Author Topic: Have you designed early life failure into a product?  (Read 5300 times)

0 Members and 1 Guest are viewing this topic.

Offline aargee

  • Regular Contributor
  • *
  • Posts: 173
  • Country: au
Have you designed early life failure into a product?
« on: August 07, 2014, 10:53:33 AM »
Or in other words,  knowingly or been forced to fit a design to a life cycle much smaller than customers/end users would expect?

I can understand that the race to the bottom in quality causes corners to be cut, etc. but designs that are planned to fail close to the end of warranty (or other milestones in a product's life)

I have never been in the design cycle for new products, I'm a field service engineer, fixing stuff or inventing kludges to fix problems. Over my many years of experience in lots of facets of the electronics industry, it *seems* like some designs I have run into have been purpose built for a lifespan of "X" whereas the customer would expect "Y" (where X<<Y).

I know there's all the evidence out there from the light bulb consortium at the turn of the century through to the printer ink marketing, I'm just curious about the people who knowingly design this stuff.

Online retrolefty

  • Frequent Contributor
  • **
  • Posts: 352
  • Country: us
  • measurement changes behavior
Re: Have you designed early life failure into a product?
« Reply #1 on: August 07, 2014, 11:30:51 AM »
Interesting question. I can't talk much about electronics products other then their real cost (inflation adjusted) to consumers has never been better. A related consumer product I can relate to is automobiles. Growing up in the 50s it seemed that most middle class+ people would look to replace their car around 3+ years or so. The last two cars I've owned have lasted me 13 and 14 years, they just seem to last longer and are more reliable.


Online SirNick

  • Frequent Contributor
  • **
  • Posts: 530
Re: Have you designed early life failure into a product?
« Reply #2 on: August 07, 2014, 01:46:28 PM »
OTOH, I drive a car that was built ten years before I was born.  ;)

Not in the industry myself, but I see aspects of products where obvious corners were cut.  One can only assume the engineers spec'd parts understanding that they would fail after so many hours of use, and that the part lifetime and warranty period were considered with that in mind.  Or that the user would likely abandon the product anyway, if it failed early.  (LCD TV and monitor PSUs particularly.)

Interestingly, I wandered into a forum thread once where the (unverified but believable) engineer of a brand-name product discussed the statistical analysis of his component choices.  Namely, PSU caps for an amplifier that were operating very close to their rated limits, where you would usually expect to see more safety margin.  The engineer spoke about how the calculated lifespan was "long enough" -- such that they would probably fail about the same time the user would be replacing the unit for other reasons.  I'm not sure if that's planned failure so much as matching a low price point to the cultural trend of throw-away products.

Offline westfw

  • Super Contributor
  • ***
  • Posts: 1213
  • Country: us
Re: Have you designed early life failure into a product?
« Reply #3 on: August 07, 2014, 02:27:59 PM »
A lot of products are built for an "expected lifetime"; it's not so much that they're specifically designed to "fail early", but they don't get the attention they would need to have a good chance of having a long life...

I've certainly written software that rendered older HW obsolete, usually just by virtue of outgrowing the memory hardware limitations of the old HW.  It wasn't so much intentional, as just something that happens w software.  And I've been on the other side of that as well "how can we pare down this new version so that it will run on the old HW."  But at some point it is VERY MUCH an explicit engineering decision: "This new SW version will NOT run on that old HW."

Online miguelvp

  • Super Contributor
  • ***
  • Posts: 2233
  • Country: us
Re: Have you designed early life failure into a product?
« Reply #4 on: August 07, 2014, 02:50:47 PM »
I think the production cost reduction run will be your main culprit. For example, Jeri Ellsworth made a video talking about her C64 FPGA Joystick, and they flew her to China because the thing wasn't working.

Once there she found out they replaced a lot of parts and cut corners, hopefully she was able to save the day and got it working. But what she designed was nothing close to the final product.

But at least it's not like the GM chief engineer that signed off that faulty switch. I still think he was told to sign it or else but it's ok GM named and fired engineers involved because obviously it couldn't be a marketing decision  ::)

Offline Joule Thief

  • Regular Contributor
  • *
  • Posts: 124
  • Country: us
Re: Have you designed early life failure into a product?
« Reply #5 on: August 07, 2014, 03:30:50 PM »
aargee, I also spent 25+ years as a service engineer for several manufacturers of medical instrumentation. Most, if not all the disappointments relating to quality have always been "introduced" into the system as a cost saving measure. Most notable was a roller assy for a tubing pump to which a $35 cost reduction was implemented. This was an attractive savings for a system that contained 3 pumps ($100 total savings per instrument in an instrument costing #25,000 USD). Unfortunately the improvement involved hollowing out the individual rollers on the roller cage, allowing the roller to collapse under the strain of the pumping action and eventually compromise the life of the roller.

You guessed it ... either a service call (avg $1000 cost) was needed for mysterious intermittent pumping issues or replacement parts had to shipped over night to the account. Eventually the original design was re-implemented and new rollers were sent to all customers.

The story I tell about my first R&D interview goes like this. First question I was asked was "We have a design choice to make in a transistor used in a power supply assembly", (remember, this is medical instrumentation large enough to require an on site service visit - the unit cannot be shipped back for repair.).

"One transistor costs $3USD and is near its design limits. The other choice is a $7 transistor which will last the life of the unit. Which would you choose? Please take some time to think about your answer before you give it".

My reply was "choose the $7 transistor".

WRONG !!!

"May I explain why I made this choice?"

"If that transistor fails for any reason in the next 15 years (life of system), an onsite service visit will be needed. This will involve buying a plane ticket for $2000 to fly across country on a day's notice, grab a $200 rental car, $300 hotel bill, at least $100 in meals and incidentals, and about a $300 dollar two way round trip for FedEx charges to ship the 50 lb part out overnight freight.  So I say spend the extra $4 to save $2500 cost to the company in service costs."

R&D response was "you are not looking at the bigger picture. We have a price point to bring this system to market and we must hit that mark".

I mumbled something about who is not seeing the "bigger picture" and told the interviewer I could never work for him under those constraints - not with the "bigger picture" I have come to see from working in the field.

Later that year I was paged out to service one of those systems. A "fix it before the sun comes up or get it out of here" scenarios. I met my wife that night. She was a technician at the hospital and saw me working on the system at 2 in the morning and let out a string of curses I can't repeat here.

We have been married going on 35 years now. That's my story and I'm sticking to it. 

« Last Edit: August 24, 2014, 08:57:45 AM by Joule Thief »
Perturb and observe.

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3892
  • Country: 00
Re: Have you designed early life failure into a product?
« Reply #6 on: August 07, 2014, 03:42:04 PM »
I still think he was told to sign it or else

Twenty years ago I would have also thought this. Or that GM management created an atmosphere of fear and obedience where no one dared to step out of line.

But since this was just ten years ago I wouldn't dismiss that the involved engineers just gave a flying fart through a rolling doughnut. The attitude about safety or work ethics in general has massively changed among young engineers.

But back to GM. Even if it was an "I give a fuck" engineer, the management is still responsible. For hiring him, for giving him that responsibility and for not properly supervising him. I.e. for getting everything wrong they are paid for.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List

Offline pickle9000

  • Super Contributor
  • ***
  • Posts: 1047
  • Country: ca
Re: Have you designed early life failure into a product?
« Reply #7 on: August 07, 2014, 03:49:29 PM »
I'd have to say more about cutting down the BOM bottom line, same effect though.

Online miguelvp

  • Super Contributor
  • ***
  • Posts: 2233
  • Country: us
Re: Have you designed early life failure into a product?
« Reply #8 on: August 07, 2014, 04:12:51 PM »
I still think he was told to sign it or else


Twenty years ago I would have also thought this. Or that GM management created an atmosphere of fear and obedience where no one dared to step out of line.

But since this was just ten years ago I wouldn't dismiss that the involved engineers just gave a flying fart through a rolling doughnut. The attitude about safety or work ethics in general has massively changed among young engineers.

But back to GM. Even if it was an "I give a fuck" engineer, the management is still responsible. For hiring him, for giving him that responsibility and for not properly supervising him. I.e. for getting everything wrong they are paid for.



http://www.autonews.com/article/20140605/OEM11/140609881/blame-for-%E2%80%98switch-from-hell%E2%80%99-falls-heavily-on-one-gm

Quote
Ironically, it was DeGiorgio who ultimately fixed the part he referred to in a 2002 e-mail as “the switch from hell.” But by the time anyone else in the company figured out what he had done, GM says the switch had cost 13 people their lives.


If he referred to the switch in an e-mail as "the switch from hell" 2 years prior being implemented doesn't that sound like coercion? 10 years or 20 years doesn't matter, look at Enron for example. Let's just say the 2000 decade is not one of our most glorious times, well this decade is not great either, maybe 20/20.


Online dannyf

  • Super Contributor
  • ***
  • Posts: 3996
Re: Have you designed early life failure into a product?
« Reply #9 on: August 07, 2014, 10:51:49 PM »
Quote
Or in other words,  knowingly or been forced to fit a design to a life cycle much smaller than customers/end users would expect?

Quote
some designs I have run into have been purpose built for a lifespan of "X" whereas the customer would expect "Y" (where X<<Y).

The two are slightly different. The first sounds closer to designing for failure within a set period of time and the 2nd closer to designing for a life of a set period.

ie. "I want the product to fail within a year" vs. "I want the product to last at least a year".

The 2nd approach is wide spread. and it is not common to see the same / similar products at different points that offer different performance (expected life included). Quality is expensive and if a customer expects a life of X, s/he should pay for life of (at least) X.

The 1st approach is rarely seen - it is mean spirited I think.

Online dave_k

  • Contributor
  • Posts: 29
  • Country: au
Re: Have you designed early life failure into a product?
« Reply #10 on: August 07, 2014, 11:18:42 PM »
I have said this of most contractors/tradies:

"There is no money to be made in fixing the problem, but there is plenty to be made in prolonging it"


Offline wagon

  • Regular Contributor
  • *
  • Posts: 82
  • Country: au
Re: Have you designed early life failure into a product?
« Reply #11 on: August 07, 2014, 11:34:32 PM »
I spend my days fixing electronic stuff.  I repair a lot of electric fence energisers : my workshop is in a small city that is a bit of a 'hub' for a lot of agriculture.  Now, I'll get to my point!

I fixed an old unit yesterday, that was over a year older than me.  It was built in 1975.  The main capacitor was still within spec, and the unit really only needed a new 47nF 'X2' cap and a new mains cord/plug.  It could well see the farmer out.  I could have sold him a new one, too.  I don't expect a new one will still be going nearly 40 years from now, since it is a lot more complicated and its little case with its thin walls is likely to suffer in the sun.  Is the new unit built with failure in mind?  I don't think so, but it does have to be sold in a market where the alternative products are, so manufacturing costs must be kept down.  Price is as important as specifications, quite often. 

With a lot of consumer electronics, like TV's, I feel that the 'market' demanded cheaper stuff, and the suppliers worked out that they just weren't selling the better stuff, so they just stopped making it!  Look where that has gotten us, now that we can only get cheap & nasty. 
Hiding from the missus, she doesn't understand.

Offline digsys

  • Supporter
  • ****
  • Posts: 979
  • Country: au
    • DIGSYS
Re: Have you designed early life failure into a product?
« Reply #12 on: August 08, 2014, 12:02:37 AM »
Sadly, even going back 30-35 yrs, a LOT of companies I contracted R+D for, often paid big bucks for designs that just made it past
their warranty. We got bonus amounts on how close we could get it. In some service departments (who we worked with), they even
ran competitions ! The one big difference to these days was that - There was NO compromise on quality / (design) reliability !
Some of the "fixes" were ingenious :-)   I occasional still get asked, but stopped doing that cr@p a long time ago. It still happens a lot though.

Edit: OK I have to confess.
It was the mid 70s (IIRC), someone had just come up with the idea to connect one of our Monroe calculators (nixie tubes) to a cash drawer !!
It was a gold mine, but they needed to ensure that it had a limited life. I made up the interface card to tie the systems together, which gave the
cashier - items, totals and date - and controlled the cash drawer. I then made up a separate diode logic pcb (100's of diodes from the nixie array)
and set up links to create a "date code". At this point the AND array would latch and turn off the drawer control.
The service techs simply swapped out with the next date code ! I still remember my bestowed hero status ... ahhh memories
« Last Edit: August 08, 2014, 12:27:20 AM by digsys »
Hello <tap> <tap> .. is this thing on?

Offline mojo-chan

  • Super Contributor
  • ***
  • Posts: 1910
  • Country: 00
Re: Have you designed early life failure into a product?
« Reply #13 on: August 08, 2014, 01:14:27 AM »
Most manufacturers give a MTBF for their parts. Designers simply pick parts that have a MTBF slightly longer than the warranty period. The EU is looking at requiring appliances to have the lowest MTBF of all components printed on the efficiency rating label.
kilobyte = 1024 bytes (JEDEC standard)
kibibyte = 1000 bytes

Offline JuKu

  • Frequent Contributor
  • **
  • Posts: 421
  • Country: fi
Re: Have you designed early life failure into a product?
« Reply #14 on: August 08, 2014, 02:06:05 AM »
Most manufacturers give a MTBF for their parts. Designers simply pick parts that have a MTBF slightly longer than the warranty period. The EU is looking at requiring appliances to have the lowest MTBF of all components printed on the efficiency rating label.
No, that would mean close to half of the products fail within warranty period! But you are right, mtbf is an important spec, products are not designed to last forever anymore.

When lightbulbs were used, it costed a little more to make them fail. Controlling the impurity of the inert gas costed something, when putting in pure inert gas would make the bulbs last much longer. In Edison museum there are lamps about 100 years old, shining 12 hours a day, 7 days a week.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf