http://home1.stofanet.dk/cfuttrup/dpc/dpc.htm
boring text
but i think it shows somewhere how air provides a lack of a 'load' ,higher Vd is required for same SPL
I just trust it to be correct,i dont bother with extreme detail
boring text
but i think it shows somewhere how air provides a lack of a 'load' ,higher Vd is required for same SPL
I just trust it to be correct,i dont bother with extreme detail
Hi all,
For a piston operating in its linear range (below cone breakup), and in the far field (essentially behaving as a point source; near field/far field transition is calculated as d^2/l, where d is the effective radiating diameter is and l is the wavelength reproduced), acceleration of a driver corresponds with SPL (cf. Danley, Toole, Geddes, Beranek, Kinsler, etc). Fundamentals of Acoustics by Kinsler, Frey, Coppens, and Sanders is a really good reference text that goes into the gory detail of deriving this.
Anyway, what that boils down to is that basically acceleration = SPL. Flat SPL over frequency means flat acceleration over frequency. The acceleration of a driver is its SPL.
What is the velocity doing over frequency? Well, velocity is the integral of acceleration over time (dV/dt = acceleration). Thus if acceleration/SPL is flat, velocity is increasing as you go down in frequency (up in time). In fact, if you double your time base (halve the frequency - drop an octave), you double the velocity.
What does position do over frequency? Position is the integral of velocity over time (dP/dt = velocity). Thus if the velocity is doubling every time you double the time (drop an octave), position is quadrupled.
Thus: for a given driver, and assuming the frequency response of the piston is flat and the SPL output is held constant, if you double the frequency you quarter the required displacement. If you halve the frequency you quadruple the required displacement.
This means if you're generating 100 dB SPL @ 500 Hz with 0.1mm of excursion, you'll need 0.4mm of excursion at 250 Hz, 1.6mm of excursion at 125 Hz, 6.4mm of excursion at 62.5 Hz, and 25.6mm of excursion at 31.25 Hz.
Obviously, SPL also scales with cone size, so that if you double the cone size you halve the required displacement. This is why we use large diameter drivers for bass - allows higher SPLs with less excursion. But you still have to consider that excursion requirements for flat frequency response increase by a factor of 4 for every octave you drop.
Dan Wiggins
Adire Audio
For a piston operating in its linear range (below cone breakup), and in the far field (essentially behaving as a point source; near field/far field transition is calculated as d^2/l, where d is the effective radiating diameter is and l is the wavelength reproduced), acceleration of a driver corresponds with SPL (cf. Danley, Toole, Geddes, Beranek, Kinsler, etc). Fundamentals of Acoustics by Kinsler, Frey, Coppens, and Sanders is a really good reference text that goes into the gory detail of deriving this.
Anyway, what that boils down to is that basically acceleration = SPL. Flat SPL over frequency means flat acceleration over frequency. The acceleration of a driver is its SPL.
What is the velocity doing over frequency? Well, velocity is the integral of acceleration over time (dV/dt = acceleration). Thus if acceleration/SPL is flat, velocity is increasing as you go down in frequency (up in time). In fact, if you double your time base (halve the frequency - drop an octave), you double the velocity.
What does position do over frequency? Position is the integral of velocity over time (dP/dt = velocity). Thus if the velocity is doubling every time you double the time (drop an octave), position is quadrupled.
Thus: for a given driver, and assuming the frequency response of the piston is flat and the SPL output is held constant, if you double the frequency you quarter the required displacement. If you halve the frequency you quadruple the required displacement.
This means if you're generating 100 dB SPL @ 500 Hz with 0.1mm of excursion, you'll need 0.4mm of excursion at 250 Hz, 1.6mm of excursion at 125 Hz, 6.4mm of excursion at 62.5 Hz, and 25.6mm of excursion at 31.25 Hz.
Obviously, SPL also scales with cone size, so that if you double the cone size you halve the required displacement. This is why we use large diameter drivers for bass - allows higher SPLs with less excursion. But you still have to consider that excursion requirements for flat frequency response increase by a factor of 4 for every octave you drop.
Dan Wiggins
Adire Audio
sorry. thanks for the correction.Roscoe Primrose said:
DanWiggins said:
thanks dan. now lets assume we know Sd, Xmax, Sensitivity, and freq.
say Sd = A cm2, Xmax=B mm, Sens=C db/1w/1m, and freq = D Hz. what would be the max SPL before linear xmax is reached? Or do we need to know more?
so far we know that SPL = C + (A x B x F sq.) /Mx where Mx is some unknown qty. in the example dan wiggins has shown Mx would be 2500 if C was 90. obviously we dont know Mx yet.
navin said:
Your "C" above is irrelevant to determining max SPL in conditions where Xmax is the limiting factor. A driver that's 85dB/w/m with a given Sd and Xmax has the same excrusion limited max output at a given frequency as a driver with the same Sd and Xmax that's 100dB/w/m, it just takes a lot less power to get the second driver to reach that Xmax...
Peace
Roscoe Primrose said:
Yep... the only time sensitivity places any limit on max SPL is if
it's so low the driver reaches it's thermal limit well before Xmax or
if the available amplifier power is not enough to drive to Xmax.
The later all to often being the case with high excursion low sensitivity drivers.
now lets assume to drivers same Sd, Xmax, etc... but one with 85db/1W/1m and the other at 100db/1W/1m.
since Sd, Xmax etc... is the same we can assume that each driver will reach Xmax at 100hz with an input of 20W. correct?
now 20W is the same as 13dbW. add 13dbW to 85db and you get a max SPL of 98db. add 13db to 100db and you get a max SPL of 113db.
where am i going wrong?
since Sd, Xmax etc... is the same we can assume that each driver will reach Xmax at 100hz with an input of 20W. correct?
now 20W is the same as 13dbW. add 13dbW to 85db and you get a max SPL of 98db. add 13db to 100db and you get a max SPL of 113db.
where am i going wrong?
navin said:
This is where you are going wrong -
"since Sd, Xmax etc... is the same we can assume that each driver will reach Xmax at 100hz with an input of 20W. correct?"
They do not reach their Xmax (or any given SPL) with equal input
power because their sensitivities are different. That's what sensitivity is all about. However for any equal excursion @ equal frequency the 2 drivers will put out equal SPL but it will take a higher electrical watts drive to the lower sensitivity speaker to
equal any excursion (SPL) of the higher sensitivity driver. Get it.
SD & Xmax are the mechanical properties that determine max SPL
capability of the speaker. Sensitivity in terms of dB/watt is merely an efficiency rating for how efficiently the electrical power input is transferred to mechanical (acoustic) power output.
Stephen D said:
sorry for being so dim....
i think we are saying the same thing....
the lower sens speaker will require more watts to reach same SPL (if it has the required Xmax) or the lower sens speaker will produce less SPL given the same watts (if both driver have same Xmax or if neither driver has reached it's Xmax).
You're both pretty much correct now - Sensitivity (actually, efficiency) tells you how much power is needed to get a given SPL level out. Xmax and Sd tell you how much potentialSPL you can get.
And of course, in the range from ~1 octave above Fb (or Fc) and down, the efficiency of the system is a function of the subwoofer box, NOT the driver. So for subs even the driver's efficiency is typically not that important. The box is.
Dan Wiggins
Adire Audio
And of course, in the range from ~1 octave above Fb (or Fc) and down, the efficiency of the system is a function of the subwoofer box, NOT the driver. So for subs even the driver's efficiency is typically not that important. The box is.
Dan Wiggins
Adire Audio
Greetings all,
I have been away and busy with other things for a while now. Sorry for the delay, but am trying to catch up.
Dan, thanks for the title. I have ordered it and plan to read it. While I work in the world of application, I am a theoretician. Your statement that acceleration is SPL bothers me. Seems to not meet both conditions of necessary and sufficient when you model complex reproduction. It also seems to require logical constraints not posted. Assumptions not made explicit. You are probably correct in your logical argument and the statement is probably correct, but I believe I need to go back to the theoretical foundations and I need to think about this for a while.
Anyway, I believe I do need to do some modeling before I conduct my experiment for this driver. Or maybe I do not need to conduct it except to verify theory. I may have missed it, but despite all the posts and how long ago the last post was, it seems that we do not have an answer to Navin’s question of SPL to Xmax. Have I missed something? Is there some specification missing that would be required to “calculate” SPL at Xmax at Fs? I believe that it would be a worthwhile goal to answer the question.
Mark
I have been away and busy with other things for a while now. Sorry for the delay, but am trying to catch up.
Dan, thanks for the title. I have ordered it and plan to read it. While I work in the world of application, I am a theoretician. Your statement that acceleration is SPL bothers me. Seems to not meet both conditions of necessary and sufficient when you model complex reproduction. It also seems to require logical constraints not posted. Assumptions not made explicit. You are probably correct in your logical argument and the statement is probably correct, but I believe I need to go back to the theoretical foundations and I need to think about this for a while.
Anyway, I believe I do need to do some modeling before I conduct my experiment for this driver. Or maybe I do not need to conduct it except to verify theory. I may have missed it, but despite all the posts and how long ago the last post was, it seems that we do not have an answer to Navin’s question of SPL to Xmax. Have I missed something? Is there some specification missing that would be required to “calculate” SPL at Xmax at Fs? I believe that it would be a worthwhile goal to answer the question.
Mark
David,
As of February 9, 2004, I had the drivers in my possession, had listened to them, tested them, and posted those tests to the thread.
While I do not like the sound of the stock unit (smeared midrange and lower treble that is too prominent), they have potential that is not yet realized. I believe I can modify them to correct many of the problems and make them a very good full range driver. I have not, however, gotten around to doing that yet. Here is the important point...I would not consider modifying this driver if I believed there was a better driver available for under 120 US dollars per pair.
Much of the recent discussion has centered around how loudly they will play at low frequencies. This is a valuable consideration when determining appropriate applications for the driver since how loudly they will play is connected to how they sound. We are just trying to work out some form of communication that is not entirely subjective.
So, question back to you...Considering how inexpensive the stock driver is, have you purchased a pair and listened to them?
Mark
As of February 9, 2004, I had the drivers in my possession, had listened to them, tested them, and posted those tests to the thread.
While I do not like the sound of the stock unit (smeared midrange and lower treble that is too prominent), they have potential that is not yet realized. I believe I can modify them to correct many of the problems and make them a very good full range driver. I have not, however, gotten around to doing that yet. Here is the important point...I would not consider modifying this driver if I believed there was a better driver available for under 120 US dollars per pair.
Much of the recent discussion has centered around how loudly they will play at low frequencies. This is a valuable consideration when determining appropriate applications for the driver since how loudly they will play is connected to how they sound. We are just trying to work out some form of communication that is not entirely subjective.
So, question back to you...Considering how inexpensive the stock driver is, have you purchased a pair and listened to them?
Mark
I'm wondering that it may be a mistake to think of these really as a full range driver (as in to be used without tweeter). It is after all listed in the woofer section at Parts Express. Perhaps the over optimistic response shown in that 1st graph they sent me has led us to expect it to be something other than it's true intent.
regarding fullranges i have a few questions.
seems that a driver that can operate from 100Hz to 20k is dificult to find (the Manger is XOed about 300Hz and the Fostex 6" and 8" and Jordan JX92 have limited hf response).
in that case there are 2 options. find a driver than can operate from the bass to about 6-8k and then augment the hf response of this driver (JX92, Fostex 103, 206 etc...) or find a driver that can operate from 250Hz to 20k and augment the bass response of this driver (JX53, TB 871, etc...). which would u prefer.
the more i think about it I would prefer the later as teh XO freq between the 2 drivers would be at longer wavelengths than the c-c distance of teh 2 drivers.
can anyone recomend drivers that have better hf dispersion & more neutral midrnage than the 871 or jx53 and can operate down to 250Hz (6db)?
seems that a driver that can operate from 100Hz to 20k is dificult to find (the Manger is XOed about 300Hz and the Fostex 6" and 8" and Jordan JX92 have limited hf response).
in that case there are 2 options. find a driver than can operate from the bass to about 6-8k and then augment the hf response of this driver (JX92, Fostex 103, 206 etc...) or find a driver that can operate from 250Hz to 20k and augment the bass response of this driver (JX53, TB 871, etc...). which would u prefer.
the more i think about it I would prefer the later as teh XO freq between the 2 drivers would be at longer wavelengths than the c-c distance of teh 2 drivers.
can anyone recomend drivers that have better hf dispersion & more neutral midrnage than the 871 or jx53 and can operate down to 250Hz (6db)?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.