Motion Constraint Loop - Is this possible?

[Anonymous]

I noticed in my assemblies that the movements locked as soon as I closed a loop of rotational movement constraints, so, as I'm beginning, the doubts came to me:

Am I doing something wrong?
Am I doing something illogical?
Am I breaking some basic mechanics concept?
Is it a bug?

I have attached a simplified model (Acknowledgments and Credits to @EdilsonMJr) that isolates and illustrates the issue:

Motion Constraint Loop simplified model
With the rotation constraint "Rotação:2" (shown in the pic) suppressed (Loop open), we have movements (by mouse or driving "Ânglulo:5"), with "Rotação:2" enabled (Loop closed) movements lock, intuitively "Rotação:2" should be enabled,  and exactly the lack of this produces wrong movements in my assemblies.

Could someone give a good technical explanation and clarify what is happening?

Every help is welcome!

Regards
GTI

[jhackney1972]

I will take a stab at it.  The two Design Center Bevel Gear sets automatically come into Inventor with driving constraints between each of them.  You are tying them together with either Rotation:1 (1.00 ul) or Rotacao:2(1.00 ul).  If you apply both of these rotation constraints you are basically locking the two gear sets internally thus locking your assembly.  If you suppress both of the rotational constrains, and suppress the Angulo:5 angle constrain, you can rotate both gear sets showing their internal rotational constrain connection applied by the Design Center. 

So you really only need either Rotation:1 or Rotacao:2 constrain to tie the two gear sets together but not both.

[Anonymous]

@jhackney1972,

Many thanks for the reply!

I'd like to stab a lot of it because this prevents my assembly from working properly. My assembly needs these two constraints enable together (Rotação:1 and Rotação:2), the system fails on the constraint that I suppress.

What you said is just a true description of what happens, and I've already seen all of thist, but if I understood correctly, that does not explain WHY.

Regards

GTI

[jhackney1972]

I was trying to explain why but either my written explanation or your understanding of it missed the point.  I will try once more by using a screencast.  In the screencast I used an entirely different constraints to hook the two gear sets together eliminating your rotational constraints.  With this I am trying to illustrate the answer to your question.

[Anonymous]

Good job! Thanks again.
It's more likely that I missed the point, it seems that my difficulty is conceptual. It's also possible that I have difficulty explaining the issue well. Please be patient. 

This "Mate:32" you created makes the gears spin in incoherent directions, excuse me if this is just for didadic purposes and curiously a second simultaneous constraint equal to "Mate:32" created between "EixoS" and "EixoI" also locks the system.

I understand that a constraint is sufficient to meshing the two sets of bevel gears, what I do not understand is why I can't have a loop of motion transmission as it happens in practice, why I can't have a second constraint by another power transmission path as it happens in practice.

I'll continue to strive to understand, do not give up.

Regards

GTI

[Anonymous]

As the screencast has the same information as your written explanation, I will comment the writing to try to be more detailed.

---

@jhackney1972escreveu:

... The two Design Center Bevel Gear sets automatically come into Inventor with driving constraints between each of them...

---

Yes, I already knew this and this was considered in this test assembly, so it is because new constraints were not made between the gears of the bevel gear sets. This is a test parameter and not an explanation why the test assembly was locked.

---

@jhackney1972escreveu:

... You are tying them together with either Rotation:1 (1.00 ul) or Rotacao:2(1.00 ul) ...

---

Yes, this is a working condition and not an explanation why the test assembly was locked.

---

@jhackney1972escreveu:

... If you apply both of these rotation constraints you are basically locking the two gear sets internally thus locking your assembly ...

---

Yes, this is a true statement and not an explanation why the bevel gear sets are locked internally.

Applying both there are no redundant constraints.

---

@jhackney1972escreveu:

... If you suppress both of the rotational constrains, and suppress the Angulo:5 angle constrain, you can rotate both gear sets showing their internal rotational constrain connection applied by the Design Center ...

---

Yes, I already knew this and this was considered in this test assembly, so it is because new constraints were not made between the gears of the bevel gear sets. This is a test parameter and not an explanation why the bevel gear sets are locked internally.

---

@jhackney1972escreveu:

So you really only need either Rotation:1 or Rotacao:2 constrain to tie the two gear sets together but not both.

---

Yes, this is a true statement and not an explanation why the bevel gear sets are locked internally.

A Motion Constraint Loop would be (in this case):

Bevel_Gear_1->Movement_Constraint_1->Bevel_Gear_2->Movement_Constraint_2->Bevel_Gear_3->Movement_Constraint_3->Bevel_Gear_4->Movement_Constraint_4->Bevel_Gear_1

They are the same number of pieces and constraints, 4.

I tested a redundant constraint made between the gears of the set and it locks the set.

Regards

GTI

[kelly.young]

Hello @Anonymous thanks for attaching the parts, much easier to assist.

I added a mid-plane that lines up between the two gearsets (Engrenagens) in my Pack & Go .zip, so be aware when you open so you don't use the ones in your library. Added this for better orientation of the Eixo pins.

I went through and deleted all of the constraints. Try to keep assemblies simple by using traditional constraints wherever possible. The only special one I used was Constrain > Motion > Rotation to link the two gear sets. 

There is a suppressed angular constraint that just brings the handle vertically flush with the front face. 

Inspect the assembly and see if that helps you out. 

Please select the Accept as Solution button if a post solves your issue or answers your question.

[Anonymous]

Thank you @kelly.young for your assist!

---

@kelly.youngescreveu:

I added a mid-plane that lines up between the two gearsets (Engrenagens) in my Pack & Go .zip, so be aware when you open so you don't use the ones in your library. Added this for better orientation of the Eixo pins. 

---

That's Added Value! So far your instructions have been very useful, I knew I should have something like the Pack & Go somewhere, but I still had not searched, you saved me time. I tested my Pack & Go and it does not work, it says "There was an error creating the folder" and "An error occurred while creating the file packngo.log, freeze and crash. Back to manual mode.

I saw the plane CENTERPLANE (mid-plane) in the gears and the "Mate:8".

---

@kelly.youngescreveu: 

I went through and deleted all of the constraints. Try to keep assemblies simple by using traditional constraints wherever possible. 

---

Instructions very welcome but it's unclear what are traditional constraints and special one.

---

@kelly.youngescreveu:

The only special one I used was Constrain > Motion > Rotation to link the two gear sets.

---

Here is my question that originated this thread, if you have read and understood the previous comments you will know why.

I know that to link the two gear sets a constrain is enough, so you've just created the constrain Rotation:1 between Engrenagens cônicas:1 and Engrenagens cônicas:2.

My question that originated this thread is:
Why a second constrain, for example, between EixoS and EixoI that could be equivalent to one between Engrenagens cônicas:2 and Engrenagens cônicas:1, lock the system ???

Before you answer that this second constrain would be redundant because a constrain Rotation:2 between Engrenagens cônicas:2 and Engrenagens cônicas:1 is equal to Constrain Rotation:1 between Engrenagens cônicas:1 and Engrenagens cônicas:2, I say, this does not respond, does not explain, just change the question. Why does a redundant constrain lock the system ???

To be more clearer I have divided the two gear sets into four separate gears in the attached assembly KY-TesteLoop-2.zip, here redundancy seems to no longer exist. I've tried to be as faithful as possible with your constrains on this.

We have:

Engrenagem cônica1<>Rotação:1<>Engrenagem cônica2<>Rotação:2<>Engrenagem cônica3<>Rotação:3<>Engrenagem cônica4<>Rotação:4<>Engrenagem cônica1

That is a closed Motion Constraint Loop (The title of this thread).

Why does the Rotação:4 lock drive AlignStrap/Drive and not lock movements by the mouse except on the Manivela? This is new in this assembly, in KY-TesteLoop-2 and TesteLoop locks everything, so I can not predict the results, for this I need clear rules. I feel like I'm almost understanding.

---

@kelly.youngescreveu: 

Inspect the assembly and see if that helps you out.  

---

Yes, it helped a lot, the assembly was left with the simplest and cleanest constraints, the gears teeth aligned, there are many good examples of good work habits to follow, after studying it by this time, now I have more control over the alignments, but this assembly does not resolve this thread. The purpose of this assembly would be just to illustrate the issue of this thread, I did not intend to use it for anything else.

Regards

GTI

[kelly.young]

@Anonymous when I mentioned traditional I meant mate/flush/concentric, just making things inline with each other. Do transitional/joints/motion after you've exhausted the generic locational constraints.

I opened the updated assembly and you separated out the gears to individual parts and constrained.

I tried to follow along, Why does the Rotação:4 lock drive AlignStrap/Drive and not lock movements by the mouse except on the Manivela? 

If you unsuppress AlignStrap/Drive it makes the handle go vertical/flush and locks everything up. You cannot Drive the constraint. If you Suppress Rotação:4 then you can drive it. I think this is because it is over-constrained.

Your rotational constraints are Yellow spins Green, Green spins Orange, Orange spins Blue, Blue Spins Yellow

If A=B, B=C, C=D, Then A=D and no need for that constraint.

When you Drive by angle constraint the rotational constraint is now unnecessary as it is being superseded by the Drive? I don't really know how to explain it, like trying to drive a car with the e-brake on. Once you take it off the freedom can continue.

Maybe you can record a screencast and show what you are trying to achieve. If language is a barrier, just go slow and circle/highlight things with the mouse, many others have done this and is visually helpful.

Please select the Accept as Solution button if a post solves your issue or answers your question.

[Anonymous]

@kelly.young,

I'll try to explain it once more, if it's not enough I'll make a video.

---

@kelly.youngescreveu:

I tried to follow along, Why does the Rotação:4 lock drive AlignStrap/Drive and not lock movements by the mouse except on the Manivela? 

---

I'll detail:

In the KY-TestLoop-2 as it was sent, without changing anything, (AlignStrap/Drive suppressed and Rotação:4 unsuppressed) if you try to drag with the mouse the Manivela (green handle) or Engrenagem cônica1 (Yellow) nothing moves, but if you drag any other moving part with the mouse, there will be movements of all moving parts including the Manivela (green handle) and Engrenagem cônica1 (Yellow) that were locked. If we have motions driven by the mouse, then why does the DRIVE command (which is enabled) on the AlignStrap/Drive suppressed not produce motions ???

---

@kelly.youngescreveu:

If you unsuppress AlignStrap/Drive it makes the handle go vertical/flush and locks everything up. You cannot Drive the constraint. If you Suppress Rotação:4 then you can drive it. I think this is because it is over-constrained.

---

Right. Why can we drive AlignStrap/Drive unsuppressed or suppressed only if Rotação:4 is suppressed?

---

@kelly.youngescreveu:

 Your rotational constraints are Yellow spins Green, Green spins Orange, Orange spins Blue, Blue Spins Yellow

If A=B, B=C, C=D, Then A=D and no need for that constraint.

---

Yes, no need for that constraint, but why declare A=D creates problems? The statement of a truth that does not contradict anything should not create problems.

---

@kelly.youngescreveu:

When you Drive by angle constraint the rotational constraint is now unnecessary as it is being superseded by the Drive?

---

No, here are situations created just to illustrate the issue that the rotational constraint Rotação:4 produces the locking and not the angle constraint AlignStrap/Drive.

---

@kelly.youngescreveu:

I don't really know how to explain it, like trying to drive a car with the e-brake on. Once you take it off the freedom can continue.

---

With AlignStrap/Drive unsuppressed (the e-brake on) we have moves by DRIVE command if the rotational constraint Rotação:4 has been suppressed,  it is an e-brake functioned in reverse.

---

@kelly.youngescreveu: 

Maybe you can record a screencast and show what you are trying to achieve.

---

For me the correct operation would be to have movements by drag with the mouse and by the command DRIVE on AlignStrap/Drive with AlignStrap/Drive suppressed and Rotação:4 unsuppressed.

That's what I'd like to achieve and I want to learn the rules to predict results.

Let me know if you still do not understand, if necessary I will make a video.

Regards

GTI

[Anonymous]

@kelly.young, @jhackney1972

Venturing into Inventor's studies, I saw a possibility to simplify the three constraints Coincidência:8, Coincidência:9, Coincidência:10 of KY-TestLoop-2 in two EQUIVALENT constraints by editing only Coincidência:9, so I can suppress the now unnecessary Coincidência:8 (unnecessary  - My God, does that also cause problems?). See KY-TestLoop-3 attachment.

Result:
Although the constraints of KY-TestLoop-3 are completely equivalent (from my point of view) to KY-TestLoop-2, the two systems now work differently, KY-TestLoop-3 now works more like or maybe equal to TesteLoop and KY-TestLoop attached earlier. That's, now Rotação:4 unsuppressed locks everything (with AlignStrap/Drive suppressed of course), it locks movements by the drag of the mouse and by the command DRIVE on suppressed AlignStrap/Drive other than KY-TesteLoop-2 that allows movements by the drag of the mouse with Rotação:4 unsuppressed and AlignStrap/Drive suppressed.

I'm starting in Inventor and it's hard enough for me to find some logic in these discrepant behaviors.

Frankly, I'm VERY FRUSTRATED. Or some generous soul explains to me the logic that is in these discrepant behaviors or I will continue to think that Inventor does not produce predictable results and is not very reliable at least in these small areas that I am exploring, not to mention the non-conflicting Rotação:4 that no one can explain why It's a Bad Boy, that this is unnecessary I already know, I may have unnecessary parts and this does not cause me problems. This is not a good way to start.

Correction:
In the comment of 04-25-2018 06:15 PM, where it reads:
...This is new in this assembly, in KY-TesteLoop-2 and TesteLoop locks everything, ...
Read:
...This is new in this assembly, in KY-TesteLoop and TesteLoop locks everything, ...

Regards

GTI

[kelly.young]

Hello @Anonymous I can understand your frustration, starting a new software is always difficult. I was looking through the parts and somewhat grasp what you're referring to, yet also confused. I get that you're trying to discern how constraints interact and need to be set up for consistent results. 

Some of what you are trying to describe is better accomplished through Dynamic Simulation. Have you looked into this yet?

The constraints are to locate and assemble parts, but what information are you wanting to extract from Inventor?

I still think a screencast showing the desired behavior would be more clear than trying to verbalize with text and references to part names. Sorry I can't be of more help as I'm not sure what you want as an end result.

Please select the Accept as Solution button if a post solves your issue or answers your question.

[Anonymous]

@kelly.young,

Again, thank you very much, you are an excellent pro.

---

@kelly.youngescreveu:

I get that you're trying to discern how constraints interact and need to be set up for consistent results.

---

Yeah, step by step and now studying movements in the assemblies, I'm studying Inventor with the intention of becoming a person who knows how to use Inventor.

---

@kelly.youngescreveu:

Some of what you are trying to describe is better accomplished through Dynamic Simulation. Have you looked into this yet?

---

A little. How can I move forward if I don't even know how to use simple constraints to get movements in the assemblies?

---

@kelly.youngescreveu:

The constraints are to locate and assemble parts, ...

---

If constraints are not to produce predictable movements then why do constraints allow movements?

---

@kelly.youngescreveu:

... but what information are you wanting to extract from Inventor?

---

So you make me think I'm talking to an artificial intelligence.

I've answered this question several times in previous comments including at the beginning of this comment.

---

@kelly.youngescreveu:

 I still think a screencast showing the desired behavior would be more clear than trying to verbalize with text and references to part names. Sorry I can't be of more help as I'm not sure what you want as an end result.

---

I'm studying Inventor and the issue is not what I want as the end result, but what the end result that Inventor produces and how I can act to make it predictable.

I do not think a screencast will change anything, you will continue to say that you did not understand and I do not believe that people as qualified as you can't understand a issue as simple and small as this.

I think I get it, maybe I'm bumping on a business strategy. I do not think you're going to answer my questions because the answer is not interesting to you.

Regards

GTI

[kelly.young]

Thanks for the compliment @Anonymous though being questioned as an AI is certainly a first, I assure you I'm a real boy, no strings on me! 

I love using screencast it is the easiest way to communicate within Inventor. I've made hundreds of screencasts if you want to hear me ramble and see me mash my esc button. If you are serious about learning the program, moving forward, it will help you easily document questions to get answers more quickly. It's not a requirement just my advice that I use and helps me immensely. 

You're talking about a visual rotational action not functioning consistently or rationally. The behavior could be a limitation but without an example it is difficult for me to visualize, sorry. 

Have you completed all of the built in tutorials from the Home screen within Inventor?

I'll relook at the thread and see if my fog has cleared. I made this guy on my desk, he's the only robot I'm affiliated with:Anycubic Kossel Delta 4.2hrs PLA

[TheCADWhisperer]

---

@Anonymouswrote:

.... how I can act to make it predictable.

  I do not believe that people as qualified as you can't understand a issue as simple and small as this.

I think I understand the problem description - you simply want this mechanism to work like it does in the real world.

To get to that stage - several basic concepts about Inventor (and other MCAD programs) need to be understood.

Inventor (and other MCAD programs) do not by default create true gear tooth forms - instead, to speed up background calculations - simplified geometry is used.  You can model true tooth forms, but since gears are generally purchased off-the-shelf the overhead of true forms is usually not worth the expense in modeling and calculation time.

To get around this and speed up calculations - artificial elements like "Motion Constraints" or Rolling Cone on Cone Joints are used in place of Contact Sets or 3D Contact Joints.  These are mathematical constructs that are theoretically perfect.

But in the real world we cannot manufacture perfect geometry.

If you examine  your simplified gears - you will note interference between the gear teeth.

I can help you set up a more realistic mechanism using Contact Sets, or better yet, 3D Contact Joints, but first you will need to edit the gear teeth to eliminate the interferences and make them more like what we would manufacture in the real world. 

[Anonymous]

My dear friend Cortana, ohh... sorry, my friend @kelly.young, 

---

@kelly.youngescreveu:

I love using screencast it is the easiest way to communicate within Inventor.

---

Okay, I agree, but my screencast, my Pack & Go, my Dynamic Simulation and many others and even I also do not work well and freeze all the time.

THEN, I suspected that Inventor could not be that bad like this and I investigated.

My Inventor is installed in a VM, I know this should not be good for it, but not enough to create all these problems, so I discovered that Windows Defender is interfering, blocking the operation of Inventor (You should solve this with Microsoft) , I turned off the WD, tested the Pack & Go and now it works WELL. Now I'm going to test the others one by one.

---

@kelly.youngescreveu:

You're talking about a visual rotational action not functioning consistently or rationally. The behavior could be a limitation but without an example it is difficult for me to visualize, sorry. 

---

I understand, you're right, but I thought you managed to replicate the issue there on your machine with the models I attached. Could not you reproduce the issue?

---

@kelly.youngescreveu: 

Have you completed all of the built in tutorials from the Home screen within Inventor? 

Not all, all of the "Path of Learning" now I'm practicing.

---

@kelly.youngescreveu:
I'll relook at the thread and see if my fog has cleared.  

---

If you can't reproduce it and if you already told me this I already would have made a video.

---

@kelly.youngescreveu:

... he's the only robot I'm affiliated with:Anycubic Kossel Delta 4.2hrs PLA

---

Is this how you would like to be materialized? You would look great! 

Regards

GTI

[Anonymous]

@TheCADWhisperer,

Many thanks for the reply

---

@TheCADWhispererescreveu:
I think I understand the problem description - you simply want this mechanism to work like it does in the real world.

Yeah, or something like the real world. I avoided using "real world" because I already know that in the assembly the real world is far away.

That's it! 

Excellent that you understood and wrote.

What is difficult to understand is more pleasurable when understood, mental challenges are the stimuli that separate the greats from the smalls.

---

@TheCADWhispererescreveu:

To get to that stage - several basic concepts about Inventor (and other MCAD programs) need to be understood...

---

Brilliant response, fantastic clarifications! This is on the level that engineers consider acceptable and even satisfactory, this brings a little comfort to my frustration.

These yours clarifications should be an initial message in Inventor or in the startup tutorials, so the beginners' expectations would be put into reality.

I had a similar debate with the SW Proteus staff because in the XXI century, in their simulation, the electronic components are "indestructible" (it does not inform when it exceeds its physical limits). The justification is because, due to complexity, this would not reflect reality. We calculate this, but the SW does not, it is difficult to accept that our machines and SW are still very primitive.

---

@TheCADWhispererescreveu:
If you examine  your simplified gears - you will note interference between the gear teeth. 

---

Yes, I had already done that check.

---

@TheCADWhispererescreveu:
I can help you set up a more realistic mechanism using Contact Sets, or better yet, 3D Contact Joints, but first you will need to edit the gear teeth to eliminate the interferences and make them more like what we would manufacture in the real world. 

---

Okay, I'd really like your help for this. Today I'm an alligator with an open mouth, just fling that I swallow. 

I already have it ready. I already have a test model where I eliminated the interferences exactly to test the "Contact Solver" but I could not see it working.

See if this attachment "Teste Contato.zip" serves to set up a more realistic mechanism using Contact Sets ou 3D Contact Joints. The models of the tutorials work mine does not work.

I tried to get a Dynamic Simulation with that attached model KY-TestLoop-3.zip and in DS it was missing parts and I do not know why.

Regards

GTI

[TheCADWhisperer]

Do not move the attached in assembly environment.

Go to Environments>Dynamic Simulation and then click the Play button on the Simulation Player.

In the Dynamic Simulation Environment you can also drag any one of the three gears - but do not drag too fast as you might overwhelm the calculations that Inventor is doing in the background.

Open the Output Grapher to get an idea of the hundreds (thousands?) of parameters that Inventor is calculating in the background.

Motion Parameter Calculations

[TheCADWhisperer]

I forgot to mention that any time you want to make manual movements or changes - be sure to go back to Construction Mode on the Simulation Player.

[TheCADWhisperer]

Here is an example mechanism that my students just finished up.

The balls are slightly different diameters and the rods have a slightly increasing angle.

The balls are sorted by size.

Well, there was supposed to be a screencast - but the forum is still broken.

https://knowledge.autodesk.com/community/screencast/0def0be0-b386-4abf-822e-570b23b6d007