Stress Documentation/Ethical problem 2

[Miner89r]

I've been given the assignment to revise the structural analysis for a large equipment tray installation. The documentation left by the previous engineer was extremely lacking and contained several errors. I told my boss the only way I could do it was to re-do pretty much everything. After analyzing it I found some poorly designed joints and sheet metal components for which I could not show positive margin. When I brought this to the attention of my lead, the answer I got was basically this: "We've already built 15 of these things and a change now would be too expensive, besides my common sense tells me that it will be OK." I suggested fairly simple band-aid type repairs to alleviate the problems but I was shut down there to. Instead I was told to use some very non-conservative analysis methods to show the installation good. Such as considering long, fairly thin beams as infinitely rigid members and constraining lug/clevis joints so that no rotation is allowed about the pin. The installation is not considered safety of flight critical but I have the feeling that it is going to cause some major headaches for someone down the line. My stress lead has 20+ years experience as an aircraft stress analyst and it will ultimately be his signature on the drawings, but my name is still on the analysis report. Does anyone have any suggestions for me or know how I might be able to CYA on this? I want to add that this is a new program and none of these things have flown yet, it doesn't look like any of the installations this group has made in the past so I am reluctant to accept my lead's "common sense says it will be OK" answer.

This might be a little off-topic for this particular forum but I know allot of the guys in this forum are stress engineers like me and that's who I wanted to ask.

 

[ewh]

Keep your resume updated. If they are going to cut corners here, you have to wonder where else they are massaging the numbers.

 

[SWComposites]

There are plenty of stress jobs right now in the industry. Ask your lead to put in writing the analysis methods and assumptions that he insists that be used (and keep a copy for yourself at home). Then refuse to put your name on the stress report, drawings, etc. If your lead wants to sign them off, then its his problem. If they don't like it and still want your name on the report, then update your resume and look for another job.

 

[Kwan]

I think Ewh and SW are on the right track. Talk with the lead about how uncomfortable this is for you. Ask him to write the method assumption letter, and then ask if it is permissible for you to write it up with his signature (not yours). I'd still raise a flag to someone though.

Here is some other options. If you are working on an FAA aircraft that will be FAA approved, talk with the DER about the issue (even if I had a letter and my number is not on the report I would consult with the DER). The DER should take up the fight for you. Also consider calling the ethics number (if you work for a large company). Of course keep the resume up to date because managers lacking ethics will cause large amounts of pain for you (keep a Pearl Harbor file).

I like contracting. I have put my foot down in the past without concern about the outcome. The next job is not far away and typically pays more! Why would I stay with company that serves out stupid pills to the management?

 

[ewh]

Contracting is always an option, especially stress engineering as SWComposites pointed out. I recently saw an ad for a contract stress position on the west coast with a rate of $90/hr+. You don't have to put up with questionable ethics (especially if the finger gets pointed at you) with todays job market.

 

[Beggar]

I'd prepare the report to contain multiple cases. In one case, use your own assumptions and draw whatever conclusions your analysis reveals. In another case, list the alternative assumptions and the relevant conclusions.

In the summary, simply state why you consider the case 1 assumptions more accurate while also giving nod to the fact that they're currently produced without evident problems.

I would certainly ask for the lead to document his desired assumptions in writing.

I would also be looking ...

--------------------
Bring back the HP-15

http://www.hp15c.org

--------------------

 

[EricZhao]

Maybe not as negative as we thought. Aircraft industry doesn't have the luxury to be over conservative. What we get from the FEA model is one thing, how the product performs in real world is another thing.

I would suggest that you describe your analysis and assumptions in more detail, and let the fellows here make some comments.

I also agree with Beggar. Document the two different methods. Incase incidents happen, you have a disclaimer as a backup. It is an over kill to search new positions just because of this.

 

[monkeydog]

Begger has a good solution. I doubt your lead will give you anything in writing. His butt is on the line, because he would have given the nod on the previous stress analysis (could explain why it is lacking and the previous stress guy is gone). You also need to elevate it up the chain. Upper management can then make the call to proceed with risk, or fix the problem (per your recommendations). They know they can pay now, or pay later.

 

[rb1957]

i'd suggest a different tack. i wouldn't write two different reports ('cause its twice the work, and one of them is worthless). i don't know how much you've talked to your supervisor about this but 20+ years expereince is most likely worth something (it could be 1 year, 20 times over). if you go through the assumptions, one by one, let him express his opinion. if you can, challenge him.

if you think this is wrong, then i'd either ...
a) ask for a different assignment,
b) complete the report, as directed, but not put my name to it,
c) annotate the report ("weasle words") so that the reader knows who made which decision (i've seen several reports which state that something was found acceptable "at the behest of the project director")

either way, unless your other experience with your company has been more positive, i'd warm up the resume !

 

[Sparweb]

Don't wave smoke and mirrors in front of people's faces.

There is doubt about the previous analysis, and the method of analysis is now in question. Load test it.

You already have some built, are building more, and the risk implied by failure often outweighs the expense of damaging a rack.


Steven Fahey, CET

 

[737eng]

As a DER, nothing tee's me off more than reading a report or stress analysis with a bunch of smoking mirrors. I tend to lose all faith in an Engineer or customer after receiving one of these reports and will either break our relationship or review every little detail everytime thereafter. Also, if you want a DER to be your mentour in the future, you want to ensure he has faith in your capabilities, otherwise he will never accept this responsibility.

The key is to show compliance with the FAR's, flight critical or not, compliance must be shown. If you can't show compliance then the rack must be retrofitted and you should ensure this gets accomplished. If it comes down to it, passing the buck, or leaving the company, is not a safe or ethical alternative, you know about it, ensure someone else with the power to do something about it, does as well before you leave. However, there is more than one way to skin a cat, ensure your analysis is not over conservative, do not over rotate until you are sure there is a problem. I have seen plenty of instances where over conservatism has caused problems. Check with other stress engineers in your dept to see if they have another less conservative approach that is still valid, and to check your model to ensure there are no errors, etc.... also we are in the age of computers, however, a good analytical hand anaylsys is still valid. Load test is also a good alternative, I have seen many a rack, galley, closet structure etc....that could not be shown good via hand calculations due to materials used, conservative assumptions, no allowable data, etc.... however, a load test proved the structure. Make sure you generate a valid test plan and get it FAA approved, and then if a DER witnesses the test, ensure he is approved to do so.

 

[EVdave13]

After 15 years in this industry, I've only ever worked at two aerospace companies where I was not told point blank to do something unethical (if not downright illegal). If you do not deal with this now you will likely have to deal with a similar situation at the next job. Work up the chain of command first, as others have said. DER consultation is good, too, especially if you get nowhere with management. But I have dealt with DER's that act the same way. Make sure someone who has the power to do something knows about it and that you document telling them. Then and only then should you consider leaving.

 

[berkshire]

I would follow the advice of Spar web and 737 Eng. and load test the part.
You may find that you were too conservative and that the part is ok.
On the other hand if the part fails, you will now know exactly where to beef it up.

A little anecdote here, Some years ago, I was working in the experimental department
Of an aircraft company in Savannah Ga., carrying out side load tests for certification on the rear fuselage of a light twin.
The fuselage /vertical spar junction buckled at half the expected load, causing a major rework of the structure
I made an off the wall comment about building it strong enough the first time.
The head of the engineering dept came back with an unforgettable remark” If you err on the side of lightness you can beef the structure up.
If you make it too heavy the fed will very rarely let you take material back out, this results in a heavy airplane”

 

[RPstress]

Frequently things that are hard to show good are actually ok. If you really can't test one with a very crude support and sandbags or something, then what EVdave13 says probably applies. While theoretically legal responsibility stops with your supervisor at the lowest, remember that when Herald of Free Enterprise went down in the English Channel they prosecuted the Cabin Boy (well, and the Captain).

As a minor aside on berkshire's topic, I remember an old guy I was privileged to work with saying how, when he worked on the Phantom, Jim McDonnell himself told them to size it all to a M.S. of -0.1. They had to test a few bits and beef some up to get it all through, but that was a sound airframe! Jim said, apparently, that the only way to get weight out is not to put it in in the first place. These days of course, we usually get told to initially size stuff to a M.S. of +0.1 minimum...

 

[EHudson]

Can you do a load test on one of the units?