Understanding Qtc, Excursion, and Power Applied

[MiniVanMan]

Okay, this is sparked by the ongoing DIYMA 12 arugments. I'm going to use my favorite toy of late, Unibox, to illustrate how Qtc, excursion and power applied all correlate to each other.

This in no way a thread designed to say one woofer is better than another, but rather to show how a misaligned driver can easily be destroyed by reaching it's mechanical limitations.

First, let's understand the limitations of WinISD. WinISD is a very popular box modeling tool, but it does not show you excursion plots, (that I know of). That's why I find Unibox to be such a valuable tool. It will show the excursion of your driver with a given amount of power, for a given alignment. WinISD does not always default to the safest for your driver. Also, going, "hey, look, I can get another 5 hz of low end extension by increasing my box size by "X" amount is not always the best idea when you don't know how it's going to affect the excursion of the driver.

So, here we go with the graphs again. I'm going to use the Dayton RS265HO as a test example.

First we'll start with a standard Qtc of .707. This is pretty much the median alignment. Good power handling, and decent low end extension.

Here's 200 watts and a Qtc of .707

http://img.photobucket.com/albums/v641/glevii/RS265Qtc7200wattResp.gif

http://img.photobucket.com/albums/v641/glevii/RS265Qtc7200wattExc.gif

Notice the driver is operating WELL within it's excursion limits.

Let's double the power to 400 watts.

http://img.photobucket.com/albums/v641/glevii/RS265Qtc7400Resp.gif

http://img.photobucket.com/albums/v641/glevii/RS265Qtc7400Exc.gif

We're still well within the excursion limits of the driver, so let's add even more power. The driver is rated at 600 watts. Let's see what happens with that.

http://img.photobucket.com/albums/v641/glevii/RS265Qtc7600Resp.gif

http://img.photobucket.com/albums/v641/glevii/RS265Qtc7600Exc.gif

Now we're starting to get up to the limits of the driver. It's probably safe to say that 600 watts is all you'd want to try and throw at this driver in a 6 liter box. By the way, that's a bit under .3 cubes. That's a tiny box.

Now, let's go with a Qtc of .5. This is a much bigger box at 21 liters (.75 cubes).

http://img.photobucket.com/albums/v641/glevii/RS265Qtc5200Resp.gif

http://img.photobucket.com/albums/v641/glevii/RS265Qtc5200Exc.gif

As you can see, with only 200 watts we're reaching the excursion limits at around 18 hz. So, with only a third of the power, we have more excursion, and bottom end reproduction. However, we also risk damage to the driver due to it being driven to it's mechanical limits. A subsonic filter would be a good idea here.

Here's what the driver looks like with 400 watts in a 21 liter box.

http://img.photobucket.com/albums/v641/glevii/RS265Qtc5400Resp.gif

http://img.photobucket.com/albums/v641/glevii/RS265Qtc5400Exc.gif

Well, we have a 3 db gain through it's bandwidth, however, we also reach full excursion by 30 hz. This is just an all around bad idea here. Will you destroy the driver. Maybe not, but you're not doing it any good. This is where WinISD fails. WinISD will just show a nice bottom end reproduction, it won't show the massive excursion the driver would have to achieve to create that response.

For continuity, here's a Qtc of .9.

http://img.photobucket.com/albums/v641/glevii/RS265Qtc9600Resp.gif

http://img.photobucket.com/albums/v641/glevii/RS265Qtc9600Exc.gif

We could drive this woofer all day. It will take a good amount of power, even above it's rated power. However, low end suffers dramatically.

When designing a box for your subwoofer, you have to know where the tradeoffs are. The RS265 models up very nicely. It requires only a small box sealed or ported. Chad states he's ran his at 21 liters with no subsonic filter, and ran into no mechanical problems. It's a testament to the durability of the driver. However, the drawback, and I hear it brought up ALL the time is the lack of output the RS series drivers have. The cone is heavy, and sensitivity has been sacrificed in the name of durability.

A good majority of the time somebody complains of a subwoofer failing, it's due to a poorly designed system. Hopefully, this thread will help at least a few people understand what's going on with their subwoofer.

[Genxx]

Great explination. I hope people read this it as it help's explain how to keep from damaging a speaker and understand the principles of overdriving a speaker. Causes, effects and possible outcome. I downloaded Unibox now I just have to figure out how to use it. Ok figured out how to use it and I wuv it. This program rocks. Thanks MiniVanMan.

[yermolovd]

winisd alpha does have excursion plots, but that's besides the point. :p

graph with a use of SS filter might be useful for people to see. ;)

this is a useful piece of info for those who need it with nice examples. thank you.

[MiniVanMan]

Since this has really started due to the debate over the quality of the DIYMA12, I'm going to post some plots for this driver. I'll say this right now. Overexcursion is NOT the problem with the driver. In a sealed enclosure, this thing is a BEAST. So, let's see if we can get to the root of the problem.

First, let's see what a standard Qtc of .707 at 400 watts yields us.

http://img.photobucket.com/albums/v641/glevii/DIY7400Resp.gif

http://img.photobucket.com/albums/v641/glevii/DIY7400Exc.gif

Plenty of excursion left. Let's bump it up to 1000 watts.

http://img.photobucket.com/albums/v641/glevii/DIY71000Resp.gif

http://img.photobucket.com/albums/v641/glevii/DIY71000Exc.gif

Damn, this thing has excursion for days.

Since we saw the RS265 ran out of excursion fairly quickly in a .5 Qtc enclosure, let's see how the DIYMA fairs.

http://img.photobucket.com/albums/v641/glevii/DIY5400Resp.gif

http://img.photobucket.com/albums/v641/glevii/DIY5400Exc.gif

Pretty good, we're at more than double the power of the RS265HO and still have excursion to spare. Bumping up the power to 600 watts we get:

http://img.photobucket.com/albums/v641/glevii/DIY5600Resp.gif

http://img.photobucket.com/albums/v641/glevii/DIY5600Exc.gif

All I can say is this driver models out as a beast. If the advertised Xmax numbers on the DIYMA 12 are correct, then this a top tier woofer.

However, there's a flaw somewhere that modeling, and testing is not finding. There's much debate over why the cone of DIYMA is flexing and breaking. The problem as we can see isn't excursion. Here's a Qtc of .4 at 300 watts.

http://img.photobucket.com/albums/v641/glevii/DIY4300Resp.gif

http://img.photobucket.com/albums/v641/glevii/DIY4300Exc.gif

Now, that's just being mean to any subwoofer out there. That's a sealed enclosure of 6.5 cubic feat. However, this sub looks like it can handle it like a champ. So where's the problem.

Let's look back at our Qtc of .707. That's a 9.2 liter box, or .32 cubic feet. That is tiny. Yet, as we see, this driver is performing very well in that enclosure.

So, from what I can tell, this driver is very versatile in a sealed alignment. Npdang very obviously had some design goals that are rarely employed in car audio, and it looks like he achieved them. Still that nagging cone problem. I'm not going to offer any theories or suggestions, as it isn't my place and it would be pure speculation. So, save those for a thread not in the tutorials section.

[Genxx]

I just wanted to say Unibox is extremely easy to use. MiniVanMan can you post the specs you used on the DIYMA for Unibox?

[MiniVanMan]

I just wanted to say Unibox is extremely easy to use. MiniVanMan can you post the specs you used on the DIYMA for Unibox?

Took them right off the front page of the site. The Qes, Qms numbers are on the graph. Just double click the graph and the numbers are below it.

[Genxx]

Ok I feel like an idiot I never noticed them before. Thanks for the point in the right direction.

[MiniVanMan]

Ok I feel like an idiot I never noticed them before. Thanks for the point in the right direction.

Hehe, I sent a PM to npdang asking where they were, so don't feel bad. :blush:

I was just smart enough to not post it publicly. :p

[Kenny Bania]

Excellent info! This is one link I found for a Unibox download : http://home20.inet.tele.dk/kou/files/unibox407.zip

Is this the one you used?

[thechris]

try pounding the DIYMA with midrange freqs. see if any modes are excited that could explain excessive stress on the cone. the linear model won't show these things unless they were included in the model.

I'd also look into stresses caused by high freq, high excursion notes that can occur in ported boxes, where Xmax can be hit at a high frequency. this implies a high cone accelleration which places high stress on the cone.

(compulsory note about "small signal" modeling)

(compulsory note about overexcursion in ported boxes.)

[MiniVanMan]

I didn't really go into ported/vented designs as one of the statements that provoked this was "subsonic filter on a sealed enclosure??? :confused: "

As for midrange frequencies coming into the equation, there are too many variables that can cause that kind of response that it's just not feasible to model out. A metal coned driver like the RS265 or the DIYMA12 can be expected to have a nasty breakup node. The RS265 breakup starts in the 700 hz range and peaks like no tomorrow at 2.5k. If you have your sub playing 2.5k then you deserve whatever happens. With a 24db low pass your sub would be 110db down by 2.5k, that's off the radar at that point. I wouldn't mind seeing frequency response data for the DIYMA, but I don't think it'll show us much. If there's a severe breakup in the 200-500 hz range, then maybe cone damage could be explained a bit. Honestly though, I expect the DIYMA to perform better across the entire bandwidth than the RS series drivers.

Most of us know not to run a sub full range and that some kind of low pass filter should be used. Again, this thread is more about the benefits a subsonic filter can have, even with a sealed enclosure.

[mvw2]

The Subwoofer DIY Page has a couple downloadable Microsoft Excel files that do a lot of the same things as Unibox as well as some different stuff.
http://www.diysubwoofers.org/prt/
Download the PORTED.XLS in the side menu. There's a lot of info available as well as tuning aids. It also shows power/excursion plots just like Unibox so you can perform the same experiments. There's a nice Comparison tab that lets you throw in two different subs or the same sub in different configurations and compare them side-by-side. You can add highpass/low pass filters as well as cabin gain and EQ. It's pretty neat.

I don't really have much of a stance on the DIYMA sub issue. From an engineering standpoint, all I can say is things you don't want happening will probably happen. It's always good to overbuild and be safe. I could only assume the margin of safety on the design is low. It works but barely. I'm sure the tolerances on the materials aren't high, so some variation in strength and imperfections will exist, either by quality of the raw material or through manufacturing processes.

As well, the real world isn't as simple as text book calculations. A lot of things we don't think about happen, unknown variables, fluctuations, etc., things that shouldn't happen but do. Some of you speak of resonances. There could be goofy things going on, stuff we can't see without looking at the sub moving through a high speed camera. The cone shape may play a role in how the air pressure forces are distributed alone the cone's surface, and there could be high pressure points or flex. It may not be even as the cone moves back and forth. There could be an issue with how the motor connects to the cone. There may be high stress points or constant flexing at particular points that could quickly fatique the material. Thin metal is thin metal. Crap happens, tearing, buckling, etc. Heck, it may not even be the cone itself but another component and its interaction with the cone. There is obviously something at work, but nobody exactly knows what, lol.

Man, I might have to pick up one of these to play with. They do certainly look like neat subs. I didn't realize their low end sensitivity was that good. I guess I never really looked at it much given the use of such a small box size. My mind goes tiny box = no low end, lol. However, your plots show an F3 of 40Hz or under which is as good as any other subs I've toyed with so far. SPL is pretty low though and I've only got 400 watts to play with. It would probably be just loud enough to blend well off that but probably not much more.

[dawgdan]

Excellent info, thanks.

It just saved me from making a mistake with a pair of Dayton Quatro 10's that I was planning to drive with about 375w for the pair in a 0.60 cubic ft enclosure (each).

It looked like each driver was hitting mechanical limits right around 45 Hz! Definitely not good. A shame too, since it seems to be a real value driver.

[demon2091tb]

It would probably be just loud enough to blend well off that but probably not much more.

Gotta disagree with this, in my tC the low end is almost too much, and as of now i have my sub at -8 on the w200 with minimal gains on 500rms to it in .9 cf, depending on genre there can be excessive low end at times, in the size box i have i have more of a problem with 70-80hz being a little lacking. Beefy midbasses and a decent powered sub should be plenty...

[npdang]

Great tutorial, thanks for posting.

Don't forget though, as others have mentioned this is a very simplified way of determining driver limits.

You have many other issues such as power and bl compression, and voice coil and suspension centering. Some drivers have a pretty severe offset in one direction that may cause damage well before the published "xmax" limit. You also have thermal issues that may need to be looked at as well (remember the Aura's catching fire).

As far as the Diyma's... I think the fragility of the driver is a bit overstated... personally, I don't think it has any problem with dynamics or playing low... at the very least you have the capability of a driver like the Adire Shiva output wise. However, I tend to tell people otherwise because we all know that "loud" is a very subjective term and some may feel the need for considerable output especially when cruising on the freeway or when their car's trunk is restrictive. Some people may have bass boost or loudness turned on and not even know, or think twice about it. In any case, I don't want to go to far off topic but I'll just say I do have test data that suggests this driver can handle somewhat near its xmax limit for hours on end without any issues, and that I'm currently working on video and some more testing that will be posted in the near future that will better show the capabilities of this driver. I think it's just utterly ridiculous some of the claims people are making..

[MiniVanMan]

Most of my tutorials are very simplified. I just remember when first starting out, how hard it was to find good information that wasn't stated in an elevated techno speak that had you running around trying to find the definitions for every term. This particular one was to show that you really need to look at more than just a response graph in WinISD.

As with anything, you can always dig deeper. I just hope that I can help give a good foundation those just starting, or at least give a jump start into a more advanced level of audio for those looking to make the leap.

I've been agonizing over finding an application where the DIYMA would suit me. It's still a possibility for my wife's vehicle, but I would need two. :o

[Genxx]

I like the simplified. Also like your tutorials where you start simple then build upon that. I feel I am at the intermediate stage so your walk through from simple to advance tutorials help's alot. I am myself am getting a real understanding of the T/S and Klippel. Keep up the good work and can you continue to build on this thread.

Some of the things npdang is talking about start to get a little over my head though. Like these bl compression, voice coil and suspension centering.:(

Thanks and keep dropping the knowledge you smart guys as some of us are trying to soaking it up.

[durwood]

Some of the things npdang is talking about start to get a little over my head though. Like these bl compression, voice coil and suspension centering.:(


Think of Voice coil centering as a teeter-toter. If the weight is balanced on each side you get a nice fluid motion. Now if the weight is not distributed properly then one side works easier then the other. So lets say it's easier for the speaker to push out then pull back back in or vice versa-that is what npdang considered suspension centering I beleive. I'm just using the teeter toter as an example so don't equate weight to cone mass, but instead to motor strength or BL. You would also get similar results if you move the center pivot point of the teeter-toter.

Example using numbers:
published xmax is 6mm one way making it 12mmpeak to peak. It might actually be 6mm one way but only 4mm the other way. Is that making any sense? Let me find an example npdang has posted.
This was his explanation and goes way more into detail.

http://www.diymobileaudio.com/forum/showthread.php?t=1001

[chad]

I was spacing out and did not see this earlier.

Yes, I did not use a subsonic for a long time but keep some things in mind here. No many mastering engineers let much below 30 cycles get in to the final product. Especially in the music I listen to. Also remember I am using around 250W. I tested the enclosure with a 1KW (a real 1KWRMS amp not a facking car amp that claims 1KW at 16V downhill, with the wind against it's back, in a wheelchair) amp in a sealed arrangement and it made NO foul noises, that was pretty damn impressive, what was NOT impressive was it's output! After I vented the enclosure I did not bring it in the house to see how power handling was altered, but I can say that you ain't gonna hurt it with 250W with no subsonic filter if you have your head screwed on straight.

Minivanman, where did you get that software? I used to use MacSpeakerz before my mac:blush:SE:blush: finally died and it was tits on for excursion plots.

Chad

[durwood]

Ya how often are you ever goign to get down to the low low end and as Npdang said the software does not take the "other" factors into play.

Here is a link for Unibox (http://home20.inet.tele.dk/kou/ubmodel.html)-it's just an excel spreadsheet

[Genxx]

Thanks for the explination durwood. It made sense and was helpful. Thanks Brian

[SQ4ME2]

Great explination. I hope people read this it as it help's explain how to keep from damaging a speaker and understand the principles of overdriving a speaker. Causes, effects and possible outcome. I downloaded Unibox now I just have to figure out how to use it. Ok figured out how to use it and I wuv it. This program rocks. Thanks MiniVanMan.

clif note to all of this is simple. "its easier to blow a speaker with too little power than with too much power".

[MiniVanMan]

clif note to all of this is simple. "its easier to blow a speaker with too little power than with too much power".

Uhhhhh, no. That's not true.

I was hoping this tutorial would alleviate some of that type of thinking. You can very easily destroy a speaker in the wrong box with too much power. This tutorial illustrates that pretty well. A clipped signal can kill a speaker as well, which is what you're referring to. This tutorial assumes power is unclipped.

[Genxx]

clif note to all of this is simple. "its easier to blow a speaker with too little power than with too much power".

That is a very generalized statement, got any data to back that up. I will call BS on your generalized statement. Here is data and other discussions that support my point and statement.

http://www.bcae1.com/2ltlpwr.htm

http://www.the12volt.com/installbay/forum_posts.asp~TID~74226~PN~1~TPN~1

http://www.rocketroberts.com/techart/spkr.htm

http://www.ofsoundmind.com/OSM2/tech/speakers/audiobasics.html

[Abmolech]

Going to have to agree.
If we know the amperes, and the voltage and therefore the ohms, what is the missing ingredient to know the effect on a network system?

Time.

Power has a this defining ingredient.

Too little power will therefore NEVER destroy a network, since thermal handling ability (time) is the defining attribute.

If you are suggesting a power amplifier used beyond its recommended limits (clipping) is a danger to a network, We need to understand how much power it is producing. A network could take clipping with considerable ease, IF the power was is sufficiently low enough, IE watts. (work done)

[werewolf]

Uhhhhh, no. That's not true.

I was hoping this tutorial would alleviate some of that type of thinking. You can very easily destroy a speaker in the wrong box with too much power. This tutorial illustrates that pretty well. A clipped signal can kill a speaker as well, which is what you're referring to. This tutorial assumes power is unclipped.
Agreed. Too little power will never destroy a speaker ... otherwise, we would all be in danger by turning volume down :p

Sure, a clipped signal can cause driver failure ... but not because the signal is clipped. Rather, because of the power delivered by a square wave, compared to a sinewave.

In any case ... it's power that kills. The trick is understanding how much power is actually delivered ;)

[Neil]

In any case ... it's power that kills. The trick is understanding how much power is actually delivered ;)

Yes, this is an absolutely critical distinction to make. Saying that clipping kills speakers is analog to saying that sine waves kill speakers, as a sine wave has a lower crest factor than music by comparison.

[werewolf]

If waveform "shape" killed speakers, we would expect that same "shape" to kill at lower amplitudes/volumes as well as higher volumes. We would expect a recording of a clipped signal, played back at low volumes, to kill speakers. Since that's clearly not the case, we can conclude that "shape" does not kill speakers.

It's the power in the waveform that kills speakers. One informative way to visualize, is what happens to the spectral (or frequency) content of the waveform under clipping. The fundamental energy is surprisingly large, and energetic higher-frequency harmonics are created as well. But again, it's not the waveform "shape" or spectral distribution ... it's the total power that kills.

[Sleepy122CID]

I've been playing with Unibox a little bit the last couple of days, and I was wondering how Qtc and box size will effect a driver's transient response / SQ? It seems that a larger sealed box will increase its sensitivity and low end response, but will there be some control over the driver lost and lead to sloppier sound? Or should it sound the same assuming it isn't overpowered and/or the proper subsonic filtering is applied?

Also, I'm a little bit confused about a couple things in Unibox. When entering an Xmax value, should it be one way or peak to peak? I'm also a little confused about the design areas (for closed box in particular - I've been modeling an Exile XT10 for possible use in my standard cab pickup). I put in a nominal power of 300W. Under the standard design heading it shows a Vb of 31 liters and a max power input of 244W for a .5 Qtc. Under the design by vb and q heading it shows a physical Vb of 10.2 liters at .706 Qtc. Even if I change the Qtc to .706 in the standard design, it doesn't match the box size and other figures in the design by vb and q area. Which one of these boxes are modeled in the response graph? Should I just mess with the Qtc until I get it to match my desired RMS power handling in the standard design box?

[Genxx]

Disregard the title. This is has some good info on QTC ect. HIC posted it in another post.

http://www.mobileelectronics.com.au/forums/index.php?showtopic=83613&hl=400hz

The vb under closed box standard design is the size of the enclosure to equal a whatever QTC you selected. For the xmax I just use the published spec. If it is 30mm xmax that’s what I enter.

31 liters is about 1.07 cu. ft.

Under standard design type in a QTC .707 and see what it says. .5 QTC seems very under dampened.IMO

On the Exile XT10 I was thinking the recommended sealed enclosure size was .3-.6 IIRC.
.6 cu. ft.=17L .3 cu. ft.=8.49L

I do not worry about the RMS power after I enter into it in under driver unit parameters section. I change the power there if I need to test the execursion limit at given power.

If I am wrong someone please correct me on this. I am no expert by any means.