"Over the shoulder" vented box

[jphutchins]

Hello,

After years of research into speaker theory I might have an excuse to build something! A friend has suggested a $1,500 (total) budget for a small PA.

I am interested in building a couple subwoofers each of which can be carried over the shoulder on a strap. Extension required is 40Hz. My current design, assuming it actually works, looks to max around 120dB with two boxes at full power. Is 120dB enough bass outdoors for 50-100 people that are all quite close to the system?

Each box would use 2x MCM 55-2421 8" woofers and has sensitivity around 90dB

Here is a diagram of the slot-ported enclosure:

Input parameters for HornResp:

Response with 40V (max after EQ is applied):

Excursion at 40V:

And port velocity from WinISD 240W input power:

In my heart of hearts I'd love an FLH (like JBell's "modular" MCM 8 bass sticks) but as a first time builder and considering the "over the shoulder" concept something sad like this seems best. Let me know if you think 120dB is enough and where I made mistakes, bad assumptions, or where there is simply room for improvement.

Cheers,
J.P.

 

[David McBean]

Let me know if you think 120dB is enough and where I made mistakes, bad assumptions, or where there is simply room for improvement.

Not sure why, but there appears to be a significant difference between the WinISD-calculated port velocity and the equivalent Hornresp results with 240 W input (obtained using the Maximum SPL tool with Pmax = 240 W and Xmax = 99.9 mm).

 

[jphutchins]

The WinISD numbers looked alarmingly low. I'd been away from the project for about 5 months and I vaguely recall struggling to get it under 30 m/s so was surprised to see 16 m/s.

It looks like I should accept the update that HornResp recommends when I open it and find that tool. Thanks for the help, your software rocks!

I'll model some wider ports in HornResp and see if I can get it under 30 m/s.

 

[David McBean]

Not sure why, but there appears to be a significant difference between the WinISD-calculated port velocity and the equivalent Hornresp results with 240 W input (obtained using the Maximum SPL tool with Pmax = 240 W and Xmax = 99.9 mm).

I worked out why there is such a difference.

The 240W WinISD input power is not held constant across the frequency range, it is simply used with Re to calculate a constant value for Eg as follows:

Eg = SQRT(Input_power * Re) = SQRT(240 * 3.29) = 28.10 volts

Setting Eg = 28.10 volts in Hornresp gives a port inlet velocity figure comparable to that calculated by WinISD.

 

[jphutchins]

Newer version has more cool stuff! OK, I've made some progress but raises more questions. Given the 40V input HornResp is showing a much more reasonable 27m/s port velocity. I duplicated higher 70m/s results using maxspl=240W and no xmax limit. Of course this shows a 135dB output since xmax is the limit in the 40-60Hz range! My question I suppose is whether or not this should concern me - of course the speaker is very efficient around the port tuning so it gets tons of SPL/W but in reality I wouldn't be feeding it more than 40V which results in 27m/s port velocity ca. 115dB. The sample window shows that it finally rises to 240W at 38Hz from a low around 10W at ca. 50Hz.

 

[David McBean]

My question I suppose is whether or not this should concern me

I will have to leave that question to someone who has more practical experience in these matters than I do .

EDIT - A quick Google search has revealed that the vent air velocity should ideally not exceed 34 m/s.

http://www.eminencedesigner.com/EDTopic14.html

 

[Rademakers]

This system does 120 dB/1 meter continuous (flat from 50 Hz and up), it consists of 2 Cubo 12's getting a total of 175 W (~87.5 W each). Fifty people is do-able, especially if they can stand around the system (as in 360 around).

https://www.youtube.com/watch?v=6o2gfNyxArI

https://www.youtube.com/watch?v=j_J04cdg3os

The sensitivty of each Cubo 12 is around 99 dB/W/m (so 102 dB/W/m for the pair). Particle velocity doesn't exceed 8 m/sec. For your application I would advise to keep 25 m/sec as a maximum and the use of a 50 Hz high pass.

Best regards Johan

 

[jphutchins]

Thanks Rademakers! I am familiar with your Cubo designs and very impressed by the results but as 40Hz is a solid requirement for me I'd have to look at your bigger ones that are not so easy to carry. I think I might have to stick with the boring concept of bass reflex + lots of power to get a portable 40Hz sub... Maybe there is a TH or FLH that fits the bill but AFAIK there is simply no way to get the kind of horn path needed for 40Hz in a sub that can be carried over 1km by one person.

Cheers,
J.P.

 

[chris661]

So you'd like to make an outdoors PA system with 4x 8" drivers for subs?
Any chance you could go bigger and use wheels, like a suitcase?

It really does depend on what you wanna do. For live music, there's a reasonable chance that a pair of 8"s per side won't be enough. I think I'd take a decent 15" per side, minimum, if any bass is required.
For the sort of thing where there's not another sound system for miles around, yours might be enough. Just. If possible, I'd take more so you're not constantly ragging the couple of drivers you've got.

FWIW, there's a lot of SPL to be gained by using a too-small cabinet tuned high. I did a boombox with a pair of 8"s that was 22L for the pair, tuned to 50Hz. Got a huge bump from 55-80Hz, and didn't notice the lack of fundamentals much. With a pair of those 15w/ch class D amp boards and an active XO, it was pretty loud 50' away, and feel-it-in-your chest loud when standing close.

Chris

 

[more10]

Second chris661. Find some proper second hand drivers. You may want to avoid the rise in response at 40 Hz.

 

[tb46]

Hi jphutchins,

Just a few comments to the MCM 55-2421 in a BR. First I would use Zaph's measured T/S parameters (unless you have measured your own). Second, a JBL style jet port might help a lot in this type of application, while the particle velocity in the center of the port(duct) is still high, the velocity at the ends can be reduced quite effectively.

I'm attaching a Hornresp Export file for a dual MCM 55-2421 BR using a duct style similar to the JBL jet duct. The duct ends see a much smaller particle velocity. To see the frequency response: go into the Loudspeaker Wizard, and set the graph to Power-Memory-Combined, then click on Filter Wizard, and adjust the necessary filter (the lower and upper cutoffs) as desired to see the simulated frequency response or displacement.

The actual response will be lower because of power compression in the drivers. My guess is that you'll loose about 2-3 dB @ an Eg=40V (displacement peak with filters is only a bit above 5mm).

Regards,

 

[xrk971]

Tb46's design here based on my PPSL band pass may work as an over the shoulder design with same drivers.
http://www.diyaudio.com/forums/attachments/subwoofers/451740d1417487191-pp-slot-loaded-sub-alpine-swr-12d2-mcm_55_2421_ppsl_sv.pdf

 

[just a guy]

I worked out why there is such a difference.

The 240W WinISD input power is not held constant across the frequency range, it is simply used with Re to calculate a constant value for Eg as follows:

Eg = SQRT(Input_power * Re) = SQRT(240 * 3.29) = 28.10 volts

Setting Eg = 28.10 volts in Hornresp gives a port inlet velocity figure comparable to that calculated by WinISD.

The difference between 16 m/s (WinISD) and 76 m/s (Hornresp) for the same input parameters is a huge and shocking discrepancy. I've only used WinISD a couple of times since Hornresp included the velocity feature so I never noticed this discrepancy before and I'm really glad you brought it to our attention.

I'm not really good with the electrical side of things and not familiar with the formulas the simulators use so I just want to make sure I understand this all correctly. With a speaker and an amp in real life the resistance varies with frequency so I assume the simulators use Eg and the resistance at each frequency to calculate the frequency response curve. Correct?

If that is correct, I assume Hornresp uses Eg and the resistance at each frequency to calculate the velocity too. Correct?

From what I understand from the quoted post, WinISD uses Eg and a constant Re to calculate velocity. Correct? This is the wrong way to calculate velocity since resistance is not constant at all frequencies. Correct?

So Hornresp calculates velocity correctly - what you see in the sim is what you would measure if you could actually measure velocity. And WinISD calculates velocity wrong - what you see in the sim is dramatically lower than you would measure if you could actually measure the velocity. Correct?

Sorry for asking so many questions but I think a thorough understanding of this massive discrepancy is vital.

 

[Brian Steele]

1. carried over the shoulder on a strap.
2. Extension required is 40Hz.
3. enough bass outdoors for 50-100 people that are all quite close to the system?

Off the top of my head, I'd say no. 60 Hz? Yes. 50 Hz? Maybe. 40 Hz? That's pushing it.

Dual neo-magnet (because weight IS an issue) 10s (or 12s) in small vented cabinets (Fb=40Hz) along with some decent DSP to boost the response at low frequencies may work. How are you planning to power these things?

 

[sine143]

dayton pa 385s in vented boxes would my cheap/small option.

 

[jphutchins]

@Tb46

Your "jet port" model is showing a 1dB greater max SPL over the original while reducing the entry/exit particle velocity from 27m/s to ca. 17m/s. The limit to max spl is displacement hitting the 8mm limit from 60-100Hz at 116dB. I am fiddling with the numbers now to see if I can keep the 40Hz response while reducing diaphragm displacement between 60-100Hz. A higher tuning as many have suggested may do the trick.

Thanks,
J.P.

 

[jphutchins]

This is fantastic to model a port in using the horn parameters. I'd like to verify the techniques so I don't get too lost.

I don't know how to get correct Atc values but in testing boundary conditions there is little difference making it quite large. Making Atc below a certain threshold causes everything to go haywire but I think my design will stay within the area where it doesn't seem to matter much.

Best regards,
J.P.

 

[jphutchins]

Throat Particle Velocity?

Modeling it in this manner has allowed me a design that I would be more capable of constructing though it would make the cabinet slightly larger.

The port exit velocity (mouth) is low, max 12m/s at max power. I'd assume that means that port entrance velocity is the same since it's the same size.

Modeling it this way allows us to see the velocity at the very narrow section of the port, now modeled as the throat. Here, peak particle velocity approaches 80m/s at 40Hz. Will this cause problems or be heard at the mouth?

Thanks,
J.P.

 

[more10]

Sorry for asking so many questions but I think a thorough understanding of this massive discrepancy is vital.

WinISD simulates a Helmholz chamber and Hornresp a pipe. You need to place the port at the end of the cabinet when simulating in Hornresp. A resonating pipe will give the air some extra speed through the port I believe.

 

[more10]

Here, peak particle velocity approaches 80m/s at 40Hz. Will this cause problems or be heard at the mouth?

I believe it will be heard. Not like having this speed at the mouth though.