My current speaker is a little bit too lean. I think when I made the cabinet, the volume was optimized for just slightly over-damped and the result is a sound that a bit lean. I would like to reduce the volume by around 7liters. I am thinking of stuffing inside the cabinet with solid foam. Will that work? It seems like the simplest solution.
The other option is putting in small sand bags and hang them by the cabinet braces.
The other option is putting in small sand bags and hang them by the cabinet braces.
over dampened typically means that the cabinet is smaller and thus the Q is higher. I think there is a lot of confusing information out there about this, which I can only guess comes from old vs new design. Many designers will call .5 critically dampened, some even go so far as to say .7 is critically dampened. They refer to this as high Q, and over dampened, then say that 1 or higher is very high Q and very over dampened. There also seems to be some confusion as to what the QTC of the sealed alignment does to the sound, descriptions which seem to almost conflict with each other.
Ok so if the box has a Q of .7 or higher is what I would call high Q, the woofers are better controlled, offering tighter bass, subjectively dryer, but with a sharper roll over and greater extension (sort of). Lower Q, lets say .5, is a sort of mid way balance, giving less cone control, a shallower roll of, but usually a lower -10 and say -20db point. When taking into account room gain, these are often more important than the -3db anechoic point.
Now if you want a smaller box because you want to raise the Q and thus make the bass tighter, better controlled, have a lower -3db point, but steeper roll off, then what you suggest is a solution. If on the other hand you actually want to lower the Q, in order to extend the bass with a shallower roll off, and give a somewhat looser (not always a bad thing) less controlled bass (Less controlled because its currently over controlled-too little overhang), then you actually need to enlarge the box.
To make it smaller you want to use something solid that takes up volume. The speaker will see open cell foam, even dense open cell foam, as a larger box. Open cell foam will do the opposite of what you want. Same for poly fill, fabric, etc. It's even possible that sand could have this effect, as I'm not really sure how air passes through sand. If it can pass through it, but would be slowed, and have some of its energy converted into heat, it will then, as I understand things, also lower the Qtc. The best solution is to simply block a portion of the box off or add lots more bracing to take up excess volume. I suppose a 2x4 or 4x4 could be used as a brace, and become your friend here.
If you actually meant that you wanted to lower the Q and have the effects of a lower Q, use foam and polyfill.
If you can give me specs on the woofer being used, and the cabinet volume, I can run simulations to see the Q, and I can run the effect of a few different dampening materials to see the effect on Q.
Ok so if the box has a Q of .7 or higher is what I would call high Q, the woofers are better controlled, offering tighter bass, subjectively dryer, but with a sharper roll over and greater extension (sort of). Lower Q, lets say .5, is a sort of mid way balance, giving less cone control, a shallower roll of, but usually a lower -10 and say -20db point. When taking into account room gain, these are often more important than the -3db anechoic point.
Now if you want a smaller box because you want to raise the Q and thus make the bass tighter, better controlled, have a lower -3db point, but steeper roll off, then what you suggest is a solution. If on the other hand you actually want to lower the Q, in order to extend the bass with a shallower roll off, and give a somewhat looser (not always a bad thing) less controlled bass (Less controlled because its currently over controlled-too little overhang), then you actually need to enlarge the box.
To make it smaller you want to use something solid that takes up volume. The speaker will see open cell foam, even dense open cell foam, as a larger box. Open cell foam will do the opposite of what you want. Same for poly fill, fabric, etc. It's even possible that sand could have this effect, as I'm not really sure how air passes through sand. If it can pass through it, but would be slowed, and have some of its energy converted into heat, it will then, as I understand things, also lower the Qtc. The best solution is to simply block a portion of the box off or add lots more bracing to take up excess volume. I suppose a 2x4 or 4x4 could be used as a brace, and become your friend here.
If you actually meant that you wanted to lower the Q and have the effects of a lower Q, use foam and polyfill.
If you can give me specs on the woofer being used, and the cabinet volume, I can run simulations to see the Q, and I can run the effect of a few different dampening materials to see the effect on Q.
PJpoes,
Thanks for the reply. The current volume is about 45 liters.
I am using MTM configuration with two Vifa PL18 6.5in. drivers. I guess what I want is a slightly warmer sound.
Here a pic 🙂
Thanks.
Thanks for the reply. The current volume is about 45 liters.
I am using MTM configuration with two Vifa PL18 6.5in. drivers. I guess what I want is a slightly warmer sound.
Here a pic 🙂
Thanks.
Warmth is a subjective term, and its often hard to know for sure what it is people want. Now the way its used in speaker design with regard to total Q is a raise in Q and thus the steeper roll off.
Please excuse the quality of the image, my normal computer is in the shop, so to speak.
Anyway, the yellow line represents your box. Its basically a critical Q box, with roughly a .5 Q. On the positive side it has a nice group delay, smooth transfer function phase, and the most extended usable response. On the negative it offers the least driver control, and so with no eq, and fed a pink noise signal, it could only handle roughly 20 watts before reaching xmax. Of course real world music doesn't contain that much even low frequency energy, but basically beyond that, say with 50 watts, then you are well beyond the speakers xmax from 50hz on down. On the other hand, from 50hz on up you are at 103 db's or so with 20 watts, which is very loud.
The Red line is with a Q of .581 and is for a 25 liter box filled with polyfill roughly 75% of the way. As you can see, this made a pretty minor change, but required reducing your box volume by over double what you want to do. You do get a little more power handling now when the speaker are fed with this unrealistic signal. It moves to 30 watts before breaking xmax, and with 100 watts, the point of breaching xmax is now more like 45hz or so.
Ok the green line represents an equal ripple alignment of roughly Q=.8 and is achieved in an 11 liter box filled 50% with fill. This is the change I would recomend in order to achieve the warmer sound you want. I think that any other change is going to be trivial at best, and this is just the nature of that vifa driver (lower driver Qts). I know this probably isn't what you wanted to hear, as it means a massive reduction in volume, but I do see it as the only solution best I can tell. Now, to achieve this without a permanent change, I would recomend cutting a solid wood brace, of sorts, that fits inside the box (if possible) and temporarily secure it so as to reduce the volume the desired amount. You could try to find extremely dense foam, such as dense cushion foam, and see if it gives the effect, but since its still a type of open cell, I think it will do the opposite.
I tried modeling the box as if the air basically was the opposite of having polyfill added, and see if that would mimic adding dense foam, but could only raise the Q to .559, which isn't enough to warm up the sound much I don't think.
Basically to reduce the volume you need to fill up the box with solid objects as much as possible, to consume some of that volume. For testing purposes, anything solid will work. Try plastic bags of sand maybe instead of fabric (maybe air won't pass through the grains of sand).
Can you measure the Qtc? Like do you have any of the various measuring devices to do this.
My next suggestion after this would be a crossover change. Making the box smaller is a start, but even that will only go so far. My suggestion there would be to change your first inductor (Q1) in the woofer lowpass crossover to a larger value. For instance if its .8mh go to .9 or even 1. Also, if you use an RC filter at the first cap point in the crossover (capacitor with resistor), you can increase the resistor value a small amount.
Please excuse the quality of the image, my normal computer is in the shop, so to speak.
Anyway, the yellow line represents your box. Its basically a critical Q box, with roughly a .5 Q. On the positive side it has a nice group delay, smooth transfer function phase, and the most extended usable response. On the negative it offers the least driver control, and so with no eq, and fed a pink noise signal, it could only handle roughly 20 watts before reaching xmax. Of course real world music doesn't contain that much even low frequency energy, but basically beyond that, say with 50 watts, then you are well beyond the speakers xmax from 50hz on down. On the other hand, from 50hz on up you are at 103 db's or so with 20 watts, which is very loud.
The Red line is with a Q of .581 and is for a 25 liter box filled with polyfill roughly 75% of the way. As you can see, this made a pretty minor change, but required reducing your box volume by over double what you want to do. You do get a little more power handling now when the speaker are fed with this unrealistic signal. It moves to 30 watts before breaking xmax, and with 100 watts, the point of breaching xmax is now more like 45hz or so.
Ok the green line represents an equal ripple alignment of roughly Q=.8 and is achieved in an 11 liter box filled 50% with fill. This is the change I would recomend in order to achieve the warmer sound you want. I think that any other change is going to be trivial at best, and this is just the nature of that vifa driver (lower driver Qts). I know this probably isn't what you wanted to hear, as it means a massive reduction in volume, but I do see it as the only solution best I can tell. Now, to achieve this without a permanent change, I would recomend cutting a solid wood brace, of sorts, that fits inside the box (if possible) and temporarily secure it so as to reduce the volume the desired amount. You could try to find extremely dense foam, such as dense cushion foam, and see if it gives the effect, but since its still a type of open cell, I think it will do the opposite.
I tried modeling the box as if the air basically was the opposite of having polyfill added, and see if that would mimic adding dense foam, but could only raise the Q to .559, which isn't enough to warm up the sound much I don't think.
Basically to reduce the volume you need to fill up the box with solid objects as much as possible, to consume some of that volume. For testing purposes, anything solid will work. Try plastic bags of sand maybe instead of fabric (maybe air won't pass through the grains of sand).
Can you measure the Qtc? Like do you have any of the various measuring devices to do this.
My next suggestion after this would be a crossover change. Making the box smaller is a start, but even that will only go so far. My suggestion there would be to change your first inductor (Q1) in the woofer lowpass crossover to a larger value. For instance if its .8mh go to .9 or even 1. Also, if you use an RC filter at the first cap point in the crossover (capacitor with resistor), you can increase the resistor value a small amount.
Doh should have waited for your response. Ok hold on, basically what I said was true, but the graphs are wrong and the size change may also change. Ok hold on, I will get you new numbers.
Ok well now the changes are going to be different, as the system Q changes in response to the port and box volume change. Ok now I would say reduce box volume by 17 liters or so, get it down to say 28 liters and retune to 50hz. Then fill with roughly 60-70% polyfill. This will reduce overall bass extension, but bring about roughly 2-3db's more energy in the 100hz range. Warmth is in the 100-400hz area I would argue, and we are having no impact on the response above 100hz, so if doing this doesn't fix it, you need to change crossover values.
Linkwitz transform would be inappropriate here and would have no impact on warmth. He didn't ask for more bass extension, he asked for more warmth, i.e. energy in the upper bass lower midrange area.
However, lets just say that linkwitz transform did do what he wanted (more bass extension), it would require a much smaller box to control xmax. Even then, it would exceed it quite quickly and easily, and when taking into account room gain, would have too much bass energy. Main speakers really don't need much below 50-60hz anyway, anything from 80hz down should be handled by subwoofers throughout the room, not the main speakers.
However, lets just say that linkwitz transform did do what he wanted (more bass extension), it would require a much smaller box to control xmax. Even then, it would exceed it quite quickly and easily, and when taking into account room gain, would have too much bass energy. Main speakers really don't need much below 50-60hz anyway, anything from 80hz down should be handled by subwoofers throughout the room, not the main speakers.
pjpoes said:
The diameter is 3in. (I think ... I need to run home to check but I am pretty sure)
That would imply roughly a 45hz tuning, so yeah, to increase the energy in that range that I discussed, you need to reduce volume a lot, say 15-20 liters, and increase the tuning to roughly 50hz. This is my suggestion anyway. Simply reducing the volume reduces extension, but does nothing to increase output in that range about the tuning frequency.
The cheapest and most simple fix may just be a crossover modification. A cheap temporary way to try would be to use an inductor of the difference in the value you want and currently have. Say you currently have a .8mh inductor in L1 (or Q1), and you want to go to a .9, then simply wire in series with that first inductor a .1mh inductor. Temporarily wire it up, if you like the sound, go for it. In fact, I would see if you have a .2mh, and then just unwind the turns until you get the sound you like. For a more permanent solution, use a 1mh inductor and do the same thing. If you don't have an inductor laying around, then just buy the 1mh in the first place.
The cheapest and most simple fix may just be a crossover modification. A cheap temporary way to try would be to use an inductor of the difference in the value you want and currently have. Say you currently have a .8mh inductor in L1 (or Q1), and you want to go to a .9, then simply wire in series with that first inductor a .1mh inductor. Temporarily wire it up, if you like the sound, go for it. In fact, I would see if you have a .2mh, and then just unwind the turns until you get the sound you like. For a more permanent solution, use a 1mh inductor and do the same thing. If you don't have an inductor laying around, then just buy the 1mh in the first place.
Here's the current xover:
L1 is .82mh. I am sort of reluctant about changing it since it may affect the freq. response at the xover point.
An externally hosted image should be here but it was not working when we last tested it.
L1 is .82mh. I am sort of reluctant about changing it since it may affect the freq. response at the xover point.
I understand the concern, but it really wont. If you consider how a crossover works, essentially the first part sets up the first order roll off, each additional part then adds to that slope increasing the order. Changing the capacitor will have a larger impact on the actual crossover point, but the inductor change will really only raise and lower the midrange level. Now mind you that would put the tweeter out of range with the midrange, and would need reduction.
Without response files I can't really predict exactly how this would work, but basically, to use a popular term here, it changes the baffle step, and the tweeter would need to be adjusted accordingly.
If you want to try readjusting the box loading, then I would recomend reducing the box volume, as I suggested, and then using a port that is 6 inches long instead of 4. This is assuming the port is 3 inches long.
my simulation suggests that the 45hz tuning isn't optimal, so I'm wondering if that is correct, but assuming the port is 3x4, then go to 3x6. If the port turns out to be 2.5 or 2.6", then the tuning is lower, more like 40hz, and that looks better, in that case I can give you the correct length. If the port is 2.5" in diameter, with no flange or flare, then go to 4.5 inches in length.
Without response files I can't really predict exactly how this would work, but basically, to use a popular term here, it changes the baffle step, and the tweeter would need to be adjusted accordingly.
If you want to try readjusting the box loading, then I would recomend reducing the box volume, as I suggested, and then using a port that is 6 inches long instead of 4. This is assuming the port is 3 inches long.
my simulation suggests that the 45hz tuning isn't optimal, so I'm wondering if that is correct, but assuming the port is 3x4, then go to 3x6. If the port turns out to be 2.5 or 2.6", then the tuning is lower, more like 40hz, and that looks better, in that case I can give you the correct length. If the port is 2.5" in diameter, with no flange or flare, then go to 4.5 inches in length.
As I understand it:
Q=0.51 is critically damped
Q< 0.51 is overdamped
Q> 0.51 is underdamped
Rob.
Q=0.51 is critically damped
Q< 0.51 is overdamped
Q> 0.51 is underdamped
Rob.
cotdt,
Visit: http://en.wikipedia.org/wiki/Damping_ratio where the relationship between the Q factor and damping ratio = 1/(2xQ) is shown.
A Q = 0.5 results in a damping ratio of = 1 i.e. a critically ratio.
b
Visit: http://en.wikipedia.org/wiki/Damping_ratio where the relationship between the Q factor and damping ratio = 1/(2xQ) is shown.
A Q = 0.5 results in a damping ratio of = 1 i.e. a critically ratio.
b
- Status
- Not open for further replies.