Martin King replied to my very similar entry in a Facebook group [DIY Open Baffle Loudspeakers - Theory, Design, and Building]
A small observation here: This regards the sub-woofer and absolute low end. The midwoofer and midrange will still benefit from a low Cms.
A small observation here: This regards the sub-woofer and absolute low end. The midwoofer and midrange will still benefit from a low Cms.
Qms = Mechanical Loss Factor
Rms = Mechanical Resistance of driver losses
Mms = Moving Mass (including air-load)
Cms = Mechanical Compliance
Kms = Mechanical Resistance of the suspension alone, its inverse is Cms. Kms is the actual stiffness of the driver, but Qms is more concerned about the losses from that stiffness (Rms) - again most of which come from the Spider).
Kms/Cms, Rms, and Qms are all dependent on excursion. In another Qms thread I mentioned that Kms at rest is the most important parameter in relation to certain "low-level" subjective results when looking ONLY at the driver's suspension behavior (..and I think there are more important things than that - principally driver efficiency, though obviously you shouldn't focus exclusively on one driver aspect to the exclusion of other important aspects).
I think you can *relate Kms to Return/Restoring force, but given Qms's equation it's not part of it (or rather it's inversely part of it through Cms) .
*similarly I believe that Mms relates to Inertia and Rms relates to Friction.
Last edited:
You can demonstrate that stiffness is not the same as resistance if you've ever experimented with a hand held bicycle pump (without it being connected to anything)
As you operate it through it's stroke, you encounter resistance and the air becomes hot. On the other hand when you put your thumb over the hole, stiffness becomes the primary effect that you notice, and the piston is restored (as long as the pump doesn't leak).
You could also demonstrate that the stiffness is reactive by taking your thumb off the outlet and pumping quickly, then reducing your pressure. The piston is restored only a small amount meaning it requires an alternating input to invoke the reactance, otherwise the whole cylinder is expelled through the resistive outlet. (This would be like putting a capacitor in parallel with a resistor.)
As you operate it through it's stroke, you encounter resistance and the air becomes hot. On the other hand when you put your thumb over the hole, stiffness becomes the primary effect that you notice, and the piston is restored (as long as the pump doesn't leak).
You could also demonstrate that the stiffness is reactive by taking your thumb off the outlet and pumping quickly, then reducing your pressure. The piston is restored only a small amount meaning it requires an alternating input to invoke the reactance, otherwise the whole cylinder is expelled through the resistive outlet. (This would be like putting a capacitor in parallel with a resistor.)
In keeping with the above ^
Depending on the design of the loudspeaker (notably long-stroke larger drivers) that the resistance from the air passing through the gap can substantively exceed the resistance of all of the suspension at higher amplitudes (from what I *remember). Obviously excursion is key with Rms and resulting Qms in this instance.
*one of many, many Klippel white papers.
Depending on the design of the loudspeaker (notably long-stroke larger drivers) that the resistance from the air passing through the gap can substantively exceed the resistance of all of the suspension at higher amplitudes (from what I *remember). Obviously excursion is key with Rms and resulting Qms in this instance.
*one of many, many Klippel white papers.
The performance of a woofer depends on volume displacement (area*displacement). The degree of increase in the resonant frequency depends on the size (the area of the partition). Below the resonant frequency, a decrease in the frequency response occurs. The return at the resonant frequency depends on the quality factor.
The pressure is limited by the limit displacement. Compensated by an increase in the area (diameter) of the speaker or an increase in the number of speakers. This is stated on the site linkwitzlab.
The nature of the transient process (aperiodic or oscillatory) and the duration of attenuation depend on the quality factor.
The pressure is limited by the limit displacement. Compensated by an increase in the area (diameter) of the speaker or an increase in the number of speakers. This is stated on the site linkwitzlab.
The nature of the transient process (aperiodic or oscillatory) and the duration of attenuation depend on the quality factor.
The wavelength at low frequencies (up to several meters) requires very large sizes of decoration (impossible in a typical residential area) or the use of woofer arrays (as in street performances) to focus
Wavelength have nothing to do with it rather we are talking about lobes and nulls. A dipole as in two poles, will generate two nulls @ 90 degree off axis on either side, since the the electric phase is 180 degree with respect to one another. The special dipole known as RiPole after the inventor Axel Ridtahler, generates only one 180 out of phase region such that you end up with largely a cardioid pattern. SLOB aka Slot Loaded Open Baffle is a derivative of that.
Scroll up to post #22 or use this link: https://www.diyaudio.com/community/threads/conflicting-ob-driver-spec.392139/page-2#post-7173012
Ah I had not clicked on the scroll down symbol...so missed his answer.
Maybe you use activefilter ? The Fs and the Xmax will rule how much linkwitz transform is needed and related to that thee max usable spl (and the nearer of the Xmax the more Thd) In the upper frequencies this is more the diameter of the driver that rules the usable bandwidth and Qts became less relvant while better Xmax is still your friend.
Many people said than for the low end in OB the Le was not criticql in the choice, the best bass was given by very large Sd 18 and more , high Xmax and heavy cone (understand more stiff). But just do not know, I just rewrite what I readed if that can help you or trigger a specialist to react if false. 🙂
So the big Dayton with low Fs are often a good choice...
Maybe you use activefilter ? The Fs and the Xmax will rule how much linkwitz transform is needed and related to that thee max usable spl (and the nearer of the Xmax the more Thd) In the upper frequencies this is more the diameter of the driver that rules the usable bandwidth and Qts became less relvant while better Xmax is still your friend.
Many people said than for the low end in OB the Le was not criticql in the choice, the best bass was given by very large Sd 18 and more , high Xmax and heavy cone (understand more stiff). But just do not know, I just rewrite what I readed if that can help you or trigger a specialist to react if false. 🙂
So the big Dayton with low Fs are often a good choice...
What is a driver without control ? ... Some choose SPL over low distortion and linearity. Its all a matter of what one prefer. One big driver or many smaller .. We have wondered far from my initial post on conflicting EBP and I am about to read patent DE19830947A1 which covers the RiPole (cardioid) woofer arrangement. I also need to spend more time on Planar Dynamic Drivers such that I can start working on my own.
For me, this thread have been resolved thanks to M. King.
For me, this thread have been resolved thanks to M. King.
(*Generally) from a HIGH Cms (not low).
The higher the Cms the more compliant the driver (and hopefully Rms is very low).
*this depends on how close/steep your midrange's high-pass is to the driver's Fs/Fb. Utilizing a midrange/midbass near the driver's fundamental resonance at an amplitude near the average spl is one of those times where you typically want a lower Cms (..to better control the driver around resonance).
Last edited:
It has everything to do with it. The wave's length is what wraps around the driver itself (and any OB, or typical cardioid leaky enclosure) and generates that declining pressure loss from in-phase/out-of-phase cancellation, and that's in-addition to the declining pressure below the driver's Fs.
Not only that - but in-room you won't get either a dipole or cardioid character/pattern much beyond a few inches from the source because of the large wavelengths involved at lower freq.s and the interaction with the room. ..though notably I'm not saying there isn't a useful result from doing so, just that the pattern at lower freq.s is typically "swamped" by room mode effects - you won't get that "focused result" you mentioned in a standard listening room at any reasonable listening distance.
Last edited:
Due to the big cone displacement of an OB and the rear out phase that will mess, I do not think one want control when choosing OB design. As a trade off I would use it only in the middle of a big room when the listening chair near the loudspeakerfor the bass... but here I know most will disagree.
Emminence Alpha15 have a problem too, is qts sky rocket when move ....min use two for side
- General reply...
Thanks for reminding me Juhazi, Klippel have been on my back burner for some time now. Need to sit with their material.
I had to go look and dig some more and open baffle is kinda new to me, so, I am not going to act an expert. If I have the wrong opinion or using the wrong tools, then this is an excellent time to adjust the course. Open Baffle / dipole loudspeaker, is kinda limiting what is at play here, because as we move from sealed to vented, incl. ported, aperiodic and transmission line, we are more and more approaching open air, which is perhaps a better as a description. So I found two debates / articles I wish to bring into the conversation to help clarify the whole Q and Cms portion of the debate. Going in, we remember that EBP isn't very useful as Martin King mentioned due to. So here we go.
#1: Post #6 from: "Qts & Cms which is high priority or what is a good combination"
Its more about the Q of the driver when in use in the application and not the driver's TS parameters Qts value.
For example, if you will be using the driver in a closed box the Q value will be elevated above the Qts value depending on the ratio of Vbox/Vas (I forget the exact relationship). I am sure you are already well familiar with the relationship between Q and the time domain response, and frequency response around resonance. So assuming one wants the in-box Q to be in the range 0.5-0.8, you need Qts to be lower, e.g. 0.25-0.5.
In an open baffle, what you have is very similar to an infinitely large box in terms of the Q of the driver in the application (an open baffle). So using a driver with a Qts that is already around 0.7 is helpful. Lower Qts values only cause a slow and premature bass rolloff. Because the dipole cancellation is causing a 6dB/oct decrease in SPL in the bass region, this can be partially compensated by using a Qts that is even higher, e.g. up to 1.4 or so. Combined with the "first order HP like" response from the dipole cancellation you get something like a third order response, and for example a third order Butterworth filter is a first order HP filter combined with a second order HP filter with Q=1.0.
High Qts driver have gotten a bad rap, and used in the wrong way they are less than ideal. In an open baffle, they do have a place. One tangential side effect is that these drivers often tend to be cheaply made all around because a small magnet/motor is used to obtain the high Qts value. So shop carefully.
Also, Cms is not an important parameter for an open baffle. This is giving you exactly the same information as the Vas parameter.
Who cares what Vas is when the "box" in an OB system is infinitely large?
#2: What is Speaker Q? (Qts, Qes, Qms Explained)
But what has speaker resonant frequency got to do with speaker Q? In order to control a speaker’s resonant frequency, we have to control damping. In the world of speaker design, designers use speaker Q, (also known as Qts or Total Q) to define how well damped a speaker is. After a speaker is designed, it is measured. Typically, it is placed into an anechoic chamber, with a microphone placed 1m away. A frequency sweep from 20Hz to 20kHz is played through the speaker and measured. Using measurement software, an impedance curve is produced below, which shows us the resonant frequency of the speaker. Here is what a typical impedance curve looks like from a measured speaker.
In the context of speaker impedance curves, the Total Q, (Qts) is a measure of the sharpness of the driver’s resonance peak. Total Q (Qts) depends on both the mechanical and electrical characteristics of the speaker. Total Q (Qts) is calculated from the electrical Q (Qes) and the mechanical Q (Qms).
What Does Qes & Qms Mean?
Qts (or Total Q) is calculated from Qes & Qms.
Qes and Qms are two quantities that quantify the suspension control of a transducer when it reaches the resonant frequency (Fs). As the voice coil of the speaker moves up and down in the magnetic gap of a speaker, a suspension design allows this movement but also controls this movement so the voice coil and diaphragm assembly do not bounce all over the place, become unstable and break.
The Electrical Q – Qes
The electrical Q of a speaker, known as Qes, is the amount of control coming from the electrical components of a speaker (the voice coil and magnet) which contribute to the suspension system.
The Mechanical Q – Qms
The mechanical Q of a speaker, known as Qms, is the amount of control coming from the mechanical components of a speaker (surround and suspension/spider) which contribute to the suspension system.
Qts is calculated by multiplying Qes by Qms, then dividing that results by the sum of the same. [source]
Qts = Qms × Qes / (Qms + Qes)
What Does Speaker Q Tell Us?
Speaker Q tells us how well a speaker is damped. A high Q value will tell us that there is low system damping. Damping is the dissipation of energy in a system over time. In the context of speakers, it is the control of the resonance frequency of the speaker.
A high Q factor means that there is a lot of resonance (the speaker is not well damped or has low system damping) and a low Q factor means that there is little resonance (the speaker is well damped). If the Q of a speaker system is too high, then the sound will be harsh and fatiguing. If the Q of a speaker system is too low, then the sound will be muddy and lack detail.
What Is A Good Speaker Q Factor?
The best Qts (Total Q) factor depends on the application. Below is a table which shows some typical Qts values and their suitability. These values are given as a general guide only as there are always exceptions to the rule.
According to some experts, a Qts (Total Q) of 0.7 is considered ideal as it balances a smooth lower frequency response with good driver damping, with a very good transient response. Again, it depends on your application. The term “infinite baffle” may be confusing if you are new to the world of speaker design. It can be taken literally to mean a speaker mounted on an infinitely large baffle or enclosure. In the real world, this is not practical, but in practice, the term “infinite baffle” typically implies a closed box which is large enough that it is not an air suspension system.
Summery.
So there we have it. Open Air is around Qts of 0.7 or higher and looking over a few drivers, Qms is in the 0.6-10 or higher... but rules are created to be broken. If we explore Folded Open Baffle which is practiced by GR Research and NaO Music & Design, RiPole and SLOB (Slot Loaded Open Baffle) as well as Aperiodic which Audio Physic is into, their, the general Qts is 0.4-0.7 which is more in-line with sealed enclosures.
GR-Research OEM drivers.
As an example, If we stop and look at the values for GR Research drivers, we can see, that spite being used in open air, the T/S values, specifically Qts and Qms, signify sealed enclosure. We do see a difference between the woofer and midrange were the midrange signify open air capability more than the woofer counterpart, which is a lowish Qts. Is this telling us something ? ...
When I look at GR Research NX-Otica and the impedance sweep, I can see that the XO between the midrange and woofer is 150Hz. This peak is 2 octaves above the Fs of 51.7Hz, and so tens of ohms is not an issue here.
Balance, its a balanced design as I see it. You don't want too high of an impedance at Fs, and further, it might tie into how we perceive sound (Fletcher-Munson- or Equal-loudness contour). With an open air design we don't suffer the box or bass-port effect on the driver, making it sluggish in the bottom end, which is evident in the time domain (Cumulative Spectral Decay). By getting away from the "boominess" or low frequency glare, the penalty is far less and the low end is perceived as cleaner and thus, we end up with a superior transient performance.
Back to Qms and Qts with some impedance.
Troels Gravesen's JA-8008 HMQ 8" has a Qts of 0.6 and Qms of 10 with an Fs impedance of 240 ohm .. Cms is 0.66mm/N but not relevant.
Danny Richi's M-165 has a Qts of 0.39 and a Qms of 2.23 with an Fs impedance of roughly 23 ohm .. Cms is 1.877 mm/N but not relevant.
And with that, I think I'll leave it here - this is not the end of these debates, but it brought me closer to
Sources:
Thanks for reminding me Juhazi, Klippel have been on my back burner for some time now. Need to sit with their material.
I had to go look and dig some more and open baffle is kinda new to me, so, I am not going to act an expert. If I have the wrong opinion or using the wrong tools, then this is an excellent time to adjust the course. Open Baffle / dipole loudspeaker, is kinda limiting what is at play here, because as we move from sealed to vented, incl. ported, aperiodic and transmission line, we are more and more approaching open air, which is perhaps a better as a description. So I found two debates / articles I wish to bring into the conversation to help clarify the whole Q and Cms portion of the debate. Going in, we remember that EBP isn't very useful as Martin King mentioned due to. So here we go.
#1: Post #6 from: "Qts & Cms which is high priority or what is a good combination"
Its more about the Q of the driver when in use in the application and not the driver's TS parameters Qts value.
For example, if you will be using the driver in a closed box the Q value will be elevated above the Qts value depending on the ratio of Vbox/Vas (I forget the exact relationship). I am sure you are already well familiar with the relationship between Q and the time domain response, and frequency response around resonance. So assuming one wants the in-box Q to be in the range 0.5-0.8, you need Qts to be lower, e.g. 0.25-0.5.
In an open baffle, what you have is very similar to an infinitely large box in terms of the Q of the driver in the application (an open baffle). So using a driver with a Qts that is already around 0.7 is helpful. Lower Qts values only cause a slow and premature bass rolloff. Because the dipole cancellation is causing a 6dB/oct decrease in SPL in the bass region, this can be partially compensated by using a Qts that is even higher, e.g. up to 1.4 or so. Combined with the "first order HP like" response from the dipole cancellation you get something like a third order response, and for example a third order Butterworth filter is a first order HP filter combined with a second order HP filter with Q=1.0.
High Qts driver have gotten a bad rap, and used in the wrong way they are less than ideal. In an open baffle, they do have a place. One tangential side effect is that these drivers often tend to be cheaply made all around because a small magnet/motor is used to obtain the high Qts value. So shop carefully.
Also, Cms is not an important parameter for an open baffle. This is giving you exactly the same information as the Vas parameter.
Who cares what Vas is when the "box" in an OB system is infinitely large?
#2: What is Speaker Q? (Qts, Qes, Qms Explained)
But what has speaker resonant frequency got to do with speaker Q? In order to control a speaker’s resonant frequency, we have to control damping. In the world of speaker design, designers use speaker Q, (also known as Qts or Total Q) to define how well damped a speaker is. After a speaker is designed, it is measured. Typically, it is placed into an anechoic chamber, with a microphone placed 1m away. A frequency sweep from 20Hz to 20kHz is played through the speaker and measured. Using measurement software, an impedance curve is produced below, which shows us the resonant frequency of the speaker. Here is what a typical impedance curve looks like from a measured speaker.
In the context of speaker impedance curves, the Total Q, (Qts) is a measure of the sharpness of the driver’s resonance peak. Total Q (Qts) depends on both the mechanical and electrical characteristics of the speaker. Total Q (Qts) is calculated from the electrical Q (Qes) and the mechanical Q (Qms).
What Does Qes & Qms Mean?
Qts (or Total Q) is calculated from Qes & Qms.
Qes and Qms are two quantities that quantify the suspension control of a transducer when it reaches the resonant frequency (Fs). As the voice coil of the speaker moves up and down in the magnetic gap of a speaker, a suspension design allows this movement but also controls this movement so the voice coil and diaphragm assembly do not bounce all over the place, become unstable and break.
The Electrical Q – Qes
The electrical Q of a speaker, known as Qes, is the amount of control coming from the electrical components of a speaker (the voice coil and magnet) which contribute to the suspension system.
The Mechanical Q – Qms
The mechanical Q of a speaker, known as Qms, is the amount of control coming from the mechanical components of a speaker (surround and suspension/spider) which contribute to the suspension system.
Qts is calculated by multiplying Qes by Qms, then dividing that results by the sum of the same. [source]
Qts = Qms × Qes / (Qms + Qes)
What Does Speaker Q Tell Us?
Speaker Q tells us how well a speaker is damped. A high Q value will tell us that there is low system damping. Damping is the dissipation of energy in a system over time. In the context of speakers, it is the control of the resonance frequency of the speaker.
A high Q factor means that there is a lot of resonance (the speaker is not well damped or has low system damping) and a low Q factor means that there is little resonance (the speaker is well damped). If the Q of a speaker system is too high, then the sound will be harsh and fatiguing. If the Q of a speaker system is too low, then the sound will be muddy and lack detail.
What Is A Good Speaker Q Factor?
The best Qts (Total Q) factor depends on the application. Below is a table which shows some typical Qts values and their suitability. These values are given as a general guide only as there are always exceptions to the rule.
| Qts (Total Q) Value | Comments |
|---|---|
| Less than 0.4 | A speaker driver that is well suited to a ported/vented enclosure |
| 0.4 – 0.7 | A speaker driver that is well suited to a sealed enclosure. |
| Above 0.7 | A speaker driver that is well suited to free-air or infinite baffle application. |
According to some experts, a Qts (Total Q) of 0.7 is considered ideal as it balances a smooth lower frequency response with good driver damping, with a very good transient response. Again, it depends on your application. The term “infinite baffle” may be confusing if you are new to the world of speaker design. It can be taken literally to mean a speaker mounted on an infinitely large baffle or enclosure. In the real world, this is not practical, but in practice, the term “infinite baffle” typically implies a closed box which is large enough that it is not an air suspension system.
Summery.
So there we have it. Open Air is around Qts of 0.7 or higher and looking over a few drivers, Qms is in the 0.6-10 or higher... but rules are created to be broken. If we explore Folded Open Baffle which is practiced by GR Research and NaO Music & Design, RiPole and SLOB (Slot Loaded Open Baffle) as well as Aperiodic which Audio Physic is into, their, the general Qts is 0.4-0.7 which is more in-line with sealed enclosures.
GR-Research OEM drivers.
As an example, If we stop and look at the values for GR Research drivers, we can see, that spite being used in open air, the T/S values, specifically Qts and Qms, signify sealed enclosure. We do see a difference between the woofer and midrange were the midrange signify open air capability more than the woofer counterpart, which is a lowish Qts. Is this telling us something ? ...
When I look at GR Research NX-Otica and the impedance sweep, I can see that the XO between the midrange and woofer is 150Hz. This peak is 2 octaves above the Fs of 51.7Hz, and so tens of ohms is not an issue here.
Balance, its a balanced design as I see it. You don't want too high of an impedance at Fs, and further, it might tie into how we perceive sound (Fletcher-Munson- or Equal-loudness contour). With an open air design we don't suffer the box or bass-port effect on the driver, making it sluggish in the bottom end, which is evident in the time domain (Cumulative Spectral Decay). By getting away from the "boominess" or low frequency glare, the penalty is far less and the low end is perceived as cleaner and thus, we end up with a superior transient performance.
Back to Qms and Qts with some impedance.
Troels Gravesen's JA-8008 HMQ 8" has a Qts of 0.6 and Qms of 10 with an Fs impedance of 240 ohm .. Cms is 0.66mm/N but not relevant.
Danny Richi's M-165 has a Qts of 0.39 and a Qms of 2.23 with an Fs impedance of roughly 23 ohm .. Cms is 1.877 mm/N but not relevant.
And with that, I think I'll leave it here - this is not the end of these debates, but it brought me closer to
Sources: