Ethical question regarding tolerances on test plan 4

[mdgates]

I'm new to engineering, so I'm looking for the opinions of some more experienced engineers regarding the following ethical concern, which I will phrase as a hypothetical.

Suppose there's an aerospace company that's trying to get a product into the market. There's a specification that defines the required performance, and a test plan that's been approved by the would-be customer as a means of demonstrating compliance with the spec. The spec requires the product perform at a temperature of 100°C, so the test plan calls for the unit to be placed in an oven at 100+/-5°C for three hours (+/- 5 minutes), then removed from the oven and operated within five minutes.

The engineering team discovers that the product will not function satisfactorily at 100°C. Engineer A suggests setting the oven to 96°C, and lowering the temperature of the test stand room to improve the product's chances of passing the test. Engineer A further suggests leaving the unit in the oven for exactly 2 hours and 55 minutes. He further suggests building a "quiescent oven" to use instead of the standard convection oven that the company has always used to test similar products.

The question is: Is Engineer A's proposal ethical?

It will be Engineer B's responsibility to execute the test. If Engineer B's boss says that Engineer A's suggestions should be followed, what should Engineer B's response be?

 

[KENAT]

Oh, you agree with MJ but no one said anything when I brought that issue up originally.:-(

I don't think the OP ever really answered the question of if the customer approved this test plan or not.

Without having seen the exact wording of the specification the 3 of us are assuming that the 100°C is really a minimum, which seems likely/reasonable, however we don't know for sure.

Certainly if this is right it seems unlikely the customer explicitly approved the test plan, in which case why try to fudge it?

Posting guidelines faq731-376

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

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[berkshire]

mdgates (Mechanical)
Do you mind telling me the name of your aerospace company, so that I can avoid buying whatever it is you are selling.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[Twoballcane]

I do a lot of testing and I go by the rule of “when in doubt, ask the customer”. You want to be as transparent as possible. Surprises like this down the road will kill your trust and career with anybody. If the customer is really hot for this product, they will either lower the spec or give you a waiver.

Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

 

[Twoballcane]

To add, for thermal testing, the chamber will never be at a constant 100 degC. The chamber will fluctuate by +/- 5 degC depending how big or small it is. The larger the chamber the more the chamber will fluctuate. So even if you lower the setting, your tolerance will go out of bounds of the spec at the low end. Now it depends who is singing it off on the customer side. If s/he was not fed, did not have their coffee yet, and in a grumpy mood, this would be a fail. If s/he has some sensibility, it would pass. For me, if I see a gross fail, maybe more than 5% drop, I would fail it.

Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

 

[Twoballcane]

I'm sorry, more than 10% I would fail it. Also, some times the spec you get is higher than the original spec the customer was suppose to meet just so the customer will have margin.

Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

 

[IRstuff]

On the converse side, there can be pathological air flows within the chamber that can result in absurdly high local ambient temperatures. We once attempted to test about 200 parts in the thermal chamber, where each part dissipated about 5W each. That was full 1 kW, and because the testing boards were tightly spaced, we later estimated that while the programmed temperature was 125ºC, the actual part ambient was closer to 170ºC, which was well past the maximum allowable temperature. Needless to say, we had substantial life failures. Reducing the number of boards, improving the air flow, etc., resulted in the expected life test performance.

TTFN
faq731-376

 

[hendersdc]

The spec can easily be 100C +/- 5C, it doesn't need to be a minimum.

The test procedure should state something to the effect of "set temperature to 100C" not "set temperature to 100C +/- 5C". The customer should have read and approved the procedure. If you are changing it you need to get their approval again.

Where did you determine that your product would fail the test? Was it during the formal testing? If so then that is a failure already, you don't get to keep trying until you pass. You and your customer would need to get together and determine what to do about the failed test, re-design or spec change.

I assume your customer will also be reviewing and approving the test results, what happens when they see the output of the temp chamber droping well below 95C?

Doug

 

[mdgates]

Thanks for all your feedback. It's clear that in this case, the spec requires that the product function at 100°C, and we know that it's wrong (and generally bad business) to propose to ship product that doesn't meet the spec.

If I were in Engineer B's shoes, I'd stick to the nominal, i.e. how the test plan should have been written, and suggest that the customer be contacted about a waiver. I'd explain this to Engineer A and the boss first, and appeal to my QE and Head of Engineering if needed. My name wouldn't go on any document suggesting that A's test plan demonstrates compliance with the specs. However, I suppose I would conduct any testing I was ordered to, even if I thought they were going to use my test data for evil instead of good.

 

[GregLocock]

Incidentally, does the test bear much resemblance to real operating conditions, or is it a high stress short duration attempt to weed out weaknesses?

Cheers

Greg Locock


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

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

 

[patprimmer]

Just a note

When your boss wants you to do something illegal or unethical, it normally ends up bad for you no matter what.

Your best chance of minimising personal consequences is to play real dumb and claim you don't exactly understand what he wants you to do.

To push the point you can even say (in writing) that you must be misunderstanding his direction as from your interpretation, it seems he is asking you to do something dishonest, so could he please give you written instructions so that you can study them and ensure you are not misinterpreting them.

Keep copies of request and reply at home. Make diary notes of verbal exchanges on the subject.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

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

for site rules

 

[JohnRBaker]

Keep copies of request and reply at home. Make diary notes of verbal exchanges on the subject.

Where I used to work, it was called "keeping a Pearl Harbor file"...

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[JStephen]

It occurs to me that you are creating a situation where your product will pass when you test it, but if any other lab ever tests it, or the customer tests it, it will fail everytime, which is not a good situation.

I would also think that as you get close to the pass/fail borderline, that there wouldn't be a sharp division, you'd just be changing the percentage pass/fail. So at 100 degrees, maybe 20% of them pass, and at 96, 80% of them pass. Either way, it's not an indicator of good reliability.

 

[IRstuff]

Warranty cost is embedded in the sell price, so you really don't want to have more than what is allocated for warranty costs, which should be less than 1.8% of sell price (

http://www.entigo.com/pdf/Cost-of-Warranty-White-Paper.pdf).

The good news is that not every user will ever hit the max operating temp condition, so if those users are only 1:10, then even a 15% failure rate would only incur about 1.5% of sell price. Nonetheless, if that 1:10 happens to be a single, large user, then you're going to get a gigantic reputation hit and a potential loss of a large customer for future sales.


TTFN
faq731-376

 

[GregLocock]

JStephen has it.

If it is an accelerated life test then you don't really know whether knocking 5 dgerees off the test temperature, and 5 minutes off the oven time, equates to a reduction of X% in real life performance, without performing additional tests. The relationship between real life and accelerated life testing is measured and expensively known, for many systems, but not in the open literature.

Cheers

Greg Locock


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

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

 

[moon161]

I have passed and failed integrity tests, and perhaps been graded on the curve as well. Power, authority, profit motives, wishful thinking, the list goes on. When you think a lot at work, integrity issues may get bypassed or ignored because so little thought is frequently required when it comes to right and wrong. Certainly very much less than self deception, just google 'Sandusky'.

I'd ask a couple 5 year olds. You have a pair (of something) and a backbone. Start your career by using as required, people will take notice.

 

[Twoballcane]

To really solve this ethical problem is to have a third party test the product per the customer spec (

http://www.dtbtest.com/).

They are not bias and will run the test by the spec. What you do in house can be pre-qualification for confidence. Once you feel good, send it out for formal qualification.

Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

Perception is reality: You may build your skill set, but what you think of it doesn't really count. The only thing that really matters is what your boss and or customers think of it. Likewise, the real you is how others perceive you, not

 

[MintJulep]

To really solve this ethical problem is to have a third party test the product per the customer spec. They are not bias and will run the test by the spec.

No.

Test labs follow test procedures.

If the test procedure is ambiguous or poorly written they will do their best, or seek clarification from whoever is paying the bill. If that clarification comes from an unethical or knowledgeable source then it's not clear what will happen.

 

[Twoballcane]

Well that is my point is that they will seek clarification from both parties particularly the one who wrote the spec (most likely the customer). The test engineer has no interest if the product passes or fails. Their job is to produce the environmental spec requirement. If it says 100 degC +/- 5 degC, that is what test engineer will dial in, no fuddy duddy or second guessing. There business relies on their accuracy to meet what’s on paper and not what the designer wants. Now shock is best effort and will depend on experience of the test engineer. Depending on the magnitude and duration (and sometimes velocity change if you are dealing with the Navy) the test will be different with every blow. Thermal and vibration can be programmed in per the spec. Random, depending on the table and fixture, can be challenging to meet the spec, but is programmable.

Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

Perception is reality: You may build your skill set, but what you think of it doesn't really count. The only thing that really matters is what your boss and or customers think of it. Likewise, the real you is how others perceive you, not

 

[Twoballcane]

I’m sorry, I meant test procured, not spec in my last two post…

Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

Perception is reality: You may build your skill set, but what you think of it doesn't really count. The only thing that really matters is what your boss and or customers think of it. Likewise, the real you is how others perceive you, not

 

[MintJulep]

The test lab has no contractual relationship with the end user.

If there is an end user representative present to witness the testing then usually everyone will defer to his or her decrees. But there is not always an end user representative present, and even when present they may not be the right person to notice that the test procedure is not compliant with "the spec". Often the end user representative is an inspector or junior engineer and all they'll look for is "approval" of the test procedure.