Which alignment for a RSS210?

[Ryan_M]

I'm starting to draft up some box drawings for a 3way I'm starting - anchored by a RSS210 obviously. I wanted floor standers and drew up the cabs roughly to see about how much space I have to work with. I've got ~45L to play with for the woofer after taking into account driver displacement and braces and such.

Now the issue... Modeling using Zaph's T/S data it looks really good using the full 45L and tuned to 20Hz to 22Hz but the problem becomes the port. putting in a 60W signal (about the max says WinISD) port velocity is huge using a 2" port getting up to 46m/s at tuning. Using a 3" port the air velocity looks much better (20m/s) but there is a first port resonance at 244Hz and I plan on crossing around 300Hz. Also the port length gets a little long so I'd have to use an elbow.

A PR seems like a good option but there's the added cost and I wanted to have curved cabs to I've got no place to put one. One other possibility is a slot port. I've heard that folding the port once or twice will break up the port resonances, any truth to that?

Any advice would be appreciated.

[diy speaker guy]

Now the issue... Modeling using Zaph's T/S data it looks really good using the full 45L and tuned to 20Hz to 22Hz but the problem becomes the port. putting in a 60W signal (about the max says WinISD) port velocity is huge using a 2" port getting up to 46m/s at tuning. Using a 3" port the air velocity looks much better (20m/s) but there is a first port resonance at 244Hz and I plan on crossing around 300Hz. Also the port length gets a little long so I'd have to use an elbow.

20 m/s is still a bit high (IMO), considering that you will probably turn it up well past xmax, at least on occasion. With a tuning that low it will only be movie effects that cause chuffing, but still...


A PR seems like a good option but there's the added cost and I wanted to have curved cabs to I've got no place to put one. One other possibility is a slot port.

You can use a PR if you like but if you don't want to you really don't have to. You might also want to consider a decreasing cross sectional area tl or mltl, like Paul K always recommends. It's a good idea when port lengths get too long in ported boxes, and/or as an alternative to putting the driver and port in the same end of a ported box, which can sometimes cause terrible lumpy frequency response.


I've heard that folding the port once or twice will break up the port resonances, any truth to that?

Not really, 1/4 wave resonances are dependent on length and shape (cross sectional area) regardless of folding. Folds can create nulls based on 1/2 wave between surfaces but they won't break up resonances per se.


Any advice would be appreciated.

MJK's worksheets. No matter what alignment you end up using I recommend a proper simulation, especially in this case since the port is longer than normal and I'm assuming you are planning on putting the driver and the port in the same end of the box. It might end up measuring a lot uglier than WinISD might have you believe.

If you don't have the worksheets I can probably do an mltl simulation for you if you provide all the box details and driver parameters. If you want a Paul K style tl I'm sure he would be happy to help, and I'd be surprised if he didn't already have a simulation for that driver. You can actually build a tl like this in a curved cab, it's a bit harder than a rectangular box but it's definitely possible.

[Paul Carmody]

I've been in this exact situation, trying to figure out which vent would be best for that driver. The software does show high port velocity for a 2" vent.

HOWEVER, I slapped this driver into a 45 L cabinet, which happened to already have a 2" dia vent installed. Well, in practice, it turned out that the 2" vent was fine. I didn't hear any chuffing (although I admit I only listened to music and watched movies--I did not run low sines through them)

Mine is just a piece of PVC and a tiny roundover. I think a 2" precision port with its giant roundover would really alleviate any leftover turbulence.

[dlneubec]

One of the benefits of a slot port is that you can size it more precisely. If you want something in between the equivalent of a 2" and 3" diameter round port, you can do so. The area of a 2" dia port is less than 1/2 the area of a 3", so there is a big change between those sizes.

[Wolf]

I used 2.5" electrical conduit for mine, Ryan. We speaker builders don't adapt just one type of pipe you know....

Later,
Wolf

[isaeagle4031]

No kidding Ben. I've got some 8" PVC in the shop waiting to be put to use. :)

I like slot vents a lot as they also add rigidity to the enclosure. But its about impossible to tune after its built

[Ryan_M]

diy speaker guy, I don't really know anything about TLs and MLTL's or what the advantages or trade offs are. I guess I'll have to do some reading. I appreciate you working one up for me as I have no idea how to design one and AFAIK the sheets are in MathCAD(?) so I"m useless there too but I don't want to put you through any trouble though as I'm not sure which way I want to go yet. Again I don't know the +'s and -'s for TLs and getting one designed to fit in a curved box might be more trouble than what it's worth - if a simple ported solution will work just as well. FWIW for HT they will be hading off the bottom end to a bunch of RSS315's but I don't think I care for subs for 'critical' listening.

Paul, what tuning did you use? Did you get a chance to crank them up?

Dan & Ben, I was considering the slot port for just that reason. I have 2.5" electrical conduit 'in stock' however it would require a port longer than what the boxes would accommodate if it were kept straight and a 90° elbow for electrical conduit would probably have a radius of 15" - again too big to fit in my box. ABS elbows are much shorter but don't come in 2.5" that I've seen.

Ben, doesn't the RSS210 in the Attitudes run up to ~800Hz? If so wouldn't that put the first port resonance right in the middle of the passband using the 2.5" port? How did you deal with it?

Thanks again everyone for your input so far!

[Wolf]

Ben, doesn't the RSS210 in the Attitudes run up to ~800Hz? If so wouldn't that put the first port resonance right in the middle of the passband using the 2.5" port? How did you deal with it?

Thanks again everyone for your input so far!

~550 Hz. I included it in the measurement/trace for the woofer, and it didn't seem to matter much in listening. I just left it. Dips aren't heard as easily as peaks.

Later,
Wolf

[diy speaker guy]

diy speaker guy, I don't really know anything about TLs and MLTL's or what the advantages or trade offs are. I guess I'll have to do some reading. I appreciate you working one up for me as I have no idea how to design one and AFAIK the sheets are in MathCAD(?) so I"m useless there too but I don't want to put you through any trouble though as I'm not sure which way I want to go yet. Again I don't know the +'s and -'s for TLs and getting one designed to fit in a curved box might be more trouble than what it's worth - if a simple ported solution will work just as well. FWIW for HT they will be hading off the bottom end to a bunch of RSS315's but I don't think I care for subs for 'critical' listening.


The difference between a regular ported design and an mltl design (for the purposes of the service I offered) is simply that a ported design considers only the theoretical helmholtz resonance but the mltl design programs calculate 1/4 wave resonances instead of a helmholtz resonance. Sometimes a simple helmholtz resonance simulation is fine and reasonably accurate, but in some situations a proper 1/4 wave model is much more accurate. Your proposed design has at least a couple of features that would be better simulated with a 1/4 wave simulator.

What I offered was not a design. I just offered to model whatever you are designing, but with software more appropriate for this design.

WRT the decreasing tl design idea, those type of enclosures can have certain distinct advantages, especially when the equivalent ported box would require a very long port. It's simple to simulate and design and simple to implement, even in a curved cab, as long as you plan ahead.

[Paul Carmody]

Paul, what tuning did you use? Did you get a chance to crank them up?


I used a piece of 2" dia PVC that was 8" long (I think this is about 23 Hz). I think there was a small roundover at the end. It had a low-n-lean sound. If I did it again, I might go for a bit more "oomph," and tune up to 27 Hz (5.5" long). This gives a subtle boost around 34 Hz.

Yes, I did crank them on various occasions. They seemed very comfortable doing LFE in movies. I couldn't hear any chuffing, but like I said I never ran straight sine waves through them.

[Whitneyville1]

How 'bout the 2" "precision port"? It's supposed to cure these things...I guess.

[diy speaker guy]

You can use the program "Flare It" http://www.subwoofer-builder.com/flare-it.htm to find out how much a flared port can actually cure. Flares do help but there's only so much they can do. The program I mentioned will clearly show the 2 issues at play here. The first is boundary layer turbulence, and this can be helped with the flared port. The second is core turbulence, and the cure for that is a larger port.

Once you understand the boundary layer turbulence issue, you can see how a high aspect ratio slot port will have problems, since so much of it's area is so close to the boundaries (walls) of the port. Basically what happens is that the air near the port walls doesn't move much.

[Paul K.]

FWIW I modeled both a tapered TL and an ML-TL having net line volumes of ~45 liters. Both were 72" long. The taper ratio I used was ~20:1, and in the ML-TL, the port was 2"D x 6"L. In both cases with a 60-watt input, which drove the RS210 to ~Xmax + 15%, the maximum air velocity in the terminus and port was 18-20 m/s. This is a bit higher than desirable, but since this peak velocity occurred at ~18 Hz, I wouldn't worry about it. F3 for both TL types was in the mid-20s. These were exploratory modelings and did not take into account the shape of the 45-liter volume; they may not be easily implementable in what the OP wants to build.
Paul

[diy speaker guy]

These were exploratory modelings and did not take into account the shape of the 45-liter volume; they may not be easily implementable in what the OP wants to build.
Paul

As long as the box is not prebuilt I'm pretty sure it would be fairly easy to implement either of these designs, as long as OP is happy with a 36 inch height.

BTW, which model looks better? It would be nice to see pics if you have time to post them.

[901Fixer]

I'm starting to draft up some box drawings for a 3way I'm starting - anchored by a RSS210 obviously. I wanted floor standers ....

Now the issue... Modeling using Zaph's T/S data it looks really good using the full 45L and tuned to 20Hz to 22Hz but.....


I'm curious why you would tune so low for music?

[Paul K.]

The taper ratio is ~21:1 and the system tuning frequency is ~21 Hz. I modeled using the specs posted by PE for this driver. The line is 72" long with the driver's center 14" from the beginning of the line. The line starts at 9"W x 8"D, and ends at 9"W x 0.375"D. Stuffing density is 1 lb/ft3 for the first 37" of the line. I've attached the system bass response (red line) for an input of 2.83 volts, showing an F3 of 26-27 Hz and an F10 of 18 Hz. With a 60-watt input (into 4 ohms), the output is ~102 dB SPL, the driver hits Xmax +15% at ~32 Hz, while the terminus air velocity peaks at ~23 m/s at 18 Hz, but is 17 m/s at 27 Hz, and less on up in frequency.
Paul


As long as the box is not prebuilt I'm pretty sure it would be fairly easy to implement either of these designs, as long as OP is happy with a 36 inch height.

BTW, which model looks better? It would be nice to see pics if you have time to post them.

[Paul K.]

Same line length of 72" and driver's center again at 14" from the beginning of the line. The mass-loading port is 2"D x 7"L and its center is located 4.5" from the other end of the line. The stuffing density is 0.75 lb/ft3 with the first 35" of the line stuffed. The line's cross-section is 9"W x 4.25"D the whole length (this may not be practical in terms of clearance behind the driver, but it was necessary to keep the line volume at ~45 liters). The port would have to be bent, of course, in order to fit in the line's 4.5" depth. I've attached the system bass response graph (red line), for an input of 2.83 volts, showing an F3 of ~23 Hz and an F10 of ~17 Hz. An input of 60 watts causes the driver to reach Xmax+15% at ~32 Hz, but the excursion decreases (as it did for the tapered TL) to a minimum at the system tuning frequency of 21 Hz, then climbing again and exceeding Xmax+15% at ~18 Hz and below. With this 60-watt input, the port air velocity hits 30 m/s at 19 Hz, but is at 17 m/s and lower at all frequencies above 30 Hz. The advantages of this ML-TL are it would be easier to build and have a lower F3, but the tapered TL will have lower air velocity in its terminus. Special note: this driver, like some other drivers, is very sensitive to the amount of resistance added in series with its coil. The resulting decrease in sensitivity is not uniform across frequencies; it primarily attenuates the output only above ~100 Hz, making it necessary to tune the system lower in order to once again flatten out the response all the way down to the knee in the curve.
Paul


As long as the box is not prebuilt I'm pretty sure it would be fairly easy to implement either of these designs, as long as OP is happy with a 36 inch height.

BTW, which model looks better? It would be nice to see pics if you have time to post them.

[Paul K.]

For the two TLs above I modeled with zero resistance, like that from a crossover inductor, added in series with the driver. Taking the modeled response for the ML-TL as an example, I added 0.5 ohms in series with the driver and made no other changes to the model. The attached response graph illustrates what I mentioned in the post on the ML-TL; the resulting attenuation is not uniform across the frequency range and would require tuning the system lower to flatten the response. Oh, just so no one gets confused, this effect will affect the tapered TL the same way.
Paul


As long as the box is not prebuilt I'm pretty sure it would be fairly easy to implement either of these designs, as long as OP is happy with a 36 inch height.

BTW, which model looks better? It would be nice to see pics if you have time to post them.

[Ryan_M]

Well I did some looking into the TL and it doesn't really work out well from a building it perspective. I'm using a box 40" tall but because there is the mid enclosure I cant start the TL at the top of the box, I actually end up ~8" short. If I understand what Paul K. is saying then once you add in some real world resistance the TL would need to be tuned a little lower which (I'm only assuming) would require a longer TL length making the situation worse. If I fold it a second time the box ends up deeper than what I'd like.

I've decided to go with the 2" precision port. Paul C. has used this woofer as a sub for HT duty in a similar sized box with a 2" port with minimal flare and didn't notice any issues - and I trust his opinion. I won't be relying on it as a sub for HT duty and in fact these will have an easier load since they will hand off to some RSS315's in this application. I checked out "Flare It" and it says I'm OK above 30Hz at max power. AFAIK there isn't much content in music below 30Hz and I probably will never get them to max power anyway. Also the port works out just short enough that I don't have to worry about elbows and the first port resonance is at least an octave above the planned XO frequency.

The 2" precision port works out well for me in a lot of ways and I'm not sure I see much point in designing in a safety margin for a scenario these will never see, though they apparently will work in such just fine. That's my armchair engineering view on it anyway...

Thanks everyone for the input!
Ryan

[diy speaker guy]

Ok, that's fine. But I still recommend simulating your proposed design with MJK's worksheets. Since it's a 3 way and you will probably crossing over pretty low it probably won't matter all that much even if the response is less than perfect, but I still recommend a proper simulation. There's a higher than average chance of lumpy response if the driver and port are both at the same end of the box. You haven't actually stated your proposed layout but that's usually how 3 ways are laid out, with the bass driver and port both near the floor.

Good luck.

[Ryan_M]

FWIW I've attached my proposed layout.

I had a quick look for MJK's worksheets, I've found some that were in mathcad. Those the ones I'm looking for?

[johngalt47]

Did you do that in SketchUp?

[Ryan_M]

AutoCAD

[johngalt47]

Cool...I use AutoCad but I usually do 2d.

[diy speaker guy]

FWIW I've attached my proposed layout.

I had a quick look for MJK's worksheets, I've found some that were in mathcad. Those the ones I'm looking for?

Yup, those are the ones, but you didn't have to go looking for them. I offered to do the simulation for you. To do an accurate model I'd need some dimensions to work with (like cross sectional area of the line, exact driver and port locations, stuff like that). I also need driver specs (including Bl and Sd) but I suppose I could find them myself.

There's probably nothing to worry about though, since the port and driver are NOT crammed into the same far end of the box, I suspected they might be but they are not. The driver and port location might not be perfectly optimum (for smoothest response) but probably not too bad as shown.