Vented vs. Sealed speaker for HT (for Vulcan)

[biff]

I thought I'd make this a separate thread. This is a continuation of the discussion in this thread, where I was cautioning the use of ported speakers when used for HT purposes. I'm not saying you can't do it, just giving a heads up to some potential issues that could arise and back it up with a graph or two. The usual disclaimers apply - I don't claim to be an expert so I'm not saying everything here is 100% accurate and if you notice something not right please say so, and If you start your house on fire or blow yourself up, it ain't my fault!

I'll start the example using the drivers that Vulcan intends to use in his next build for surrounds, information on that design can be found here. This design uses the Dayton DC130BS-4, but I might use some different drivers along the way to emphasize some points.

In a typical HT setup the surrounds and center channel, etc. are crossed over at ~100Hz, letting the sub take over the base duties. Some receivers let you adjust that frequency and even let them run full range. Here's a couple response curves using the DC130 in the two box alignment suggested in the write-up, Sealed in white and ported in blue.




There isn't a lot of difference in the plots but the ported design does have an f3 (where the output is down 3dB and is accepted as the point where the system is rolling off) about 25Hz lower than the sealed version. Seems like a no brainer, you can get a little more bottom end and maybe set the crossover frequency to the subs a little lower. But lets look at what kind of power this setup can take. Again sealed in white and ported in blue. This graph shows how much power the driver can take before exceeding it's Xmax limits, as in the maximum mechanical travel it has before you start smashing it's innards.




Some important stuff to note here, both drop off pretty steeply below 150Hz. The poted design has a rise again where the port output helps augment the driver output but then drops like a stone to about 1/4 of that of the sealed version an we're only talking a couple watts here. So by the response graphs it looked like 100Hz was OK for the sealed version and maybe the ported would have been OK down to 80Hz but looking at these graphs a 150Hz crossover point seems more suited unless the system will only be used at very low volumes. But even then anything with some power behind it below ~40Hz especially for the ported version would cause physical damage to the speaker. And the reason I'm using HT for an example is that there's lots of signal down this low where music (in most cases) has next to nothing below 30Hz. So considering that example - in order to protect the speaker well against infrasonic content you'd want to crossover at at least 100Hz but 150Hz would be better - which pretty much eliminates any bass advantage the port gave you in the first place. Again if you're only using these at low volumes using only a dozen watts or less then you might be OK but even still you need to protect either design from content below 40Hz.

For a different view point I'll model the 7" Seas CA18RNX that I use in my build linked to above. Here's the response curve for typical sealed and vented enclosures.




Ported looks much nicer reaching nearly 30Hz lower, so lets look at the maximum power curves.




Much nicer looking than the curves above but the ported version will quickly self destruct below 30Hz, fine for music bad for HT where agian a lot of content below this - and you won't have time to jump up and turn the volume down before damage occurs. The sealed version though is very happy to take the rated wattage value of the speaker (referred to as it's thermal limits) all the way down. So even though it won't produce frequencies as low as the ported version at least it is still protected, though you would still want to cross over to the sub before the natural roll off of the driver.

Hope that helps at least someone. Again if you see any errors or gaps in my explanation please let me know.

Ryan

[vulcan500rider]

Thanks for the explanation, Biff. Can you explain graph #2 a little bit for me, though? It looks to be Hz on the horizontal axis, but I'm not 100% sure what the vertical represents, nor where exactly the "danger zone" begins for the driver. Actually, I'm not sure I really understand the vertical on the first graph either; if it's dB, how is it that we're working in negatives? Sorry, I'm just a little confused.

Also, I did a bit of looking at my HT Receiver. It claims that, with 6 ohm loads, it's rated for an RMS of 90w @ 120-20000 Hz (speakers only, not sub). Does this mean it crosses over to the sub at 120Hz? It also says that in surround mode it can max at 140w/channel RMS--though that's with less than 10% THD, rather than the 1% quoted for the 90w spec above.

[biff]

Originally Posted by vulcan500rider

Thanks for the explanation, Biff. Can you explain graph #2 a little bit for me, though? It looks to be Hz on the horizontal axis, but I'm not 100% sure what the vertical represents, nor where exactly the "danger zone" begins for the driver. Actually, I'm not sure I really understand the vertical on the first graph either; if it's dB, how is it that we're working in negatives? Sorry, I'm just a little confused.

Sorry, didn't notice that the graphs don't have the axis labeled. The horizontal axis is always frequency. the vertical axis for the 'transfer function magnitude' is in dB and the vertical axis for the 'maximum power' graphs is watts.

You're probably thinking of dB as in sound pressure level (SPL) but dB is simply a logarithmic ratio. Wiki it if you want to see how it's calculated. For SPL its a ratio of the sound pressure level where 0dB is defined somewhere as some very low volume - I think I've read somewhere that 0dB(SPL) is the equivalent to a mosquito buzzing from 10ft away. The one used most often in this kind of stuff is plus or minus 3dB. A 3dB difference is the equivalent of a doubling and -3dB is half, so 3dB is twice as much as 0dB and 103dB is twice as much as 100dB. In the magnitude response graphs the -3dB point is where the output is half of the reference point of 0dB. For an example look at the first graph, the sealed design is down approx. 7dB at 70Hz and the ported box is down about 4dB. So if both designs were given the same signal strength the ported box would be twice as loud as the sealed design at 70Hz.

The maximum power graphs show how much power in watts the speaker (as in the driver itself) can handle at a given frequency. Speakers have two basic limitations, one is the thermal limit and one is a mechanical limit. The thermal limit is how much power the speaker can take before you start over heating the voice coil, this is the wattage rating you see on the spec sheets. The mechanical limit is how much the speaker cone can physically move before you run out of suspension travel, or smash the voice coil into the back of the magnet structure. In the first graph you see the lines flat spot at 30W, this tells you that even though the speaker isn't moving to it's mechanical limits you can't put anymore power in or you could start to cook the voice coil. Where the line dips below this you've reached a point where the speaker is moving at it's mechanical limits even though you may not be anywhere near it's rated power of 30W

Referring to the second graph, the power handling of both designs drops fairly steeply below about 150Hz. And at 80Hz either design can only handle about 8 Watts. The power handling of the ported design then starts to rise again as you go lower towards the tuning frequency of the port. This is because the port is 'controlling' the motion of the drivers cone so it is actually moving less - you don't lose any SPL though because now the port is contributing some of the bass as well. As you go below the tuning frequency the port loses control of the speaker cone movement and the speaker cone can flop around as if there was no box at all, this is why the power handling drops like a stone as this point in the graph and is only about 1/4 of that of the sealed design, though both are pretty low at this point. Looking at these graphs I think the 120Hz crossover your receiver has is about right to protect them from any serious bass... if you use them at high volume. And if you're crossing at 120Hz you're pretty much not getting any benefit from the extra bass the ported design was supposed to give you.

I used the 7" drivers I have as another example. You can see in the response graphs that they play very nice with a ported box, but looking at the max power graphs the line drops off much more severely than the DC130's example below the tuning frequency. Though the sealed design is happy to take it's maximum rated power all the way down. If I were using these for HT, the sealed would be safe running full range but there wouldn't be much point since they can't produce much below 100Hz. The ported looks much nicer for this application but I would absolutely have to make sure they were crossed over at about 40Hz or a serious bass hit would destroy them but looking at the response graphs I'd probably cross them at 80Hz anyway since there response starts to drop not much below that. This example doesn't have the midbass droop in power handling like the first example but if I tuned the box lower to try to play the speaker lower before the port loses control of the cone movement then that droop would start to show up.

Originally Posted by vulcan500rider

Also, I did a bit of looking at my HT Receiver. It claims that, with 6 ohm loads, it's rated for an RMS of 90w @ 120-20000 Hz (speakers only, not sub). Does this mean it crosses over to the sub at 120Hz?

That seems to be what they are saying.

Hope that clears some stuff up. If you have anymore questions let me know.
Ryan

[vulcan500rider]

Thank you for taking all the time to explain this. I read your reply over about 5 times, over the course of a couple of days to make sure I actually followed it--I may be a little slow, but I think I finally understood all of it.

I'm definitely not going to bother going for the ported version of the surrounds. You've convinced me that there's no feasible benefit involved, as my receiver will crossover before any benefit is seen--thankfully, as there could be damage to the speaker otherwise. One thing I'm not sure on now, however, is whether I can use these speakers at all... If I'm understanding right, these will max the voice coil at 30w, and my system is RMS of 90w per channel. Now, I realize that I won't be putting 100% power into these at any time, but that still seems awfully low. Would it make much of a difference to take the tweeter into account?

Again, thanks. Your explanation helped me feel *slightly* less out of my depth...

Sean

[biff]

Don't get too hung up about watts. You's be surprised at just how few watts it takes to get loud. You need lots of watts when trying to produce bass but since you would be eliminating all bass duties from this speaker by crossing them over at 120Hz all that 30W that the speakers can take goes into mids and highs. Here's a plot of those DC130's with a 120Hz high pass filter fed with 30W... I believe this is at 1m away. (SPL in dB is the vertical and frequency is the horizontal)





Now look at the chart about 2/3rds the way down this page and compare. A jack hammer operating about 1m from your head is ~100dB. They'll be fine for surrounds. Mind you you won't get the big dynamics as if you had a big 3way floorstander sitting behind you crossed over a lot lower.

BTW you don't need to take the tweeter into account, the actual energy that high up - where the tweeter will be playing - is pretty low. You'll max the woofer before you max the tweeter.

BTW(#2) I looked into the specs in the software I used for modeling the 7" (the CA18RNX) since the results above looked too good to be true. And they were too good to be true. This software uses the term Xmax (the limit of mechanical movement) to be one way only (which is standard), not total movement. Someone entered the total movement distance in this spot so the software thought the driver has twice as much travel as what it actually does. When I remodeled the design with the proper data that power handling dip in the upper bass was there too. If you'd like to see the revised graphs I can put them up, but the point is just to let you know they aren't some 'super' driver.

OT: one thing you could do depending on your needs/wants/budget is use one of the MTM's you already have built as a center channel and split the other one to make the surrounds. You have to buy another tweeter and a few more XO components though. Then you could build mains with better speakers. The tritrix are a very good speaker for the money but they are an ultra budget design. If you'd like better SQ for listening to music you could look into building higher end mains using better drivers, without wasting what you already have.
.... just a thought

Ryan

[vulcan500rider]

I wasn't so much worried about maxing the tweeter. I was actually thinking that if the woofer had a max of x power, the tweeter consumed y power, and my receiver supplied z power, then the actual power to the woofer, x = z-y. Therefore, the amount of power the receiver is feeding into the woofers would be less, and there would thus be less chance of maxing them out. As you noted, however, because of the minimal amount of power it takes to produce a high frequency vs a low (which makes perfect sense, had I thought about it) means that y is going to be quite minimal when compared to x.

I'm actually a bit relieved to see that the other speaker you had graphed isn't such an enormous step above the drivers I'm using. I had chocked up some of the difference to size (7" vs 5 1/4), which I assume means that a correspondingly larger room for speaker movement is left--though I may be mistaken in that assumption. However, that still didn't seem to quite add up.

To your OT, what I had actually thought about doing is taking advantage of the recession destroyer (or whatever exactly the name is) package again from partsexpress, which gives me parts for 2 MTMs. Then I could just buy one more tweeter, some extra crossover parts for the MTs, and I've got a center and two surrounds for about $300 CAN (I think--I still have to run all the numbers again). My problem is that I'm not really sure what I could replace the drivers with that wouldn't a) be significantly more expensive and b) require a completely different crossover that I'm really not capable of designing at this point.

Honestly, I'm also not sure how much benefit I'd gain from changing them out. The TriTrixs were such a step up from the Sony HTIB that I started with that it's hard for me to really imagine stepping up again...<laugh> I'm sure the audiophiles feel different, though <laugh>

Sean