Joint Issues with Scissor Jack

[Drewpan]

@TheCADWhisperer 

I have been following your YouTube video series about the Scissor Jack from here:

Scissors Jack Video 7 of 8 - YouTube

When I come to setting the Sub-Assembly so that it rotates and the Screw passes through the Nut, there seems to be an issue where the animation goes crazy and Nut and Assembly seem to go the "wrong way". My Handle also seems to have decided to teleport away from the rest of the project.

If I manually move the parts they seem to be linked and jointed correctly. Joint 15 seems to be the issue. I have carefully watched the video numerous times and I have played with negative values and reversing the joint but I still cannot seem to get it working so that when I rotate the handle, the screw tracks correctly and the arms lift and lower properly.

I have attached my Fusion File this time. It is almost identical to yours - I have attempted to follow you exactly but there have been some places where this is not the case. There is one other error I am aware of where the sheet metal has not worked. I know it is because there are artefacts from the cutout before folding. I am not worried as it should not affect the rest of the design.

When I get this working I intend on finishing the video course and redrawing the Project from scratch for more practice.

If you could please assist as I am quite keen to do the last video on analysis to see how this actually works.

Cheers

Andrew

[davebYYPCU]

I'll chime in, I did this one a while ago, and without reference to anything but your file, I got to say, why Cylindrical Joints, where you need Revolves?  Probably not learnt to set type for each joint, and cylindrical carries over from last used.

Screw sub assembly, change Cylindrical 2 to a Revolve and Cylindrical 3 to a Revolve, will bring the handle back into position as a bonus.

There is also a cylindrical for the Screw Clearance Rivet, would be a Revolve as well.

Might help....

[TheCADWhisperer]

In Dynamic Simulation (which Fusion doesn’t have) Revolve Joint results in a mathematically redundant joint.  One Revolution Joint is enough to remove Translation DoF.  In the real world …. I’ll explain more tomorrow. Check back as technically Cylindrical is wrong too.

[TheCADWhisperer · Accepted solution]

@Drewpan 

What is the purpose of this unconstrained line in your arm sketch?

What is the purpose of this geometry in Upper Left Arm?

Upper Left Arm appears to have some interferences?

Before testing kinematic operation I prefer to examine the mechanism for interferences.

[TheCADWhisperer]

@davebYYPCU 

A month or so ago someone here wanted to create "perfect" sheetmetal part with merged faces.

I argued that in the real world every dimension has a tolerance associated with the dimension and at combinations of upper/lower limits of the tolerances we end up with tolerance "stackups" that result in interferences in part/assembly.  Therefore we should consider each and every feature at min/max condition at the design stage.

By representing the real world as closely as possible at the design stage we do not end up with the rather a obvious issue of 33 powerful rocket engines "blow-torches" blasting at a concrete slab over a shallow water table (essentially a pressure vessel).  That would be ignoring more than a century of pressure vessel experience.  We might still miss some things - but we should not miss the obvious.

Now I assume there is a door in the room where you are reading this...

If you modeled the door assembly in Fusion 360 - you could model with only one hinge at the bottom with a Revolution Joint.  Adding additional hinges and Revolution Joints does not change the operation of the kinematics.

So why do we usually have two or three hinges?

The arms and linkages of the jack form what is actually a computationally complex kinematic chain (in the real world).

Let's simplify this down to just one of the arms...

We know that we cannot manufacture perfect parts and in addition for motion to occur without binding we need a certain amount of clearance...

I have increased the size of the hole in the arm for illustrative purposes.

Let's analyze the real world Degrees of Freedom (DoF).

1. The arm can move side-to-side (Translate) along the axis of the pin until it contacts the base.

2. The arm can rotate around the axis of the Pin - our intended motion to do the work of the mechanism - but if we add a Revolution Joint this artificially removes the Translational degree of freedom violating #1.  On the other hand, a Cylindrical Joint permits Translation and Rotational DoF.

3. Now back to the hole that I enlarged for illustration.  The arm can also twist a very slight amount until the sides of the holes meet the cylinder of the pin.  Almost everybody misses this twisting DoF.  (Edit: Actually a Rotational DoF, Twisting is another possible issue.)

In the digital world we tend to overconstrain our mechanisms removing DoF that actually exist in the real world.

We can idealize our mechanism for the purposes of digital analysis - but we must not loose site of the fact that we did something that might adversely effect what we think we know.

Going beyond mere kinematic analysis into closed loop linkages analysis of dynamic motion loads, we need to be very very careful about how we idealize the mechanism for analysis without over constraining.

Autodesk Inventor Professional includes a tool for Mechanism Status and Redundancies (overconstrained DoF) analysis.  

In mechanics this is named after a particular person in history - but I forget his name.

In Autodesk Inventor Professional - Dynamic Simulation Environment we have more Joint types and can graph a plethora if information about each Joint...

It turns out that constraining this mechanism such that we don't get an overconstrained Redundancy warning is quite challenging to do.  In fact, the bane of our existence in Dynamic Simulation is resolving this issue.  This was one easy to understand mechanism that my students likely had hands-on experience with and illustrated the issue nicely.

We approach assembly constraints (Joints) in a different frame of reference when we know our Design Intent is to go beyond mere kinematics.   Working from this frame of reference - we did the best we could in Fusion 360 with a limited set of Joint types.  And knowing these limitations I might have been haphazard in my assignment of Joints.

Inventor 2023 Help | Mechanism Status and Redundancies Reference | Autodesk

[davebYYPCU]

It was the door hinge/s that makes sense, of your previous announcement, I use one joint and Rigid group, a lot, but the file I reviewed,

I was finding alignment issues, (not going far enough back for those geometry issues)

there are people in here recommending Cylindrical, to resolve joint alignment conflict.  (without fixing them, Yikes - my mind jumping into that rabbit hole)

I later revisited the referenced video, and noted you just say Joint, and type was not mentioned, ( not critical of this, but new guys are gunna get, last type used if not awakened - especially now they divided the Dialogue Box into two tabs, and, you explained earlier the target audience.)

Cylindrical in this context, needs immediate realistic limits, (commented) that prevents the handle issue, and avoids an unnecessary introduction of (expensive) contact sets.

I am also fond of the animated line drawing as an investigation tool.

I think we are on the same page, thanks.

[Drewpan]

Thanks for your reply,

The original purpose for this line was a fudge to make it easier to draw the construction line around the circumference. There is a construction line from either side of the cutout but this is straight. Later in the video you drew a line that went to the circumference and since mine was missing I had to add it back in.

There is no purpose to these geometries, they are accidental. They occurred because we copied one of the arms and then modified the width of them. We did modify the sketch also which should have modified the extrusion cut. What I should have done was check it had worked properly but I assumed I had copied the video correctly and missed this until much later as by this time they were hidden by other parts. I am pretty sure this is the reason I couldn't make the sheet metal parts because of this.

I had missed these interferences so I will go back and re-echeck them all. These sorts of checks are not second nature to me yet as I am still learning. As you previously stated - these videos are fairly advanced and you would expect your students to know this stuff.

Cheers

Andrew

[Drewpan]

Hi,

Ok, so I have removed the spurious geometry and the unconstrained lines; fixed the extrusions and the sheet metal now is working; removed all of the interferences; and lined everything up. At this point I am still unable to get the Sub-Assembly working correctly to rotate and move the screw properly. The animation goes wild still and I am not sure why or what it is attempting to do. For some reason also, the handle Joint is seated correctly but the handle comes astray - again not sure why this would happen.

Andrew

[davebYYPCU · Accepted solution]

In message 2 above, I said, 

Screw sub assembly, change Cylindrical 2 to a Revolve and Cylindrical 3 to a Revolve, will bring the handle back into position as a bonus.

There is also a cylindrical for the Screw Clearance Rivet, would be a Revolve as well.

They were not changed in this file.

The handle disappears because the cylindrical joint allows the slider part to do it, without sensible limits of travel, change to Revolve fixes it.

Cylindrical 14 ??  Can be a Revolve, but Revolve here as is will fail, is lined up with something.... and still works because it is cylindrical.

Might help.....

[Drewpan]

Hi,

After making those changes the model is now working correctly.

I think I may have been caught with the "if you don't change it you get the last type by default" with the Joints.

Thanks to you and @TheCADWhisperer for you help.

Cheers

Andrew