[WIP] Telescoping boom excavator

[ord]

On 11/16/2020 at 10:38 PM, howitzer said:

Is there any way to make the boom support narrower? The problem is the gearing for axial rotation, which seems to be pretty much impossible to fit in between the turntables in any narrower configuration than this:

You could mount the boom axis above the luffing axis (rather than on the same plane)? This looks like how it's done in the Gradall excavator photo you attached. This could also lend itself well to controlling luffing by linear actuators, if you want to go down that path. 

Great project btw - it has a lot of similarities with an industrial robot arm. Well, I guess it kind of is a robot arm... on tracks 

[howitzer]

4 hours ago, ord said:

You could mount the boom axis above the luffing axis (rather than on the same plane)? This looks like how it's done in the Gradall excavator photo you attached. This could also lend itself well to controlling luffing by linear actuators, if you want to go down that path. 

Great project btw - it has a lot of similarities with an industrial robot arm. Well, I guess it kind of is a robot arm... on tracks 

Maybe I should try to build a prototype of this kind of model. Though I fear that it might be pretty ugly even if workable solution, and I suspect it would force the boom position too high up.

[ord]

5 hours ago, howitzer said:

Maybe I should try to build a prototype of this kind of model. Though I fear that it might be pretty ugly even if workable solution, and I suspect it would force the boom position too high up.

I admit, there is something nice about having the axes in the same plane. If you were to control it with linear actuators you could do away with the lower turntables and have it mounted on pins, I think, to streamline it a bit. I wonder how luffing is driven in the actual machine... 

[howitzer]

5 hours ago, ord said:

I admit, there is something nice about having the axes in the same plane. If you were to control it with linear actuators you could do away with the lower turntables and have it mounted on pins, I think, to streamline it a bit. I wonder how luffing is driven in the actual machine... 

They use hydraulic cylinders, which are much thinner and shorter relative to boom size than LA's would be in Lego version, and it also helps that in the real machine it's possible to choose geometry much more freely...

[Rudivdk]

The more I look and think about that arm, the more intrigued I get... It sure is a nice choice, and a rather challenging project.

One thing that comes to mind for me is if you could make the bucket LA operated too. It will probably end up slightly affected by the boom rotation, but it might be worth a try. You could use one of these to transfer the powering from the superstructure into the extending part of the boom. And since you have a turntable on both sides of the boom, you could also think about routing functions towards the boom from each side (extension, rotation, bucket movement and than there is even 1 possible function left, quick release mechanism for the tool maybe?...).

Although all of this will probably blow the boom out of proportions...

I'll be following this build!

[Mr Ogel]

A cool model. I still have questions about the concept itself. From the pictures on Gradall I would assume the telescopic arm is resting on another arm which is lowered and raised by hydraulic like at a normal excavator not by a rotational mechanism. Also I read the extending would be also done by hydraulics -maybe this would be an easier way to follow?

Anyway I would assume your gearbox is not fitting into a supercar, that's a pity...

[howitzer]

On 11/23/2020 at 11:42 AM, Rudivdk said:

The more I look and think about that arm, the more intrigued I get... It sure is a nice choice, and a rather challenging project.

One thing that comes to mind for me is if you could make the bucket LA operated too. It will probably end up slightly affected by the boom rotation, but it might be worth a try. You could use one of these to transfer the powering from the superstructure into the extending part of the boom. And since you have a turntable on both sides of the boom, you could also think about routing functions towards the boom from each side (extension, rotation, bucket movement and than there is even 1 possible function left, quick release mechanism for the tool maybe?...).

Although all of this will probably blow the boom out of proportions...

I'll be following this build!

I thought about that but decided against it due to the mechanical complexity of routing power for LA's through the turntables. I currently plan to have the axial rotation to be powered from both sides, with one side turning the front part and the other turning the rear part of the boom support. I also don't have any of those sliding 8T gears but they indeed do open up some possibilities so maybe I should get some...

15 minutes ago, Mr Ogel said:

A cool model. I still have questions about the concept itself. From the pictures on Gradall I would assume the telescopic arm is resting on another arm which is lowered and raised by hydraulic like at a normal excavator not by a rotational mechanism. Also I read the extending would be also done by hydraulics -maybe this would be an easier way to follow?

Anyway I would assume your gearbox is not fitting into a supercar, that's a pity...

I actually didn't do much research into how the real machine works as I'm now just making an attempt to build a machine with all the important functions, disregarding looks and mechanical authenticity. The boom on the real machine is indeed luffed with hydraulics, I'm not sure about the extension but hydraulics would be logical there too. Axial rotation is driven by geared turntable.

This gearbox is extremely unsuitable for a car, as it has 4 inputs and 8 outputs with a selector for choosing the outputs in sets of four...

[howitzer]

A couple of progress photos. 0,33l beer can for scale:

[howitzer]

Another progress photo. This time I added the undercarriage:

I don't have the medium sprockets for the treads (waiting for BL order to arrive) so I couldn't add the treads themselves yet but otherwise I believe the undercarriage is complete and works fine.

---

Now the structural and functional parts are mostly complete and there's only minor things to add like hosing/wiring and some additional support for the boom gearing beside the cosmetics like exterior and a cockpit. And of course programming the interface for controlling this thing.

[howitzer]

So, I got the thing assembled enough for testing, and all the functions seemed to work fine, except the axial rotation of the boom. I could hear the gears straining and while one direction of turning worked end-to-end, the other direction didn't but the resistance got periodically too big and gears skipped. It appeared that the two sections of boom support went easily out of phase, so to speak, instead of turning together, which caused a buildup of friction.

So I set out to design a new mechanism based on worm gears.

The first problem was of course finding out how to actually drive the turntable with worm gears, and it appears that there's no good way of driving the 60T turntable with the old smaller radius worm gear, so the larger radius one must be used. I guess I could solve this by getting two of the old type 2 turntables, but that didn't appeal to me at this point so I started building support for the worm gear with the type 3 turntables. The half-stud distances became soon a real pain in the butt as the connectors allowing for half-stud translations are few and limited in directions.

In the end I managed to make something that actually works, pictured here:

The result is an ugly mess of connectors and weak connections, and I would really like to make it sturdier and cleaner but it works much, much better than the old design. There's also the problem of needing 2 worm gears to drive it and I only own three, third of which is required in the boom telescoping (note that it's missing from it's place behind the motor in this photo, I had to borrow it for testing) so I'll have to take the third one from inside the superstructure, as I've used one of these worm gears in there also. There's also other gearing corrections that have to be made, the boom raising/lowering is far too slow (geared down twice with 8:1 worm gears plus 12:60 turntable gearing).

Which brings me to another matter: I began to work on the exterior and while I don't have a photo, I simply don't like it. The superstructure is too high and bulky, and I'm seriously thinking of dismantling it and rearranging the motors and gearbox so that the height of the superstructure is reduced. Here's an unfinished CAD model for how I think the design could be improved:

Much of the support is of course missing, and not even all the gears are in place but you get the idea. Instead of having the pairs of selectors stacked, I planned to have them side-by-side, with motors in a row. The black double-angled beams mark the position of the old design's maximum height, with carriage and boom in the same place as in the previous version. One thing of note here is that I would drive the gear selector with a dumb motor as it's smaller and fits easily in the place, but it has the downside of not being able to be controlled for rotations, which makes it harder to use here. I might have to build some kind of limiter there too, to make sure the selectors stay in phase.

[Glaysche]

2 hours ago, howitzer said:

So, I got the thing assembled enough for testing, and all the functions seemed to work fine, except the axial rotation of the boom. I could hear the gears straining and while one direction of turning worked end-to-end, the other direction didn't but the resistance got periodically too big and gears skipped. It appeared that the two sections of boom support went easily out of phase, so to speak, instead of turning together, which caused a buildup of friction.

So I set out to design a new mechanism based on worm gears.

The first problem was of course finding out how to actually drive the turntable with worm gears, and it appears that there's no good way of driving the 60T turntable with the old smaller radius worm gear, so the larger radius one must be used. I guess I could solve this by getting two of the old type 2 turntables, but that didn't appeal to me at this point so I started building support for the worm gear with the type 3 turntables. The half-stud distances became soon a real pain in the butt as the connectors allowing for half-stud translations are few and limited in directions.

I've been working on a robotic arm that went through many of these same challenges.  I actually tried a 60t turntable with 1L worm gear, 56t turntable with 2L worm gear, and a variety of other things.  I was not happy with any of the solutions.  What I eventually ended up doing was to have 2 12t gears driving a 60t turntable.  This ended up reliably rotating in both directions and was a pretty reasonable amount of structure needed to make it all work.  Here are some pics:

 

This is the robotic arm.  The joint of interest here is the rotating joint at the top.

 

Zooming in on that, I have two 12t gears driving the turntable:

 

Here's a pic with the arm removed:

 

These 2 12t gears are attached to 28t gears which are driven by an 8t gear seen at the top middle.  This localizes all the high torque to just those two axles and the rest of the gear train is relatively low torque and easier to deal with.  These two axles needed really good axle support but once that was built, it worked pretty well.

[howitzer]

5 hours ago, Glaysche said:

I've been working on a robotic arm that went through many of these same challenges.  I actually tried a 60t turntable with 1L worm gear, 56t turntable with 2L worm gear, and a variety of other things.  I was not happy with any of the solutions.  What I eventually ended up doing was to have 2 12t gears driving a 60t turntable.  This ended up reliably rotating in both directions and was a pretty reasonable amount of structure needed to make it all work.  Here are some pics:

<snip>

These 2 12t gears are attached to 28t gears which are driven by an 8t gear seen at the top middle.  This localizes all the high torque to just those two axles and the rest of the gear train is relatively low torque and easier to deal with.  These two axles needed really good axle support but once that was built, it worked pretty well.

Thanks for the idea, somehow I never actually thought about actually using the 12T gears in parallel to the turntables. The first version had them driving it in perpendicular, and while that was mechanically simple and easy to build, the gears ended up slipping too easily. I might have to attempt this way too.

Nice robotic arm by the way!