4x isobaric

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi everybody,

I have found info on the use of two drivers in isobaric coupling, but what happens (in terms of sensitivity and box design math) if I use 4 drivers? Two should be facing out into the listening room, and the other two should be hidden inside the box, with isobaric coupling to the outter two.
The drivers are Peerless SLS-8 drivers, nominal 8ohm with about 87dB sensitivity each.
My plan is to put connect two drivers in parallel (giving 4 ohm), and then connecting two of these sets in series, giving 8 ohm again.

If I have understood it correctly, two drivers in parallel (mounted isobaric) end up with the same sensitivity as a single driver. (if this correct?)

If so, I am splitting the power (with two "4-ohm pairs" in series), but getting double the cone surface. Where does this lead me?
Can anybody tell me the overall sensitivity of this system?

Thanks,
Jennice
 
It will be the same as a single driver.

Isobaric is a waste of money and cone area. You'd be better off making a single sub with two woofers (or even all four) of the desired box size - then make a linkwitz transform to get the Q you want.

You are better off getting an amp that can handle 4 ohms than run with a limitation to 8 ohms loads anyway.
 
Ron E said:
It will be the same as a single driver.

Isobaric is a waste of money and cone area. You'd be better off making a single sub with two woofers (or even all four) of the desired box size - then make a linkwitz transform to get the Q you want.

You are better off getting an amp that can handle 4 ohms than run with a limitation to 8 ohms loads anyway.
Ron E
Could you tell us why isobaric is a waste of time?
On another thread I posted some questions regarding help with my JBL 15" subs and it was suggested to me to run them in an isobaric configuration (also no room for two subs it had to be one enclosure)
I made a makeshift isobaric loaded protoype and so far the results are very promising.
You can look under the thread help me make my decision on new subwoofer or not.
As you will read my initial plans were to change my sub for a Parts EXpress Titanic 15 or something similar.

I'm interested to read your views on this if not for anything more but to learn more about this crazy addiction we all have.
Thanks Dave.
 
he didnt say it was a waste of time, only cone area and money. Cone area is wasted because you used two drivers and get only the cone area of one, and time is wasted because you spend twice as much on drivers.

But, you get double the themal power handling and half the enclosure size, plus canceling of some distortion and a slight increase in linear Xmax.
 
...and the madman returns :)

Hi folks...

Thanks for the replies. I was not planning to make it as a dedicated sub, but as the low-end of a 3-way. (Don't worry, I got someone to help me with the X-over).

Thus, I would use 8 speakers in total, 4 for each side.

If I understand it correctly, using 4 drivers (2 isobaric sets) in each enclosure, it would give me the sensitivity of one driver, but with better power handling?
Am I correct, that this setup would have lower displacement (cone movement) for a given SPL than a single driver (since there are two cone surfaces to move the same air as one would have to do)?

I have penty of these drivers around. I have also considered using the 4 drivers per enclosure, all facing out (2 front, 2 rear). This would give better efficiency - but how much more?
Also, it would need a rather big enclosure - which isn't very compliant with a good WAF ;) .


Jennice
 
...update

Hi again... just a quick update.

I plan to use the bass drivers with a HDS134 or HDS164 for the mid's, and a scanspeak 970k or 930k series for the top.

If I crossed at about 200 - 250, I am concerned about the displacement of the HDS134 (+/- 3.5 mm).

The HDS134 is supposed to sound better, but the HDS134 will certainly be better to handle the power at these frequencies.

Ron E stated that the sensitivity of 2x 2isobaric would be the same as one driver. This would suit the mid-driver well, as they both have about 87dB sens.

The tweeter it at about 89, so it would need a little series resistor, but I wouldn't "waste" a lot of energy this way (only in the tweeter).


What do you guys think.

Jennice

Edit: I just realized that maybe this shouldn't have been posted in the sub-section (although my primary question was about the bass drivers), but it's too late to worry now I guess...
 
As the gentleman said, a waste of money because you pay for two drivers and only get one, and a waste of cone area for the same reason. IT may be a valid way to make some drivers you already have work for some application where you need a smaller box, but IMO if you are starting from scratch your money is better spent looking for a driver that will work in that application in the first place.

You could also call it a waste of power because you need twice the power to get the same SPL. Isobarik is not a free lunch. Depending on how it is done, it doesn't always save as much space as you think bacause if it is not clamshell (which looks ugly) you have a tunnel to build.

Since the OP has a bunch of these lying around and is presumably making a multi-way with passive crossovers, it should work OK, but I think just using 2 active drivers speaking to the outside world in parallel might work better in the end for baffle step compensation purposes..... it is really hard to say as there was not enough detail given.

For a straight powered sub application, it is much better to use a Linkwitz transform and use all the cones for making noise. You can build small box with a Q of 1+ and transform it to the Q and F3 you desire. You will get better max SPL's and less distortion for a given SPL because each cone will only be moving half as far.

You should be able to take the 8"ers up to 500Hz with no problem if you want to. Cuts down on midrange IM distortion.
 
Hi all,

What would happen if I placed all 4 drivers facing out of the enclosure (useful driver surface)? I would mount 2 drivers in parallel, and two of these sets in series, getting back to 8 ohm.

I'm fairly new to this, so please comment on my math:
2 drivers in parallel should give +3dB for "doubled surface", and +3dB for half impedance (double power), resulting in a 6 dB gain.

If I take two of these sets in series, I would loose 3dB to reduced power, but gain 3 dB due to doubled (again) surface. If I am correct, this should give 6dB gain over the single driver, while having a better power handling. Is this correct, or am I wrong (if so, where)?

Jennice
 
As far as I know, you have that correct. You should gain 3db for doubling cone area once (2 drivers), another 3db for doubling again (4 drivers), and none from impedance since that didn't change. Which gives you a 6db net gain. Depending on frequency and driver layout, though, you can have lobing issues with that many drivers. I'm not sure what exactly happens to the power handling. Anyone know if it double or quadruples over a single driver?
 
Issue A:

azrix said:
Depending on frequency and driver layout, though, you can have lobing issues with that many drivers.

Maybe it's a language thing (english isn't native to me), but I don't understand this part. Is there another way of describing the phenomenon you're referring to?

Issue B:
You refer to the impedance not being changed... so what matters really is the "available" membrane area (factor relative to a single driver), and the power drawn from the amplifier (factor relative to power into a single driver)?

Jennice
 
Frequency lobing is when the off-axis response is alot quieter than the in-axis response.

This is not a concern when designing a subwoofer.

Since you keep the impedance at 8 ohms, then yes, you only need to consider the membrane area and the power drawn from the amplifier.
 
Hi Simon5,

If I was to use 4 drivers all facing out (useable surface area), I would try to design the X-over at the baffle step frequency, so that the rear facing drivers would assist in making a (close to ideal) point source, radiating all 360 degrees.

Maybe I didn't explain that part originally... my plan was to either make a smallish enclosure with 2 x 2 drivers isobaric, or make use of all 4 drivers, 2 facing front, and 2 facing rear.

I think it was Dave (Planet10) who makes the baffle step compensation for a 2-way with an addidional woofer that is mounted at the rear. The rear driver would have a simple LP filter, letting the driver play as the baffle step reduces the apparent output from the front driver.

Jennice
 
simon5 said:
Frequency lobing is when the off-axis response is alot quieter than the in-axis response.

Not exactly, at least that's not what I meant by lobing. What I was refering to is when you have phase cancelation at certain frequencies and certain angles off-axis. For example, say you have a speaker that's -3db at 30deg off axis at a certain frequency, -20db at 45deg off axis, and back up at -6db at 60deg off axis. If you were to plot this kind of response on a polar chart, you'd have three teardrop shapes, or 'lobes', that connect at your point of origin. This polar response will change and become less of a problem as you go down in frequency. It's mostly an issue with lines arrays of dynamic drivers. But, it can happen with only two drivers if they're trying to reproduce the same frequency and are far enough apart.

simon5 said:
This is not a concern when designing a subwoofer.

No, it wouldn't be, but if you're using a large woofer as a bass driver in a speaker design, which I believe is the case here, you can have problems if you crossover high enough and use more than one or two drivers. It probably isn't going to be an issue in this case, so forget I said anything :) . Just trying to be thorough.
 
Jennice said:
Maybe I didn't explain that part originally... my plan was to either make a smallish enclosure with 2 x 2 drivers isobaric, or make use of all 4 drivers, 2 facing front, and 2 facing rear.

Oh, I didn't realize this. This is generally refered to as a bi-pole arrangement, I believe. You can run the front and rear speakers in phase, so that all the drivers are going out as the same time, or out of phase, so that when the front speakers go out, the rear ones go in. That's about all I know of that. If you are going to do that, you might look at mounting the woofers on the sides of your cabinet so that you can have a more narrow front baffle. By doing that you might also be able to get away with larger drivers, if you want.
 
I figured I'd let the drivers play in phase if I were to go for the "4 drivers facing out" solution. This could compensate for baffle step I suppose, giving the 6dB gain which would otherwise be lost.
(Isn't baffle step -6dB).

I have thought about placing the drivers on the side, but wouldn't that mess up the sound stage? If placed side-ways, I thought the drivers should only play below 100Hz or so (where you start hearing directivity)...?

Jennice
 
azrix said:


No, it wouldn't be, but if you're using a large woofer as a bass driver in a speaker design, which I believe is the case here, you can have problems if you crossover high enough and use more than one or two drivers. It probably isn't going to be an issue in this case, so forget I said anything :) . Just trying to be thorough.


I appreciate all the input I can get :)

My plan was to cross somewhere around 2-300 Hz if I went for the 4-driver-out solution, depending on the baffle step, and hence the width of the cabinet.

However, due to the use of the room (and spouse!) the speakers wouldn't be too far from the wall, so maybe baffle-step isn't a real issue here. (in real life considerations)???

Jennice
 
frugal-phile™
Joined 2001
Paid Member
Ron seems to have a decided dislike for isobariks...

I went thru an isobarik phase... the design has a place and i really enjoyed those speakers.

I also think the design brings more to the party than just 1/2ing the box and paying the price. The extra layer of cone isolating what's on the inside from the outside tends to make the boxes sound less boxy for one.

and it allows me to put 4 12" (a lovely semi-vintage Foster) into a 42 litre box -- a pair of 6 ft^3 boxes is something i don't want to contemplate.

push-pull push-push

dave
 

Attachments

  • push-pull-push-push-fosters.gif
    push-pull-push-push-fosters.gif
    8.5 KB · Views: 308
Yes, 3dB max SPL gain and 2x power handling over an isobaric pair. As already mentioned, with small boxes the space taken up by the 'hidden' driver and/or it's connecting tunnel becomes significant and you often hardly end up with a smaller box and you have wasted all that money on an extra driver. IMO isobaric only comes into real use if you refuse to use any active EQ and your box is bigger than 75 litres. I go with the LT every time.
 
I have played with WinISD, and come up with compromised at about 20 liter for isobaric (plus some extra for the spacing between the drivers), and 80 liter for 4 drivers. these two should give the same response/roll-off, but at 4 times the enslosure size!!!
The ideal flat roll-off happens at 41 liters and 164 liters, respectively. There's no chance that I'm going to build speakers this size.

Considering that I want to make a sub-inclosure inside the main box, containing the mid- and tweeter driver, I would end up in excess of 100 liter with the 4 "visible driver" option.

Actively compensating for an error that could be avoided with a different speaker design, is not a thought I like.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.