Wednesday, July 9, 2008

TBs TechReview 02 – TECHNIC PF, Linear Actuators – Part I

Yesterday a few PF new elements hit my workbench… ;)
Among them, two Linear Actuators (LAs) with respective brackets and one of the new PF lights.

So, I decided for a review to the Linear Actuators.

The first impressions were:

  • LAs are proportionally thinner than I’ve imagined from the several images seen before, which is positive and more realistic for most of the existing machines you may want to build.

  • As an introductory note, each LA has a simple and standard interface in its base, where you can find the access to a driving axle-hole (orange part) and a simple connecting interface. This connecting interface is built so and adapter part (bracket) could fit to adapt LA connection type or geometry and apply for different use cases. We may also speculate if there will be in the future, a small PF motor which adapts directly into the LA base/driving interface... :)

    I got two different of these brackets for each LA, which was a surprise for me. Until then I imagined to actually exist just one (the one we have seen for months on all previous Excavator images).
    But in fact there is a second and different one, which I imagine to be used at Telescopic Handler (8295) B-model, whose instructions are not yet available.
    Because these new information became all available at almost the same time, I was able to check it appears at inventory listing from 8295 building instructions, but it is not used with the main model.

The leftmost bracket is included twice with 8294 and once with 8295. For the rightmost bracket, it appears just once at 8295 (presumably for B-model as stated above).
Hopefully (at least I wish it...) we will see some sort of Service Pack at S@H, with at least one LA and one bracket of each, in the future.
The rightmost adapter looks to have been designed, to adapt directly in front of the XL and M PF motors. Just great! :)

The Linear Actuator characteristics:

Despite being something easy to imagine for those with some technical knowledge, first thing I was willing to do, was opening one to look at the internals and show here at TBs how does it looks inside and how does it really works (kill your curiosity, and of course mine also)… :P
Don’t worry… I’ve four of them, at the moment and more on the way. :)
Below you have the resulting images.

Attention: Don’t try this yourself unless you are prepared to loose one LA!...
The disassembling process is destructive, mainly because some parts are strongly glued (if clicking at some of the images below to see them larger, you will see still some glue residues on the 'Dark Bluish Gray' base, which I wasn't able to remove completely).

Despite the appearance that disassembling the unit is a manageable issue, it happens no to be the case. At least not without significant damage!

LAs look solid and robust, with both plastic and metallic parts being strong, very nice looking and well finished, as usual for the LEGO parts.
Despite this, the assembly internal parts are able to shake to some extent, which denotes some internal space gaps existing. This naturally allows keeping the operating friction at minimum levels, but I believe it would have been possible to minimize this internal spacing even further and give the LA a more solid touch.

The LA consists of a metallic piston rod with a terminal screw, which runs inside a plastic cylindrical case. Together they act like a worm gear setup.
Like any worm gear arrangement, LAs behave:
  • Reducing the rotational speed at the input, to increase the torque (linear force in this case) at the output.
  • The direction of transmission (input shaft vs output shaft) is not reversible, due to the greater friction involved between the rod terminal screw (worm) and the plastic case where it turns (worm-wheel). This has the advantage to eliminate any possibility of the output driving the input at any load circumstances.

There isn’t any kind of internal clutching mechanism, which has to be somehow achieved externally for part safety, every time a LA is driven by an external motor.
Meanwhile, I discovered the info above is not correct at all. LAs have in fact internal clutching (more about this, on the part II for this review, still to come).

Both LA piston head and base fixing points are similar to one of the recent 1x2 thick liftarms, with pinhole and axle-hole. This design option for a mixed connection type, makes it more versatile and suitable for several applications/uses.
Also the brackets that fit into LA base use the same solution, which makes them very useful as well.

The plastic piston head (1x2 liftarm like) is strongly attached to the metallic rod, which I didn’t want to disassemble and break into more parts (despite the unit being already lost…). :(

The LA driving interface is an axle hole on its basis (the orange part) which also maximizes its applicability, in opposition to an eventual protruding axle, which was for me the most probable expected solution till certain point in time.

If not fixed, the piston rod tends to freely rotate as the driving axle gets turning. Mainly because the low internal friction, mentioned above.

Physical dimensions:

From end-to-end the LA measures 11L when retracted and 16L when extended. This means the rod to extend about 5L or 40mm (LA linear extent).

On the diameter side, LA cylinder measures 15mm i.e. nearly 2L.
As commented in the beginning of this post, this makes the LA realistic and relative well proportioned. This was one of my fears when I saw the first images available, because it looked not realistically proportioned. Although IMHO, from the technical construction point of view, it could have been even thinner making it to look further with the real thing.
However we must realize that LEGO natural dimensions and intrinsic spacing required for pin or axle-holes impose some limitations here, which lead into the actual design.

The metallic rod diameter, measures ~5mm which just a bit above any TECHNIC whole diameter, if it is of any importance...

In order to achieve the full LA extent, the driving axle-hole needs to turn 26 full revolutions.
To be more precise it is in fact around 26,5 full revolutions. However it would be not significant trying to bet on precise number, as you can always force a bit at both ends which also makes some variation among readings.
Anyway this means, if using a direct motor drive and without considerable load, that a motor should be running approximately for:

To be continued…

Edit 1:
Corrected above the statement where I wrote that LAs don't have internal clutching.
In fact they have! More to come on the review 2nd part.

Edit 2:
The link for 2nd part of this review was added.

Last Update: 2008.Aug.03 11:39 GMT


Anonymous said...

That's hell of a review Conchas, congratulations. How much do you think they'll be on Bricklink?

Unknown said...

Did I get it right that if the piston is not fixed, when turning the input, the pistol will just rotate without retracting?

Theo said...

Thanx for the review. Waiting for part II :)

TechnicBRICKs said...

Answering by order,

@lego adam
Don't guess, but it shouldn't take to long to know... As soon the inventory becomes available I'm sore there will appear several to sell.
However I believe the prices will start relative high, and will drop significantly after some time.
Hopefully S@H will have this part to sell also as an individual PF part in the future, which will increase the offer and put additional pressure on the BL price.

yes right! or at least you can't determine exactly how much it will retract and rotate.
don't forget internal friction is low for the white tube to rotate, but the screw acts like a worm, so the longitudinal friction between the screw and white "tube" is relative high, which causes this effect.
However I don't think it is an important or negative effect. Who will want a piston not fixed somewhere?

I think you well need to wait sometime.
Next is the excavator (8294) and for part two I have some ideas that may require extra time to prepare.

Unknown said...

Thanks for your reply!
I`m asking because you can`t make a grabber hand with it without building some kind of frame to prevent rotation.

TechnicBRICKs said...

Ok! I see your point.

Carlos Ribeiro said...

Great work Conchas, i can't wait for mine #8294, thanks for the review.

Anonymous said...

Thanks a lot for saving life of one of my (still to come) linear actuator! Waiting eagerly for part 2 where I hope to learn everything about torque limiting and maximum push/pull force...

Parax said...

conchas, Have you seen the new diff from 8297 yet?

Parax said...

Interesting how you report 26 revolutions, Mine goes closer to 27 at 9675 degrees. I wonder if there is any variance.. (dont let the tip rotate while winding)

Thomas J Avery said...

Coolness. Many thanks for making the sacrifice. You've answered our burning questions about the part's innards.

Now, I'm wondering how stable the part is (as a whole) under high loading? With as much "play" as you've described, the whole thing could be prone to buckle easily, especially closer to max. extension (think column buckling). Hopefully I'm wrong.

It looks pretty well built for what it is. Glad to see real metal in use :-) If only the "driving tube" was metal too!

Brian Davis said...

It seems fairly stable to me, and has a firm "stop" feel to it at the ends of travel (if the piston is not free to rotate, for course). Not sure what the stripping torque could be, but I'm also favorably impressed by the LA's. Now I just need time to use them... actually, the free rotation might find some interesting uses down the road, it's a "feature" I wasn't expecting, but it lends some additional possibilities to these units. Like a shaft rotation only producing an extension if and only if a "guide" element or limit is engaged (mechanical logic).

Anonymous said...


As Parax said, the revolutions are closer to 26 and a half.

TechnicBRICKs said...

Hi guys,

answering one by one. ;)

Yes, I've seen the new differential too! ;)
I hope to write something in a further post about this, but the main issue is to verify if the new design allows blocking like its predecessor or not. From the images I saw, it looks like it doesn't which we may consider a design drawback.
Hope at least that the new design is really effective and better avoids gear slipping.

For the LA revolutions, you are right! They are above 26, in fact around 26,5 according to my measures, but just decided to present an integer number as it can't be a very precise measure (anyway I'll change the post for a more precise information).
I don't believe there are significant variations among parts, but you can always force a bit on both ends, so the reading may vary slightly.
Which method did you use to measure?

Yes it is very well built, but it buckles a bit when extended.
However since it is supposed to work in a linear way, it shouldn't a problem.
Sure in some applications (even on the Excavator) there are forces with components perpendicular to the movement axle, but under reasonable loads, I don't believe they represent a problem also.

The "problem" is that you can determine exactly how much it will rotate and extend when the piston is free. It will always depend on factors that affect internal friction conditions, like the operating inclination.

TechnicBRICKs said...

I meant " can't determine...".

Brian Davis said...

True, but... picture an LA in an upright position, with a small weight on the top of the piston. If the weight/piston is free to turn, the internal friction will spin (not lift) the piston/weight combination rather reliably. The moment the weight/piston combination is prevented from spinning, it will be forced up.

This was not a behavior I would have thought of, except having seen Kevin's pneumatic logic, and working with mechanical logic elements myself. As a "weakness", it doesn't bother me in the least - as you point out, in most applications, the piston is fixed to a structure and it's not an issue. and as a potential alternative usage, it intrigues me.

TechnicBRICKs said...

For sure, under controlled conditions like the scenario you described, it can be taken advantage from this unexpected behavior.

Parax said...

Re: "Which method did you use to measure?"

Rigged guide bars to prevent rotation and connected it to an NXT motor and let it run until the end, whereabout I pressed the button to stop it, then manually turned the motor to find the end point while the NXT displayed angle on screen, I'll post a pic of rig on BL. here

Parax said...

Brian, Perhaps some kind of governer device? as rotation is impeeded it extends to a stop switch?

TechnicBRICKs said...

What a setup! ;)

It's like a rotation sensor.
Doubt it would be reliable enough. Because for moments there may happen extension even without rotation being exactly impeded.

Alex Campos said...

One Linear Actuator gives its life in the name of science... rest in peace, son, your sacrifice wasn't in vain. ;-)

Now, I think the only measurements needed for LA's are the comparison with pneumatics: LA's most obvious competitor in terms of field of application. But, like I discussed with Conchas some time ago, the disparity of actuating systems won't make that very easy.

Anonymous said...

Wow! Thanks for doing this as it saves my curiosity.

BlueToothKiwi said...
This comment has been removed by the author.
BlueToothKiwi said...

Thanks for the review, Conchas - very nice!!

I took the liberty of using one of your pictures in my blog post today on TheNXTStep blog about LA's and how you can use it with a MINDSTORMS NXT.


TechnicBRICKs said...

Be my guest Tim. :)
Absolutely no problem, using material from TBs!

Actually I'm post processing photos and videos for the LAs review, next part. ;)

Anonymous said...

Thanks for sharing your review on that product.

linear actuators said...

It looks simple device but I'm sure that device is very useful.

Related Posts Plugin for WordPress, Blogger...

© 2007-2014 TechnicBRICKs
TechnicBRICKs contents may be sporadically updated, if the authors finds further relevant info about a certain post, or content/spell mistakes. Hence please don't be surprised if you find few changes at later visits, relative to a previous read.

TechnicBRICKs often shows other peoples' creations and/or images. We always try to credit the author(s) and link to their main publishing website, and if possible with their name in real life.
Since this is not always possible, we request that if you find something here that is yours or from someone you know, you leave a comment on the respective post and claim the authorship.

TechnicBRICKs is optimized for Firefox 16.0 and 1600x1200 resolution displays or wider.

LEGO® is a trademark of The LEGO Group of companies which does not sponsor, authorize or endorse this blog.
LEGO, the LEGO logo, the Brick and Knob configurations, the Minifigure and MINDSTORMS, are registered trademarks of The LEGO Group.
Original LEGO images are copyrighted by The LEGO Group and are used here in accordance with their fair play policy.
You can visit the official LEGO® website at