Engineers, Quality Control, and thinking people - a dilemma

[Specop 007]

Originally posted by: hypn0tik
Mind if I ask what you do for a living?

You're obviously not an Engineer based on your comments, so you probably believe that all the numbers you see on products you buy don't have any tolerances associated with them.

Fortunately for you though, I'm in grad. school (just started my M.A.Sc in EE). You don't have to worry about doing business that employs me for a couple years.

Surely.

Currently I do...Oh yeah..Engineering! And I've done QC in the past.

Gee, aint that ironic.

Let me point out what the problem is. Your trying to argue numbers which for a college guy is perfectly acceptable.
Now look at the business side.
The customer gave exact specs on what he wanted. -1.1 volts, +/- .1 volt. The product is out of spec. Period. End of story. You can play the numbers game all day long, but let me ask you, what do you think will happen when the customer gets the part and tests it prior to usage?
Will the customer accept the part? Business isnt just about crunching numbers, its also about your customers. This product is, quite simply, not what the customer ordered.
Yes, its close enough it may very well work. Or, it could work enough of the time that they in turn pass it on.
But what if its just enough, juuuust enough out of spec to be a bit flaky. What if its an airbag control unit, and just what if the out of spec unit fails? Whos ass is on the line for the soccer mom who slammed head first into the windshield? YOURS. Because YOU shipped an out of spec component out the door.

Dont think it matters? Remember that shuttle that went up years ago, and didnt come back down?
Our parts were spec. Period.
Know those antilock brakes in your car?
Our parts are spec. Period.
Know those heart monitors in hospitals?
Our parts are spec. Period.
Laser guided missles?
Are parts are spec. Period.

Yes, most likely this part could ship without a hitch. But do you want to risk a pissed off customer on the phone demanding you eat the cost and give him a replacement?
Do you want to risk a customer who tests it, sees it will still work but questions your products and never buys from you again?

Engineering is more then JUST hard numbers. Technically, you may very well be correct. Hell, maybe the poor schmuck in charge of ordering the component doesnt know how to properly submit specs for a quote. Who knows? Now, role play and pretend your the customer. How would you feel if you ordered -1.1V +/- .1V and got .87, or .95 or .99? Now, would you pass that kind of part on to your customer?

But do you want to risk it? Do you want to risk a lost customer, a mad customer or worse?

**EDIT**

And just so we're clear, technically from a numbers point you ARE right. From a strictly numbers standpoint I do agree with you.
But, having been "you" (saying its good), having been the QC guy saying it isnt and having been the guy getting his ass chewed by the customer I would fail it. I'm just hoping, with all due respect, that in the future your open minded enough to think beyond the simple numbers and think of the customers expectations and the possible implications of having a part that doesnt meet the customers intended specs and fails, even if the customer didnt properly state the exact specs. So please dont think of me as trying to be all high and mighty and "put you in your place" as thats not my intentions at all.

Good luck with your education and career.

 

[torpid]

Originally posted by: hypn0tik
But is it company policy?

Standard engineering practice is to express your results with the same precision as the problem statement. Pretty much any person in science will tell you that.

I think you and I might be posting based on differing definitions here. I was merely suggesting that it is common to measure in 10x precision, not necessarily to deliver a product down to 10x precision. That is because you want to have a clear idea of the "fringes". If it is -0.97 and does not vary under any circumstances, the "results" would still be that it falls within spec. But it still is quite useful to measure down to the 10x precision. If you measured at 1x you might see it be a constant -1.0v even if in reality it's wavering heavily between -0.950 and -1.049.

 

[hypn0tik]

Originally posted by: Specop 007

Originally posted by: hypn0tik
Mind if I ask what you do for a living?

You're obviously not an Engineer based on your comments, so you probably believe that all the numbers you see on products you buy don't have any tolerances associated with them.

Fortunately for you though, I'm in grad. school (just started my M.A.Sc in EE). You don't have to worry about doing business that employs me for a couple years.

Surely.

Currently I do...Oh yeah..Engineering! And I've done QC in the past.

Gee, aint that ironic.

Let me point out what the problem is. Your trying to argue numbers which for a college guy is perfectly acceptable.
Now look at the business side.
The customer gave exact specs on what he wanted. -1.1 volts, +/- .1 volt. The product is out of spec. Period. End of story. You can play the numbers game all day long, but let me ask you, what do you think will happen when the customer gets the part and tests it prior to usage?

Will the customer accept the part? Business isnt just about crunching numbers, its also about your customers. This product is, quite simply, not what the customer ordered.
Yes, its close enough it may very well work. Or, it could work enough of the time that they in turn pass it on.
But what if its just enough, juuuust enough out of spec to be a bit flaky. What if its an airbag control unit, and just what if the out of spec unit fails? Whos ass is on the line for the soccer mom who slammed head first into the windshield? YOURS. Because YOU shipped an out of spec component out the door.


Dont think it matters? Remember that shuttle that went up years ago, and didnt come back down?
Our parts were spec. Period.
Know those antilock brakes in your car?
Our parts are spec. Period.
Know those heart monitors in hospitals?
Our parts are spec. Period.
Laser guided missles?
Are parts are spec. Period.

Yes, most likely this part could ship without a hitch. But do you want to risk a pissed off customer on the phone demanding you eat the cost and give him a replacement?
Do you want to risk a customer who tests it, sees it will still work but questions your products and never buys from you again?

Engineering is more then JUST hard numbers. Technically, you may very well be correct. Hell, maybe the poor schmuck in charge of ordering the component doesnt know how to properly submit specs for a quote. Who knows? Now, role play and pretend your the customer. How would you feel if you ordered -1.1V +/- .1V and got .87, or .95 or .99? Now, would you pass that kind of part on to your customer?

But do you want to risk it? Do you want to risk a lost customer, a mad customer or worse?

**EDIT**

And just so we're clear, technically from a numbers point you ARE right. From a strictly numbers standpoint I do agree with you.
But, having been "you" (saying its good), having been the QC guy saying it isnt and having been the guy getting his ass chewed by the customer I would fail it. I'm just hoping, with all due respect, that in the future your open minded enough to think beyond the simple numbers and think of the customers expectations and the possible implications of having a part that doesnt meet the customers intended specs and fails, even if the customer didnt properly state the exact specs. So please dont think of me as trying to be all high and mighty and "put you in your place" as thats not my intentions at all.

Good luck with your education and career.

I agree with you entirely that it is about the customer. In fact, my initial thought was "Why not simply ask the customer and save yourself the argument?". However, I assumed that they decided not to call and ask (for whatever reason), and hence this discussion. In fact, this is probably the worst thing they can do. If they call, they are admitting that the product is slightly out of spec. If they don't call, they are trying to pass off a product that is on the border. Both cases don't bode too well for them, which is why they might not be contacting the customer.

Now, I obviously don't have the experience that you do so my views are simply based on my education and what I was taught. I may very well be wrong, but I merely posted my opinion since that's what the OP was asking for.

I don't know the full details, but if someone was to blame, it would be the engineers for designing for exactly -1.0V. As I stated earlier, I think they should have designed for something lower than -1.0V. However, at this stage, that doesn't matter.

Edit: Regardless, I think this is a pretty interesting discussion and it actually has given me some insight to the QC perspective. Thanks for your kind words in your edit.

 

[Beller0ph1]

Why don't you ask the customer?

 

[TuxDave]

If they use the sig-fig argument and the customer's specifications you'll get.

-0.95-->-1.04 OK for -1.0V
-1.05-->-1.14 OK for -1.1V
-1.15-->-1.24 OK for -1.2V

So essentially you end up with a product who's output voltage can be anywhere from:
-0.95V to -1.24V.

If that was the case, I'm curious why the customer didn't just give a single voltage with a tolerance value.

So yeah, I say that in the end, the customer probably unintentially gave the wrong impression of his spec for his product and what you have right now isn't what he wants.

 

[mrrman]

As a quality person myself, it fails unless the designs allows for a +/- tolerance which is determined by the end user /MFG of the product

 

[Justin218]

I'd say it is most likely okay, but just get a number and a tolerance in the future

 

[RadiclDreamer]

Depends, if the customer truely needs it to be EXACT, then it fails. If they can do with a very small difference then its a pass. But based on the info

F

 

[Specop 007]

Originally posted by: hypn0tik

I agree with you entirely that it is about the customer. In fact, my initial thought was "Why not simply ask the customer and save yourself the argument?". However, I assumed that they decided not to call and ask (for whatever reason), and hence this discussion. In fact, this is probably the worst thing they can do. If they call, they are admitting that the product is slightly out of spec. If they don't call, they are trying to pass off a product that is on the border. Both cases don't bode too well for them, which is why they might not be contacting the customer.

Now, I obviously don't have the experience that you do so my views are simply based on my education and what I was taught. I may very well be wrong, but I merely posted my opinion since that's what the OP was asking for.

I don't know the full details, but if someone was to blame, it would be the engineers for designing for exactly -1.0V. As I stated earlier, I think they should have designed for something lower than -1.0V. However, at this stage, that doesn't matter.

Edit: Regardless, I think this is a pretty interesting discussion and it actually has given me some insight to the QC perspective. Thanks for your kind words in your edit.

I agree, the best thing is to call the customer! If the part needs to be reworked, it may cause a delay. The customer is in the best position to know if the current product will properly meet the design goals they have in mind or if it needs to be redone. For all we know the part is for a TV remote and it'll work between -.5V and -3V, or it could be for a mission critical part / life saving part and it simply has to be within the -1.1 +/- .1 volt without question and the end customer will take a delay rather then risk a potentially flawed part.

But trying to second guess what the customer needs and send a part that doesnt meet whats on paper is just asking for unneeded trouble. Generally, applications with mission critical / life saving components are properly expressed on the quote sheet, but again personally I would not want to try to guess the application and guess if the part is good.

Like I said though, if I could give one piece of advice and have it stick its to keep in mind the part is for a customer. You need to have his expectations in mind when testing the final product, and keep in mind the ramifications of shipping a product that may not meet his expectations even if he perhaps didnt properly submit the specifications on the RFP.

 

[JRich]

What's the customer spec? Over here we run by our control limits which are way tighter than the customers limits. If the part is out of our controls, but it's within customer spec we send it on.

 

[oznerol]

It is not for something basic like a TV remote - it is actually a military device.

It is standard QC process to measure 10x the required precision. It is also company policy to do so, I believe that is actually the ISO standard.

The -1.0V is actually the low-end of the customer required tolerance. However, we previously inquired to the customer and our own ENG department agreed that -1.0V was adequate.

It is essentially a question of whether or not -0.97 = -1.0.

I am not involved in this project, so I don't know exactly why the customer is not contacted. However, I do know that the deadline is the end of this week and a lot of money is involved to get it shipped out on time, and contacting the customer and getting a certified response can take a while - especially if an engineering change notice is required to documentation - so that is most likely the reason.

My opinion, and the opinion of the ENG team here, is that it is passable, as the customer lists precision to one tenth, as listed in the spec. The big issue is not a customer problem, but rather a company problem - as we obviously do not have a standard procedure for rounding in gray areas such as this.

Product passes all other tests, and I believe it just went through, and passed, emissions testing. So this ridiculous stray voltage issue is holding up everything.

 

[Indolent]

Originally posted by: ducci
The big issue is not a customer problem, but rather a company problem - as we obviously do not have a standard procedure for rounding in gray areas such as this.

Product passes all other tests, and I believe it just went through, and passed, emissions testing. So this ridiculous stray voltage issue is holding up everything.

I'm going to guess you're not going to say what company this is? Just curious.

 

[Rogodin2]

Fortunately for you though, I'm in grad. school (just started my M.A.Sc in EE). You don't have to worry about doing business that employs me for a couple years.

You've quite the ego lad. Put a cork in your irrlevant mouth.

-0.03v is enough to invalidate the standard.

Rogo

 

[gooseman]

I say it fails.

 

[irishScott]

Are these theoretical numbers? If so, then I think your golden. In any case, I'd ask the customer. If asking the customer is not feasible, then I agree with the sig-fig argument.

 

[Triumph]

I expect that to mean that the voltage value should be 1.1V, +/- 0.1V. So I'd call it out of spec, but why don't you just ask the customer what they meant?

 

[Triumph]

Originally posted by: ducci

It is essentially a question of whether or not -0.97 = -1.0.

hahahahahaha

did no one else find this funny?

 

[dug777]

Originally posted by: Triumph

Originally posted by: ducci

It is essentially a question of whether or not -0.97 = -1.0.

hahahahahaha

did no one else find this funny?

Woof!

 

[LS20]

why not call and clarify?

the #1 stressed in PM is communication

 

[Vette73]

Originally posted by: FoBoT
shouldn't the original specs have included tolerances?

i blame the project manager for getting vague requirements, it is always the project managers fault

To me it seems they kinda did. -1.0v to -1.2v
So not only is it near the bottom of the spec it is outside the hard spec number.

My wife works for a chip maker and she is caught between MGT and design of a chip. They can recall chips and cost millions of dollars or tell customers that XYZ part of the chip does nto work so you lose XYZ function.

Needless to say it may have come down to well its close to spec so send it out it shoudl be OK. Now its not. Very close to what the OP is talking about. For me fix it now before you ramp up production so there are no recalls or problems AFTER the fact. And if there is problem they can say WE said -1.0--1.2v NOT -0.97.

 

[Skacer]

Head Engineer probably wants it to pass because it's his ass on the line if the part is delayed.
Head of Quality probably wants it to fail because it's his ass on the line if the part is shipped but the quality fails.

This is why your company has a "Head of Quality".

 

[Joemonkey]

from a quality engineer where i work:

I have had that argument many times. It is not clearly defined in any standard, however the basic theory is that you measure with 10x more accuracy than your tolerance and make a judgement of inspec vs. out of spec. based on the actual 10x greater resolution number. If you round your answer after you measure with 10x greater accuracy you may as well have used a lower resolution measurement to begin with.

 

[Specop 007]

Originally posted by: ducci
It is not for something basic like a TV remote - it is actually a military device.

Ouch. Ship an out of spec part on a .mil contract?

Give it a shot, see what happens.

 

[Gibson486]

Originally posted by: Specop 007

Originally posted by: hypn0tik

Originally posted by: Train

Originally posted by: hypn0tik
Significant figures!

They required a spec -1.0V. In this instance, they only need accuracy to within a tenth of a volt. If they needed a higher accuracy, they would have specified 1.00V or 1.000V. Clearly they didn't. In this case, -1.0 means that the voltage could be anywhere from -0.95V to -1.04V. Since -0.97V falls within the acceptable range, it's good to go.

If they had only specified -1.0V, I would agree with you. Howerver They said "And Lower" pretty much making it one sided. The significant figure is meaningless since he also said -1.3V was acceptable.

The "and lower" has no bearing on the fact that they specified to an accuracy of 1/10th of a volt.

It's really an elementary scientific principle. You can't specify your result more accurately than what you start out with. In this case, -0.97 is more accurate than -1.0. You should express your result with the same accuracy. In this case, -0.97 -> -1.0, which is indeed the required specification.


On the other hand, it is extremely poor engineering practice to design for something right at the boundary. Their engineers should have designed for something LESS than -1.0V (-1.2 for example) so that they'd have a tolerance of 20%. Regardless, this is not the issue we are looking at.

What company do you work for.
I want to make sure I never do business with a company with shady employees selling out of spec stuff.

I have to agree here. If it is below spec, it is below spec. No excuses. What ever engineer says other wise should not be in that field. Try that in a military contract and see what happens. I remember when i was in charge of testing limits for chips. I had one chip that was 1 mV out of spec and i passed it because were running out of time and needed to get a shipment out asap. I never heard the end of it after. I almost got my head torn off.

 

[Demon-Xanth]

I'd say the person that wrote the spec failed.