edited:
There's one trick you can use to get lower F3 with given driver.
In previous example parasitic resistance had the effect of raising Qts. If you deliberately add some more resistance in series with the driver (to raise Qts even further) , than max flat alignment will give even lower F3.
Again, Alpair 10P:
Fs=42,4Hz
Vas=30Liters
Qts=O,33
Let's add additional 1,2 Ohm resistor in series with driver. Together with parasitic resistance of 0,5 Ohm it gives total series resistance Rs=1,7 Ohm.
Qts will rise to 0,405 and max flat alignment gives:
F3=42Hz
Fb=42,4Hz
Vb=30,5Liters
Note you need even bigger box now, but desired F3 will be reached. Also, some amp power will be lost in resistor.
This method also represents a kind of bass boost (passive type). In simple words you are trading quality for quantity.
And it has its limitations, not suitable for drivers with already high Qts, if series resistance needs to be to high, or if box volume needs to be to big in relation to Vas.
New Qts value with series resistor can be calculated HERE
In previous example parasitic resistance had the effect of raising Qts. If you deliberately add some more resistance in series with the driver (to raise Qts even further) , than max flat alignment will give even lower F3.
Again, Alpair 10P:
Fs=42,4Hz
Vas=30Liters
Qts=O,33
Let's add additional 1,2 Ohm resistor in series with driver. Together with parasitic resistance of 0,5 Ohm it gives total series resistance Rs=1,7 Ohm.
Qts will rise to 0,405 and max flat alignment gives:
F3=42Hz
Fb=42,4Hz
Vb=30,5Liters
Note you need even bigger box now, but desired F3 will be reached. Also, some amp power will be lost in resistor.
This method also represents a kind of bass boost (passive type). In simple words you are trading quality for quantity.
And it has its limitations, not suitable for drivers with already high Qts, if series resistance needs to be to high, or if box volume needs to be to big in relation to Vas.
New Qts value with series resistor can be calculated HERE
Davor, thanks so much. I will take some time to go over this but very good I can calculate this myself.
How does it sound at 60-70db? This is the volume I listen at and was told the Scanspeak need to be played louder than that before it comes alive.
This looks like a red flag. A speaker is not a kite, it doesn't drag along the ground until you get the airspeed up. There could be some distortion involved in making that statement.
chuck55,
By "coming alive" would it mean "feeling the bass"? At higher SPL it is likely that you will feel more bass and also hear more of the other frequencies, as everything is louder. 🙂
For certain wide-band (aka full range) drivers there is a hump in the mid-range and some of the upper frequencies, and that can give the listener a perception that more mid-range details are coming through at lower SPL. And if the speaker is not strong in mid-bass and bass, there will be less masking of the mid-range from the lower frequencies, again creating the perception of more mid-range details even at lower SPL.
My experience with the Scanspeak 15W8530 is that it plays fine at lower SPL... but being a modest sized cone, there is more slam and punch as the volume is cranked up (as expected).
At lower SPL to enjoy more bass, it is likely that you will have to bump up some parts of the lower frequency range using EQ/DSP.
Hope this helps. 🙂
By "coming alive" would it mean "feeling the bass"? At higher SPL it is likely that you will feel more bass and also hear more of the other frequencies, as everything is louder. 🙂
For certain wide-band (aka full range) drivers there is a hump in the mid-range and some of the upper frequencies, and that can give the listener a perception that more mid-range details are coming through at lower SPL. And if the speaker is not strong in mid-bass and bass, there will be less masking of the mid-range from the lower frequencies, again creating the perception of more mid-range details even at lower SPL.
My experience with the Scanspeak 15W8530 is that it plays fine at lower SPL... but being a modest sized cone, there is more slam and punch as the volume is cranked up (as expected).
At lower SPL to enjoy more bass, it is likely that you will have to bump up some parts of the lower frequency range using EQ/DSP.
Hope this helps. 🙂
Yes, that has been by experience as well.
As I've never heard the Revelators, or similar Audio Technology flex units, I can only project from the specs what I think I might hear. I also spent a lot of time perusing them and that Revelator 15W8530 in particular. I was told the 38mm voice coil "Takes tons of power to sound right and the larger VC is more power hungry."
I would imagine, but not know, a large voice coil will require more current (and sound volume) to build up enough flux density to get moving without compressing the dynamics (lack of life). Not just from Fletcher Munson curve.
I can imagine it is more than just mid-range being emphasized. That Revelator MM is 13 grams and a force factor of 7 TM. Compare that to a TB's 4 grams / 5.4 TM or Seas W15LY001 6.28 gram 4.6 TM. So the Revelator will take a lot more power to move that heavier mass and while the momentum of the heavy cone will give more bass, it also can not move fast enough to give as much mid-range detail.
So I understand why you asked before what I had for an amp as that kind of driver needs the current to sound right.
It would be interesting to compare the TB to Revelator on appropriate music, which would not play to either driver's inherent strength, at 50 / 60 / 70dB. I would liken it to playing the Devore O/96 which has a 10" driver to a 4" speaker. Is no wonder they call it "colorful but not colored" as I have found that style of speaker misses the fine detail but its strength is that it picks up the colors of the music better. I don't know. Can only guess.......
chuck55,
I now have the Revelator in a 3 way crossed around 320 Hz to an 8" woofer. It gives very good details in the mid-range. I think just looking at the weight of the cone and assuming that the lighter cone will give more detail is an oversimplification... the SOA Purifi drivers are state of the art and have cones that are heavier than the Revelators. An underdamped cone can also lead to more resonances and actually have more non linear distortion products, IMD and in reality less detailed.
A lighter cone will need less power to move; if the magnet strength is similar, a speaker with a lighter cone is likely to have a higher sensitivity rating.
I now have the Revelator in a 3 way crossed around 320 Hz to an 8" woofer. It gives very good details in the mid-range. I think just looking at the weight of the cone and assuming that the lighter cone will give more detail is an oversimplification... the SOA Purifi drivers are state of the art and have cones that are heavier than the Revelators. An underdamped cone can also lead to more resonances and actually have more non linear distortion products, IMD and in reality less detailed.
A lighter cone will need less power to move; if the magnet strength is similar, a speaker with a lighter cone is likely to have a higher sensitivity rating.
If the Alpair 10C metal cone is anything like Markaudio-Alpair6 - harmonic distortion resonances at 800Hz and 2kHz I'm leaning towards the Markaudio Alpair-10P Paper Cone 6" Full Range. Maybe less clean and punchy sound but smoother.
I've been reading other threads here and everyone seems to recommend the Alpair over the Jordan but comparing to Jordan-JX92S FR and distortion the Alpair has more distortion and less flat F-R. Zaph's site sure was a wealth of information.
I've been reading other threads here and everyone seems to recommend the Alpair over the Jordan but comparing to Jordan-JX92S FR and distortion the Alpair has more distortion and less flat F-R. Zaph's site sure was a wealth of information.
Alpair 10.3? Very similar voicing to the metal A6.2, goes lower, but the A6 better at mids/top.
Do note that the A6 you point to is only similar in size and uses the same basket. But the A6.2m smokes the old A6.
I went thru 3 sets of JX92s, couldn’t believe that the one i was hearing had significant ringing issues that had me ove them on. The Eikona is better, but still has the same issues.
We did a shootout at teh last VI diyFest between Eikona, Alpair 11ms (stock), and A10.3eN. The last outimaged the other 2, otherwise A11ms was best, Eikona last.
dave
Do note that the A6 you point to is only similar in size and uses the same basket. But the A6.2m smokes the old A6.
I went thru 3 sets of JX92s, couldn’t believe that the one i was hearing had significant ringing issues that had me ove them on. The Eikona is better, but still has the same issues.
We did a shootout at teh last VI diyFest between Eikona, Alpair 11ms (stock), and A10.3eN. The last outimaged the other 2, otherwise A11ms was best, Eikona last.
dave
Speaker distortion is not a significant indicator of how it will sound. We are not especially sensitive to regular small amounts of low order distortion.
Some response variations can be fixed with a little EQ, some less so.