OB theory questions

0 Members and 1 Guest are viewing this topic. Read 4077 times.

JohnLL

  • Jr. Member
  • Posts: 37
OB theory questions
« on: 3 Mar 2012, 05:29 pm »
These are theory questions, but I am going to pose them using an example. Lets say you want a two way OB with one driver being a 15” H frame. Obviously the larger the top panel the lower you can cross over but the more visually imposing.

Let’s say you chose a 17”x17”x.75” ply for the top panel and mounted a 4” driver dead center.  Edge models the baffle step at ~ 600 Hz. I wish I could use cad, but I am going to have to ask you use your imagination for following alternative.

Suppose you cut a .75” channel in 4.5” diameter PVC and then cut the ends at 45 degrees.  You mount this PVC on each side of your 17”x17” baffle similar to a picture frame. In profile the baffle is still 17X17 overall and the ply that is still visible is 9”x9”. Looking down (in plan) you see a 4.5” PVC, 2.25” radius on the front side of the baffle and the same on the back.

1.   I think (or hope) the driver sees the second example as equivalent to a 37”x37” flat baffle. Edge models the baffle step at ~ 250 Hz. The math is (4.5 X 3.14) -4 =~ 10”. The “4” is the inside diameter of the pipe.  10” X 2 =20”+ 17”=37 “.  Is my assumption re the baffle step correct?

2.   I am hoping the 2.25” radius nearly eliminates baffle diffraction, front and back.  The only baffle diffraction software I am aware of models the radius is at the edge of the baffle, not proud of it. Therefore the software is only suggestive. What are your thoughts?

3.   I think I have created some sort of wave guide-for better or worse.  It cannot have any effect on high frequencies.  The throat is pretty open and it flares rapidly so I would guess its overall effect is minimal. Thoughts?

I would be happy is share how I might actually build this, but lets first discuss whether it makes sense. Thanks In advance!

Rudolf

Re: OB theory questions
« Reply #1 on: 3 Mar 2012, 08:44 pm »
Sorry John,

but I won't understand what you are explaining without some visualisation. There are too many ways for me to misunderstand your words. No CAD needed, just some scribble.

Rudolf

opnly bafld

  • Full Member
  • Posts: 2412
  • 83 Klipsch LSIs
Re: OB theory questions
« Reply #2 on: 3 Mar 2012, 09:40 pm »
If I understand he wants to put 4.5" pipe on the sides of his baffle.

    pipe
p            p
i  baffle    i
p            p
e            e
   pipe

JohnLL

  • Jr. Member
  • Posts: 37
Re: OB theory questions
« Reply #3 on: 4 Mar 2012, 12:15 am »
Thanks opnly bafld ! You found a simple way to convey what I could not. What opnly bafld drawing shows as baffle is 9"x9". I got a little too fancy partially describing how it might be built. John

JohnR

Re: OB theory questions
« Reply #4 on: 4 Mar 2012, 01:57 am »
From what I understand, the area of the baffle with the pipe on it would be 18x18". The extra "thickness" due to the pipe won't appreciable affect the low-frequency response.

My suggestion for something to try would be to put the pipe so it only goes out the back... let's try...



Since you're using a full-range driver, the radiation at the back will get pretty chaotic above a couple of kHz anyway... or, here's another idea:




JohnLL

  • Jr. Member
  • Posts: 37
Re: OB theory questions
« Reply #5 on: 4 Mar 2012, 04:59 am »
Great image JohnR! I tried messing with cad before I posted but it would have taken me forever to get it. The essence of my first question is when does LF turn from 2PI to 4 PI. Does sound follow the counters of the baffle or the extremes of the length and width? I believe from your post (maybe incorrectly) that you believe it is the extremes that rule. I am not arguing that but I want to confirm my understanding. In my imagination significant curved surfaces on the edges of a baffle lower 2PI just like wings on a baffle. Not so? It seems "obvious" to me, and I cannot imagine how I could be wrong but......   I still want your and other opinions.

JohnR

Re: OB theory questions
« Reply #6 on: 5 Mar 2012, 12:48 am »
The essence of my first question is when does LF turn from 2PI to 4 PI.

I'm sort of struggling with that question as I have never thought about it that way. Since you ask, though, I wonder if it explains why the modeled dipole peak never seems to be as high as the idealized dipole model. Need to think about that more... but away from home this week so can't run Edge. However, since you are using Edge, have you tried the check box to tell it to model open baffle? Clicking that on and off will compare closed box vs open baffle quickly, I think that will help with your question.

Quote
In my imagination significant curved surfaces on the edges of a baffle lower 2PI just like wings on a baffle.

I think it will make a difference, but not as much as you are expecting. Here is (I believe) your proposal, with the front radiation in blue and the rear radiation in red. The difference between the two path lengths determines the amount of cancellation. I'd guess that the effective increase compared to not having the pipes would be about 2.5" - so effectively the baffle width would be about 22". Best way to find out would be to make a prototype and measure it ;)



mcgsxr

Re: OB theory questions
« Reply #7 on: 5 Mar 2012, 01:59 am »
By memory, and I am not a science guy so forgive me if I botch it, we are verging on territory that a member named JohninCR used to play around in.

I recall he raised eyebrows using pool noodles on the edge of baffles to claimed positive effect.

Myself I have not played around with diffraction reduction measures like this, but I would recommend a search on some posts of John's from around 2-5 years ago here on AC.

Rudolf

Re: OB theory questions
« Reply #8 on: 5 Mar 2012, 02:44 pm »
The essence of my first question is when does LF turn from 2PI to 4 PI. Does sound follow the counters of the baffle or the extremes of the length and width?

In a dipole there is no baffle step (turn from 2PI to 4 PI), because there is a driver at both 2PI spaces - albeit with opposite phase. From my experience differences between front and back radiation of a driver don't really start below 400 Hz (or even 800 Hz).

I believe you need to look into http://www.linkwitzlab.com/models.htm. No need to understand the equations, just try to follow the pictures without the math.  :D
If you want to worry about something, it would be the point, where the baffle width exceeds the wavelength of the frequency in question. That's were the radiation pattern starts to change from the dipole 8 to more exotic petal patterns.

You will find a concise description of the edge diffraction process at http://www.linkwitzlab.com/faq.htm#Q8. I had to read it four times, before it finally sank in, but after that it pretty much explained all my questions.

Regarding the effect of your PVC pipe JohnR's picture is spot on from my POV - as always. Probably you should be aware that sound does not travel in straight lines, but in spheres. Every time such a sphere hits a border or an edge, it starts a new sphere - which interacts with the original one. Very difficult to illustrate properly, but that's how sound can move around a corner.

Rudolf

JohnLL

  • Jr. Member
  • Posts: 37
Re: OB theory questions
« Reply #9 on: 5 Mar 2012, 03:49 pm »
Thanks for the responses! Rudolf I am aware of your work concluding that baffles need to be sized by frequency. This leads to 3 or 4 way speakers. I am most definitely not saying you are wrong. Linkwitz is way over my head.

JohnR on reflection I think you are probably correct. I toyed with OB a year ago and eventually decided my room was too small. It was your recent paper outlining the advantaged of ob subs near field that caused me to reconsider. Thank you.

I guess ob will have to wait until I have a more appropriate space or maybe I might combine ob bass with a sealed box for the HF driver. Best to all. John

JohnR

Re: OB theory questions
« Reply #10 on: 5 Mar 2012, 10:37 pm »
Didn't mean to discourage you. The hybrid might be a good combo in a small space although I haven't personally tried it yet. What size is the room and what are the likely speaker placement from walls and distance to listening position?

It was your recent paper outlining the advantaged of ob subs near field that caused me to reconsider.

Thanks :) I have another one about to come out that looks a bit more closely at how distance to listener / from wall affects bass response.

JohnLL

  • Jr. Member
  • Posts: 37
Re: OB theory questions
« Reply #11 on: 6 Mar 2012, 12:17 pm »
Thank John-I am not discouraged. My man cave-ette is a spare bedroom. My better half does not care for my classical music and so near field (and ob bass) enables me to play it louder to my ears.
I am interested in what you discovered. I think you can get some of the new small drivers to go 70-75 hz F3 sealed. In a small room it would be hard to have a ob sub positioned so that you could actually see the ob driver. I am not sure that is important crossing over that low. At this point these are just idle thoughts.

mightym

Re: OB theory questions
« Reply #12 on: 1 May 2012, 01:17 am »
JohnLL,

I'm not shilling for "Dave" at planet 10 here, but since you have a small room and listen ~"nearfield" anyway, have you considered some sort of FAST alignment like the afore mentioned's "Tyson" design?

I was just reading this thread as I try to catch up after being out for a while.....It makes some sense to me, even though it's not OB.....

Just a thought....besides there just weren't enough John's in here....;)

John

zmyrna

Re: OB theory questions
« Reply #13 on: 2 May 2012, 04:31 pm »
I have the same "pipe" idea actually.
For convenience, I will use them only on the vertical edges.
And I plan to fill them with sand.
So I am hoping benefits on:
1. edge diffraction
2. baffle dampening