Randomly pulling components off the assembly line
 

Comment: My husband and I both were told in class story about bad business
practices in college 7 years ago that we can't find again.

The story goes something like this:

A electornics (perhpas tv) manufacturer decided to cut costs on
manufacturing. They developed a policy of sending supervisors on the
manufacturing floor and randomly pointing to resistors or other small
components, and asking "What does this do?" They'd then pull a component
and if the TV still worked, the supervisor would then tell the whole line
to remove the component.

The company was able to sell their tvs cheaper than anyone else, but in
the long run had problems with reliability. The brand lost face in the
market.

My husband's boss used the first part of this story saying this was a good
practice, but had not heard the part about reliability problems that was
the point of the story as told to both my husband and I. We'd love to
find the story again and find whose version is correct.

I don't know if the story was true or not, but a similar joke was always told by Lotus founder Colin Chapman. Early Lotus cars were noted for interesting and original engineering, as well as being very light, but were often plagued with component failures. The joke was that Chapman would start taking parts off a new design, one at a time, till it fell apart. Then he'd put one back.

As far as the OP goes, I doubt this was ever a practice, at least not at random. Surely R&D would perform some sort of work to determine what could be taken out of a product.

-RB

I have heard of companies doing this sort of thing with regard to non-critical components. It goes something like this "If we halve the length of the power cord, we save a few cents per item. Is that going to stop people buying it? If not then do it". I have never heard of anyone randomly removing parts though. The stories I have heard have always involved parts that would still perform the same function, but not be noticed by the user (such as the power cord, or cheaper lower tolerance parts used where the tolerance isn't critical etc).

me

This sounds like a myth to me. Assembly lines are created very specifically. Parts are ordered. Products are designed and tested. But hard to say that something like this has never happened.

sounds to me like someone trying to explain why PCBs (printed Circuit boards) have obvious places for capacitors, resistors etc, yet are missing them. take a look at a lot of your electronics and you'll find these "missing capacitors". I'm not in electronics, but i'm guessing that they were in the initial plans for the electronics, but were found to be superfluous or were even (GASP) a mistake by the designer. or a remnant from an older version, or something else.

also i highly doubt they'd do that at the manufacture floor level, it'd be done in the design and testing phase.

I had always figured that the motherboards were created with the capacity to have all sorts of functions added, just as vars often have little panels where certain options belong if you get that option as part of your purchase. It is easier, cheaper, and more reliable to build a few basic forms which can have features added into places already prepared for the features than to have a unique build for each set of options.

The result being, of course, a less-than-full Lotus.

Also, I doubt that someone on the assembly line would be able to give a meaningful answer to "what does this do?"

Still ... corporate America is filled with examples of management interfering with and changing processes that they don't understand, usually to the detriment of the company (then of course getting a bonus for lowering costs, while those below hir get dinged when the process no longer works!).

Sounds about right.:fish:
<--Lotus owner

Actually, mine has very few issues.:)

I'm sure you can remove a part or two from many complex devices without immediate bad effects, but that doesn't mean it would be a good idea!

As A Turtle Named Mack guessed, it is because they print a single board with all the added features and just for add the components for the added features to the lower priced models.





According to legend, a manager in my company needed to order a new fleet of vans to replace the current (at the time) fleet that was aging and beginning to be costly to maintain. So he noticed that the vans were usually ordered with an anti-rusting agent on the undercarriage. He decided to order the new fleet and saved the company millions. He was promoted for his finiancial genius.

His sucessor was later fired for spending billions to replace the whole fleet after the floors rusted out and the vans did not last as long as the previous fleet did.

Yeah, I have no problem believing that. What I've never been able to figure out is if the people who make these decisions (in the case of your example, the people who promoted the first guy and fired the second) are just that glaringly stupid, or if they're aware of the ridculousness of their decisions but figure that by the time the decisions catch up with them, they'll be long gone.

I found out about a similar situation with a former employer. It turns out that there was a task not getting done in a certain department. It was a simple, but time consuming task, and the person doing the task would have very little time for anything else (sorry for the vagueness, but I don't want anything identifiable here), and the company was in a hiring freeze and were forbidding the use of temps due to budget reasons.

So, for want of a $30,000 a year temp, this task - which would have made the company $1,000,000 a year! - was not getting done.

Even when this was explained to upper management, the response was "it's not in the budget."

It isn't just about adding or subtracting features for lower priced models. Circuit boards with additional space and traces are a relatively cheap hedge against "function creep" - adding functions and features as time goes on. Not only that, but some devices may be removed because discrete components get replaced with integrated circuits during the life of the product. That's actually a cost savings - not changing the board design but merely the components on it.

Lotus is sometimes said to be an acronym for Loads Of Trouble, Usually Serious. The story relates to Lotus's time in motor racing - Formula One in particular. If the car was in one piece at the end of a race, it was obviously too heavy and if it had fallen apart it was obviously too light. Thye ideal was for it to cross the line then collapse into a pile of dust.

Another somewhat similar reason for curcuit boards to have spaces for components that aren't installed (or "populated" as the say in the trade) is that integrated circuits can often operate in different modes, or enable/disable certain features, by installing "pull up" or "pull down" resistors on particular pins. For example, you might enable some feature by putting a 50 ohm resistor between pin 1 and 3.3V, or disable it by putting a 50 ohm resistor between pin 1 and ground (in digital terms, you're setting that input to a one or a zero). Typically designers create two places for resistors to account for either scenario, either "just in case" (ooops, we actually needed to have that feature turned on. time to get out the soldering iron, remove that resistor from R21 and move it to R22), or possibly to allow for two slightly different versions of the same product.

This one turns up in a SF Novel "Wreath of Stars" by Bob Shaw. In his novel someone said that he designed a radio. The company he worked for took the prototype, and then turned it on. Then they turned it off and removed one component and turned it on again. This was done until it stopped working. Then they put the last component back and that was the radio they manufactured.

The novel itself dates from the 1970's so the story in the Op is an old one.

Original idea couldn't be true
 

At least in regards to electronics, this story could not possibly be true. In any electronic device, removing any one part virtually always either stops it from working or affects the performance dramatically. There could be a very few parts that could be removed, without affecting most devices, but then perhaps 1% of them wouldn't work, and that kind of situation is totally unacceptable. The chances of hitting any such part, in a random selection, are small. And, the whole exercise is unnecessary; if an engineer thinks a part is redundant, he can tell simply by looking at the circuit diagram. Duh!

Furthermore, removing a single part would require changes in a huge amount of paperwork for ordering and manufacturing. This just isn't done. Instead, engineering changes wait until there are enough to justify a major revision. The only exception might be flaws that are serious enough to warrant the effort.

Finally, electronic devices are not tested in assembly. So if someone were to pull a part, he wouldn't know its effect.

Why are some printed-circuit boards apparently missing parts? Because a single board design is used for several models of the device, to save money. It's commonly done.

This is a believable story to me. I've seen in private corporations and military service where the accomplishments of a small team are credited to numerous people up the chain, so something like this would be very possible.

In a somewhat unrelated story, I recall how a hard drive manufacturer a few years back (late 90s) was behind on production prior to the Chirstmas season, so they made the decision to put bricks in the boxes where hard drives would go and sort it out after the holidays when they had production capacity to spare. Sales didn't suffer and the bricks were replaced whenever anyone complained. I can't find a reference to it, but I remember it from the news.

After a longer than expected delay I finally managed to find a copy of the book to verify this.

Bob Shaw. 'A Wreath of Stars'. 1976. Pan Books, pg 96

Obviously the tale has been around for a long time.

Reminds of a halogen reading light my parents bought some years ago. Worked really well so they bought a matching one from the same company some time later. The second one would work for about 30 min then shut it self off for a while. Then turn back on for several minutes and then off again and repeat. We noticed the switch was getting hot, very hot. Everything about the two lamps were identical, except the light switch. The newer one had a switch with a much lower amp rating than the old one. Could not imagine and electrical engineer choosing a switch that just meets the amp usage of the light bulb alone.

Makes me wounder sometime about how thing are chosen to save money. Switch "B" is 2 cents a unit cheaper then old switch "A". Does switch "B" work in light. Yes sir, installed it and light turns off and on. That is all we need to know, make the change.

I can almost guarantee you the the decision was made by some mid-level manager with no one from engineering involved.

I totally agree. A production line worker (or even supervisor) wouldn't know what a component did. It's not his job to know.

Well there will be some basic testing, but not the kind that would necessarily test every component, or test all functionality. And testing would be an automated part of each sub assembly. If you're testing at the very end then you're doing it wrong.

Components like resistors cost 2 cents each. The cost saving is negligible.

It's possible that the oveheating switch was of a type suitable for its purpose, but as with any component some just come out of the factory defective.

If the lamp maker bought a batch of 10,000 switches it's inevitable at least a few will go wrong.

That should probably been watts not amps since the lights had dimmer switches. I believe that is how they are rated.

Now to try and remember my electrical courses that I took some 20 years ago.

The problem with using a 300 watt dimmer switch with a 300 watt bulb is that is not the real wattage. Watts = Amps * Resistance^2. The light is 300 watts plus one needs to add in the wire and the dimmer switch. As the dimmers switch warms up, it also increases the watts used do to the extra resistance. More load mean more heat and more heat mean more load. Things keep getting hotter until something has to give. It is always good to go a little bigger with these type of things to make sure bad things do not happen.

Another example of the same type of thing causing problems. I borrowed some floor sanders from by brother. They were rated and 15 amps and used a 15 amp plug. Used them for any length of time and they will trip the circuit breaker. The 15 amp circuit breaker while rated at 15 amps can not handle the loads for long periods of time. Plug it into a 20 amp outlet and problems go away. Brother when I talked to him about it, said he has the same problem with them all the time.

Or, it could be China made. THe Chinese are excellent at making cheap stuff, but quality control is not their strong side. We've run out of steel? Use pot metal instead! No screws left? Just use a screw of roughly the same type and force it!

There has been reports on automatic fuses in Sweden that's made in China, where, on opening them, it was discovered that the fuse components were gone, and it was just an ordinary switch. Many electricians now test all the automatic fuses by shorting them in a test rig, just to be safe.

WARNING: Three years since the last post...

I may have found a real life case of this on youtube in the form of a video on youtube that rather ironically dates from the year (2012) this discussion was started.

1999 Daewoo color TV with missing filter capacitor (from the factory)

I'd expect that that is an example of a error in manufacturing more than a deliberate attempt to make a product cheaper by leaving off "unnecessary" components.

Agreed that it would be more a mistake than deliberate attempt, you could change that heatsink to a much cheaper one and save more money. But watching the video, it actually appears to me that there was something soldered in there, and subsequently removed, and the solder cleaned up, (whoever did it did a good job). The hole didn't look as clean as it would be new to me. (the fact that the holes are not uniform in size, and the fact that there is no heat distortion shows it was a good job, heck you can see the repair done just beside it that wasn't as well done)

Also that's what a sub $1 part? Why do that when it would cause a lot of returns for saving $1 per unit? I don't think it'd be worth it.

I'd always heard the practice went back to Earl Muntz, Mad Man Muntz, late 1940s, early 1950s.

See:

http://www.smecc.org/mad_man_muntz!.htm

I was re-reading the thread, and someone mentioned power cords being shorter to save costs.

I've noticed it, in particular since the price of copper has gone up around 2009+ (it's the only history I can find).

My 4 year old coffee grinder has one of the shortest cords I've ever seen (not sure if it's even 2 feet?), newer toasters are similar.

OY

Kitchen appliance power cords have gotten shorter for safety reasons.

Seems to be rather spotty. Some are still long, some are very short. No rhyme or reason.

OY

Which backfires when shorter cords lead to the use of extension cords.

The idea here is that newer homes have more outlets around the kitchen area. My nearly 20-y old house has I think 8 outlets around the kitchen (mostly where over the counter).

OY

You may be right there, but recently I found two videos (Apologies in advance for the 'Steve Irwin' style of presentation.) about the repair of failed computer monitors, in the first there are a number of positions for uninstalled components drawn onto the circuit board. In the second case it's clear that a cost saving measure was taken after the board was drawn out with cheaper components being substituted at production.

EEV Blog 347:Bad Capacitor LCD Monitor Repair

EEV Blog 365:ESR Meter - Bad Capacitor Monitor Repair

Often the same printed circuit board is used for more than one product, hence the "uninstalled" components. The missing components are for an upgraded version.

The failed capacitor problem is definitely the result of using cheap components, though the cheap components were supposed to be within spec. (https://en.wikipedia.org/wiki/Capacitor_plague) There are even rumors of industrial espionage and capacitor makers stealing patented or proprietary formulas from other makers then saving money by omitting key components of the electrolytic mix. The caps were cheaper but they failed after only a year of two in use.

Thanks for the addtional info, I've watched a number of teardown/repair videos and always wondered about the empty spaces on circuit boards. The basic story has been around a while, but things like this tend to act to re-inforce it.

One last YouTube video, pay attention to what is said starting at 6:24...

"Fake" Dr. Dre Beats Mini Bluetooth Speaker Teardown - For Fun!

...nice to catch an Urban Legend...

Hold on, what I noticed was he opened a Pepsi with an old pull tab. I haven't seen those in years Where did that come from?

Hipster grocery store. But they sold out.

He said he was in Shanghai, China...

Quite possible they still use the old style pull tabs.

OY