Making drawing and taking measurements 3

[Mech5656]

Hello, Engineer.

I am currently working on creating a gasket drawing for a heat exchanger cover and need some guidance on obtaining precise measurements. I have attached a picture of the cover, showing the shape and features required for the gasket. My question is: What is the best and most accurate method to measure the partitions and zigzag areas to accurately replicate them in AutoCAD for the gasket drawing?

 

[tygerdawg]

Define "accurate" and what is your required tolerance.
Is it really necessary to produce a gasket with exact same outline as the narrow zig-zag webs? Seems that would not be stable.

A simple method I have used to recreate such items with reasonable accuracy is:
[ul]
[li]take a well-positioned photo image (that may even include a ruler or other calibration object)[/li]
[li]insert the image into your CAD modeler[/li]
[li]scale the image to correct size[/li]
[li]use CAD drawing tools trace over the required edges and develop a CAD sketch as needed[/li]
[/ul]
Else for extreme precision you could have it measured with a CMM.

I recently saw a neat trick on YouTube where an engine rebuilder made a gasket by overlaying a sheet of gasket material onto the mounting flange and then used a small ball peen hammer with light strokes to hammer out the shape of a workable gasket. Clever low-precision idea.

TygerDawg
Blue Technik LLC
Manufacturing Engineering Consulting

http://www.bluetechnik.com

 

[3DDave]

For setting up to take a photo - get the camera as far from the item as possible to eliminate perspective and other distortions common in lower cost cameras. Cameras with lenses that are in the sub-$10,000 range are low cost for this purpose. Take a photo of the object and then take a photo of graph paper at the same distance and orientation as the object surface to check for pincushion and barrel distortion. These distortions can be removed/corrected with a photo editor.

Alternatively, if the item can be placed on a flat-bed scanner, that can do a very good job for reasonably flat items.

 

[JStephen]

I'd want to see what the existing gasket looked like, to be honest.
If the gasket zigzagged to match the metal, it might be a pain to get it aligned when assembling.
Could that area just be a rectangular strip as wide as the overall zigzag?
Alternatively, could you just set the part on a sheet of gasket and trace around it? Or set paper on it and trace the edges?

 

[mfgenggear]

RTV is a great invention. I used the ball peen hammer method it works.
When 100 or 1000 gasket have to be made steel rule die. Be fore that talk to the techs who have to assemble it. Nothing worst than slinky gasket that nearly impossible to line up, only salvation is add form a gasket or grease to hold it in place.

 

[fel3]

3DDave…

I think you are far too pessimistic about the cost of suitable lenses for this purpose. Macro lenses are very highly corrected optically to produce a flat field and they exhibit very little to no pincushion or barrel distortion (the many distortion charts that I have seen in online review tests for macro lenses from several manufacturers bear this out). Modern macro lenses exhibit this behavior from closest focus to infinity, so they are suitable for more than just macro/close-up photography. Most macro lenses are also very affordable.

For example, I have three Olympus/OM System macro lenses for my Micro Four-Thirds kit: 30-mm f/3.5 (currently on sale at Amazon for $200), 60-mm f/2.8 ($400 sale price....I recently bought one for my oldest granddaughter's upcoming 16th birthday), and 90-mm f/3.5 ($1,300 sale price). The "full-frame" equivalent focal lengths for the same field of view are 60-mm, 120-mm, and 180-mm, respectively. If I was to do the OP's shoot, I would use my 30-mm because it has the widest field of view, which makes it easier to cover a large object like this gasket from a reasonable distance.

In addition, the mirrorless interchangeable lens cameras that I am familiar with all perform geometric distortion corrections in camera for their brand of lenses, at least on JPEG files. For example, a few months ago I bought an Olympus 12-mm f/2.0 lens for doing star trails and wide-field tracked exposures of the night sky. I knew that this lens had quite a bit of barrel distortion, but I wasn't worried about it, especially for my main usage. The barrel distortion clearly shows up the RAW files, but the JPEG files are fully corrected to virtually eliminate distortion.

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[hydtools]

Use a rectangular piece of sufficient length and width to cover the big zag.

Ted

 

[DAVIDSTECKER]

May be print out a big piece of Graph paper, place the cover on it and trace it.
It's a lot easier to measure it when it's all on the same plane and solid surface.

 

[Brian Malone]

Go with DAVIDSTECKER's and JStephen's suggestion: use paper to get a relief image on a flat media and take your detailed measurements. Choose appropriate holes that you can measure their diameter and center-to-center dimension(s) on the actual part to compare to your paper pattern to verify you laid the paper in a stable position and it did not shift significantly while you were doing the geometry transfer/trace work.

 

[mfgenggear]

Well that is basically doing the ball hammer trick. Really tho my opinion the simplest method will be to take the cover and take measurements with coordinate measurement machine. Inspect and record. With those measurement create a steel rule die and stamp it out. What ever is the final configuration. regardless if it is like a wet noodle. Mechanics have been dealing with installation for decades. I either glue one side or adhere it to with grease if allowed or compatible sealant.

 

[durablack2]

I would trace the edges on paper, make the model, then print out a drawing 1:1 to confirm your model is the same. This will work to a certain level of accuracy. Looks like this is extruded material then sliced to width. The the mesh section at the top with the smaller holes is welded in?

Mechanical Engineer