[rick57]

I will almost certainly get it, but other than Martin's project 7 www.quarter-wave.com/Project07/Project07.html  anyone have Open Baffle worksheet experience so far to relate?

Thanks

[nullspace]

Hi. I've used it, and it very closely approximates what I get in-room. It's not comprehensive in that it doesn't take into account every last nuance of your listening environment, but it's the best out there and does a fine job to boot. I highly recommend it.

Regards,
John

[MJK]

It's not comprehensive in that it doesn't take into account every last nuance of your listening environment

Yet.

but it's the best out there and does a fine job to boot. I highly recommend it.

Thanks, I appreciate the positive feedback.

Martin

[Rudolf]

Those OB worksheet did less for me than I expected - and more, all at the same time.

I found that they can´t (not yet) account for all important variables in the room, so results still have to be taken with care. But most competiting programs would not even allow for that.

What I did not expect was to get a great learning tool with them.
By fiddling around with the different variables I got a much better understanding of what causes what and where to head to for a desired result.

So you might not get that precision landing with those sheets, but you certainly will know which direction to go.

In short: highly recommendable IMHO

[nullspace]

Martin,

Have you seen http://www.musicanddesign.com/roomgain2.html on room gain and dipoles? There are some folks over on diyaudio who confirm the results; my current OBs couldn't make it to 40hz with a tailwind, so I can't comment personally. I'd be interested to hear your opinion and whether or not room gain as described could be integrated into your worksheet.

Regards,
John

[MJK]

John,

I had not seen that analysis before.  I did a quick read and can offer the following off the cuff comments.

1. His monopole souce has to be a sealed enclosure, a TL or BR is a dipole souce at low frequencies below the cut-off.  So in some ways I think he is talking apples and oranges.  My Alpha 15A drivers have a fs of about 40 Hz, so for an equal comparison a sealed system would need to be used that is tuned to the same 40 Hz.  For many of the small two way sealed systems most DIYers build I don't think they get to 40 Hz.  For a sealed sub that goes down to or below 40 Hz I think that the major reinforcement is due to floor or corner loading.

2. His assumption is that the room is tightly sealed so that it can be pressurized.  I personally have an isolated rectangular room in my house with a door that I use for my stereo and I do not believe by any stretch of the imagination that it is even close to being tight enough to build pressure.  There are so many leaks that I don't believe I will see the effects he is simulating.  So his analysis is probably for an extreme condition that most people will not have in their homes.  So one has to ask are the conclusions valid?

3. I believe that my OB system easily reaches down to 40 Hz, it is a big baffle.  I state this based on my analysis of the OB and my analysis and correlation test results for a few TL's that I have used in the same room.  So I am judging the 40 Hz based on a few years of experience in that room with systems that measure flat to 40 Hz.  Assuming that the OB calculations are correct I can see a smaller baffle with a woofer and full range driver getting to 50 Hz easily if the system is designed right.

4. As for what is discussed on forums, I don't believe any of it. 

That is my take,

Martin

[Bob in St. Louis]

As for what is discussed on forums, I don't believe any of it. 

Except the Off Topic stuff.....That's all true.

Bob

[JohninCR]

Regarding the link to JohnK's info about the lowest limits of dipole
bass, I think I understand why dipoles roll-off so steeply below
the room's on axis resonance, but I'm having trouble with why a
dipole would have a peak there.  It makes sense that the front
wave would have the same peak as a monopole at that room mode,
however, shouldn't the rear wave share the same peak and create
a net null as shown lower in frequency?

[nullspace]

JohninCR,

I have no idea.

Martin,

Thanks for your comments, and thanks again for the helpful MathCad worksheets. I've attached a sim of what I'm hoping is my summer project -- Fertin 20EX and Madison Executioner X21 on a baffle 26"x42". If I don't finish working on the back deck first before starting another speaker project, SWMBO will tear me a new one and since painting weather here in NJ is about at an end until next year, it'll be April at the soonest before I get started on the speaker prototype in earnest.

Regards,
John

[Rudolf]

I had not seen that analysis before.  I did a quick read and can offer the following off the cuff comments.

Martin,
I believe you should take the time to reread that article (part 1 and part 2) unhurriedly. Obviously you did not yet find all the information needed for a balanced view. I really would appreciate to see a more in-depth judgement of yours regarding John´s article/findings.

1. His monopole souce has to be a sealed enclosure, a TL or BR is a dipole souce at low frequencies below the cut-off.  So in some ways I think he is talking apples and oranges.  My Alpha 15A drivers have a fs of about 40 Hz, so for an equal comparison a sealed system would need to be used that is tuned to the same 40 Hz.  For many of the small two way sealed systems most DIYers build I don't think they get to 40 Hz.  For a sealed sub that goes down to or below 40 Hz I think that the major reinforcement is due to floor or corner loading.

In most any listening room this room gain difference would only show well below 40 Hz (longest room dimension > 4 m). So maybe you weren´t confronted with this phenomenon anyway.

2. His assumption is that the room is tightly sealed so that it can be pressurized. 

Figure 3 in John´s article accounts for a quite large opening in the side wall. Actually a room can only be pressurized as a whole at very low frequencies. As soon as the largest room dimension is lambda/4, if not before, we have to think in pressure gradients across the room to draw valid conclusions.

3. I believe that my OB system easily reaches down to 40 Hz, it is a big baffle.

John talks about "the woofer was placed in what would be considered a typical postion, with the response computed at a typical listening position." Listening to a dipole in the very nearfield you will hear ANY frequency the driver can deliver - regardless of room size limitations. With the transition to farfield room reflections will take over, leading to the complex dipole response we know. Can we veryfy (or falsify) his simulations without knowing those positions?
 

4. As for what is discussed on forums, I don't believe any of it. 

It´s not the medium that has to be mistrusted, but the participants 

Sorry to be that much off topic, rick

[MJK]

John,

Very impressive response.  Can you tell us the crossover and boost configuration you used?


Rudolf,

I reread the article and still have the same concerns about the assumptions. I think assuming the room is sealed is way off the mark so in my opinion the room gain due to pressurization is grossly over estimated.  Also, with respect to the opening modeled, no details are provided that tell the reader what acoustic impedance boundary condition was used.  If the opening has the acoustic impedance of a rectangular piston in a baffle I would expect the resistive component to be very high, dominating the problem and standing wave would be significantly attenuated or even eliminated.  I guess I just do not agree with the plotted results or the discussion.  Everybody should form their own opinion.

Martin

[rick57]

nullspace
That's encouraging to hear, thanks

Rudolf,
> you might not get that precision landing with those sheets, but you certainly will know which direction to go
A canny observation. 
Imo it's ok to be off topic if it's interesting, as is here. (Initially I was "excited" by JohnK's work, but while I think he's done interesting work in several audio areas, wrt dipoles & cardiods etc, I've become somewhat sceptical re his assumptions). 

> It's not comprehensive in that it doesn't take into account every last nuance of your listening environment
MJK: Yet.

Martin, any particular area(s) in mind? 

Cheers

Rick

[MJK]

Rick,

I am always looking at upgrades to my MathCad worksheets.  If you look back over the past few years you can piece together the trail of upgrades.

With respect to the OB worksheets, I have a JAES paper that seems to take the dipole modeling done in programs like the EDGE (which is essentially the same algorithm as my worksheets) and goes to the next level.  I need to finish working out the math and program it.

With respect to other worksheets, I am upgrading my horn theory so some more horn worksheets may come this winter.  I have sketched out the method for analyzing the BIB geometry up against a rear wall and in a corner.  I would like to upgrade the baffle shape options and the ability to model toe in of the speaker.  There are a few more areas I am interested in exploring.

I have no schedule or priority on these updates.  To be honest, the response to the new worksheets has been a bit underwhelming.  So I have decided to go back to working on whatever interests me at the time and release new worksheets when I am happy with the final results.  No schedules, no promises, no special interests projects, no more outside work for others, no more self induced pressure, and significantly more fun.  I have also cut way back on my forum participation.

Martin

[nullspace]

Martin,

High-pass is first-order at 132hz. Low-pass is first-order at 28.5hz with 15db of boost, and out-of-phase. With first-order filters, I'm planning on PLLXO's, a cap for the high-pass and a RC filter for the low-pass. The Madison pro audio woofers are 21" across and have 18mm of xmax & plenty of power handling.  I'm cautiously hopeful the asymmetrical xover works out in practice as well as it models. I have some concerns about amplifier integration, with SET on the top and pro audio on the bottom, but we'll see.

If you'd like, I'd be more than willing to email you my worksheet so you can take a closer look.

Regards,
John

[rick57]

Martin,

> No schedules, no promises, no special interests projects, no more outside work for others, no more self induced pressure, and significantly more fun. 

Fair enough!

re project 7 you said:
“One goal was to position  . .  at least one of the woofers near the floor boundary.”

Do you think there is either reinforcement (or interference) from woofers mounted close to the floor?

Regards

Rick

[MJK]

John,

I would be very interested in seeing your simulation.

Rick,

There are definitely advantages putting any driver near the floor.  First it pushes the floor bounce null much higher in frequency.  If the woofer is crossed over below this frequency you can avoid the associated null.  For a mid or full range the driver starts to become directional so the depth of the null is reduced.  Second, the driver will "see" a mirror image below the floor which at low frequency will effectively increase the infinite baffle quoted SPL from the manufacturer.  The real woofer and the phantom woofer will combine to increase the SPL, in reality woofer is radiating all of its SPL into a smaller acoustic space.  The downside is that it will probably excite room modes more efficiently.

Martin

[JohninCR]

Martin,

Regarding dipoles and room modes, I have a question.  I have always accepted
the common belief that dipoles stimulate the on axis room mode, since their
radiation is concentrated on axis.  Only recently have I had dipole output that
can reach down to my on axis mode of 27hz, but I still haven't been able to
stimulate that mode.  I started thinking about why, and at first I dismissed it
as being the room, since the front and rear walls aren't parallel (my 21'x12' room is
actually 21' deep along 1 side wall and almost 22' along the other), but a monopole
sub has no problem stimulating all 3 primary room modes in my mostly concrete
room despite some bass trapping strategies during construction.

My lightbulb lit up after reading JohnK's info about dipoles being limited below the
room fundamental.  While the front wave and it's reflections reinforce and build
where the on axis room dimension = 1/2 wavelength, the rear wave should act in
the same manner, at the same frequency, along the same axis, to create a null,
not a peak.  Am I missing something in my simplistic visualization?  Primary room
modes are pretty simplistic in a rectangular room.

[Bob in St. Louis]

John, that's the kind of post that really "throws me for a loop". 
Any good places to learn about the fundamentals of what you just said?
(Don't explain here, I don't want to hijack, just a link maybe?)

Happy Thanksgiving

Bob

[Rudolf]

... While the front wave and it's reflections reinforce and build
where the on axis room dimension = 1/2 wavelength, the rear wave should act in
the same manner, at the same frequency, along the same axis, to create a null,
not a peak.  Am I missing something in my simplistic visualization? ...

John,
waves readily interpenetrate without too much disturbance. Think about it: If every wave peak would immediately be equalized by the nearest wave trough, there wouldn´t be any waves at all.

In real life there even isn´t a sound "wave". Speaking correctly, sound is about a series of alternating compression and rarefaction areas travelling through the air. It is perfectly ok that a compression area (wave peak) coming from one side of the room and a rarefaction area (wave trough) coming from the opposite side meet (not "clash") in mid-air and continue their ways quite unfazed.

What really bothers those compression and rarefaction areas are "hard" surfaces. They are "bounced off" from the room walls. And a loudspeaker cone could sustain those areas (if moving in phase with them) or delete them (if moving 180° out of phase relative to them).

A dipole cone moving at the frequency of the lowest (1/2 wavelength) room mode would be perfectly in phase with the travel of the reflected compression and rarefaction areas.

Rudolf

[JohninCR]

Rudolph,

Thanks for the reminder that the waves have to be travelling in the same direction
to affect each other, so the rear wave has to either travel around the baffle or
reflect off the wall first making the phase relationship different than directly out of
phase as they travel together in the same direction.  This gives us the same net
wave as the original wave fronts and something that builds at the modal frequency.

Now I just have to rethink whether I understand JohnK's theory that dipole output
falls sharply below the fundamental room resonance, something I plan to test as well.