General Part Discussion

[2GodBDGlory]

Just now, gyenesvi said:

Such a flexible part is an interesting idea, and even the solution to make it using existing pieces, though I also agree that it might end up having an invisible effect. I guess for it to have a visible effect, it should be rigid when connected, so that it actually transfers force from one wheel to the other.

I have been thinking how much stress it would actually mean on parts. Obviously when disconnected, no stress. When connected, and the suspension would be pressed by both wheels at the same time, it would mean no stress either, as there is no difference between the two sides. So it would potentially mean stress when one tries to make it articulate. For a manual model that would mean tilting it left/right and pressing down, which is not common to do for prolonged time. Even in that case, the system would try to lift the wheel opposite to the tilt, so maybe that would not be too much stressful either. As for rigid parts, I imagine the left-right connector to be built from short axles and connectors. Do you think this would have more stress than a driveshaft in an RC model? I believe the weak spot could be the connector part, which would have to be some kind of sliding part (like a linear clutch).

I don't think it causes any problems in practice--even using axles as torsion bars for tracked vehicle suspension is commonly done in MOCs without damage. I think it'd just be a philosophical issue for TLG, though, since it requires twisting parts that theoretically aren't supposed to be twisted.

[allanp]

On 2/4/2024 at 4:18 PM, Divitis said:

I am afraid the problem with anti roll bars and Lego is that they work by bending themselves. So unless a specific piece comes out that is designed for torsion, I don't think a regular part will ever be put under such stress in an official set.

Any time an axle is used to transmit torque, like in every Technic set, it will be under the same stress, and it will twist slightly. The only difference here is by how much and for how long. Under full articulation, with one wheel all the way up and the other all the way down, the anti roll bar will visibly twist but will also bend back as soon as the obstacle is cleared. As long as it's not twisted too far (within the bounds of elastic deformation), and as long as it's not held in that position for too long, it'll spring back like any other Technic axle being used to transmit torque. Just don't display your model in a fully articulated pose! I would be more concerned about the pieces used at the end of each axle. A pair of stacked thin lift arms will crack where as a full stud width cross hole (such as in the 2x4 L beam) will be fine. It'll likely not have any noticeable affect to someone pushing it over obstacles, but it would be nice to have for the sake of realism, if the real vehicle has anti roll bars if course.

[gyenesvi]

2 hours ago, 2GodBDGlory said:

I don't think it causes any problems in practice--even using axles as torsion bars for tracked vehicle suspension is commonly done in MOCs without damage. I think it'd just be a philosophical issue for TLG, though, since it requires twisting parts that theoretically aren't supposed to be twisted.

I was actually thinking that this should be a somewhat different case. For a torsion bar, you would use a longer axle to allow it to twist, as that's the goal. But here the goal would actually be to not allow it to twist much, but rather transmit the movement from one side to the other, in order to force the wheels to move up/down in parallel. That's why I'd say I'd rather use short axles and 2L / 3L connectors. And hence, the stress would be on the part where you have the disconnectable connection of the two sides I guess.

2 hours ago, allanp said:

Just don't display your model in a fully articulated pose!

Exactly :) Or disconnect it for that purpose..

2 hours ago, allanp said:

I would be more concerned about the pieces used at the end of each axle. A pair of stacked thin lift arms will crack where as a full stud width cross hole (such as in the 2x4 L beam) will be fine.

Yes, I agree with this, as those parts/axle holes could also end up having some stress on them.

[2GodBDGlory]

1 minute ago, gyenesvi said:

I was actually thinking that this should be a somewhat different case. For a torsion bar, you would use a longer axle to allow it to twist, as that's the goal. But here the goal would actually be to not allow it to twist much, but rather transmit the movement from one side to the other, in order to force the wheels to move up/down in parallel. That's why I'd say I'd rather use short axles and 2L / 3L connectors. And hence, the stress would be on the part where you have the disconnectable connection of the two sides I guess.

True, I guess it depends on how much you want to tune the suspension towards stability or articulation, and Lego could go either way

[Thirdwigg]

I actually find that that part does not do too well with twisting axles. After a certain amount of torque the axle will "skip" 90 degrees in the axle slots.

[Divitis]

This is how a torsion spring looks like.

All that Lego needs to do is putting one into a 'case' with axle holes.

Then, we could even have different versions (soft, hard and so on) that match the rigidity of the shock absorbers. 

[Jundis]

16 minutes ago, Divitis said:

This is how a torsion spring looks like.

The talk was about torsion bars

19 minutes ago, Divitis said:

All that Lego needs to do is putting one into a 'case' with axle holes.

You mean like this? :-P

@Milan Maybe we can chop out the last 10 or so posts into the "Part Discussion" thread.

[Divitis]

17 minutes ago, Jundis said:

The talk was about torsion bars

If you put that sort of spring into a bar shaped case it becomes a bar, no?

 

18 minutes ago, Jundis said:

 

You mean like this? :-P

An axle going through this piece rotates the same amount left and right. So it would effectively lock the suspensions together. The spring would only make it stiffer.

Connecting axles to the left and right of the torsion spring would have the output one rotate less than the input one, which is the point of the system. I think.

If someone knows better please advice. I'm no engineer and don't want to pretend.

[Mikdun]

5 minutes ago, Divitis said:

If you put that sort of spring into a bar shaped case it becomes a bar, no?

Torsion bar is just a bar/rod made from proper grade of steel - it does not have any loops.

5 minutes ago, Divitis said:

So it would effectively lock the suspensions together. The spring would only make it stiffer.

That's more or less how the stabilizer/anti roll bars works - they link movement of both sides of suspension, so the body rolls less.

Edited February 6 by Mikdun

[Divitis]

3 minutes ago, Mikdun said:

That's more or less how the stabilizer bars works - they link movement of both sides of suspension, so the body rolls less.

Help me learn. They are not intended to be rigid though, they are supposed to absorb part of the movement. So the hockey piece wouldn't work.

 

6 minutes ago, Mikdun said:

Torsion bar is just a bar/rod made from proper grade of steel - it does not have any loops.

Of course you're right. I guess I was thinking of springs because that's something that Lego already uses. So it would be more easily 'legal' for them to implement than a bending piece of metal.

[Jurss]

Some longer existing technic axle would work just fine

[Mikdun]

10 minutes ago, Divitis said:

So the hockey piece wouldn't work.

If you connect one side to the housing ant the other to the spring hole it will work.

[Divitis]

6 minutes ago, Mikdun said:

If you connect one side to the housing ant the other to the spring hole it will work.

Genius!

Anyone has implemented it in MOC yet?

[Lego Tom]

2 hours ago, Divitis said:

This is how a torsion spring looks like.

All that Lego needs to do is putting one into a 'case' with axle holes.

Then, we could even have different versions (soft, hard and so on) that match the rigidity of the shock absorbers. 

A torsion bar is a straight steel-alloy bar that twists to absorb shocks, such as hitting a pothole. There is no "spring" involved, per say other than what twisting the bar provides.

[msk6003]

1 hour ago, Divitis said:

Genius!

Anyone has implemented it in MOC yet?

Not a torsion bar type but someone used it on suspension. link

[dhc6twinotter]

1 hour ago, msk6003 said:

Not a torsion bar type but someone used it on suspension. link

I may be wrong, but It looks like it’s being used for return-to-center steering.  Some people did that before the servo motors were introduced.  

 

[2GodBDGlory]

Just now, dhc6twinotter said:

I may be wrong, but It looks like it’s being used for return-to-center steering.  Some people did that before the servo motors were introduced.  

 

I'm pretty sure he's using them for both steering and suspension, after taking a closer look

[gyenesvi]

5 hours ago, Divitis said:

Help me learn. They are not intended to be rigid though, they are supposed to absorb part of the movement.

3 hours ago, Lego Tom said:

A torsion bar is a straight steel-alloy bar that twists to absorb shocks, such as hitting a pothole. There is no "spring" involved, per say other than what twisting the bar provides.

I'm not proficient in the subject and the naming of things either, but my impression from offroader videos I have seen is that the function of this kind of torsion bar is to limit the body roll / axle flex. Examples I have seen were a straight bar (no springs) that connects the two sides of the suspension (either independent or live axle), and aims to limit its flex. Its purpose is not to provide springiness or dampening, for that the regular suspension parts are there. It is connected when driving on roads / higher speeds to limit body roll and improve cornering. So it would allow some flex, but not too much, it would try to keep the two sides at the same height. However, when going off-road, at slow speeds, it should be disconnected (the bar can simply be disconnected in the middle) to allow for max axle flex.

So when it comes to implementation in lego, I think using springy parts or even plain long axles would allow too much flex and would loose the point. That's why I'd aim for some rigid connection, built from shorter axles and axle connectors, especially if it needs to be disconnectable, it must be split in two parts somewhere. I believe the slack in the parts (like pin connections) would allow for enough flex when the two sides are connected, and could still force some parallel movement of the two sides. My first idea would be to use a new 2L driving ring to connect two 2L axle connectors in the middle; not sure if that would be strong enough though, maybe too fragile..

[2GodBDGlory]

14 minutes ago, gyenesvi said:

So when it comes to implementation in lego, I think using springy parts or even plain long axles would allow too much flex and would loose the point. That's why I'd aim for some rigid connection, built from shorter axles and axle connectors, especially if it needs to be disconnectable, it must be split in two parts somewhere. I believe the slack in the parts (like pin connections) would allow for enough flex when the two sides are connected, and could still force some parallel movement of the two sides. My first idea would be to use a new 2L driving ring to connect two 2L axle connectors in the middle; not sure if that would be strong enough though, maybe too fragile..

I'd agree that that's probably the best way to implement that kind of sway bar with Lego parts, but I really doubt that Lego would build a function that depends on normal parts twisting. Even if it works perfectly, it seems to me that they always treat all non-specialty parts as if they were completely rigid, in theory.

That idea using a driving ring to connect two axle connectors is one I've used on my Ram TRX MOC a while back, and I'd say it works well! I'm not sure it would be reliable enough for use in a Lego set, though, since it depends on having the sway bar in the right position for connecting, and bracing it might be hard with only legal techniques. (I think I might have used a short 4mm tube between the axle connectors, which Lego wouldn't do)

[howitzer]

On 2/5/2024 at 9:23 PM, allanp said:

Just don't display your model in a fully articulated pose!

I believe this is the reason TLG hasn't officially used axles as torsion springs, people would inevitably display the models in articulated poses and then complain about permanently twisted axles...

[gyenesvi]

I just made a quick sketch for this and built it; the two sides are not connected in the middle (of course, it could use more bracing structure, but I don't think it would be hard). Using the old 2L driving ring seems a better fit here as that one has its ridges inside all the way long.

Disconnected state:

Connected state:

It actually feels pretty strong as I tried to flex it with my hands, but has some amount of flex that looks just enough to transmit movement between the two sides. The actual amount of flex could be adjusted by varying the length of the axles used.

2 hours ago, 2GodBDGlory said:

I'm not sure it would be reliable enough for use in a Lego set, though, since it depends on having the sway bar in the right position for connecting

I think this would not be a problem, since in the resting position of the suspension, the two sides are automatically leveled and the connectors become synchronized, easy to move the switch.

11 minutes ago, howitzer said:

I believe this is the reason TLG hasn't officially used axles as torsion springs, people would inevitably display the models in articulated poses and then complain about permanently twisted axles...

It would be good to see what happens in such a pose. I could even believe that in a connected state it does not let the suspension articulate, and simply lifts one wheel. So when somebody wants to pose it with much articulation they will naturally have to disconnect it to look good :) Even if displayed permanently in a connected state, I don't think the 2L and 3L axles would twist.

Edited February 6 by gyenesvi

[Divitis]

This is the clearest video I've seen on the topic.

My impression is that the amount of force needed to put their bar in action seems to be more than what would come naturally out of turning.

[Lipko]

10 hours ago, gyenesvi said:

It actually feels pretty strong as I tried to flex it with my hands, but has some amount of flex that looks just enough to transmit movement between the two sides. The actual amount of flex could be adjusted by varying the length of the axles used.

Doesn't it bend more than it twist? Acting as a transverse leaf spring?

[SNIPE]

Note: that the hockey spring part can break very easily. I have 3 and one is broke

Edited February 7 by SNIPE

[Jundis]

11 hours ago, Divitis said:

This is the clearest video I've seen on the topic.

My impression is that the amount of force needed to put their bar in action seems to be more than what would come naturally out of turning.

Nice video, but I don't get why the bar need the thin 2L liftarms...?