[6thplanet]

What impact does FS of a speaker have on the low end response in a U frame set-up? Also I've read the MK paper on U/H frames, in the equation he lists, how it L relitive derived? And is "c" a constant? I'm tring figure out the max depth of the Sides based on upper x-over frequency. I'm running my woofer up to 300 Hz, his demo was based at 200, with a depth of 7.5" I'm guessing to push the quarter wavelenth resonate up pass my goal it'll be shorter, but by how much?

[holdent]

In Martin's paper L is the depth of the U or H-frame and as he notes "The depth was selected to push the first quarter wavelength resonance above the desired crossover frequency of approximately 200 Hz" (http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf).  "c" is a constant and is the speed of sound (344 m/sec). 

U & H frames have a characteristic peaking response (see Figure 5) which he tames by selecting an appropriate 2nd order xover to flatten the peak (Figure 1). 

I'm not sure what you are referring to when you wrote "L relative".

[6thplanet]

Sorry, mis quote, what is " r " effective  in the "L" effective equation. From his example I couldn't find where that came from.

[Rudolf]

AFAIK "r" effective is the radius of a circular disk with the same surface area as that of the H frame. This implies that the H frame cross area is not stretched too much from quadratic.

Rudolf

[6thplanet]

Wow, is there another equation for doing that? I tried to get the radius of a square using Pythagorean Equation from a 16" x 16" sqaure as noted and I get 11.3". How he arrives at 9" I don't get?
If I use 400 Hz as a safe frequency for my 300 Hz x-over I get a 33.9" wavelength. Quarter that to 8.475", so can I assume my side length not to be any longer than that?
Also I'd still like to know how FS effects low freq. exstention. I see that he (MJK) states that length doesent effect low end efficiency, it's the baffle size. So would a lower driver FS help in that respect, too?
Thanks for all this help!!

[Rudolf]

Wow, is there another equation for doing that? I tried to get the radius of a square using Pythagorean Equation from a 16" x 16" sqaure as noted and I get 11.3". How he arrives at 9" I don't get?

16x16 = 256
256/3.1414 = 81.49.. (where 3.1414 = Pi)
Square root of 81.49.. = 9

[6thplanet]

Ahhh, thanks buddy! Well I messed around with the U frame a bit, then turned that into an H frame. Wow. I went big at 23"x23", then trimmed that to 22" at the bottom and 16" at the top, 7.5" depth front and back. Incredible! This is my first venture into OB stuff and this H frame thing is awesome. I can't believe I've been building boxes for what seems like no reason... I'll prolly trim it up some more and see how far I can go and still get the bass response I have right now. Cool stuff!

[SteveRB]

...wow.

[Rudolf]

I'll prolly trim it up some more and see how far I can go and still get the bass response I have right now.

I see, that you have already adapted to the spirit of this forum
Martin's papers will have shown you, that there is an upper and a lower limit for each H frame. Make it deeper, and you move the lower limit to lower frequencies. But at the same time the upper limit moves lower too - and a resonance peak develops at the upper end.
I always found that making a H frame no deeper (front to back) than it is wide, keeps that resonance peak from becoming an issue.
Your actual H frame begs for a second driver mounted below the installed one.

Rudolf

[6thplanet]

I'd like to add another woofer but these were buyout deals and I only bought 2 (still kicking myself in the rear for not buying more) As for the H frame, what's up with N frames? Can sticking a wide baffle at a ~45* angle and slim up the width of the front, still yielding the same results as a wide baffle/ width combo? Any thoughts on that?

[Rudolf]

I'd like to add another woofer but these were buyout deals and I only bought 2 (still kicking myself in the rear for not buying more) As for the H frame, what's up with N frames? Can sticking a wide baffle at a ~45* angle and slim up the width of the front, still yielding the same results as a wide baffle/ width combo? Any thoughts on that?

I've built N and M frames. They behave - surprise, surprise - like a crossover between H frame and W frame. Going lower, but suffering from a pronounced resonance peak, which HAS to be equalized. In the end I would go for a half W frame or a "ripole", if width is an issue.

But you know that you can build a H frame much slimmer than yours?!

Rudolf

[6thplanet]

N framed, slimmed the baffle to 14x17, installed at a 45* angle, and now the the actual opening is 10" at the top and 12.5" at the bottom. Still has exhalent low end but, as noted it gained in the midbass. I damped the rear and a bit in the front to control the midbass. Then I raised the gain of the neo8 to match up with the woofer. Sounds good, I'm liking this set-up, not to mention the neo8 is a smooth sounding planar!

[MJK]

I think your U, H, M, N, W, or ripole designs would become more efficient if you limited the cross-sectional area to a square just slightly larger than the driver's OD. Then set the enclosure down on the floor to maximze bass reinforcement. In this case smaller might be better and deeper.

[6thplanet]

I originally had the woofer down low for that purpose, but since the planar only plays to 500Hz the woofer ( crossed @ 300Hz ) has to fill in the rest. The low mount woofer was drawing the imaging down because of the distance between the drivers, hence I had to bring the woofer up closer to the planar. Also, the surface area of the baffle dictates the lowend response, right? Is there any way to physically measure any distances to dictate the lowest frequency it can produce? I still havent heard what driver FS might coralate to as far as low end cut off.

[MJK]

I originally had the woofer down low for that purpose, but since the planar only plays to 500Hz the woofer ( crossed @ 300Hz ) has to fill in the rest. The low mount woofer was drawing the imaging down because of the distance between the drivers, hence I had to bring the woofer up closer to the planar.

Have you considered mounting the woofer in a small H frame down near the floor and then putting the planer directly above but tilted back so the listener is on axis? Or tilting both the H frame and planer back so the listener is on axis? This keeps the drivers close, gets bass reinforcement from the floor, and may help eliminate the floor bounce. I am surprised that you had imaging problem, my drivers tend to be fairly far apart without any issues as long as the crossover frequencies are low.

Also, the surface area of the baffle dictates the lowend response, right?

It is one of the variables that determine the low end response.

Is there any way to physically measure any distances to dictate the lowest frequency it can produce?

If your H frame cross-section was close to the driver area a prediction could be made based on the depth of the H frame. With your very large area the model breaks down and I am not sure how accurate any calculation would be.

I still havent heard what driver FS might coralate to as far as low end cut off.

I don't understand the question.

[6thplanet]

  With the woofer down low the c-c distance was right at 34-35". The x-over is 500Hz on the planar, 300 on the woofer, so I used a wavelength of 400. The 400Hz wavelenth is 34", so I was right on the max which I guess is why it was drawing the sound stage up & down. This is relative to where I sit also, it might not have been an issue if I were say further from the speaker. Bringing the woofer up solved that issue.

  I see your point with the lowering everything and tilting it back, I thought of tring that at a later stage of experimentation. Let me ask this, would you tilt the whole H frame or just the baffle inside the H frame? I ask because if I cut an angle at the bottom, the back opening would be less than the opening in the front. Or maybe just angle the planar...

  The FS question has to do with the driver. If a speaker has a 40Hz Fs, how would it differ from a driver with a 25Hz Fs in the same H frame if all other specs were the same. I'm also under the assumtion that drivers with a Qts higher than .50 are more suited to the OB frame set-ups, this true?
 
  Lastly, thanks for all the feedback guys!! This is/has been a neat experience having never played with this type of enclosureless design. Its all a new and exciting adventure for me.

[MJK]

I see your point with the lowering everything and tilting it back, I thought of tring that at a later stage of experimentation. Let me ask this, would you tilt the whole H frame or just the baffle inside the H frame? I ask because if I cut an angle at the bottom, the back opening would be less than the opening in the front. Or maybe just angle the planar...

The FS question has to do with the driver. If a speaker has a 40Hz Fs, how would it differ from a driver with a 25Hz Fs in the same H frame if all other specs were the same. I'm also under the assumtion that drivers with a Qts higher than .50 are more suited to the OB frame set-ups, this true?

I would build another smaller H frame with a bottom and tilt the whole unit back.

In my opinion, higher Qts drivers work better in dipole systems. But you cannot just look at fs and Qts, you need to look at the entire system including the crossover. A driver with an fs between 40 and 50 Hz and a Qts around 1.0 is very easy to work with, you can go passive or active and get bass down to 50 Hz. Once the fs drops below 40 Hz the design becomes more difficult, passive may no longer be an option if efficiency is a concern. Active plus EQ may be required. You need to do the design work to get the optimal solution for any driver, I don't think a general rule of thumb statement is possible.

[6thplanet]

Cool, I have no problem trying different designs. I don't get why a lower Fs would need EQing? Seems like you'd want to EQ a driver with a higher Fs to achieve FR below that point. Anyway, I'll give the smaller tilted frame a go and see what happens...thanks!

[JohnR]

I think the point is that with higher Fs there be too much eq needed to get low. With lower Fs it's feasible but you still need eq to compensate for dipole rolloff

[6thplanet]

Any reason an H frame can't be done with a tube? Like converting a small kick drum by mounting a baffle in the middle? Any weird resonances ya think compaired to a square/rectangle?