39 replies to this topic

[syd-monster]

Question!

Out of the two which has the lighter weighing cone (Mms) & higher Bl.
Midbass "punch" and what a true midbass needs is speed & kick; the weight of cone and ability to respond/react to the electrical/magnetic fields is critical. A lighter cone & higher Bl is the one to choose, if its down to just these two.
This more often than not is what seperates the everyday midbass's from the ones that really rock and do it effortlessly and often to a degree, with less distortion & cone break up.

Edited by syd-monster, 01 February 2012 - 11:36 AM.

[Big_Valven]

Off the top of my head Syd, that'd be the Focal.

Muzzy, very informative post, especially factoring cabin gain in approximately as well. I wonder what the difference is at Xmech compared to Xmax for the two drivers? I've found the focals to be very well behaved under significant compression - ie it reaches xmech much more gracefully than the Alpines. It doesn't take much volume on the open road to start moving out of linear xmax and factoring in that most 6.5" drivers do get pushed fairly hard in this situation, it's a worthy consideration for real world daily listening especially if you wind up the wick a bit.

[muzzy66]

syd-monster, on 01 February 2012 - 11:35 AM, said:

Question!

Out of the two which has the lighter weighing cone (Mms) & higher Bl.
Midbass "punch" and what a true midbass needs is speed & kick; the weight of cone and ability to respond/react to the electrical/magnetic fields is critical. A lighter cone & higher Bl is the one to choose, if its down to just these two.
This more often than not is what seperates the everyday midbass's from the ones that really rock and do it effortlessly and often to a degree, with less distortion & cone break up.

This in theory might be a big consideration, but not sure how much importance it will play in this specific application. I think the speed of the driver will be affected less by the weight of the cone, and more by the damping charactereistics of the driver in relation to the enclosure. In this case I would predict the Focal would have a more even (and lower overall) group delay then the Alpine due to the higher Q being more suitable to the 'enclosure'.

In my past experience I've always found that the first flaws I will pick in a car are usually large FR disparities (which are usually immediately noticable regardless of output level), followed by excessive distortion (which is easilly missed to start with, but can get very irritating long term). I don't think that actual driver speed really going to be that big an issue on driver as small as a 6.5", but when you put a driver in an innapropriate enclosure the combined effects may cause a slow and laggy response.

That said this is mostly assumed, so I can't claim I can scientifically prove any of it!

[Domza]

Big_Valven and Muzzy 66 I think you would be correct in predicting that the Focal would have a lighter cone.. When I listened to the both speakers I did find the Focal speakers to be more resolving of details as opposed to the Alpine, but weather or not this is attributed to the weight of the cone I'm not too sure as I am no expert. But I have made my desicion based on not only these findings that Muzzy 66 has provided me with but also with "real world" listening evaluations with non other than...... My ears!! To me I think the Focals are more transparent and at the same time have reasonably good dynamics in which this leads me to say that it's a very musical speaker on a number of fronts. I also think that my Boston G21244 sub will blend perfectly given it also has a very musical nature and in my opinion very well suited for my application (SQ).

Regards,
Domza.

[Big_Valven]

Echoes my sentiments from listening. I'm sure you'll be quite happy with them. Oh, and don't be afraid to give them gobs of power

[syd-monster]

BTW, so im clear all the other factors that Pete, BV, etc have mentioned are just as important. (Particularly Qts, infact provided the driver plays reasonable safely in the Freq range I intend to use it for & fits the install, Qts is my first deciding spec with others following it.) its just that for midbass in particular Mms is decidedly critical in that we want thump like a sub, but snap like a mid... ah... cake and ear it too! Hence the compramise... hence the what can I gain from what specs these drivers display... with this given information what can I choose...
Above ALL that though is listening is the primary thing. I used to think all the answers were in specs, RTA & Freq responce graphs, distorion and wave plots, when in reality the best equipment to judge a speaker with is still ourselves...

[Domza]

Big_Valven, on 02 February 2012 - 04:07 PM, said:

Echoes my sentiments from listening. I'm sure you'll be quite happy with them. Oh, and don't be afraid to give them gobs of power

Yeah I'm sure I will be extremely happy with my set up, and yes I will have at least 250 watts RMS on tap for my midbass so I think this will be quiet ample for the Focal's.

Regards,
Domza.

[simplesq]

syd-monster, on 02 February 2012 - 04:19 PM, said:

BTW, so im clear all the other factors that Pete, BV, etc have mentioned are just as important. (Particularly Qts, infact provided the driver plays reasonable safely in the Freq range I intend to use it for & fits the install, Qts is my first deciding spec with others following it.) its just that for midbass in particular Mms is decidedly critical in that we want thump like a sub, but snap like a mid... ah... cake and ear it too! Hence the compramise... hence the what can I gain from what specs these drivers display... with this given information what can I choose...
Above ALL that though is listening is the primary thing. I used to think all the answers were in specs, RTA & Freq responce graphs, distorion and wave plots, when in reality the best equipment to judge a speaker with is still ourselves...

I couldn't agree with you more Chris ...
Quite often we forget what a true midbass is supposed to do: IMO a true midbass will NEVER be boomy, instead it should be lively, snappy, have decent speed and attack but also have low frequency extension. In doing this all your bass will be upfront with the sub only filling in that lower fundamentals. I cannot comment of the aforementioned Focals but I can comment on the Flagship Beryliums & as far as I'm concerned they fall short as they tend to be a little boomy and do not have decent low frequency extension.

[muzzy66]

Boomy low end is a very common issue I've found in car installs, but I think it's got as much to do with the install as it does the gear in most cases. On paper (or at least in simluation) the Be from the Kit 7 plays as low as just about any ~7" driver out there, with good extenson down to about 40hz in theory. I think a lot of people struggle with the sub -> midbass crossover and this usually is the main culprit for boomy midbass from my findings. Not all though

I really don't find that boomy midbass is an issue with speed or cone weight though. I've found it's most commonly an issue with either a large FR peak somewhere in the lower midbas (usualy around 40-60hz) or ugly resonances caused by installational imperfections. I think that some careful tuning will usually help clean this up!

Just what I've found from my experiences

[RustyP]

I have a couple of questions related to the theory behind the midbass drivers suggested much further above, mostly by muzzy66 ^^
I hope someone on here with more knowledge than myself can enlighten me on the parameters further!?

It is said the Scanspeak W18/8535's are an excellent midbass driver, however:
1. You mention the resonant frequency (Fs) indicates how low the driver can play. The 8535's value of 26Hz seems insanely low however, and actually lower than my subwoofer ( ;-_-). Is this an issue? or is it going to be insignificant if you highpass it around 80Hz or so anyway? To word it another way I guess I'm asking if the fact it resonates lower than it will be playing means it will be clean to it's 'lowest' or if it will actually cause issues with trying to equalize/Xover with the sub?
Further to this, if you are going to HP the midbass at lets say 80Hz (for argument's sake) then is there going to be much of a difference between an Fs of 50Hz and one of 40Hz, or is it sufficiently far lower than 80Hz that it shouldn't affect it too much?
2. You also talk about Qts being important, with around 0.5 being OK and 0.7 optimal. The 18W/8535 is rated at 0.38. It almost looks based on your explanation that this is too low to make good sound, however, apparently it does? I take it there is a caveat to your statement that would allow the 0.38 to be OK? Is it the Fs/Qts being <<100 that still allows this to be OK even if the Qts is relatively low? Or is it that this speaker in question is just unsuitable for car doors, and the lower Qts means it is better suited to a different installation?

I'm merely using the W18/8535 as an example btw, to illustrate questions I have with the reasoning behind your statements regarding the T-S characteristics. Hopefully the SQ gods out there can help improve my understanding!
Thanks =)

Edited by RustyP, 18 February 2012 - 08:59 AM.

[syd-monster]

RustyP, on 18 February 2012 - 08:39 AM, said:

I have a couple of questions related to the theory behind the midbass drivers suggested much further above, mostly by muzzy66 ^^
I hope someone on here with more knowledge than myself can enlighten me on the parameters further!?

It is said the Scanspeak W18/8535's are an excellent midbass driver, however:
1. You mention the resonant frequency (Fs) indicates how low the driver can play. The 8535's value of 26Hz seems insanely low however, and actually lower than my subwoofer ( ;-_-). Is this an issue? or is it going to be insignificant if you highpass it around 80Hz or so anyway? To word it another way I guess I'm asking if the fact it resonates lower than it will be playing means it will be clean to it's 'lowest' or if it will actually cause issues with trying to equalize/Xover with the sub?
Further to this, if you are going to HP the midbass at lets say 80Hz (for argument's sake) then is there going to be much of a difference between an Fs of 50Hz and one of 40Hz, or is it sufficiently far lower than 80Hz that it shouldn't affect it too much?
2. You also talk about Qts being important, with around 0.5 being OK and 0.7 optimal. The 18W/8535 is rated at 0.38. It almost looks based on your explanation that this is too low to make good sound, however, apparently it does? I take it there is a caveat to your statement that would allow the 0.38 to be OK? Is it the Fs/Qts being <<100 that still allows this to be OK even if the Qts is relatively low? Or is it that this speaker in question is just unsuitable for car doors, and the lower Qts means it is better suited to a different installation?

I'm merely using the W18/8535 as an example btw, to illustrate questions I have with the reasoning behind your statements regarding the T-S characteristics. Hopefully the SQ gods out there can help improve my understanding!
Thanks =)

Im sure Muzz/Pete will add. In general, all specs let you do is choose the best suited driver for the best suited application. Ca

However, let clarify why those specs are still good for a midbass.
Any midbass that can play in the lower regions, an Fs of 29Hz is low, is of benefit but its not necessary that it do so.
The "flexibility" of this low Fs just means you will have greater frequency responce coverage, which lets you pick more different x-over points (between it an a sub), less likely to distort down low and probably handles higher power levels.
A higher Fs isn't a negative as such, and there is a likely hood that the driver also plays higher and that it reaches a higher top. e.g. I recently aquired some Beyma 10" midbass, with an Fs of 55Hz, not very low, but more than what I need from a true dedicated midbass, the beauty is that they play right upto 4khz!!! so I will be covering plenty of the frequency range with 1 driver.

The low Qts suggests that it needs to be controlled, or in other words an enclosure. Lower Qts also sometimes means higher sensitivity, so it probably needs less Wrms to get loud.
Just on those figures, Id say this midbass is suited for a tuned enclosure in a two-way system... great for a house/shelf monitor speaker. Not so good in a door or kickpanel or an unknown volume or if tuned at all.
However! some people go to the lenght of building tuned enclosures on their doors (see 2Loud2Old's commodore as a GREAT example) and end up with incredible results!!
A high Qts ussually means a driver that can control itself better/more so and handle more power. Its also means that a tuned enclosure is not important and infact a high Qts means it can play in what we call infinate baffle. "IB" is a very rewarding setup in that it can be quite natural sounding. Im personally a big fan of IB.

All of the above is basically, the right driver, in the right type of install, executed correctly.

Edited by syd-monster, 18 February 2012 - 09:35 AM.

[Pulse-R]

As above, low Fs is a good indicator that the suspension is not overly tight. But it is only a small part of the story.
if you feed it too much power it will still distort. best to use the crossover selection and amplifier to control the cone.

It also means you aren't trying to cross it over in the resonance region, allowing smooth impedance curve in the pass band.

[muzzy66]

RustyP, on 18 February 2012 - 08:39 AM, said:

I have a couple of questions related to the theory behind the midbass drivers suggested much further above, mostly by muzzy66 ^^
I hope someone on here with more knowledge than myself can enlighten me on the parameters further!?

It is said the Scanspeak W18/8535's are an excellent midbass driver, however:
1. You mention the resonant frequency (Fs) indicates how low the driver can play. The 8535's value of 26Hz seems insanely low however, and actually lower than my subwoofer ( ;-_-). Is this an issue? or is it going to be insignificant if you highpass it around 80Hz or so anyway? To word it another way I guess I'm asking if the fact it resonates lower than it will be playing means it will be clean to it's 'lowest' or if it will actually cause issues with trying to equalize/Xover with the sub?
Further to this, if you are going to HP the midbass at lets say 80Hz (for argument's sake) then is there going to be much of a difference between an Fs of 50Hz and one of 40Hz, or is it sufficiently far lower than 80Hz that it shouldn't affect it too much?
2. You also talk about Qts being important, with around 0.5 being OK and 0.7 optimal. The 18W/8535 is rated at 0.38. It almost looks based on your explanation that this is too low to make good sound, however, apparently it does? I take it there is a caveat to your statement that would allow the 0.38 to be OK? Is it the Fs/Qts being <<100 that still allows this to be OK even if the Qts is relatively low? Or is it that this speaker in question is just unsuitable for car doors, and the lower Qts means it is better suited to a different installation?

I'm merely using the W18/8535 as an example btw, to illustrate questions I have with the reasoning behind your statements regarding the T-S characteristics. Hopefully the SQ gods out there can help improve my understanding!
Thanks =)

Good and valud questions, and something I'm going to find out for myself soon enough.

Generally every driver I have modelled that has a low q has a 'smiley face' effect from about 40hz to 200hz where they dip quite significantly in the middle. The result is that they are quite significantly down at 80hz and 100hz, where a lot of that midbass 'kick' comes from. This seems to be the case for every such driver, even the Revelator and the Seas L16RN (though the effect is fairly modest in these). Drivers with a low Q and a high FS seem to make this even worse, because they also don't play very low...meaning that from 200hz they start start rolling off and never come backup.

The interesting thing is that for some reason the 8535 is the only driver I've modelled out that doesn't adhere to this rule. According to the most recent datasheet it has an FS of 26hz and a QTS of 0.38 - it does still have a very minor 'smiley face' effect, but because it has so much output below 200hz the dip caused by the low Q isn't enough to cause it a major dip. It being up getting a 2dB sub-bass around 20hz-30hz instead of the typical 'smiley-face'. For example, here are the modelled results for an 18W/8535 in a car door, with cabin gain:

300hz: -0.05dB
250hz: -0.11dB
200hz: -0.24dB
160hz: -0.45dB
120hz: -0.85dB
100hz: -1.20dB
80hz : -1.38dB
60hz : -0.71dB
50hz: +0.90dB
40hz: +2.00dB
30hz: +2.20dB

As you can see according to the simulated numbers, the 8535 never dips by more then 1.5dB at any point from 20hz to 300hz. Now lets compare that to the SB Acosutics SB17NRX35-4 which has a slighty lower low Q (0.32) and a low (but not as low) FS of 35hz. This driver shows a very pronounced 'smiley face' effect:

300hz: -0.29dB
250hz: -0.49dB
200hz: -0.80dB
160hz: -0.35dB
120hz: -2.30dB
100hz: -3.10dB
80hz : -3.70dB
60hz : -3.50dB
50hz : -2.45dB
40hz : -1.70dB
30hz : -1.80dB

Looking at these numbers, the Scanspeak 18W/8535 is +/- 1.38dB from 50hz to 300hz which is fairly linear, and may not even be that audible. The SB Acoustics is +/- 3.7dB from 50hz to 300hz which is a very large variation, and will definately be very audible. The very low FS on the can gives it a very strong bottom end extension which (combined with cabin gain) brings up that 80hz region where the big dip usually would occur. The downside is a fairly singificant 2dB moutain in that 30hz-40hz subbass region which you may need to deal with...but the important thing is that output is strong and fairly linear throughout that whole midbass region.

So it seems the Scanspeak 8535 (and a select few other DIY drivers) can somewhat make up for the low Q issue (at least in terms of low end frequency response) by having very impressive low end output, but 98% of drivers with a low Q will not be so capable. The big plus side with the Scanspeak 8535 is that it uses the SD-1 motor, a world class motor design that allows incredibly low levels of harmonic and intermodule distotion in the bass and mdibass (and even the midrage). Compared to other drivers of the same size, the 8535 has few peers as a pure bass drivers go...with it's only real limitation in a car being that 30hz-40hz peak of 2dB. I don't think it's too big a sacrifice given what the driver is capable of.

Now lets compare that to the Focal 165KR, which has a nice high Q for a car door (0.74) but also relatively high FS (73hz):

300hz: +0.31dB
250hz: +0.41dB
200hz: +0.40dB
160hz: +0.20dB
120hz: -0.48dB
100hz: -1.38dB
80hz : -2.40dB
60hz : -3.38dB
50hz : -3.29dB
40hz : -4.14dB
30hz : -6.20dB

Here you can see that the Focal is easillythe most linear of the three in the upper midbass from 160hz to 300hz, and from 120hz down there is a constant but very smooth rolloff. This is largely due to the driver having a high Q, which allows the simluated car door to act more like an optimal enclosure then it does with the other drivers. But the clear downside is that due to the fairly high FS, the driver just doesn't have the ability to produce much output below 120hz. By 100hz it's already 1.5dB down, but is still usable. By 80hz it's already nearly 3dB down, which significant enough to make me not want touse it that low.

Finally, lets look at the Rainbow W165 Germanium - a driver with that optimal combination of a high Q (0.81) and a relatively low FS (54hz):

300hz: +0.01dB
250hz: +0.01dB
200hz: 0.00dB
160hz: -0.02dB
120hz: -0.11dB
100hz: -0.30dB
80hz : -0.27dB
60hz : +0.28dB
50hz : +1.07dB
40hz : +0.75dB
30hz : -1.11dB

It's very intereting set of results, because the SB17's FS of 35hz is still very low for a 6.5" driver, but it suffers very badly from the 'smiley' effect. On the other hand the stupidly low 26hz FS of the 8535 (actually lower then a Morel Ultimo 10) seems to boost the low end enough to bring that 'smiley' area up a couple of dB, and in the process almost eliminates the smiley effect. The interesting thing is that the Scanspeak 15W Revelator and Seas L16 (both 5.75" drivers) both have an FS around 35hz and a Q of around 0.4, and they both handle the 'smiley' effect reasonably well, though not quite as well as the 8535. This is brings me to the conclusion that there is a clear relationship between qts and FS when it comes to determining a driver's ability to perform in a car door. With a low FS and high Q being desirable, I would typically suggest that working out the fs/qts is a good starting point, with the lower the value the better. Lets look at the fs/qts values of the drivers I've listed so far:

Rainbow W165 Germanium: 54Hz / 0.81 = 66.66
Scanspeak 18W/8535: 26Hz / 0.38 = 68.42
Seas L16RN/SL: 36Hz / 0.46 = 78.26
Scanspeak 15W/8530K01: 32Hz / 0.41 =78.04
Focal 165KR / 6KPR: 73Hz / 0.74 = 97.71
SB Acoustics SB17NRX35-4: 35Hz / 0.32 = 109.38

Interesting? These results would suggest that the Germanium would be the best overall peformer as a midbass, followed by the Scanspeak 8535 -> Seas L16 -> 15W Revelator (barely) -> Focal KR -> SB17. We already confirmed that the SB17 suffered more from the smiley face effect and also had the second worst extension (after the Focal), so it's safe to conclude it was the worst performer in this group. The Focal was nearly perfectly linear above 120Hz, it fell off fast after that, giving it the worst low end extension of all the drivers - so second last position makes sense. The Seas L16 and 15W Revelator results were not posted but they model out almost identical to each other, and very similar to the 8535 (with slightly less low end extension and slightly more 'smiley' effect) - their number and position makes perfect sense. The Scanspeak 18W had the best low end output by a good margin but from 60z to 300hz it's output was not as controlled or linear as the Germanium, so second best makes sense. The Germanium was modelled out the best overall, with the best ballance between extension and linearity in this group, so it's rank at the top makes sense too.

Some people might say that this is just a coincidence, and there are other factors in play on top of just these three 'specs'. Lets test that theory too. Are you aware of Rainbows Kick-Bass models? They are advertised as a model that provides better upper midbass performance at the sacrifice of bottom end extension (they are not recommended to be used below 80hz). Lets look at their TS specs compared to the regular Profi which has the same motor, cone, basket, surround, everything - simply different TS specs:

Rainbow W165 Profi Kick: (fs=67Hz, qts=0.80, fs/qts=83)
Rainbow W165 Profi (fs=53Hz, qts=0.62, fs.qts=85)

Notice the fs is lower on the Profi (which should give better output down low) and the qts is higher on the Profi Kick (which should give better output >120hz). The overall fs/qts values are similar, so I'd expect the both drivers to have a pretty even low-midbass vs high-midbass ballance. I've yet to see these results and am modelling it as I type this, so lets see for ourselves if the prediction comes true:

165 Profi:
300hz: -0.14dB
250hz: -0.20dB
200hz: -0.32dB
160hz: -0.50dB
120hz: -0.93dB
100hz: -1.43dB
80hz : -1.76dB
60hz : -1.54dB
50hz : -0.73dB
40hz : -0.79dB
30hz : -2.17dB

165 Profi Kick:
300hz: +0.01dB
250hz: 0.00dB
200hz: -0.02dB
160hz: -0.08dB
120hz: -0.34dB
100hz: -0.79dB
80hz : -1.27dB
60hz : -1.76dB
50hz : -1.61dB
40hz : -2.46dB
30hz : -4.66dB

Ok, well I think that pretty much sums it up. A few years back I had a owned the Profi Phase Plug set, while a friend owned the Vanadium Kick drivers. We installed the Kicks in one door, Profi in the other door and did a L/R comparison. We found that the ?Kick had ever so slightly more midbass detail, but that the difference was very subtle. The modelled results reflect not only my claims about the TS specs,but also our subjective findings at the time. The two drivers are pretty equal overall, with the Kickbass model giving slightly better midbass output from 80hz upwards, while the regular model gives better output from 60hz down. Now you can see why Rainbow recommends the Kickbass models to be runwithan 80hz crossover - not to protect the speakers divers, just because they don't give optimal output below that point.

So, many things to consider when choosing a midbass for your car doors. What we know by now is that really want a driver that has a high Q ( 0.7 - 0.8 ), a low FS (< 60hz), good motor, and you probably want some decent excursion to (> 5mm) for power handling purposes. Sadly I've yet to find such a driver that meets all of these criteria, so you will usually be making a compromise somewhere. My first step is to look at fs/qts, and then when I've narrowed do the field that way i ask myself what's more important to me - upper midbass 'kick', or a low midbass extension? If it's the former then go for the drivers with the highest Q values. If it's the latter, then go for the ones with the lowest fs.

Hopefully some of the info in this post will give you some insight into the effect of different values on a midbass driver in a car door - just remember two things:
1. The above is only true for a 'car door' type application - drivers with a high Q need a lot of air behind them to work optimally so they work well in a car door or infinite baffle, but they work poorly in modest sized sealed enclosured
2. These are all just simluations and not real world measurements!

Edited by muzzy66, 18 February 2012 - 07:04 PM.

[RustyP]

Omg You are a god!
Thanks heaps guys! I think I'm well and truly starting to understand this stuff now!
Much <3

[muzzy66]

P.s.

I've just updated the post with impoved grammar and some extra information, if you can be bothered going through it all again!! The extra info is only really in the last half of the, so you don't need to read the whole thing!