reverse engineering question

[tj2057]

I have an assembly that I am having a problem trying to constrain a couple parts in.

These assemblies get wrecked a lot and it is our job to repair them.

My customer has bought a new assembly and asked me to draw it up, so that when we receive the wrecked units, we have some dimensions to go by.

The new assembly has been drilled out for clearance in a couple areas, looks like the mfg. used a hand drill to get the thing to work.

I have modeled the part as we received it, but if I constrain one of the holes to another hole where the pivot bolt goes, it will not let a part of the assembly rotate like it is supposed to.

There is something wrong in between the geometry of the holes in relation to each other that I cannot seem to figure out.

Any help would be appreciated.

I am going to post the assy in the customer files section

I have used pack and go and it is about 40mb in size.

[tj2057]

I cannot post or even create a message in the customer file section. The "new message" tab is greyed out.

[graemev]

Without seeing the file(s) all I can suggest is to reduce the number and types of constraints to a bare minimum. Also check that the moving piece or assembly is not grounded.  Many a novice will constrain assemblies the same way that the physical parts are assembled. For instance, a loader bucket that pivots on pins either side.  If the geometryfor those pins is not exactly concentric and coaxial, the bucket will not rotate.  An insert done on only one side, however, should suffice.  (Your exact circumstances may vary and for a variety of reasons.)  Sometimes there needs to be a constraint on the other pin for other reasons, in which case a mate between an axis and a center point may be more effective than mating axis to axis or using an insert constraint.

If your model has compounding linkages (A links to B links to C, etc.) try suppressing one constraint at a time until the assembly moves as needed.  It may not cure the problem, but may help in diagnosing the conflict.

[JDMather]

---

@tj2057 wrote:

I have used pack and go and it is about 40mb in size.

---

Open each part file.
Drag the red End of Part marker at the bottom of the browser to the top of the feature tree hidding all features.
Save each part file with the EOP in a rolled up state.
Right click on the filename (or folder containing the files) and select Send to compressed (Zipped) Folder (if you do the entire folder make sure there is no extraneous stuff like OldVersions, styles.....

You must get below 1.5 Meg the entire zipped folder or individual zipped parts to attach here.
If you can't get it that small the limit is 10Meg over here http://www.augi.com

[tj2057]

Thanks for the response Mr Mather.

I know where the conflict is, I'm just not sure on how to resolve it.

There are multiple linkages that have to work together and I just don't have the layout of the holes correct.

I will try to shrink it up to upload the files.

[mcgyvr]

---

@tj2057 wrote:

Thanks for the response Mr Mather.

I know where the conflict is, I'm just not sure on how to resolve it.

There are multiple linkages that have to work together and I just don't have the layout of the holes correct.

I will try to shrink it up to upload the files.

---

So fire up the measure tool. Figure out whats off and adjust the models.

[tj2057]

I have attached a mock up.

As I stated, it looks like the mfg. has used a hand drill to open up the holes to make this work.

I am not sure how to go about getting the geometry of the holes correct.

I modeled everyhting as close to the sample as possible.

[tj2057]

attached pdf of the project

[JDMather]

You have wayyyyyyyyyyyyyyyyyyyyyy too many dimensions and no logical use of symmetry about the origin for the part sketches.  This makes it verrrrrrrrrrrrrrrrrrry hard to find problems and fix.

Do Nottttttttttttttt use autodimension tool.

Dimension these parts like the machinist out on the shop floor would expect - classical orthogonal dimensioning as done on documentation drawings.

While I take a look at this thing - you might read this document http://home.pct.edu/~jmather/SkillsUSA%20University.pdf

[tj2057]

I understand. I did use the auto dimension. I created sketches from the main assy and copied them to create the mock up. The base parts are drawn up like a machinist would make them. I can go through and do it over for you if you would like. I do appreciate your help.

Mike

[JDMather]

No problem - I will simply convert to Inventor Blocks.

One of the holes needs to be moved - they will never line up.

Is the location set in a manufacturing drawing?
Can you revise the drawing (have authorization)?

The easiest way to determine the correct location is to project from one part to the other.

[tj2057]

Yes,

We have complete freedom to change whatever is required. We are not the mfg. of this part, we are just asked to repair them when they get damaged. This part is a tilting step ramp used to gain access to the lid on the top of rail tank cars. My customer regularly fails to raise the ramp before moving the train. These units get really wrecked and we have to try and salvage them with some consistency between units.

[JDMather]

---

@tj2057 wrote:

 My customer regularly fails to raise the ramp before moving the train.

---

Sounds like a business plan - for you! 

[pcunningham1]

That looks like a company called Carbis. I bet your customer could get 3d cad files from them.

[tj2057]

You are correct. It is a Carbis unit. I am not familiar with that company, but most won't release cad data. I imagine that's why my customer asks us to repair them...I am pretty sure they would suggest to buy new units after seeing a wrecked one.

[JDMather]

I think for this to work the 3 lines and 2 lines have to be and remain parallel during motion.

Right now they are off very slightly.

The holes will have to be moved to keep these lines parallel.

I'll try to work up a solution if I get a chance.

What I'm not sure about at this point is if the crossbar line holes have to fall on the two "vertical" lines for all three "horizontal" lines to remain parallel. (once the holes are moved so that they are in fact parallel - which the aren't at present)

[tj2057]

I created a sketch like you did, however, I didn't check the angles like you have shown. That's a good point. I agree, they should be parallel. On my sketch I drew circles at the center points of the pivots at the diameter of the hole spacing. The handle hole spacing and the cross bar hole spacings are 24". The hole spacing on the lower plate is 23 7/8".

I have actually measured those dimensions several times on the actual parts.

They are using 3/8" bolts in 7/16" holes, and have opened some of them up with a hand drill. So it will work, it's just not correct.

[JDMather]

There will be pleny of clearance in actual parts that it will work - but for Inventor constraints to work the hole spacing must be exact.  I hope to have a revised model later today.  This one actually is a good teaching aid for use of Blocks and reverse engineering mechanisms where precise information isn't known and must be figured out from the geometry information that is roughly known (to the best of our measuring ability).

[JDMather]

This simple block example has me questioning the parallel theory.
If the links truely are different lengths - I have a bit more work to figure this one out.

[JDMather]

OK,
I know it can be done now - just need to set up the correct hole locations.

I got a solution figured out - I will post it after I let my students chew on this one a bit.