No
https://www.bcspeakers.com/en/products/lf-driver/18/8/18SW115
https://www.bcspeakers.com/en/products/lf-driver/18/4/18SW115
Same flux density. It is the same motor.
No, force is measured in Newtons, not Tesla meters. You are looking into incomplete underlying data that participate in generating force. Unfortunately this piece of data can be misleading upon first glance.
If we are talking about motor foce, it is given that we attribute it to input power. We compare forces at same input power, namely 1Watt (for each driver compared) Unfortunately this Bl number can appear higher for weaker driver and vice versa.
B&C 18SW115_8 pushes 13,20 Newtons per 1 Watt at the cone.
B&C 18SW115_4 pushes 14,33 Newtons per 1 Watt at the cone.
...And that´s more in my book.
No, Refer to the links I provided.
Indeed. It is the same motor. Indeed the difference can be negligible, but if one has an amplifier to drive the driver to its full potential, and there is no other issue in the decision, why would he be taking weaker motor driver...
Indeed. Why would we be comparing drivers at different power inputs though? We do that at same input power.
No, not everything. Bunch of correlated, but not causal specs that don´t make up the output of the driver in the real sense.
Only case where 18SW115-8 would heat up less is with the same voltage, and at that point it would play less, bluntly being characterized as a less playing weaker driver if we don´t look at the underlying parameters and situation.
As counted before, the 4Ohm version pushes more Newtons at the cone per Watt of input power. That would result into higher accoustic pressure. In order to match the output of these two drivers, the 4Ohm version would need to get less power, and given the same physical and material properties, it would heat up less at that point.
No, and no need to. Physics works. I do not need to taste crap to know it is not tasty. Yet it is not supported by math and physics. This problematics is. Some manufacturers arguably did this, and came with the same results and opinions.
Most heat on the speaker is generated by the resistive load. It is so stupid, but if you remove the resistive load, you get less heat. And so low impedance drivers are on the rise in quite a few high end PA designs for this particular reason along others. Case in point, run an electrical kettle on a 0,2mm wire and on 1,5mm wire. The first one will cause fire. More resistance, more heat.
I hope I was successful with the explanations.
Last edited:
The 4 ohm version would be wound with less turns of larger gauge wire. It is likely a better, more optimum FIT for the magnet gap compared to more turns of smaller diameter wire, driven with the square root of two more voltage.
Looks like about 8% more..
I was using older spec sheets from ~2011 when I first purchased B&C drivers.
The smaller gap probably was a typo error.
B&C uses the equivalent "input power", 2v (4 ohm) and 2.83v (8ohm).
Using the new spec sheets, the 8ohm version still produces more acoustic pressure (SPL@1w/1m) than the 4 ohm version.
That said, the difference in nominal impedance to Re is ~16% in favor for the 8 ohm version, which would accounts for the sensitivity disparity in spite of the 8% increase in force.
Yes, you were.
The 4 ohm version probably uses almost 29 grams more wire.
I was aware that BL can be less on a lower impedance driver, yet still impart the same (or more) force, but don't know the math.
What is the formula you used to convert BL to Newtons per 1 Watt at the cone?
Thanks,
Art
Many high end designs people opt for the 4Ohm drivers because of additional motor strength, that can be exploited in a demanding design. In other cases, it might be more than 8%. It really makes a difference if the only action to do is to choose. Not a world of difference, but heck, better is better.
Yes, nominal impedance is not a best measure.
The formula is
F [Newtons] = "BL" x I; where current I is whole √ 1/R, don't know how to write it with keyboard. P [1W] = R x I^2 and you go from that, getting current I for the force equation.
It's a static force, which can be less relevant in real world due to the nonlinearity of suspension behavior in frequency and position, but it means something, and simulations with different motor strength support that.
What simulation doesn't give you is the suspension stifness progression.
I measure that separately. Still not very standarized and normalized, but it gives me a picture. In my test, for example RCF drivers spend much less power on suspension, usually drawing 120-170VA at Xmax, while B&C speakers eats in ballpark of double.
In certain cases it matters. It is stupid 150Watts here, hyperventing there, whatever design decision there, more motor efficiency there, and it is good 500Watts of difference in power input. A decibel is good decibel. Add it to port compression, and now you have three decibels.
These things matter...
Yes, nominal impedance is not a best measure.
The formula is
F [Newtons] = "BL" x I; where current I is whole √ 1/R, don't know how to write it with keyboard. P [1W] = R x I^2 and you go from that, getting current I for the force equation.
It's a static force, which can be less relevant in real world due to the nonlinearity of suspension behavior in frequency and position, but it means something, and simulations with different motor strength support that.
What simulation doesn't give you is the suspension stifness progression.
I measure that separately. Still not very standarized and normalized, but it gives me a picture. In my test, for example RCF drivers spend much less power on suspension, usually drawing 120-170VA at Xmax, while B&C speakers eats in ballpark of double.
In certain cases it matters. It is stupid 150Watts here, hyperventing there, whatever design decision there, more motor efficiency there, and it is good 500Watts of difference in power input. A decibel is good decibel. Add it to port compression, and now you have three decibels.
These things matter...
Hurrication,
The 18SW115-8 BL is 30.3Tm, squared is 918.09, divided by 6.5Re=141.24
The 18SW115-4 BL is 26Tm, squared is 676, divided by 4.1 Re=141.24=164.87
What units are 141.24 and 164.87?
Art
Some other good formulas: https://www.klippel.de/fileadmin/kl...rature/Papers/Green Speaker Design Part 2.pdf
Hi ! Cedia is over, so I'm hoping you can talk some more about your driver and the end product 🙂
Maybe 233C could be a great title for a sequel of some sort 🤔
Ack! True true, the simulators have constant voltage assumed to equal power, however the impedance curve changing means different real power. I'll have to let that percolate in my brain.
I gotta get a simulator up and running. I still suspect that generating more sensitivity at low frequencies for subwoofers is not much like trying to extend low bass in a fullrange.
wow, very nice ! congratulations on the release !
thats a very shallow box, and the VC seems to be in the front of the driver, enabling the shallow depth? seems like it has a good deal of excursion too, how much linear and mechanical does it have?
You have me very interested if you do end up selling to the general public.
70% BL was just a hair over 20mm one way, but it needs some dimensional tweaking because when pushed really hard the cone will smack the motor. Probably around 25-30mm of mechanical xmax once that's tweaked.
Right now, this specific sub will be sold by Tru as it is their project, so that will be the place to find one once they are in production.
Right now, this specific sub will be sold by Tru as it is their project, so that will be the place to find one once they are in production.
My mistake, substituted the "Minimum Impedance" of 4.1 ohms for the Re of 3.3 ohms.
The 18SW115-4 BL is 26Tm, squared is 676, divided by 3.3 Re=204.8 (whatever that unit is 😉 )
Given the scales, I feel I am back to my original interpretation (maybe not in this thread or even this forum*). A large variation in physical parameters makes little difference to output at the lowest frequencies, just 1-2 dB. So I feel like the really dominant thing is simply the cone area.
What happens if you take that Dayton UM15 and re-run the simulations changing only the Sd to an 18" and 12" and 10"? (Yes in real life that is nonsense without changing the other parameters but let us journey to Perfect Theory Land for some moments 🤔 )
*I can't remember at the moment, I think it was AVSForum, someone kindly modeled a couple very different 8" to go two units back-to-back in 1.3 cubic feet. The curves looked really similar even though one was IIRC a quite expensive LaVoce more-like-a-midbass and the other a much less expensive Eminence.
The real number to use is really closer to 4.1 than 3.3. 4.1 includes eddy current losses which are real, albeit at a higher frequency above resonance. You sort of have to extrapolate what it would be at fs, given that it tends to rise with frequency starting at 3.3 at DC, going up to 4.1 at whatever frequency “Zmin” is at, and continuing to go up with frequency above that.
You get more realistic sims if you use Zmin instead of Re, even for small signal. And as you heat up it just keeps on climbing.
I would have assumed Re stands for R effective, the part of the overhung voicecoil actually at work.
Surely Bl should as well only include the only the part of the voicecoil actuallty at work in the gap.
Otherwise any comparsion between different speakers with different ratios of overhung to gap depth would be unfair.
Especially in comparsion to an underhung coil.
No?
Surely Bl should as well only include the only the part of the voicecoil actuallty at work in the gap.
Otherwise any comparsion between different speakers with different ratios of overhung to gap depth would be unfair.
Especially in comparsion to an underhung coil.
No?
Last edited:
In terms of comparing Bl to motor force for different impedance drivers, one needs to use the same units for both, Re (3.3 ohms) rather than Zmin (4.1 ohms). Zmax at Fs is 60 ohms..
The minimum impedance in a cabinet will be lower than Zmin at Fb.
What it rises to with heat is another consideration, as the 8 ohm version will loose more motor force to heat than the 4 ohm version.
No, Re is the DC resistance of the voice coil, in ohms.
BL is the product of magnetic flux density in the voice-coil gap times the length of wire in the magnetic field, in Tesla-Meters.