AudioCircle
Industry Circles => GR Research => Topic started by: bdp24 on 15 Aug 2012, 07:48 am
-
I was going to do this in a PM to you Danny, but I figured someone else may find the info usefull, so here goes.....
I am preparing to put together a pair of OB H-Frame subs using the GR 12" drivers and Rythmik plate amps, and your answers to these questions will help me do so in a more informed manner. Here's what I would like your thoughts on:
- I saw how DonS put together a pair of H-Frames under his Geddes speakers, using the subs as stands for the speakers. The H-frames were sitting upright, the two drivers in each in vertical alignment. This made me wonder.....since H-Frames are dipoles, with nulls to either side, they can be placed close to side walls without penalty. Can H-Frames be placed laying down, one side on the floor, the two drivers aligned horizontally, also without penalty? I ask because it occurred to me that a pair of H-Frames could be used as stands for a pair of old Quads. Ignoring the matter of sub placement in regard to room modes, what are your thoughts on this idea?
- The GR 12" OB drivers have an impedence of about 12 ohms, and therefore provide an optimum load to the Rythmik amp when used as a trio, as the amp is designed to have maximum power output when looking into a four ohm load, the combined impedence of three GR drivers. This alone increases the output of the amp. Using three drivers also increases the output of the sub simply because of the fact that three drivers move more air than do two. Is this all true, and therefore three drivers per amp is preferable to two? Is there a penalty to be paid for using three drivers per amp? Is 370 watts for the three drivers enough, or would you recommend the 600 watt Rythmik amp?
- In building the H-frames, does the distance of the box walls from the driver make a great deal of difference? For instance, two inches vs. four? How about the depth of the boxes?
- For flush mounting of the drivers, what should the diameter of the hole for the basket be? The depth? And the through hole?
Okay, that does it for now. Thanks---Eric.
-
- I saw how DonS put together a pair of H-Frames under his Geddes speakers, using the subs as stands for the speakers. The H-frames were sitting upright, the two drivers in each in vertical alignment. This made me wonder.....since H-Frames are dipoles, with nulls to either side, they can be placed close to side walls without penalty. Can H-Frames be placed laying down, one side on the floor, the two drivers aligned horizontally, also without penalty? I ask because it occurred to me that a pair of H-Frames could be used as stands for a pair of old Quads. Ignoring the matter of sub placement in regard to room modes, what are your thoughts on this idea?
Yeah, no problem on any of that. You could make a whole line of them across the floor.
- The GR 12" OB drivers have an impedence of about 12 ohms, and therefore provide an optimum load to the Rythmik amp when used as a trio, as the amp is designed to have maximum power output when looking into a four ohm load, the combined impedence of three GR drivers. This alone increases the output of the amp. Using three drivers also increases the output of the sub simply because of the fact that three drivers move more air than do two. Is this all true, and therefore three drivers per amp is preferable to two?
This is what they were designed for.
Is there a penalty to be paid for using three drivers per amp?
Using three in a vertical stand arrangement gets a little tall for a monitor above them. No problems using three vertically or horizontally though.
Is 370 watts for the three drivers enough, or would you recommend the 600 watt Rythmik amp?
370 is enough.
- In building the H-frames, does the distance of the box walls from the driver make a great deal of difference? For instance, two inches vs. four? How about the depth of the boxes?
Yes. It is a balance of several concerns. And it depends also on how high they play.
Added depth adds greater front to back separation, loads the drivers and increases output. However, long panel lengths are easily excited and can easily resonant, and that can be a really big deal.
If the driver plays up very high then the panel depth needs to be kept short to prevent a cavity resonance of the shorter wavelengths that would begin to propagate inside each area.
For our Super-V design about 6" to 6.5" on each side of the woofer was optimal and worked great.
- For flush mounting of the drivers, what should the diameter of the hole for the basket be? The depth? And the through hole?
Make it easy on yourself and just surface mount the driver. Then it just needs an 11" through hole. You can't see them that well deep into an H frame anyway.
-
Great, thanks Danny. So for best sound (with maybe giving up a little possible output), the rule for the size of the boxes is "the smaller the better", the baffle being just big enough for the drivers and the depth no greater than necessary?---EJ.
-
Great, thanks Danny. So for best sound (with maybe giving up a little possible output), the rule for the size of the boxes is "the smaller the better", the baffle being just big enough for the drivers and the depth no greater than necessary?---EJ.
If it does get smaller (or not as deep) then there is less front back separation, less loading on the drivers, and a little less SPL.
The example of what I am using in the Super-V is just right for that application.
It could get deeper if it wasn't playing up as high, but then it better be as rock solid as a piece of granite because those woofers put a lot of pressure on those side panels.
-
If it does get smaller (or not as deep) then there is less front back separation, less loading on the drivers, and a little less SPL.
It could get deeper if it wasn't playing up as high, but then it better be as rock solid as a piece of granite because those woofers put a lot of pressure on those side panels.
Danny,
How do the drivers react on a simple straight baffle (no side pieces), say, 14"-16" wide. I was planning to do a simple mock up on a flat panel when I get the the kit, similar to image below. I will have to wait a month or two before I am able to build the spec enclosure. I assumed the SPL would be reduced and the woofer system would fall off an octave or two higher. But, I also thought the servo amps could help compensate. Thoughts?
(http://www.audiocircle.com/image.php?id=42772)
-
If it does get smaller (or not as deep) then there is less front back separation, less loading on the drivers, and a little less SPL.
The example of what I am using in the Super-V is just right for that application.
It could get deeper if it wasn't playing up as high, but then it better be as rock solid as a piece of granite because those woofers put a lot of pressure on those side panels.
I'll be crossing over at no higher than 100hz, so a baffle of 14" square for each driver and a cube interior depth of about the same (7" in front of and behind each driver) should be a good compromise, eh?
-
I'll be crossing over at no higher than 100hz, so a baffle of 14" square for each driver and a cube interior depth of about the same (7" in front of and behind each driver) should be a good compromise, eh?
Yep, very good.
-
Danny,
How do the drivers react on a simple straight baffle (no side pieces), say, 14"-16" wide. I was planning to do a simple mock up on a flat panel when I get the the kit, similar to image below. I will have to wait a month or two before I am able to build the spec enclosure. I assumed the SPL would be reduced and the woofer system would fall off an octave or two higher. But, I also thought the servo amps could help compensate. Thoughts?
The mass being moved by a pair of servo subs would just about jerk something like that over, or at least move it across the floor.
And the output would be significantly reduced. The front wave and back wave will cancel each other out too quickly, and be canceled out (to some degree) by the movement of the baffle.
-
Got it. I'll book the shop time early.
-
I have two 12 inch servos on an 18 inch wide open baffle but have not fired them up yet to see how much bass they deliver. I bet it will go into the low 30s. If you did three 12s with two side by side on the bottom and then one above in the middle you would no doubt have killer low bass. This would have to be at least 27 inches wide at the very bottom. My speakers use three layers of highly refined 3/4" MDF with green glue in between for the baffle and base, and the brace in the back is three layers of 13 ply birch ply also using green glue. These 85 lb. speakers will be very stable. Check out the preliminary proto pics on my website. I will post their low frequency response in a week or two when I measure.
-
You won't get any low bass at all. There is a guy on diyaudio.com who has a setup with like 16 18 inch woofers in OB, a WALL of subs, just a monstrous setup, and even with all that he only gets to 20hz.
http://www.diyaudio.com/forums/multi-way/142015-my-open-baffle-dipole-beyma-tpl-150-a-118.html#post2922246
-
Actually the servo system will force the drivers to play flat to 20Hz or less (depending on where you set the extension filter), and it will do that in on any baffle or even in free air. However, the output level will be significantly less on a flat baffle.
-
Actually the servo system will force the drivers to play flat to 20Hz or less (depending on where you set the extension filter), and it will do that in on any baffle or even in free air. However, the output level will be significantly less on a flat baffle.
What would a "U" alignment do for the servos?
Thanks,
Ed
-
What would a "U" alignment do for the servos?
Thanks,
Ed
A U frame does much of the same thing that the H frame accomplishes, but most that I have seen use a large un-braced rear section that is very resonant. They really need to be very well braced and very heavy to overcome the pressures in the U.
-
Hmmm... this opens up some interesting design ideas. What do you figure the depth should be for a u-shaped enclosure? Would 12-13" be ok or too much? Or how about a slightly flared U almost like a horn.
-
Hmmm... this opens up some interesting design ideas. What do you figure the depth should be for a u-shaped enclosure? Would 12-13" be ok or too much? Or how about a slightly flared U almost like a horn.
The H frame divides the chamber in half and relieves a lot of pressure that a U frame will have on the side panels. It is a much better solution.
-
The main predictor of how low you can go for a given volume is the front to back separation between the drivers. A flat baffle of say 12 inches gives you a separation (or D) of 12 inches. Six inches to the side at the back, six inches to the center at the front.
If you add a twelve inch side piece (an H baffle) you add another six inches for the waves to travel to the back of frame, 12 inches to get to the front, and another six inches to the inside again.
In effect you are tripling the baffle width. You would need a 36 inch baffle to have the same capability as a 12 inch wide and deep H baffle.
The downside is that the deeper you make the H frame, the lower the dipole peak and other resonances.
A U frame has the same effect, but the resonances will be lower again because it acts acoustically twice as deep as the H Frame. Some recommend trying to damp this.
Many have indeed used H-Frames on their side with Quads and other dipoles.
As for using three drivers, you'll lose the force cancellation opposed drivers give and sub will shake.
I'd go to four a side, or bigger drivers if you want to stay OB.
But if you really want deep loud bass you're better off using a sealed sub like the Rhythmik to come in below 40Hz. The OB subs may play to 20Hz, but they won't play loud.
-
Fellas,
Sorry to interrupt here but.....an H-frame dipole woofer is not a force-cancelling configuration. It matters not if you have one driver or sixteen, all cones are moving in the same direction during operation. A "W"-frame configuration is where 100% force cancelling is realized since both cones are always moving in the opposite direction from each other. I've seen a few "hybrid" configuations where the drivers are orientated 90 degrees with each other to yield a smaller width, which yields a "hybrid" force cancellation.
You do realize some even-order distortion reduction by "inverting" alternate woofers in a "H"-frame configuration, (because one cone is moving towards the magnet and the other away) but not vibration reduction from force cancellation.
Yes, there are (minor) pressure differences created in a dipole woofer configuration, but relative to closed boxes or vented boxes these are miniscule. I don't see any problem in damping wall cavities, but remember that (most of the time) these systems are low-passed well below frequencies that would excite these panel resonances sufficiently. Some dipole configurations are extreme...Ripole...and then you might start to see significant pressures created.
It's not correct to say a servo system will force a dipole woofer configuration to play flat to 20Hz in any condition. We're talking about two different things here. The filtering schemes in the sub-woofer amplifiers are 'equalization', but the servo configuration is for 'distortion reduction' and doesn't counter dipole front/back cancellation....directly.
Regarding the original query.....A third driver will yield 3.5db more output capability than two drivers. IMO, that would be the most important reason to do it.....vice distortion, resonances, or other issues.
Cheers,
Dave.
-
Davey is correct. There is no force cancellation in a H frame. Both drivers are (and half to be) in phase with each other even if one is inverted.
The W frame does not give you 100% force cancellation either. Since the woofers in a W frame are leaned over to about 45 degrees of vertical. They need to be facing straight up and down and opposing to get you complete force cancellation. The latest Orion uses this configuration. Unfortunately this keeps you from being about to brace the side walls as the whole area has to be open to allow a woofer to be installed through the hole.
And damping the wall cavities is a BIG deal. You don't even have to hit the resonance frequency of the panel to have a problem. The woofers themselves put a LOT of pressure on the side panels. And the flexing of the side panels can be a very large distortion. I have heard so much panel flexing in an Orion that it completed colored the bottom end.
It's not correct to say a servo system will force a dipole woofer configuration to play flat to 20Hz in any condition.
Yep, it will still play to 20Hz easily. I will whip one out and take some free air measurements as soon as I can.
But if you really want deep loud bass you're better off using a sealed sub like the Rhythmik to come in below 40Hz. The OB subs may play to 20Hz, but they won't play loud.
Actually our OB subs will play very loud and a pair of them are easily as loud as a woofer in a sealed box. And that is to 20Hz.
-
Well, when I referred to a "W"-frame configuration I meant completely parallel baffle plates. So, a "W" that looks like this "|_|-|_|". Sorry about that. That configuration does have complete force cancellation.
You can skip the free-air measurements, I believe you. Your driver may indeed play easily to 20Hz, but it's because of equalization, not the servo configuration.
A sealed box woofer will indeed play louder than an OB configuration......all other things being equal. It will take you four OB subs to equal the SPL capability of one sealed-box woofer.....all other things being equal.
Cheers,
Dave.
-
Well, when I referred to a "W"-frame configuration I meant completely parallel baffle plates. So, a "W" that looks like this "|_|-|_|". Sorry about that. That configuration does have complete force cancellation.
I think this is what most people refer to as a W frame:
(http://gr-research.com/images/obtest.jpg)
But what you are talking about (like in the latest Orion) will have complete force cancellation.
You can skip the free-air measurements, I believe you. Your driver may indeed play easily to 20Hz, but it's because of equalization, not the servo configuration.
The servo system uses an extension filter much like a transform circuit. It monitors the driver and makes corrections that will maintain a linear response.
A sealed box woofer will indeed play louder than an OB configuration......all other things being equal. It will take you four OB subs to equal the SPL capability of one sealed-box woofer.....all other things being equal.
Check again what I said. And one pair of our OB servo subs in a H frame has about the same output as our sealed box servo sub (one woofer) It is pretty close though.
-
A pair of "H"-frame subwoofers (four drivers total) will have the same potential SPL capability as a single (one driver only) in a sealed box. (The same drivers being used.) That's what you meant? That, I agree with.
The latest Orion woofer scheme is not the first to utilize the force cancelling configuration. That dates to about twenty years ago (for Linkwitz designs) and I've seen drivers in that configuration from other folks way back when I was a young pup. About 35 years ago. :)
Your angled baffle OB system looks great. That configuration trades off force cancellation for reduced height and larger cavities. Depending upon the users floor construction (how vibration insensitive it is) that tradeoff might (or might not) work well.
I think you're missing my point regards equalization and servo configuration. No problem.
Cheers,
Dave.
-
A pair of "H"-frame subwoofers (four drivers total) will have the same potential SPL capability as a single (one driver only) in a sealed box. (The same drivers being used.) That's what you meant? That, I agree with.
No, my in room response measurements show that two of our SW-12-16FR's (not 4) hit about the same SPL as a single SW-12-04 in a sealed box.
The latest Orion woofer scheme is not the first to utilize the force cancelling configuration. That dates to about twenty years ago (for Linkwitz designs) and I've seen drivers in that configuration from other folks way back when I was a young pup. About 35 years ago.
Yes, I was just using the Orion as an example of one that might have been seen.
Your angled baffle OB system looks great. That configuration trades off force cancellation for reduced height and larger cavities. Depending upon the users floor construction (how vibration insensitive it is) that tradeoff might (or might not) work well.
It was originally built for height reasons, and it was just a test box used with various upper drivers.
The problem with it was that the woofers placed a LOT of pressure on the side panels. I eventually added a small brace in the front and lined it with No Rez. That was just so that it could be used in testing. What a design like this really needs is side panels that are at least 1.5" thick and lined with No Rez.
The H frame fared much better as the panels front and back were smaller and had a divider between the two woofers. So it was much more solid and sounded cleaner. I still recommend 1.5" thick walls and the No Rez. It definitely improves the sound.
In the Super-V's there is a rear box for each woofer that is partially overlapped by the side panels that are 1.25" thick. And it too really needs the No Rez.
(http://www.gr-research.com/mis/SV5.jpg)