Linear Actuators VS Pneumatics

[daniele]

I prefer pneumatics BUT with this, you don't have a precise control. I think pneumatics need a more precise valve.

[timslegos]

I prefer pneumatics BUT with this, you don't have a precise control. I think pneumatics need a more precise valve.

Absolutely couldn't agree more. With a lot of the MOC's I make, I need precise control, which is often why I use LA's.

tim

[allanp]

Oh what a lovely thread to bump, and using one of my favorite rants too

I forgot to meantion a couple thing in that particular speech, namesly that two pneumatics in parallel will stay in sync with each other whereas LAs can go out if one of the clutches slip making them work against each other and sapping power. Pnaumatics also look more authentic with their hoses running to them and their shiney rods! They even sound cooling with their cool pffft sound and compressors almost sounding like engines. I agree that improvements can be made to the valves so that they are more proportional (also having the ports on the rear of the valve would be better than on the side because many valves could be stacked side by side) and obviously we could really do with having longer cylinders.

[Aris]

I made a test in the past with a simple construction(similar to the one made for the LA using a scale) to measure the power of the pneumatic cylinder and the result was quite incredible having a force of 2.5 Kg

Edited May 11, 2020 by Aris

[allanp]

Technicly you could say the LAs are just as powerful, but that's when you have the motor stuck directly on the end of it (so no drive trains to create friction), and that's just boring!

[kieran]

Both have there uses but for me its pneumatics, that said I never got on with the first version with only one whole in the cylinder.

[Aris]

Indeed, the best pneumatic cylinders are the ones having round base, which means that these are lighter than the previous version with the square base.

[jorgeopesi]

As my avatar says it is obvious that I prefer the linear actuators, they are strong enough, biggs, accurates and it's great fun to make the transmission, not to put the L.A. beside the motor .

[hoeij]

Has anyone tested an actuator with an L motor? Does that produce enough torque to make the

internal clutch start to slip?

[Zerobricks]

Has anyone tested an actuator with an L motor? Does that produce enough torque to make the

internal clutch start to slip?

M motor can do that, dunno why L wouldnt be able to do so.

[ShaydDeGrai]

I used to teach a robots class where we'd often prototype things in LEGO parts before heading to the machine shop. We had brick built linear actuators at that time (no official part available yet) and the new parts are clearly stronger and smaller than the ones we used to build, but for the types of models I was assigning, it was pretty clear that you needed both. Pneumatics did some things very, very well. LA had other fortes. Sometimes the mechanism itself was less important that the form factor and we just went with what was easier to fit/mount.

LA had the advantage of smooth starts and stops, controllable velocity and precision placement down to the size of gear tooth; Pneumatics, not as good on these fronts.

Pneumatics fit into tight places, faster to react, faster to reach max/min extension, controllable strength (via air pressure in the reserve), lighter (results in less torque at the base when articulating multiple joints) and more vibration tolerant.

LA's need either embedded motors or a drive transmission to move them. This adds bulk, weight and complexity to the design. In the case of LEGO models it also raises the question of batteries and snaking power cables.

Pneumatics need pressurized air, this means a reserve, hoses and some way to recharge the reserve dynamically (we used to use geared motors to drive a brick built compressor for our tanks, giving us the worst of both worlds: hoses, wires, tanks, motors, drive chain _and_ batteries...)

LA can lock their position solely by the force of friction through a worm gear. This can be both a blessing and a curse. If you want a model to hold a pose without draining batteries or bleeding off air pressure, it's a big win. If your model is subject to external forces (possible large ones) it's a good way for things to break off or even snap.

Pneumatics need active pressure to hold a pose, but can also be bled off quickly if need be. If a pivot point is over torqued, safety valves can trip, bleeding off the air and allowing the model to move with the external force in a timely fashion rather than fight against it and risk breaking something.

And finally, for the true geeks out there, pneumatic valve networks can be configured into the mechanic equivalent of NAND gates and PLAs so with enough valves, tubing and air pressure you don't even need a NXT brick to control things. I suppose I could probably build an actuator-based relay-like system with the addition of a few non-LEGO conductors, but it would never be as compact as the valve based implementation.

[Aris]

Indeed.To be honest, both pneumatic cylinders and LA have their special uses. In case you need fast movement(by using air tanks) you should use pneumatics(but you are able to gain slow movement by using just small air pumps). In addition, if you need a function having only slow and stable movement you should use LAs. However, it depends on the weight that your invention has to lift and the point of lifting it.

[hoeij]

Pneumatic is more realistic, but is not 100% realistic because it uses air, which is a lot more

compressable than the oil used in true pneumatics. And you definitely notice this.

Even so, 8868 is awesome.

However, 8043 is even more awesome, and remote control works better and more precise with linear actuators

than with pneumatics (to pick up a pencil with 8868 requires a good bit of practice, 8043 is much easier

to operate).

[88high]

I am more of a LA person, mainly because they are more precise on their movements, and can be controlled compactly via remote. No auto-pumps and switching systems to control! However, there are a few downsides...

the first is the drivelines to power he LA's... They can be hard to incorperate with a compact model.

The second is price...

A big LA costs around $7.00-$10.00 USD... Times around 4 or 5 or so (the amount that you should have should be at least 4 or 5)The small LA's cost around $2.00-$5.00 USD and you hould get around 2-3...

However they dont need tubing and switches and air tanks, so you dont need to buy that...

Thats my opinion...

[Aris]

Mate, I agree with you up to a point.Both pneumatics and LAs have drawbacks. Look at my robot's picture.I wouldn't ever use 4 linear actuators instead of 4 pneumatic cylinders in order to make this robot hold a full glass of water weighing 600g tightly.Pneumatics give me greater confidence as for heavy things such that.So, I would definitely use LAs for light models.For heavy robots that I like building are not suitable.I love the precise movement of a LA but they usually do not serve my needs.

Here is the video of my first big robot enjoy

[AKM_76]

I'm here from the 2014 discussion redirection, and just wanted to say that I personally have zero interest about pneumatics.

[DrJB]

The main 'reality' with pneumatics, whether in Lego or in real life, is that it is very difficult to do precise position control. Most pneumatic (industrial) applications I have seen have an actuator either fully closed or fully extended. There is nothing in between for a very simple reason. One cannot do position control (easily) as that specific position depends on how much load there is on the transducer's end, simply because of air's inherent compressibility. Think of the doors in buses, pneumatic cylinders are used to keep the doors either fully open or fully closed, not in between. So, asking for precise control from pneumatics is a challenge ... way bigger than Lego.

Edited November 15, 2013 by DrJB

[Hrafn]

The main 'reality' with pneumatics, whether in Lego or in real life, is that it is very difficult to do precise position control. Most pneumatic (industrial) applications I have seen have an actuator either fully closed or fully extended. There is nothng in between for a very simple reason. One cannot do position control (easily) as that specific position depends on how much load there is on the transducer's end, simply because of air's inherent compressibility. Think of the doors in buses, pneumatic cylinders are used to keep the doors either fully open or fully closed, not in between. So, asking for precise control from pneumatics is a challenge ... way bigger than Lego.

Interesting. That doesn't seem to be an issue in hydraulics like those in many construction vehicles - is it an issue in pneumatics because the air is compressible while liquids are essentially incompressible?

If Lego made a proportional pneumatic controller (especially one that connected to the PF system!) I think it would still be a valuable addition, even if it wasn't all that precise and the motion depended not only on the controller's setting but also the load on the cylinder. Any proportional control would be nice to have, even if imperfect.

[Kumbbl]

the precision of lego pneumatic depends IMHO (of course a.o.) on the smoothness and the right power of the air-supply.

The best examples are the airtank-with-manual-pump supported Front-Loader 8459 and the compressor driven legend 8868

8459: here the air from the pump is stored first in the airtank and if you then open one of the valves the full air pressure overpowers the cylinders which results in this not controllable movements which oscillates between either fully retracted or expanded --> this is a bad example

8868: it seems that here the air-supply has exactly the right power and constant flow - at least in my model i can controll all movements (apart from the claw), means slewing, boom-movement very smooth and quite precisely... --> this is a very good example that lego pneumatic is not per se unprecise...

Conclusion: the smoothness and the control-ability of Lego pneumatic depends not least on the air-supply...

Ovwerall: i would not see this topic so dogmatic: both - LAs and pneumatic - have its pros and cons... personally i prefer pneumatic (mostly because its "reality" and the reason mentioned by allanp: "...two pneumatics in parallel will stay in sync with each other whereas LAs can go out if one of the clutches slip making them work against each other and sapping power..." this is a real and big goodie IMHO) but i also appreciate the complexity of drivetrains like in the 8043 excavator - which BTW works very smoothly and well for me - even just powered by M-motors - here LAs probably are overall the better solution... but one thing must be mentioned: LAs without PF is a real nogo - driving LAs by hand is a playabilty nightmare (e.g. the 8265 Frontloader is the best example - a quite good model when motorized - a bad model when not)

so IMHO: both have the right to exist - both offer great and versatile possibilities to build great models

Just my 2cts

[AlmightyArjen]

I love pneumatics and do a lot with them. See also my YT channel. The system has more power that LA's but it cannot be regulated. You could regulate the speed of a cilinder by squeezing the tube to the cinlinder a bit but that's it.

I don't think Lego will ever make regulated valves for Pneumatics: it means that you need some kind of precise pressure control also and that will be to large/expensive...

Edited November 15, 2013 by AlmightyArjen

[DrJB]

Interesting. That doesn't seem to be an issue in hydraulics like those in many construction vehicles - is it an issue in pneumatics because the air is compressible while liquids are essentially incompressible?

That's exactly it. Oil is essentially not compressible. Many hydraulic installations, e.g the loader arm of a backhoe, the operator sends enough oil to move the arm to a given position and then he can stop there. This, he can do no matter how much load the bucket is carrying. There are safety precautions/protocoles preventing him from doing this, but that's another discussion.

[Junpei]

<snip>

LAs without PF is a real nogo - driving LAs by hand is a playabilty nightmare (e.g. the 8265 Frontloader is the best example - a quite good model when motorized - a bad model when not)

so IMHO: both have the right to exist - both offer great and versatile possibilities to build great models

Just my 2cts

I agree. Pneumatics are much harder to integrate with RC PF in a small space, the opposite of an LA. I honestly think that unless in a sizeable model, if you're going to make something manual, make it pneumatic. 8455 is a good example. The pneumatic cylinders move so many functions, the playability is outstanding, but imagine making it RC. It would be impossible. If you used LA's, though, you could probably cram at least a few functions in that are RC. But LA's are just unrealistic, and like Kumbbl said, they have a high risk of coming out of sync, and they are much less powerful. So I think that for models that are not huge, if you want RC, go RC and use LA's. But I and many others would use pneumatics. If you want pneumatics, go pneumatic, and ditch the RC. I honestly would like a manual pneumatic set much more than an RC set with crappy LA's.

So screw LA's, even if you have to sacrifice RC, pneumatics are the way to go.

But really, LA's are OK with me I guess. They're good for RC, and I'll use them for something like a crawler's steering, or a gearbox, things like that.

Other than that though, for something like a backhoe, they simply suck. Take 8069, for example, which is, unfortunately my only Technic set. Anyone who has ever owned an 8069 knows how painful it is to first sync the LA's to build the front bucket's arm, and more so how painful they are to operate, as well as the bucket tilting mechanism, there is so much backlash and unnecessary sh*t that you can barely operate it. Now imagine that with pneumatics. It would be easy, more powerful, and more playable. The ridiculous mechanism would be no more and would open up more space for routing tubing and placing switches, etc. It would work faster, and overall just better. There would be the realistic pneumatic tubing all over the model, just like in real life. Beautiful.

So my verdict: GO PNEUMATIC!!!

[schermvlieger]

I prefer pneumatics too, for all the reasons mentioned by twentyleggedhen

[AlmightyArjen]

In my opinion still the best Pneumatic set ever! And it was even my first

[Kumbbl]

In my opinion still the best Pneumatic set ever!

:thumbup: :thumbup: by far... and - as described in my previous posting (#69) - with very smooth and controlable movings - even the slewing works very smooth - the crane can be slewed exactly to a certain position...