Why 10-25 times net power?

[KitemanSA]

Oh god, not quats.

"We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard..."

"... and we can spend lots of money on them. Oops, did they hear me? No? Whew!"

Now you know the rest of the quote!

[alexjrgreen]

and we can spend lots of money on them. Oops, did they hear me? No? Whew!"

Now you know the rest of the quote!

Not this office - did you want the one down the hall marked "ITER"?

Quaternions are only more expensive in mental effort. NASA reached the moon without them, but now uses them routinely because they're more stable in calculations.

Some of the instabilities in plasma may turn out to be more tractable using quaternions.

[chrismb]

The guys here working to simulate a polywell would really have appreciated your help...

Thanks for your positive comment, and those made by others over the last week since my post.

Does that mean Ph.D.? If not, it implies a nonstandard education

non-standard education.

I have an M.Sc. in mechanical engineering from the University of Alberta. The topic was microscale droplet deformation in an electric field, with focus on dynamic effects. It was supposed to be a prelude to modelling of electrowetting actuators in lab-on-a-chip devices

I've also done some work on roughness-dependent dynamic contact angles, quite good work I thought but I hit too many 'unconventional' buttons along the route for it to stick and in no small part due to holes from a non-standard education, this I freely admit. But I felt there was substantial merit in my dynamic, entropic considerations of Wenzel along with designs arising from that for improvements to the Wilhemly technique to accommodate true 3D 3-phase contact line interactions, rather than just the 2D contact line that is usually considered normal to the propagation vector, plus kit that can accommodate interactions where the advancing and receding angles vary by <1 degree and so cannot be assessed visually. I also conducted an experiment on 'navile' contact angles with 3 liquids and noted hysteresis in those, which suggests it is more than just roughness, but that may or may not have been experimental inaccuracy but the conventional interpretation insisted that it was just experimental errors so I wasn't given the opportunity to explore that further. I believe there are more fundamental entropic considerations in these systems.

And many years ago I did some contract research on multi-phase fluid flow (I constructed an improved Taitel-Duckler model with a dynamic interfacial fricition factor, which gave a calculated minimum holdup for slug growth of 0.359..).

I do not seek to compete on qualifications, however, and never will. You would win in a pure paper chase, but this is not my motivation. You, as most professional researchers, will make steady solid contributions which, one hopes, will add to the body of knowledge steadily and you may be on watch when something interesting happens. Whereas 'my kind' will make significant quantum differences, or none at all.

there can be holes in a nonstandard education, particularly in terms of basic principles...

We could revisit that, but I think it finished. I am not sure I ever questioned, nor countered, 'standard basic principles' nor was it my intent.

I just wish you were a little more interested in the 'succeed' part re: Polywell...

This is the bit I thought I should just polish up a little. I am interested in Polywell's success and will flat-out celebrate if it gets anywhere, but I just don't think it lies in some of the current speculation. Not sure what else there is to do in the meantime so I do not seek to discourage anyone from doing so, but there has to be some proven mechanistic evidence generated for me to get a grip on. At the moment the debates seem to be as much about the debators getting a grip on each other, rather than the problem, and that's not really my thing. As you can tell, I've had too much of that in my life already.

Experimental evidence is king and Polywell is just flat out of any traceable experimental evidence, but I'll be looking out for any good simulation work as well. If either come along, either on this forum or elsewhere, I'll put my two-penny-worth in.

best regards,

Chris MB.

[MSimon]

Experimental evidence is king and Polywell is just flat out of any traceable experimental evidence, but I'll be looking out for any good simulation work as well. If either come along, either on this forum or elsewhere, I'll put my two-penny-worth in.

Yes it would be nice to have data. However, we can infer a few things from what has happened so far.

1. The review panel has not said: it is fundamentally worthless
2. The Navy thinks more data is required

[IntLibber]

and we can spend lots of money on them. Oops, did they hear me? No? Whew!"

Now you know the rest of the quote!

Not this office - did you want the one down the hall marked "ITER"?

Quaternions are only more expensive in mental effort. NASA reached the moon without them, but now uses them routinely because they're more stable in calculations.

Some of the instabilities in plasma may turn out to be more tractable using quaternions.

Scripting rotations in the LSL (linden scripting language) used in Second Life scripted objects uses quats, it is very hard to visualize how a quat will look, vs euler.

[alexjrgreen]

Scripting rotations in the LSL (linden scripting language) used in Second Life scripted objects uses quats, it is very hard to visualize how a quat will look, vs euler.

It's easy enough to convert quaternions into euler angles when you need to:
http://en.wikipedia.org/wiki/Conversion ... ler_angles

The extra stability in calculations makes quaternions worth the effort.

[93143]

chrismb: Thanks. That clears things up a fair bit.

We could revisit that, but I think it finished. I am not sure I ever questioned, nor countered, 'standard basic principles' nor was it my intent.

At this point I'm not sure it wasn't based on a misuse/misunderstanding of terminology. I am content to leave it where it is.

Anyway, I'm glad we could reach an understanding...

[IntLibber]

Scripting rotations in the LSL (linden scripting language) used in Second Life scripted objects uses quats, it is very hard to visualize how a quat will look, vs euler.

It's easy enough to convert quaternions into euler angles when you need to:
http://en.wikipedia.org/wiki/Conversion ... ler_angles

The extra stability in calculations makes quaternions worth the effort.

According to the wiki page on quaternions, it is claimed that use of euler angles per se does not cause gimbal lock issues, it is poor logic in the way some programming calculates rotations in euler angles.

Which is logical. It shouldn't matter if straight up is 1 and -1 for straight down versus +/-90. Poor logic is going to get you gimbal lock either way.

[alexjrgreen]

According to the wiki page on quaternions, it is claimed that use of euler angles per se does not cause gimbal lock issues, it is poor logic in the way some programming calculates rotations in euler angles.

Which is logical. It shouldn't matter if straight up is 1 and -1 for straight down versus +/-90. Poor logic is going to get you gimbal lock either way.

A better description of gimbal lock is here http://www.gamedev.net/reference/articl ... le1095.asp

[DeltaV]

and we can spend lots of money on them. Oops, did they hear me? No? Whew!"

Now you know the rest of the quote! :wink:

:shock: Not this office - did you want the one down the hall marked "ITER"?

Quaternions are only more expensive in mental effort. NASA reached the moon without them, but now uses them routinely because they're more stable in calculations.

Some of the instabilities in plasma may turn out to be more tractable using quaternions.

Scripting rotations in the LSL (linden scripting language) used in Second Life scripted objects uses quats, it is very hard to visualize how a quat will look, vs euler.

That's because you're not using the "angle-about-axis" representation of a quaternion:
normalized_quaternion = [[scalar, axial _vector]] = [[cos(angle/2), axis_unit_vector sin(angle/2)]]
(see S L Altmann, Mathematics Magazine, Dec 1989)

Note that it takes two complete rotations (720 degrees, 4pi radians) for a quaternion to return to its initial value.

A normalized quaternion that starts as
[[cos (0/2), u_hat sin (0/2)]] = [[1,0]] = +1
for zero rotation angle will become
[[cos (2pi/2), u_hat sin (2pi/2)]] = [[-1,0]] = -1
after one complete turn (or any odd integer number of turns) and return to
[[cos (4pi/2), u_hat sin (4pi/2)]] = [[1,0]] = +1
after two complete turns (or any even integer number of turns), axis u_hat fixed throughout.

This corresponds exactly to what happens in real life:



http://jesseenterprises.net/amsci/1975/ ... 12-fs.html
http://www.evl.uic.edu/hypercomplex/html/dirac.html

The result of combining two arbitrary 3D rotations (arbitrary angle magnitudes and axis directions), represented by normalized quaternions, is
[[c, C]] = [[b, B]] [[a, A]] = [[ba - B.A, bA + aB + BxA]]
where rotation [[a, A]] occurs first, followed by [[b, B]] to yield resultant [[c, C]].

This is much easier than Euler angles.

(Edit - replaced dead link)