Mechanical Reverse Engineering 18

[designmr]

This is more of an OPINION thread than a question. I am posting it because I always get excellent comments here.

If you were asked to "REVERSE ENGINEER" something, what would that mean to you?

I've dealt with several engineering companies during my work experiece, and this is the first I had to get into "reverse engineering", so looking for opinions.

Thanks again in advance.

 

[GregLocock]

At the simplest level, copy the dimensions and materials of a part or system. That is rather common.

Rather more entertaining is when your prototype exhibits a certain characteristic and your competitor's does not. Then you try to reverse engineer their solution. As an example of that, the accepted wisdom was that to improve isolation in a car you should stiffen the structure, and soften the isolators.

Cursory examination of a competitor's car showed that they could not possibly be doing this, yet their noise performance was pretty good.

So we did an experiment comparing our strategy, their strategy, and a few other things, and learned a great deal.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[MikeHalloran]

I used to work in a shop that made marine exhaust risers, which appear to be a very simple product. They are not really simple at all.

The primary objective of my job was to make our product work better. I learned a lot by examining and testing competitors' products that we were replacing, conducting similar tests on our own products, and doing comparative anatomy.

I kept a huge spreadsheet that carried a rough math model of every part we built, and another spreadsheet with test data from every sea trial we conducted, of ours or anyone else's stuff.

I was continually refining the math model spreadsheet, and it eventually got good enough to accurately predict the performance of our product, so the tool got better, and by making incremental changes and measuring the results, our product got better, and our process got better.



Then I was replaced by an id10t, but that's another story.


Mike Halloran
Pembroke Pines, FL, USA

 

[IRstuff]

Reverse engineering, in its purest form, is about figuring out how something was "engineered," i.e., it's not only knowing the physical characteristics and form, but also why certain things were done the way they were. The latter is the most difficult and often fruitless endeavors, since it's often the case that we wind up throwing up our hands and conceding utter lack of understanding of why something was done.

TTFN
faq731-376

 

[rakuday]

Hi,

If you don’t have drawings of a part, but have a physical component and would like to make improvements, then RE is the best approach.

To quote an example, several years back I had worked on a project for an automobile company. Their engine block had vibration problems. They did not have the drawings or CAD model to conduct the FE analysis as the castings were very old. So we had to digitally scan the physical part and generated new CAD models. Complete FEA was done on the CAD model, the problem was identified and the casting was modified to address the issue.

Hopefully, below links should help.

http://www.npd-solutions.com/reoverview.html

http://en.wikipedia.org/wiki/Reverse_engineering

Thanks

 

[dhengr]

Designmr:
Reverse engineering is pretty common in many engineering endeavors. There isn’t much sense in wasting the time to reinvent the wheel each time you want a round, rolling, weight supporting thingy. Of course, you do have to be careful of patent infringement in the process. In point of fact, much of what we do has been done before, with minor variations. You know, a shaft is a shaft, a simple beam is a simple beam. Many times the effort is to make it better, less expensively, stronger, less fracture critical, etc. And, there are also plenty of times when someone’s idea is out-n-out stolen, with insignificant variations, just to be able to say it’s a new idea, I’m not stealing your idea. And, you and your company have to weigh the ethics of doing that, from case to case.

 

[arunmrao]

Ask the Chinese engineers. They are engaged in doing this every day. Simply said,reproduce the competition's part including dimensions,material properties and packaging.



_____________________________________
"It's better to die standing than live your whole life on the knees" by Peter Mayle in his book A Good Year

 

[MadMango]

I think reverse engineering means different things depending on the industry you're in. I take it as taking something apart, usually a competitors product, figuring out:
how does it work
why does it work
did they meet all the standards and regulations
why did the other guys design it this way

You do learn a lot during the process, and sometimes discover that your way is the best approach, or their way doesn't work with your processes. The results of reverse engineering can go to improve your own design "best practices", to being used as ammunition for your sales force to talk up your products and/or talk down the competitions.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the best use of these Forums?

 

[Cockroach]

I do this on almost a daily basis, to reverse engineer a part. Typically my clients have the need to get around a patent and therefore, need a second solution to a particular design in order to enter the market. My engineering practice is not limited to only getting a competitor's product, taking it apart, copying the components, putting it back together and manufacturing a copy that is equal to the first. It also necessitates having an understanding of patent law, the industrial process and framing the limitations of a particular design.

Not only China reverse engineer. Japan undertakes this same practice in order to miniaturize, improve upon, or forward a particular technology as an improvement. So it is an evolutionary endeavor. I point out that South Africa also did this in order to circumvent international laws on armaments, as did certain European countries during the 30's. There are other examples.

Reverse engineering is the black art. It can be a bad thing, but typically as an academic exercise, a good thing for the economy. Or so I believe.

Regards,
Cockroach

 

[Jboggs]

In my experience, the term "reverse engineer" is something you usually hear from non-engineers who think "all you gotta do" to duplicate some complex mechanism is measure all the parts and "draw it up". Sounds simple, huh? Well you can measure something all day long but that doesn't tell you what the required precision is in the original design. Or the dimensional or geometric tolerances. It can't tell you what the exact material is, or any heat treatment, or the precise finishes, or fits and clearances. But then I guess that stuff isn't really important.

 

[TheTick]

One of the pitfalls in reverse engineering is that you might also reverse engineer a competitor's mistakes useless artifacts.

 

[tomwalz]

Down to the chcmical analysis of the material and the production process that produced the material. All to answer the question "How did they do that?"

Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

Good engineering starts with a Grainger Catalog.

 

[designmr]

Always great replies, thank all.

 

[IRstuff]

I think some confuse "reverse engineering" with "copying." The Chinese products are currently "copying," wherein they duplicate a product, warts and all. The Japanese "reverse engineered" back in the 60s and 70s; they then built upon that foundation and improved and enhanced their products. The Soviets "reverse engineered" Intel 8080 processor in their MiG-25s, wherein they had to design a substitute microcode memory because their semiconductor processing couldn't manage process the Intel microcode, as-is.

It's the "why did he do that," and "why he didn't do this," that spells the difference between reverse engineering and copying.

TTFN
faq731-376

 

[EngineerErrant]

One of the pitfalls in reverse engineering is that you might also reverse engineer a competitor's mistakes useless artifacts.

Truth, and it's hilarious when it happens. Another name for this phenomenon: "ham-butt problem," after an amusing anecdote:

A mother is preparing a ham for a family dinner, and cuts off the butt of the ham as she always does. Before she throws it out, she realizes she doesn't actually know the purpose of removing the ham butt; she just learned it from her mother. She phones her mother to ask why. "Why, I never really thought about it. My mother always did it..." She then calls the grandmother and asks, and is answered with a laugh. "Well, dear, I had such a tiny pan..."​

Also a danger: missing crucial details. Torakusu Yamaha brilliantly copied the components of an organ and produced his own "working" model...with horrific intonation. It wasn't until he learned music theory and tuning that he was able to start producing quality instruments.

"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." -Scott Adams

 

[limey1]

Seems to be several angles on this. My take is take a prototype, or a rev 0 print. Cobble, raunch, hack, adjust, challenge, then once you have it just about as good as you can get it (Cost / quality / time). Then make it and have the drawing made up to reflect the final part.

 

[ajack1]

To answer the original question my understanding of reverse engineering is you take a product and produce a design from it rather than taking a design and producing a product from it.

 

[TheTick]

...minus any patented features you may be infringing upon.

 

[ajack1]

I am not sure the definition of reverse engineering, or my understanding of it, has anything to do with it is legal or not.

 

[IRstuff]

Speaking of duplicating worts; I worked at a microprocessor company, and we worked on a prototype processor. A bunch of the original designers left the company before we completed the debugging; so their CMOS version of the processor came out about 6 months after we completed the debug, and the exact same signal was inverted on both chips. The only sad part about the whole affair is that it was a military standard instruction set architecture, and was OBE by processors like the 80386.

TTFN
faq731-376