2-way omni woofer selection

[puppet]

I've got a couple basic question regarding woofer selection for a "kithara" knock-off.

The tweeter will be the heil amt ... cantilevered over the woofer. The first driver I considered was the Seas h1411 .... it's well behaved response looked pretty easy to work with. A lot of researching omni woofers shows a significant drop in SPL from the woofer(s). With a sensitivity of 89db ... I'm second guessing the Seas driver for this project. I can't see attenuating the amt that much (95db down to 82db probably??)

Enter the Eminence Deltalites ... the 10, 12 and 15". Am I on the right track thinking I need the increased sensitivity these drivers provide? .. in this application?

After looking these drivers spec's over it actually looks like the 15" might be a better choice. It's response graph appears to be the most even while maintaining it's stated SPL over most of it's range. Going with this driver would put the xo point in the 1khz area. I think the amt can cope with that .. though not 100% sure.

The 12" model isn't out of the question either ... the 10" however may not have the bass extension I'm looking for. The 12 or 15" will go into a BR enclosure ... volume hasn't been determined yet. These drivers seemed to be well liked on the forum and it's interesting to see the variety of enclosure volumes recommended for them.

I've seen B4 alignments suggested along with up to a 1/3rd larger (volume) enclosures recommended here. Eminence recommending even small boxes in their spec's as well. ... interesting and confusing all at the same time. I'll be doing some experimenting with that later.

Can a guy get good performance in a 2-way omni with the deltalites ... or am I out in left field? The SPL loss is a big factor in my thinking at the moment. The response of the drivers I think can be worked with ... as long as I'm not being too optimistic with crossing this 15" driver @1k ??

[dlneubec]

Obviously Heil was a brilliant guy, but the concept seems flawed to me. How do you resolve the conflict between the omnipolar pattern of the upfiring woofer, which is in phase in all directions, with the dipole pattern of the Heil tweeter where the rear output is out of phase with the front output.

If we assume the the ouput from the woofer is in phase with the output of the tweeter to the front, then the output from the back of the tweeter has to be out of phase with the woofers omni output. So, as you get closer to the crossover pointsthere will be increasing cancallations between the drivers and essentially a full cancellation at the crossover point toward the rear. I would think that would make for a pretty wonky power response and really unpredictable off axis response.

It would seem that the tweeter would almost change from monopole output at the XO point, due the out of phase cancellations with the woofer, back to dipole as you move away from the XO and the woofer would go from monopole at the XO to omnipole away from the XO. You'd have lots of reflected sound in all direction from the woofer and to the front and rear from the tweeter, to very little rearward output from both drivers at the crossover point. How would you voice such a speaker?

[puppet]

Thanks for responding Dan. I've spent a lot of time reading about your omni projects. They are one of the reasons I started to second guess the seas option for a woofer.

Heils' amt may be flawed as you say. It's not something I've given much thought to while listening to them when paired normally with direct radiating enclosures. I assumed .. perhaps wrongly .. that an omni woofer arrangement wouldn't be all that different from DR with regard to the frequency range of the woofer (specifically, it's radiation patterns). It is my understanding that lower frequencies tend to "wrap around" regardless. A null would/does occur at the rear in either case ... unless I'm not totally grasping the concept.

Question is if the rearward null @xo becomes a deal breaker. My thought process was that a xo point of say 1k for a 15" woofer would be around the point of driver beaming. An offset tweeter arrangement would position the amt toward the rear of the driver below (along the line of what Heil did with his kithara).. essentially veiling 10-15% or so of the woofer drivers output rearward anyway. Coupled with a fairly large side null the amt's exhibit, 270* of woofer output should be OK performance wise. It may be that rearward room treatment would need special attention ... won't know for sure till it's put together I guess.

The rough drawings done in sketch-up provide the tweeter with vertical and lateral movement relative to the woofer. I had hoped that that feature might help to address some performance issues as it was being tested out and perhaps afford sound shaping on a per-room basis.

Thank you for your input Dan ... keep up with your great projects too!! I especially like the tube array omni you built.

[dlneubec]

Yes the longer wavelengths are essentially omni, but that is below the baffle step where the driver is playing in 4pi space. I was assuming an XO in the 1khz or higher range with the Heil, which is a wavelength of a about a foot at the low end. A monopole woofer is unlikely to be within a foot of the rear output of the heil, unless it is firing up like this one. Perhaps if it is at the very top of the enclosure, maybe it could wrap around and over the top, however that seems unlikley. With the upfiring driver immediately below the dipole tweeter, there is little doubt in my mind that nulls would occur.

You also shouild consider how wavelengths shorten as frequency rises and what acoustic slope you are targeting. Let's say you have a LR/12db/octave acoustic low pass on the woofer, at 2khz it is only down 18db and the wavelengths are closer to 6" at that point. Clearly steeper acoustic slopes would be helpful with this approach as it limits the overlap area where confilcts could occur.

I'd say you are doing the right thing by making the tweeter adjustable vertically and laterally.

Another idea is to place a concave cone shape above the woofer (see the Duevel speakers) and set the AMT on top of that. If you are able to size it appropriately to addresses your crossover point and slopes, that might help increase the distance the soudn has to travel to wrap back around to the other driver. It might also help negate the effects of woofer beaming and maybe allow a higher XO point.

I'm a believer in trying differet things, experimenting and pushing the perceived audio limits, so I don't want to discourage you. I just wanted to be sure you were aware of the potential problem ahead of construction. That way you might have some alternative solutions ready, should problems occur. I do know someone who tried an upfiring woofer into a dipole mid. I warned him that it could be problematic and it did turn out to be. He ended up putting a shelf of sorts between the drivers and then I think he used fabric and perhaps some stuffing to attenuate the rear dipole output in order to get something that was more workable.

I also believe there are many means to audio nervana, not just one good solution, as we so often hear on DIY forums.

I look forward to seeing your progress.

[dumptruck]

It's hard to tell from the photos if the Kithara does have the acoustic center of the Heil directly above the woofer or offset a bit. Either way, if you scroll down this page you can see that they went in a different direction after that design, with the woofer angled forward:
Heil Oskar AMT

[puppet]

Hi dumptruck .... it's offset. A guy had many pic's of a pair for sale in "america's hat" ... definitely offset to the rear at least 1-1/2 to 2".

I've seen the other designs from precide ... and read enjoythemusics review(s) of the kithara. If they moved away from that design I think it would be hard to say why exactly ... it was Oskars' baby. They did use an low grade woofer in that kithara (betting it was a one off eminence) It looks really close to some offered by Orange County speakers (their "Hi-Fi" drivers) ... but I'm only guessing here.

I thought a higher efficiency driver might mate better with the amt .. especially in an upfiring position.

[DeonC]

Hi guys

How about pointing the woofer downwards, with the front of the cone radiating into the box, and the rear of the cone into the room. With a drivers like the DeltaLites the AMT can then be mounted on top of the magnet (they have neo magnets, so they should be inherently shielded). Decware had a similar omni project, and there is one posted in another omni thread on the forum using a Pioneer 10-inch woofer (downfiring into a sonotube, iirc). This approach could solve a number of problems, not least of which would be dispersion (no need for a wooden cone above the woofer). Looks and WAF would probably suffer though.

Enjoy,
Deon

[puppet]

I found another fellas amt measurements on the .net .... he too meant to xo from a pair of 8" drivers to the amt @around 1-1.2k. According to the graphs he posted, the amt has a sharp dip from 1k to just before 2k ... 2k on seems fine. He tried it anyway and did eventually add in a mid. He also touched on what he perceived as a di-pole cancellation issue with the amt itself in the 1-2k range ... something I've never even thought about. That in itself may be the reason for the response drop in that range. He tried compensating for this but didn't have any luck ... bit the bullet, dropped in a mid woofer and was happy.

That pretty much throws a wrench into the 1k xo scheme I had hoped for. Funny how those seas (h1411) woofers keep coming back into the picture.

[dumptruck]

Originally Posted by DeonC

Decware had a similar omni project

Speaking of that, you never know, they might be willing to sell you a pair of their inverted-cone ERR mids.

[puppet]

I sort of doubt that dumptruck ... I wouldn't do it. Funny thing about that driver though ... on ebay recently there was a pair of french made drivers (vintage '60's I think) that were inverted. Magnet/VC inside the cone with a (4) post support to the rim and a typical frame above (outside like normal). So, cut away the frame and there it is. Might be some truth to the old saying that pretty much everything has been tried "before".

Honestly, building an omni that can absolutely rock hard is the goal. I've mocked up the design physically and love the sound sensation. Lacking instruments I'm going to have to error on the side of matching drive(s) efficiencies (or trying to) on paper. What sounds good with Santana and Maynard F. will get the nod.

If Dan is still here .... did you experience any spl benefit sandwiching those drivers in your first omni project. I wondered about that and stumbled on the forum thread you started .. but you didn't mention that aspect specifically. Were there cancellation issues with the woofers pointed at each other? For my use I was thinking about a foot apart ... if i ended up with the seas woofer(s). What was your overall impression on that driver arrangement?