I have a idea of lacing a wheel so that all spokes on one side are leading and all spokes on another side are trailing. It would look cool, but affraid of braking a weak hub shell because it's twisted with quite a lot of force.

Any comments on that? I havn't found anything from internet about that.

Not going to work for 3x. Think about the spokes crossing over each other.

What hub? # of spokes? Type of spokes?

I think you understood me wrong somehow, there will be no spokes crossing each other anywhere.
32 regular spokes on Nexus 3-speed internal geared hub.

I've seen a narrow-spindle front road hub get well on the way to twist straight off in a build like that. Supposedly you can use the same length spokes as for an ordinary 3X build. It gives you a kinda-sorta radial look w/o quite that nasty load case on the flanges.
Feels a bit dodgy for a rear hub, as pedalling torque would have all spokes on one side lose a bit of tension simultaneously. But for a symmetrically dished wheel that migh still be manageable.

Not possible to use only leading or trailing spokes on one side. The spokes would have no tension.

Edit: well maybe you could get tension but I don't think you'd get very far down the road with that build.

If the pattern is what I'm thinking, you could never get it tight or true. there would be no opposing force on either side of the hub to keep it in position.

Getting tension doesn't seem to be a problem....

As for durability, who knows? for narrow spindle/high flange - none at all.
For low flange wide spindle, supposedly good enough.
The guy who built that one ran it as a front, and at slightly lower the usual spoke tension. For whatever usage profile he was using it was apparently good enough.

Sure there is. One side tries to rotate the hub clockwise, the other side counter-clockwise. Plenty of opposing force. Main issue during build is said to be that you kinda have to sneak up on it in small increments, as the hub is pretty much free to flop around until the torque begins to be noticeable.

But why, just for aesthetics?

the one thing i have noticed about making interesting spoke patterns is that no one notices them who isn't another wheel geek.

+1

This is a bad idea : this spoke patern puts a lot of torque on the hub. I could easily see this hub shell snap in the middle.

+1 bikinfool

Well, the OP wanted to use it for a IGH, with that diameter to the hub shell it'd probably be plenty strong. Not saying that the probable survival of the hub shell is enough to make it a good idea though...

Probably works OK on a front wheel if you have a beefy enough hub shell. But on the rear think about what will happen while pedaling: as soon as the hub puts torque on the wheel, you'd be adding tension to one side while releasing tension from the other, and the rim would shift from side to side. Brakes would rub with every pedal stroke.

doing all leading, all trailing would be like making a bridge with an "S" shaped middle section, with the supports only at the tips. It's balanced when there's no load on it, but once the load is put on anywhere outside the centerline, it'll twist out of shape.

how about 3 leading, 3 trailing? (or 2, 2 for a 32h)
or crow's foot...

True. I didn't even realize spokes crossed each other until I started racing and paying attention to which equipment I was buying.

If it was a good idea it would have been done. There have been metal spoked wheels for at least 120 years.

Sounds pretty silly to me... maybe you could invent a pattern where 1/4 of the spokes on each side went the other way, but that still wouldn't be a real good idea.I would have been about five.

But then, I'm forever being puzzled by people's mechanical blindness... how anyone can look at almost any part of something as simple as a bike and not immediately understand how it works is beyond me.

God damn, seriously read the previous replies someone did it, jesus christ.
I would echo the other concerns of the dish constantly changing if it were a rear wheel.

how about 1/3?

https://www.sheldonbrown.com/power_wheel.html

But then the OP's gonna have to invest in some real good brakes and pads, or else he'll never stop

custom wheel.

if the manufacturers jump on it, then that would mean it's somewhat viable and not another repeat of spinnergy rev.X or mavic r-sys wheels.

for an all leading one side, all trailing other side, I would expect the wheel to fold in on itself when the hub snaps in half.

I'm perplexed by it as well. I've always been a curious boy, staring at the lawnmower or vacuum cleaner after mom or dad used it, trying to figure out how it worked. Yet, it never occured to me until age 12 to take a close look at the pattern spokes were in. I had played with cranks, put the chain back on, etc. but not for years did I think to analyze why the spokes didn't just go straight out like a wagon wheel.

Aside from the issues of unneccesary stress on the hub it seems obvious that as soon as you apply the brakes one set of spokes will unload while the other winds up, moving the rim to one side.

That will be one incredibly squirrelly bike.