Definitely, If there were 8" high sensitivity drivers (>94 dB) that played from 50Hz to 1500Hz then without issues then I would love to have one!!
that driver seems to be close:
http://www.jantzen-audio.com/download/PD_JA8008_Oct07_v1-f.pdf
unfortunately a bit too low Qms
http://www.jantzen-audio.com/download/PD_JA8008_Oct07_v1-f.pdf
unfortunately a bit too low Qms
'Optimum' cone size is going to depend on system design and performance goals and where and how well the performance tradeoffs are made. If you want to use a 21" to 1k, you are going to need to design a special driver, and it will still beam lower because of size, but using a more conventionally sized, say 15" to 500Hz is usually no issue at all for pro oriented drivers. Some of the dedicated sub drivers don't look like they'd be all that good that high, but there is a dearth of measurements on this as most applications have little use for FR data that high.
The inductance need not be all that high with good motor design; my AE TD 15's have 0.3mH and very good Le(x). There are plenty of other pro drivers with good Le and still have high sensitivity and power handling, due to good motor design and features such as shorting rings.
I could not agree less.
Could you explain your logic behind this comment.
A lot higher than that Doug, as you well know. Maybe you meant 1k?
I haven't used 21's, mainly 15 and 18" but I would otherwise agree with good drivers 350Hz is no problem at all for an xover, and for some drivers, much higher is possible.
John @ AESpeakers makes great drivers. Even the AV series have decent measurements up past 500Hz and those are high end subwoofer drivers that can play from 15Hz to 500Hz...that is sweet.
Can you please elaborate more on that ? AFAIK, there are speakers with crossover in the 100Hz region (Linkwitz' Orion), what would be the big NO-NO there ?
I am very very new to DIY speakers. But woudn't bass driver size also be dependant on room size? My living/audio room is about 4meter wide, 6 meter long on 3 meters high.
What I'm trying to say is that I think that bigger is better for a woofer when you play in a very large room or outdoors. Indoors or in smaller rooms there might be other issues as well.
What I'm trying to say is that I think that bigger is better for a woofer when you play in a very large room or outdoors. Indoors or in smaller rooms there might be other issues as well.
Yes, there is always speakers with all kind of XO's and drivers. I do not criticize any speakers. To me, 100 Hz XO sounds like speakers and a subwoofer.
Hi, Brett.
Not sure what you mean by "logic".
Here is a deffinition of baffle diffraction step loss: BDS
This formula gives you the start point for the transition from a driver radiating from 2π space to 4π space that is -3dB.
f3 = 4560 / Wb (width of baffle in inches)
A 20" baffle yields an f3 of about 228 Hz. That loss or step function kind of hangs in the middle of the mid bass region, so you are left with a discontinuity in mid bass response.
A smaller driver with a smaller baffle could, in theory, be located at the bass crossover frequency. So, if you used a 8" driver with a 10" baffle you end up with an f3 baffle step response of 456 Hz.
That may be a good choice for a crossover point, where you can adjust driver sensitivity between the woofer and a midrange to compensate for the step loss. With a 15" driver with a wider baffle, 228 Hz is not such an ideal choice for that crossover point.
The actual step loss rate at the f3 point is theoretically 6 dB an octave, but the room has a part in that and the actual loss may be closer to 3 dB per octave.
You probably knew all of that, but I like to blather and that was my "logic" with the statement.
Nice answer, thanks for helping the guy that is new to DIY. 🙄
In any size room is all about displacement, the bigger the room the further you are back and SPL drops approx. 6 dB for double the distance. It becomes more about the max SPL potential/Sensitivity of a driver when the distance increases.
There is just nothing that can do displacement like a quality large driver. You need multiples of smaller drivers but that is a compromise to me also.
If WAF is not any issue then there are 12" and 15" woofers that serve this purpose better then 8" woofer combinations. Ignoring line array solutions.
Thanks, so if we have a baffle that is 16" wide we would just acommodate this loss in the Crossover design? Ie. shelfing filter to match the -6dB slope starting at 285Hz.
Last edited:
Well, yes, if that is where you want to crossover. I am no expert, but there are a lot of balls to juggle in any design. Filter order, phase considerations, polar and vertical lobbing, and the effect of locating a crossover point in regions of the audio spectrum are some of the issues besides step response. Additionally, it is hard to predict exactly the slope of the step because the room gain plays a part in this.
As mentioned before, mid bass is generally a region where you want as much continuity as possible. The same goes with the region for human voice. Ideally, you try to set those crossover points where they least conflict with the audio spectrum that we tend to listen to most critically.
However, your proposal is one good solution. A better solution might be a filter between the amp and preamp. This is better for the DIY builder, but not as portable as it would be if it was already rolled into the crossover and/or through clever driver selection. Speaker manufactures, if they address the diffraction issue at all, tend to compromise and try to deal with it in the speaker design. However, one "size" does not fit all.
I tend to try to minimize the number of passive crossover components as much as possible to keep distortion and driver isolation from the amp as low as possible. Yes, an active crossover is the best if you can do it, too.
Another benefit for the preamp/amp filter is that you can make it variable with pots so that you can dial it in to work with your room. Every room is different and the amount of step loss varies with the room and the speaker location in that room.
Other tricks to reduce diffraction effects include rounding the corners of the cabinets as much as possible and locating the driver off-center of the baffle.
I chose a somewhat different approach by making the baffle shape a trapezoid.
It's pretty easy to predict where the step response begins in the spectrum compared to predicting the magnitude of that step.
Last edited:
I do not think its a matter of where you want to crossover. BSC is a normal thing period and it has to be handled with in the bass drivers region. The point about larger baffles is that the -6dB drop is down in the lower frequencies and needs to be handled. The more narrow the baffle the more you push that Baffle drop out of the range of the woofer.
In passive designs this just takes more work in active designs is very simple, I love active designs because they are simple.
In passive designs this just takes more work in active designs is very simple, I love active designs because they are simple.
Last edited:
Yes, that was what I was saying. My point is that if you can make the baffle step and the crossover point align you can (or may-be-able is a better phrase) to manage both at the same time.
With a larger woofer, as you pointed out, that pushes the crossover point lower (if you want it to coincide with the baffle step) into a region of the spectrum that may not be a good choice for the performance of the system. That can be a compromise for a larger woofer versus a smaller woofer.
Does that mean that it can't be done? No, just about every imaginable combination exists either in the commercial market or some DIYer's home. Everything is a trade off and larger woofers have good attributes and some attributes that make the design less than perfect. Baffle step is just one such attribute that needs to be considered.
That is not a slope per se : when going from 2Pi to 4Pi you loose 6dB, once, no more no less.
The slower the 2Pi/4Pi transition (ie the bigger the baffle radii compared to the considered wavelength), the slower and smoother the -6dB transition "slope".
And then at some lower frequencies the wall behind the speakers comes into play (the closer the wall the higher the frequency), and you are back to 2Pi.
And after that (or even before in a small room) the whole room will also enter the game and that whole 2Pi/4Pi thing will not have any meaning anymore.
Last edited:
I've read all your posts, and will comment once as I hate huge quotes in replies. I understand BSC. So, your earlier statement "A 15" driver with a baffle width of 20" puts the baffle step at about 200 Hz, which is not a good point in the spectrum for a step response nor a crossover point" states that 200Hz is 'bad' but so far you have given no reason why other than a generic explanation of BSC. That was what I was asking.
As you have no idea of the design compromises in any given project, what I wanted to know was how can you make such a blanket statement? I see just as many compromises in say using 4 8" drivers on a baffle with a 450Hz BSC transition region, as I do with a 15" driver on a baffle with a BSC at 200Hz. If fact with all my own designs, 200Hz would be below the xover, leaving the mid unaffected, and leaving a large, capable 15" to do the heavy lifting below. I also see no particular problem with not having the BSC at xover - design choices and compromises.
- Status
- Not open for further replies.