Reverse engineering XBL2 (1 Viewer)

[Bill Fagal]

"He that tooteth not his own horn, the same shall not be tooted." --John L. Lewis

In case anyone still has any doubts...I was right!

I just came across United States patent application #20030133587 by Hyre, David E., and Wiggins, Daniel C. It's an interesting read--check it out!

From their disclosure, here is a pic of the heart of the matter--twin gaps with a VC hung between them!

**As I'd suspected, there's also mention of a 3-gap permutation. (Can you say CES 2004 mystery monster driver motor? You heard it here first!! )**

If you got enough magnet/pole cross-section, four gaps aren't out of the question.

The only thing they show that I didn't figure into my model is the channel they cut in the pole piece...which actually doesn't make a lot of sense to me. Without faraday shielding, that extra channel seems like it would cause Le to vary quite a bit with excursion, and it also looks like it robs the VC of some heat sinking.

Dan, what was the thinking on this? (If you're ready to share yet, that is!) Or do production XBL2 motors look more like my model and skip the pole channel?

Bill

 

[ChristopherE]

Hey, Bill...

I actually had the pleasure of seeing the FEA for this monster that Dan has been slowly leaking about, and discussing potential XBL^2 permutations...
Not a CAD drawing or anything of the like, but the FEA-derived BL plot that he's posted the center section of... an impressive plot, without competition, for sure.

I just wanted to make a comment about your prediction...
Why say you only 3 gaps?

Granted, one of the specs that I've always been curious about (and never find much discussion on) is with regard to flux dissipation as you start to build such massively tall top plates, obviously as you add more and more gaps, you have the potential to get thicker and thicker. Even the 2-gap Brahma top plate is huge.

If flux dissipation were a concern, it would seem to create a scenario where the gap closest to the magnet would have potentially more flux than the gap closest to the basket... which could be solved by widening up the gaps progressively as you moved toward the magnet.. but that seems like a waste of some good flux, doesn't it?
Another strategy that could be used would be to actually use magnet slugs to space the gaps apart from each other...
Creating a multi-gap XBL2 by layering numerous traditional top plates separated by magnet slugs to create the spacing desired.

That would be more my guess for this monster...
Multi-gaps (Dan's mindset was definitely not one that led me to believe his interests stopped at 2 or 3 gaps ), separated by magnet slugs, to keep the "B" of the BL high.
After all... did you see the actual BL number, from the plot that he's leaked so far?

 

[Bill Fagal]

Hi Chris,

Good to see you on DIYHTF!

Yeah, I think I saw that center section of BL plot you're talking about. It's a whopper. But as cool as the motor will be, the real story here, I believe, is going to be the suspension Adire has come up with to put all that horsepower to ground, so to speak. A linear suspension with that kind of travel would be revolutionary. My suspension designs are my most closely held secrets, so I'm eager to see what they've done (and I hope it's not what I've been working on!).

But getting back to motors (since they're such fun), I agree with you that the CES motor may be a 4-gapper--hard to say from the shape of the plot, which I believe (conspiracy theory on) they smoothed to keep us from counting the humps/gaps.

Regarding flux dissipation, I used to wonder about that, but about a year ago I ran some simulations that showed it's not much of a factor. Providing the flux-carrying circuit isn't saturated away from the gap, the flux distribution is determined almost entirely by the gap spacing. This is because an unsaturated circuit actually presents negligible impediment to flux flow, no matter how far it takes it from the magnet.

One of the first simulations I did to resolve this question in my own mind was of an XBL-like motor with multiple top plates spaced with additional magnets, essentially identical to what you're describing above, and (at least the way I laid it out) it seemed to be a non-starter. The spacer magnets caused the flux to largely bypass the lower gaps, making for a very top-heavy distribution. That was my brief education in the question of flux dissipation. I'll see if I can dig up that old sim and post it.

Back to Dan's motor, of one thing I am quite certain: it will have a very fat pole piece. However many magnetic gaps there are, they all have to feed their flux through the pole, and judging by those BL numbers, that's a lot of flux. I wouldn't be surprised to see at least a 6" VC on this baby. In fact, I'd be surprised to see much less.

Which begs the question: How much will it weigh?

 

[michael-e]

Neat stuff,and heres me thinking during my DC circuits class,,

BH curves,saturation, co ercive force... so boring

I see it CAN come in handy!

FEA rocks-when people can use it :-D

 

[Hank Frankenberg]

Well, they beauty of neodimium is the tremendous flux density per volume. Just the ticket for powerful "slugs".
Dan Wiggins (imitating the new California governor's Austrian accent): "I've got the POWUH!"

 

[Bill Fagal]

FEA rocks-when people can use it :-D

Amen! And it's especially nice when it's free!

Bill

 

[ChristopherE]

Now that CES is over, I thought I'd raise this thread back up... since the cat is out of the bag with respect to the Parthenon, and all the "how are they going to do it?"

Staying on topic on this thread...
I am curious as to why Dan chose to use the dual-gap, dual-coil approach to reach 80mm, as opposed to using say, a single coil, with a triple or even quadruple gap?

I'm going to play with my FEA tools here to see if I can figure out what's traded off.

Maybe Dan will even pop in and comment, obviously he's been through this exercise before.
And if I can't figure it out, I'll Email him anyway.

My guess, initially, is pure expense... more gaps, more magnet, more money, for essentially the same performance.

 

[DanWiggins]

Chris,

Essentially cost, etc. Besides, multigap/single VC has been done for quite a while (Brahmas et al). Multiple gaps/multiple voice coils is pretty new, so I thought it time to show it.

Note that the gap geometry of the Parthenon is the same as the Tumult, but with the addition of the second voice coil, we're able to get ~2X increase in linear stroke. Not to mention that we get more cooling (more surface area of the VC). Le does increase, but for a true subwoofer, that's not too bad (and it's still reasonable).

Now, if you want to get really stupid, dual VC quad gap can get rediculous - like 170mm one way. How does 13" peak to peak sound?

Dan Wiggins
Adire Audio

 

[Seth_L]

I don't know Dan...

I myself like really stupid.

 

[Rory Buszka]

I think we're set with enough motor technology to last us the next ten years. Now we need to work on suspensions to harness all this power.

 

[DanWiggins]

Already taken care of... Showed a little suspension at CES that was good for 8" peak to peak linear range. That should work for a while...

Dan Wiggins
Adire Audio

 

[Craig Treusdell]

I must have missed those pictures....

 

[Seth_L]

I doubt you missed it Craig. You might have just not realized what you were looking at. Dan's new driver doesn't need a conventional surround and "suspension".

 

[DanWiggins]

It has a suspension, just not what most people would recognize. Check out the pictures of the Parthenon from CES. See those coppery colored spirals? Those are the spider and tinsel leads, all wrapped into one (pun intended). Extremely linear, easy to make, monster stroke.

Dan Wiggins
Adire Audio