Hello all,
A few years ago, I started to be intrigued by the parameter called "Rms" (mechanical resistance), which is sometimes indicated by the manufacturer, and expressed in kg/s; I still am, hence this post on this excellent forum I discovered a few months ago.
Then the German engineer Bernd Timmermanns - who is responsible for the DIY-speaker magazine (Hobby-Hifi) - started also to speculate about it and concluded that, on the basis of his empirical experience with probably hundreds of speakers designs, a lower Rms factor corresponded usually to a cleaner, dryer and more articulated bass (in the adequate enclosure of course).
I then started to look at the various datasheets available for 15" woofers, this in order to have a standard point of comparison, and because I decided I would give it a try with the lowest Rms-15" woofers I would find among the 20 or so manufacturers then available on the Internet (subject to other relevant parameters such as Fs, Qys, Vas etc. suiting my needs).
I could not always get the Rms figure in question but an Italian engineer working for a reputed speaker manufacturer explained that it could be easily calculated from the more commonly available parameters by using the formula Rms=(2Pi x Fs x Mms)/Qms Interestingly, for a very few manufacturers who published all these parameters including Rms, the formula did not work!
This formula seems to be little known (at least, I've never seen it mentioned elsewhere), and it allows to establish interesting correlations, e.g. low moving mass, low resonant frequency and high Qms tend to lead to a low Rms. Those who looked into the Rms issue usually explain the lower Rms as a result of soft suspension and spider, limited air-compression effects behind the dust dome, better dissipating voice coil material and voice coil winding technology etc. etc.
My enquiry ended up with a couple of manufacturers who manage to produce 15" woofers with an Rms around 1 kg/s or less! I run tests with a 400liter BR enclosure with 2 such woofers per channel. Besides other qualities, what strikes me is indeed the bass resolution, which is very noticeable at reasonable or lower levels, if I compare it with various other designs I've built or heard (using hard PRO-suspensions) and for which the power level had to be turned up to loud levels to get an interesting "bass". The notable exception being a few (but not all of them) horn loaded systems.
In another post, the Qms parameter was discussed http://www.diyaudio.com/forums/showthread.php?threadid=120505&highlight=
but has anybody else looked into the Rms issue as such ?
(Sorry if this has already been discussed elsewhere; my search ended up with too many irrelevant posts. Should I have missed something, hints welcome!).
A few years ago, I started to be intrigued by the parameter called "Rms" (mechanical resistance), which is sometimes indicated by the manufacturer, and expressed in kg/s; I still am, hence this post on this excellent forum I discovered a few months ago.
Then the German engineer Bernd Timmermanns - who is responsible for the DIY-speaker magazine (Hobby-Hifi) - started also to speculate about it and concluded that, on the basis of his empirical experience with probably hundreds of speakers designs, a lower Rms factor corresponded usually to a cleaner, dryer and more articulated bass (in the adequate enclosure of course).
I then started to look at the various datasheets available for 15" woofers, this in order to have a standard point of comparison, and because I decided I would give it a try with the lowest Rms-15" woofers I would find among the 20 or so manufacturers then available on the Internet (subject to other relevant parameters such as Fs, Qys, Vas etc. suiting my needs).
I could not always get the Rms figure in question but an Italian engineer working for a reputed speaker manufacturer explained that it could be easily calculated from the more commonly available parameters by using the formula Rms=(2Pi x Fs x Mms)/Qms Interestingly, for a very few manufacturers who published all these parameters including Rms, the formula did not work!
This formula seems to be little known (at least, I've never seen it mentioned elsewhere), and it allows to establish interesting correlations, e.g. low moving mass, low resonant frequency and high Qms tend to lead to a low Rms. Those who looked into the Rms issue usually explain the lower Rms as a result of soft suspension and spider, limited air-compression effects behind the dust dome, better dissipating voice coil material and voice coil winding technology etc. etc.
My enquiry ended up with a couple of manufacturers who manage to produce 15" woofers with an Rms around 1 kg/s or less! I run tests with a 400liter BR enclosure with 2 such woofers per channel. Besides other qualities, what strikes me is indeed the bass resolution, which is very noticeable at reasonable or lower levels, if I compare it with various other designs I've built or heard (using hard PRO-suspensions) and for which the power level had to be turned up to loud levels to get an interesting "bass". The notable exception being a few (but not all of them) horn loaded systems.
In another post, the Qms parameter was discussed http://www.diyaudio.com/forums/showthread.php?threadid=120505&highlight=
but has anybody else looked into the Rms issue as such ?
(Sorry if this has already been discussed elsewhere; my search ended up with too many irrelevant posts. Should I have missed something, hints welcome!).
Rms is the lumped mechancal resistance of the moving assembly. Looking at the formula
Rms=(2Pi x Fs x Mms)/Qms
is very misleading because Rms is a fundamental physical parameter and Qms is derived from that (and other things). This is like saying if we lower the 0-60 time of a car the horsepower goes up.
Another reason why this formula is misleading is that both Fs and Qms are functions of Mms (and other things). That is, the quantities on the right side are not independent variables. In the loudspeaker model, Mms and Rms should be thought of as independent quantities, though in practice changing one may change the other (though not in a rigidly set way). For example, you could take any fixed driver and fasten an additional 50 grams of something (clay, hot glue, whatever) on the cone on the outside- this would up the moving mass but hardly change the mechanical resistance at all. This is consistent with your equation because raising the moving mass while keeping other physical entities constant also lowers Fs and raises Qms.
As for correlation with subjective performance, personally I think there are far more important issues than Rms. Rms only significantly effects the linear behavior of the speaker near resonance and of the nonlinear distortion causing mechanisms, Rms variation is a very small factor (compared to Cms on the low end, Bl everywhere, and Le on the high end).
Rms=(2Pi x Fs x Mms)/Qms
is very misleading because Rms is a fundamental physical parameter and Qms is derived from that (and other things). This is like saying if we lower the 0-60 time of a car the horsepower goes up.
Another reason why this formula is misleading is that both Fs and Qms are functions of Mms (and other things). That is, the quantities on the right side are not independent variables. In the loudspeaker model, Mms and Rms should be thought of as independent quantities, though in practice changing one may change the other (though not in a rigidly set way). For example, you could take any fixed driver and fasten an additional 50 grams of something (clay, hot glue, whatever) on the cone on the outside- this would up the moving mass but hardly change the mechanical resistance at all. This is consistent with your equation because raising the moving mass while keeping other physical entities constant also lowers Fs and raises Qms.
As for correlation with subjective performance, personally I think there are far more important issues than Rms. Rms only significantly effects the linear behavior of the speaker near resonance and of the nonlinear distortion causing mechanisms, Rms variation is a very small factor (compared to Cms on the low end, Bl everywhere, and Le on the high end).
You may do it yourself
I mean, do some mods on the driver. Silicon oil can soften the suspension to some extent. And spider can be further softened by some cuts. I've tried on 2 fullrangers and 1 guitar driver, not much change, though. Only a few Hz drop in fs, IIRC. Maybe I should do it more aggressively....
If used on OB, a proper baffle size provides extra air load to the cone, and this help dropping fs significantly. The fs of my woofers (Eminence SigmaPro18) is rated at 28Hz, it's measured as rated in free air and on small baffle. In larger baffle (my center sub), it drops to 19Hz.
So, the fs of more commonly used pro 15's is often rated 30-some, it should be dropped to 20-some when properly loaded. That should be good enough.
All t/s parameters are in compromise. Low fs means heavy cone or soft suspension. And they mean low efficiency or fragile - both are no good in pro sound usage.
And actually they are not that bad.
Interesting data CLS, thanks for sharing. How big's your baffle? I forget the OD of your mid horns.
Personally I'm in the low Rms camp but having poked at the T/S maths quite a bit I've found it more or less impossible to say if it's really Rms which matters or just that drivers with lower Rms tend to correlate with other attributes that favor better reproduction of transients. My guess is much of it is Mms but I've not worked with enough drivers to do more than speculate.
Personally I'm in the low Rms camp but having poked at the T/S maths quite a bit I've found it more or less impossible to say if it's really Rms which matters or just that drivers with lower Rms tend to correlate with other attributes that favor better reproduction of transients. My guess is much of it is Mms but I've not worked with enough drivers to do more than speculate.
My baffle of main channel is 55cm wide x 60cm high, which is very small (narrow) for an 18", thus not much loading.
OTOH, the baffle of center sub is 152cm wide in total x 60cm high, for 2x 18". The whole width of it is folded slightly to form a recessed V shape and completed width is 136cm. In addition, there's some more area of top plate, and it's located very near the wall, thus much more air loading to the drivers.
OTOH, the baffle of center sub is 152cm wide in total x 60cm high, for 2x 18". The whole width of it is folded slightly to form a recessed V shape and completed width is 136cm. In addition, there's some more area of top plate, and it's located very near the wall, thus much more air loading to the drivers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.