Skram vs. C2E - Subwoofers info research

Measuring rear rejection has to be done properly, and that’s a problem if you’re in the near field where cabinet shadowing will be occurring.

The first thing is to ensure that the rear position is equi-temporal with the front one. And that also means finding the acoustic centre, if you want to check near field rejection. Then you have to go a decent distance from that, at least a few metres, in both directions. You’ll find that any box of a dual 18” size has 3-5 dB near field rear rejection like this thanks to the cabinet’s shadowing. That's also present on a TH118XL and is what I see in the last imgur link too, but it isn't particularly exceptional.

I think confusion maybe stems from people not realising that most of the larger manufacturers refer to far field data, as this is where the vast majority of listeners to PA systems are going to be.

Anyone who has setup a wall of horn subs will notice this too, but while impressive up close it isn’t often very useful in real life once you get to distance.

The best thing is still to measure in the far field, where it matters. 10 metres (which is what the polar maps are generated at in my model) is a good indicator for real, useful rejection. The model uses the boundary element method, which solves the wave equation in 3D as a partial differential equation called the Helmholtz-Kirchhoff integral - so all cabinet geometry effects are included, unlike an isophase lumped element model like Hornresp.

You’ll see on the animated heat map of output that close to the cabinet there is a drop of 3-5 dB in the colour allocation, but this disappears in the far field. This effect is also created primarily by the cabinet’s size and shape, including the size of the radiating area on the front, regardless of internal layout. I spent 7 months researching this last year, with real-world measurements to qualify plenty of models. There’s always plenty of room for more investigation, but I the only thing I can see about the Paraflex alignment which could cause rear rejection is the fact that a lot of the low and upper corner response has phase offsets created by the stacked series resonant pipes. You can see a similar but less exaggerated effect in the response ripple of a straight transmission line, and those are modelled very well by a BEM model of sufficient mesh resolution.

Cardioid system designs such as gradient or end-fire have to be made with consideration as to whether you want a ‘clean stage’ or to avoid upsetting the neighbours many miles away. It is a case of ‘pick one’ usually, and only physical acoustic circuits like the Fulcrum patent or the Danley BC achieve both.


As an aside, I appreciate you guys popping out of the FB group to share the data. I genuinely think a lot of these concerns stem from the somewhat insular nature of the community and the use of Facebook for it. Sometimes the fast-moving chat and flood of cabinet drawings on the HOQS can come across like some magic bullet has been found, even if that isn't the intention. Plus, the data on there isn't yours and isn't easily shareable outside. What would you do if the group gets Zucc'd?!

We've all been excited and inexperienced at gathering worthwhile data, but learning happens a lot faster if things a presented more as questions and discussion across a broader range of communities.

Ricci wrote: Jay michali wrote: Great conversations happening here, thanks everyone. It's too bad we aren't closer. I'd talk you into letting me borrow a few cabs for a full workup over the summer. I wonder if there are any C2E cabs near me in the US...Perhaps something worth looking into as I'm planning to get back at it this year.

I have nothing to contribute but wanted to thanks everyone for your input, that's a brilliant thread.

 Hello everyone!   

  Considering how the topic of some rear-rejection or a directional effect as observed with the HOQS Paraflex subwoofers has become a focus of this discussion it is helpful for us to have more real world data for us to work with, so with that in mind here (below) are some contributions from our friend Patrick Sander in Munster Germany ... 

 These are multiple outdoor measurements taken at distances ranging from 1m to 10m.... 
 The subject is a single Paraflex Golden Formula C-2E 1x18 subwoofer in a field .....Fundamental tuning of this cabinet is 35hz... This cabinet's outer dimensions are 24" x 48" x 36" ..

 At each distance he first measured with this cabinet facing the microphone on-axis, and then without changing the position of the mic (or measurement settings) a second measurement was taken after turning the cabinet around (facing away from the microphone) while keeping our cabinet located in the very same place ...  

First at 1 meter:



Then at 2 meters:




Here we are at 4 meters:





At 6 meters, still looking great here:



 Then finally at 10 meters (almost 33 feet) :

 



   As you can see this amount of rear-rejection which we have in these charts coincide fairly well with what we see in the chart at the link which Jay Michali shared in this discussion earlier .......

 Jay's Paraflex subwoofer is of a different design being a Type "A" layout based Paraflex as opposed to the GF C-2E Paraflex which Patrick has,  so you can see that the dip & bump features are in different locations but the amount of rear rejection is very similar . .

In case anyone missed it earlier here is the chart which Jay had kindly provided:

 

So the effect has been measured out to 10 meters, and is still pronounced at 10 meters .. .
  
 We cannot yet say for sure if it diminishes beyond that point but nevertheless 10 meters is almost 33 feet and many stages aren't even 33 feet deep, so this is in fact a useful feature of HOQS Paraflex subwoofers, and it is one which has already been used to a practical advantage during shows & events, even by some who have responded in this discussion .. .     

A few examples of reports include a happily acceptable amount of bass energy on stage despite the array of Paraflex subwoofers being located directly in front of that stage, and another report included avoidance of noise complaints from a certain direction (as opposed to other more conventional cabinets which had previously been used in that particular challenging event space generating more complaints from neighbors without producing any more SPL on the dance floor ..)... 

 A combination of consistent measured data & consistent reports from experienced people whom we trust is extremely valuable to us...   After all these fellows are in the trenches with our boxes..... They come from various backgrounds & other forums (some are from Speakerplans apparently and are still members) ..... They have various levels of experience but some of them are seasoned industry veterans who have worked with many more PA cabinets in the field than I ever did, so contrary to whatever perceptions there may or may not be this community is quite the opposite of "insular" (if that word was used in the derogatory sense, but of course not, since there is no place for that sort of thing in a technical forum) .... In other words what they have to say holds a lot of weight, and is much preferred over conjecture from afar .. 

  We do encounter skepticism all of the time and I can totally understand why...... I would probably be skeptical too if coming from an outside perspective ......   Of course it doesn't help that Akabak is not predicting the sort of passive-cardioid or rear-rejection effect that we are measuring, and it was one of our own developers by the name of Tom Armitage-Giles (also of The UK) who was the first to point this out to us a while back when the predicted pattern in his Akabak simulation of a Paraflex subwoofer did not even remotely resemble the reports .......  Not sure if anyone ever figured out why that is (note: it was inconclusive and probably never even hit the public feed for that reason),  however, Tom is fully brilliant so I don't doubt his Akabak skills.....  

Toasty, if you don't know Tom already you two should connect,  he is a good guy, highly recommended   



Yes, the Facebook group is fast paced which can often be fun but posts do get buried quickly and we also realize that it is not always easy to navigate and find what you want, unless you know exactly how to go about it and even then is a little tricky at times ........ Yes, being Zucc'd would be terrible, and that is always a risk with Facebook so we definitely don't want to "have all of our eggs in one basket" so to speak    ........... We do love our Facebook group while also staying aware of the issues and have taken some measures, but ultimately a new platform would be best and that will take time (work on it has already begun) .. 

Here's some comparative data, since talk is cheap. I forgot my own rule of pictures = thousand words!

DAS EVENT-218A ported subwoofer, at 3-metre distance in a very large hall with Schroeder frequency of approximately 30 Hz. Traces taken with separate matched mics front and rear, impulse matched to acoustic centre:


Vertical stack of 3x Danley TH118XL tapped horn in landscape orientation, 4-metre distance outdoors. Separate mics again, front and rear, distance matched to the acoustic centre:


Note the overlapping phase traces at front and rear positions, in both sets of data? If the ‘delay locator’ or acoustic timing reference is kept the same between measurements (turn off the automatic ‘set offset to IR peak’ in REW) then this is an easy way to make sure you know the physical distance offset is from the acoustic centre of the cabinet. Just move the mic or box back and forth, until the phase overlaps. You might be surprised where the mic ends up!

Also, here's an example of a tuned cardioid system, same location as the other TH118XL trace but with an extra rear-facing sub added and tuned for maximum rejection in the far-field at the lowest notes, which were causing issues off-site and at other stages:


Edited by toastyghost - 11 February 2022 at 1:15pm

Sinai Sound wrote: Talking in depth about speakers! On speakerplans! What dark magic is this!!! When it's freezing cold, and dark at 7pm at night, I'm taking one of these into an open space somewhere with Kyle

I know, right?! it’s a shame our demo stock of battery-powered 4x 1000W DC amplifiers won’t arrive til mid March, because that opens up a lot of outdoor space well away from boundaries. Generators aren’t the one as the noise contaminates the data.

Also, I just updated my last post to clarify that you have to keep the timing reference consistent between measurements. By default, REW will automatically shift that on each new trace.