Dual skar12 paraflex C CRAM

[David McBean]

Hi maxolini,

I would be very interested to know the sequence of events that led to the stepped segment screen layout shown in Post #83 (copy attached below). For some reason the S2S, S3S and S4S throat area labels for segments 2 to 4 are missing.

(The stepped segment screen layout shown in Post #89 is also not correct in that the S4S throat area label for segment 4 is missing, however I have found the problem in this case and the bug will be fixed in the next update).

Kind regards,

David

 

[Booger weldz]

I think that screen is well over a year old version of ‘horn response‘ David?

 

[LORDSANSUI]

I mean, the paraflex C model he found, MMJ was modeling using TH with open rear chamber, while you and me modeling it as PH1, so, worth maxoline compare the results, maybe one is more accurate then the other.

To make it more visual, see attachments.

The box are exactly the same, but the way you define each segment and the way you arrange them are different, each model will present some small issues like:

For TH: What is the position and the size of S4?
For Paraflex: How to make H7 stepped at S8?

 

[Booger weldz]

Neither it would seem. The results (as drawn and in many older ‘paraflex models) get tainted with the excessive size (‘Q’) CrossectionAl area of the too short sides resonator in comparison with the ‘fundametal‘ (long) side. Fix that and it ‘works’ much better By filling in the 3/4 harmonic issue ‘perfectly’ . (Qw basics)

the actual distance to exit and from both closed ends is what ends up in the impedance plot. especially with appropriately sized chambers that aren’t allowing akward discontinuities ina system that ‘shares’ its guts and the resonances if it’s smooth and ‘flat’ ish in the REAL response?

redraw/resize that and sim to see this in greater detail/ further understanding
so it’s ‘correct’ with what is measured when it’s done that way? that design isn’t promoting an ideal relationship between resonators in parallel And continues to create confusion and phase issues that don’t have to exist.

That resonator is ‘too big and too short’ to ever match up with the 3/4 harmonic issue created in the long side. They will never ‘match/compliment eachother’ ..

 

[Booger weldz]

http://www.fonema.se/qpipe/qpipe.ht...sQUij2-feN_TtdRrty7IfqV-jacY996mHXJwUYF2DfyEg


horn response continues to show the discontinuity. It manifests in ways you can estimate by simulating the exit as 0.1 instead of trying to split all that up (as parts of either, or?) hard to say, the real issue is the internal size and lengths that dont ’work’ anyhow(???) and until they do, it’ll never pan out ‘perfectly’.

build back better 😝

 

[David McBean]

I think that screen is well over a year old version of ‘horn response‘

Thanks Bw, if that is indeed the case then it's probably the reason why the screen appears as it does. Mystery solved .

 

[maxolini]

View attachment 1149313


’perfect’ being no trace of the 3/4 harmonic and cancelation notch from the long sides 1/4 wave lengthfundamental

Ok , and that mean?? Cause what's the harmonic at 3/4 and what's the 1/4 fundamental?
I mean what freq?
The fb?

 

[Booger weldz]

http://hyperphysics.phy-astr.gsu.edu/hbase/Waves/opecol.html#c1

 

[maxolini]

So 3m is the golden length of the pipe... Horn?

So if I aim for a 3m length that will put me in the sweet spot and the folding is the key so to have a box that do not resemble a tall locker but more like a big fridge ...

I am struggling to get the areas of the drawing in to horn resp, any tips to calculate S1,S2....S5
Crossectional areas?

I wish that HR can be linked to freeCad so when you slide the controls on the speakerwizard it change the dimensions on the 3d plan, then generate the 2d drawings and cut list.

I really admire who can generate all by hand, cad and sketchup are tools to draw but you need to thinker the measurements and areas by yourself.

So basically you tweak your dream SPL target in HR then sweat blood to translate that in to a live cabinet dimensions , folding, bracing, cut list optimization.

All the great DIY designs were done around 2009 2014.... And looks like we hit the point of "you can't reinvent the wheel"

 

[Booger weldz]

Horn response allows you to polish up the designs of the past and tweek them to specific TS parameters. No wheel to reinvent. It’s all there already. It’s 1/4,3/4,5/4 …. Use it or dont🧟

 

[maxolini]

by 1/4 ,3/4 and so on you mean you can use all X/4 scheme ? in which X equals and odd number ?
1/4 , 3/4 , 5/4 .... and so on ? of your target lower response you want to hit , like a 30hz flat cabinet ?

 

[maxolini]

What is that little piece of wood there ? does it really needs to be there ??

what about this

 

[maxolini]

ok so , so more questions....
this is not draw to scale ( the drivers )??

where is S1 and S2 on the plan
cause i do not know when using 2 drivers one in top of the other, HR treats them as one bigger driver ?
so S1 must be in the middle of the 2 , instead of the middle of the driver #1 ?

 

[Booger weldz]

join FB messenger…. This form of communication will only continue to confuse/fail everyone

 

[maxolini]

join FB messenger…. This form of communication will only continue to confuse/fail everyone

Booger,
I respect you point of view however i do not agree with it,
if your statement was true, then this forum have vanished from the Web since FB messenger or any other form of IM.

but is here to stay

 

[LORDSANSUI]

This panel is increased to create the stepped joint, I mean, the throat of H3 segments is bigger then H2 mouth segment.

Sometimes this different acoustical connection can improve the response.

The opposite of stepped segment is to have the segment mouth = next segment throat a kind of "continuity".

As always suggested, you need to run simulations to see the results and confirm the benefits or not.

 

[maxolini]

Ic,

Horn Length increased from 244 to 277 cm
i tweaked the HR from the original CRAM to a bit longer path length to get rid of that spike

i increased the red boxed segments and decreased the blue boxed segments
also gained more lower extension
i dont care above 100 , 90 Hz for a sub

since the CRAM looks easier to build i think this will be my 1st sawdust experiment
double 12 CRAM box...
just i need to confirm that the below looks good with you guys

 

[Booger weldz]

s1-s2 and s6-s7 are the driver distance (to center of the pair)from the closed ends of each resonator in ‘ph1‘ mode which is the most accurate.

if you fold a 90 cm resonator up three times(270cm) and attach it to another( 90 cm ) resonator in parallel you have only to center the drivers at 45 cm to find a nice/simple in phase with no discontinuity in the enclosure As seen @ ~ 96 hz in your sim (the 3/4 harmonic of the 1/4 (32hz) hiccup that’s ‘missed’ the mark as shown )


what you have created looks great(close enough to the hypotheticsl ‘perfect’ )

 

[Booger weldz]

Booger,
I respect you point of view however i do not agree with it,
if your statement was true, then this forum have vanished from the Web since FB messenger or any other form of IM.

but is here to stay
View attachment 1150637

join FB messenger…. This form of communication will only continue to confuse/fail everyone

I’m just trying to make it possible to show you the videos of my fusion 360 simulations of real (built/measured) . You can’t do any of that in this forum. If you want to understand the stuff and make it possible for people to communicate in ways that technology has progressed into simple .?

i added the photo app thanks to you helping me. Maybe you need to progress/adapt to ‘more’ (?) as well. It’s up to you, no disrespect or worrues 🙏🏻

its important to avoid any phase discontinuity/sizing issues in these parallel qw resonators and understanding that is your friend. Otherwise you might not like the result and effort/wood/$$

they can ring, go silent or be way to hot up top if not careful(some of this can be exposed/seen already with horn response by shuffling how you ‘explain’ the converge to exit in an effort to expose any ‘weird’ that will exist at some drive levels or the real result.

@USRFobiwan knows this area of concern quite well . He is a wealth of really good/ tested info on these particular designs and a huge help as he explains things better/straight up than most

 

[maxolini]

what you have created looks great(close enough to the hypotheticsl ‘perfect’ )

Lets Page @USRFobiwan and see what he thinks of this
i just want to start the actual sawdust when all the brains here agree that it is worth to cut some wood

cause it happens that you end up with something then somebody come up with a better or improved solution and you said to yourself
gezzzz,,,, i have must waited some...

like littlemike said on post #85
https://www.diyaudio.com/community/...15-or-jbell-2-cube.394208/page-5#post-7229989

" Any sub is better than no sub "

Measure your results, and accurately measure both SPL and Impedance, otherwise you do not know what you have. There are cheap ways, there are more expensive ways, but ALL of them are cheaper than the drivers and amps you're gonna use. The mics and meters won't lie, but they may hurt your pride.

Pick one, buy some wood, and make some sawdust. Any sub is better than no subs, and if it is not what you wanted, you can always design and build something better.