LED lighting and planned obsolescence, intentional or not.

[Psi]

This is just for the LED lighting planned obsolescence and regulation discussion that was happening in the superconductors thread that I moved to here, as it was off topic for superconductors.

[Psi]

I was commenting on the false statements you appeared to be making.

Ok, fair enough, I should have said "most" instead of "all". I know there are some LED lighting manufactures that sell good products.
I was not suggesting that manufacturers are having design meetings on how to get their devices to fail just outside of the warranty in order to sell more. Or they're testing them to ensure they do fail soon after warranty.
I'm saying they are choosing their product design requirements based on their chosen product lifespan and they are doing that process wrong. Lifespan requirements in the design process should be determined by a cost/benefit analysis from the users perspective, not the sellers perspective. If you can add $0.05 of extra parts to a product with $2 BOM cost and that will make it last twice as long then you should do that. Not doing that means you are designing it to fail prematurely.

I hate reading stuff like this.  You essentially claim that "most" manufacturers actually go to the trouble of designing the product to be a time bomb, rather than simply minimizing the product cost, which sets the minimum selling price.   Yet, You offer zero evidence of anything you say.  What $0.05 worth of parts will make an LED bulb last twice as long?

Whatever.   People believe what they want to believe and there is little in the way of facts or evidence, that will change their minds.

That is exactly what i'm not saying. I'm saying they are doing it as a side effect, not on purpose, at least not directly.
For any product you need to come up with an expected life-span requirement that you want to achieve so you have something to check against and verify the design. The lifespan requirement is often decided by competitors products and by the market.

Instead of looking at that question with an eye towards what's best for the product and best for the consumer they are looking at it as an exercise of lowering BOM cost and maximizing profits.

In most cases the cost to make it last 5x longer is not even $0.05, it's free, just change the value of a resistor to run the LED at 50% brightness so the temps are all under control and the bridge rectifier is not running at 105C and not cooking the electrolytic cap and LEDs.  At 50% brightness you might need 2 of them or you might need a larger version to light your room, but that is a different issue.

The main problem is that to sell LED lights with proper lifespans the cost would be higher as you might need more of them, and companies selling them would lose money because the public are stupid and will usually buy the cheaper product and save a little cash even if it only last 20% as long.

This is why we need legislation against planned obsolescence, either intentional or unintentional.

[gnuarm]

I was commenting on the false statements you appeared to be making.

Ok, fair enough, I should have said "most" instead of "all". I know there are some LED lighting manufactures that sell good products.
I was not suggesting that manufacturers are having design meetings on how to get their devices to fail just outside of the warranty in order to sell more. Or they're testing them to ensure they do fail soon after warranty.
I'm saying they are choosing their product design requirements based on their chosen product lifespan and they are doing that process wrong. Lifespan requirements in the design process should be determined by a cost/benefit analysis from the users perspective, not the sellers perspective. If you can add $0.05 of extra parts to a product with $2 BOM cost and that will make it last twice as long then you should do that. Not doing that means you are designing it to fail prematurely.

I hate reading stuff like this.  You essentially claim that "most" manufacturers actually go to the trouble of designing the product to be a time bomb, rather than simply minimizing the product cost, which sets the minimum selling price.   Yet, You offer zero evidence of anything you say.  What $0.05 worth of parts will make an LED bulb last twice as long?

Whatever.   People believe what they want to believe and there is little in the way of facts or evidence, that will change their minds.

That is exactly what i'm not saying. I'm saying they are doing it as a side effect, not on purpose, at least not directly.

Ok, then we agree and this discussion is over.

For any product you need to come up with an expected life-span requirement that you want to achieve so you have something to check against and verify the design. The lifespan requirement is often decided by competitors products and by the market.

Instead of looking at that question with an eye towards what's best for the product and for the consumer they are looking at it as an exercise of lowering BOM cost and maximizing profits.

In most cases the cost to make it last 5x longer is not even $0.05, it's free, just change the value of a resistor to run the LED at 50% brightness so the temps are all under control and the bridge rectifier is not running at 105C and not cooking the electrolytic cap and LEDs.  At 50% brightness you might need 2 of them or you might need a larger version to light your room, but that is a different issue.

The main problem is that to sell LED lights with proper lifespans the cost would be higher as you might need more of them, and companies selling them would lose money because the public are stupid and will usually buy the cheaper product and save a little cash even if it only last 20% as long.

This is why we need legislation against planned obsolescence, either intentional or unintentional.

I was with you until you started talking about the absurd idea of legislating "planned obsolescence".  Now, you have gone off the rails.

[Psi]

I was with you until you started talking about the absurd idea of legislating "planned obsolescence".  Now, you have gone off the rails.

We can't keep manufacturing trash and filling up landfills with throw-away devices, or devices that might as well be.
Well... I guess we could, but it would trash the earth and probably kill us eventually. It's not something anyone wants.

I guess it's possible companies might start to do it on their own if public option shifts enough to change purchasing habits and affects their profit. But if that happens it's likely to cause new laws anyway.

[Haenk]

That is exactly what i'm not saying. I'm saying they are doing it as a side effect, not on purpose, at least not directly.

This is why we need legislation against planned obsolescence, either intentional or unintentional.

I'm certain they *are* doing this on purpose. Imagine your light bulbs last "forever". Selling you a full set of bulbs is a one-time-business, after that all manufacturers can close their doors, as there is no need of further supply ever again. And there is hardly any tech cycle, requirin new "hardware".

Forcing a longer lifespan howevery is a very good thing though, as this is eventually toxic electronic waste, which should be avoided.

[gnuarm]

That is exactly what i'm not saying. I'm saying they are doing it as a side effect, not on purpose, at least not directly.

This is why we need legislation against planned obsolescence, either intentional or unintentional.

I'm certain they *are* doing this on purpose. Imagine your light bulbs last "forever". Selling you a full set of bulbs is a one-time-business, after that all manufacturers can close their doors, as there is no need of further supply ever again.

So the only possibilities are inferior products that fail as soon as the warranty is up, or products that last forever? 

I think I need to stop reading EEVblog.  This place is insane!

[AVGresponding]

That is exactly what i'm not saying. I'm saying they are doing it as a side effect, not on purpose, at least not directly.

This is why we need legislation against planned obsolescence, either intentional or unintentional.

I'm certain they *are* doing this on purpose. Imagine your light bulbs last "forever". Selling you a full set of bulbs is a one-time-business, after that all manufacturers can close their doors, as there is no need of further supply ever again.

So the only possibilities are inferior products that fail as soon as the warranty is up, or products that last forever? 

I think I need to stop reading EEVblog.  This place is insane!

Your inability or unwillingness to understand or recognise context is quite annoying, much like talking to a smart-aleck child.

Modern manufacturing business models rely on replacement sales to sustain them, which should be pretty much obvious to anyone; once you've reached market saturation, you either live off replacement/upgrade sales, or you die. Apple's iPhone is a very good example of this.

Incandescents aren't a great example for commercial and industrial lighting models as they haven't been used in large numbers for decades. The costs of fluorescents was well understood, in terms of installation, running, and maintenance costs. LED lighting has thrown this for a loop; we've effectively been sold a pup, where promises about longevity haven't proved reliable, and because of the heavy integration of LED lamps into fittings, the replacement costs have risen hugely, particularly in installations that incorporate architectural or decorative luminaires.

Domestic customers are less affected, since most of their lighting uses some variation of ES or BC lamp units, which makes them quick and easy to replace when they fail.

[Psi]

That is exactly what i'm not saying. I'm saying they are doing it as a side effect, not on purpose, at least not directly.

This is why we need legislation against planned obsolescence, either intentional or unintentional.

I'm certain they *are* doing this on purpose. Imagine your light bulbs last "forever". Selling you a full set of bulbs is a one-time-business, after that all manufacturers can close their doors, as there is no need of further supply ever again. And there is hardly any tech cycle, requirin new "hardware".

Forcing a longer lifespan howevery is a very good thing though, as this is eventually toxic electronic waste, which should be avoided.

yeah, I could be wrong but I don't think most are doing it on purpose at an malicious or intentional level, like i don't think they're having meetings about how they can improve their policy of planned obsolescence, or withholding bonuses if the engineering teams products last more than 25% longer than the warrantee length when tested internally.  (yeah I know stuff like that actually happened for incandescent lamps a long time ago).

I thinks 90% of it is just chasing profit and planned obsolescence comes out of that as an unintended consequence, which they don't mind because they sell more product. I'm sure they know they are effectively doing it but just don't talk about it. It's a convenient side effect that they have no incentive to fix.   A new player enters the market and copies their competitors because doing anything else makes their company non-viable.

It's ultimately the publics fault for buying cheap junk, or saving a few dollars now when it will cost a lot more later.
But you cant easily make people less dumb, so the only real option is regulation.  I would love to see critical thinking and problem solves skills be a required class at school though.

[tautech]

This is just for the LED lighting discussion that was happening in the superconductors thread that I moved to here as it was off topic for superconductors.

Recently with new SMD tweezers:

First real use other than mucking around.....

Was gifted 2x 1.5m LED ceiling fixtures, not working.
They use a 75W rated slimline 220VDC smart driver and after after finding its datasheet became aware the driver is likely not the problem and instead the LED panels.
https://resources.tridonic.com/PDB/Ressource/GroupPdf/en/Driver_LC_75W_100%E2%80%93400mA_flexC_lp_EXC_24265.pdf

~100 3V 1W 3535 SMD LED's are in parallel pairs in each fixture and ST42 quickly identified several suspect ones that wouldn't come ON with its 3V LED test.
ST42 was perfect for this    something my old ST3 can't do.

Now to find some 3V 1W 3535 SMD LED's online......

It's interesting the driver supply current can be set by way of configuration resistors and after reading the datasheet I don't believe the driver is at fault however the LED's may have been driven too hard after finding several in each fixture had failed.
Pushing the envelope to gain best bang for buck (brightness) certainly is the manufacturers fault rather than adopting for a more conservative configuration. 

Still waiting for 3535 LED's.....have any Psi ?

[DavidAlfa]

What do you call planned obsolescence?
My Ikea bulbs have been working for almost 10 years. Only recently one started flickering.
Going for $1 bulbs is not planned obsolescence, you know the parts inside have John's Goodenough QA.
Bought a chinese bulb as a temporal fix, lasted 4 days 

[Psi]

So the only possibilities are inferior products that fail as soon as the warranty is up, or products that last forever? 

I think I need to stop reading EEVblog.  This place is insane!

I wish i could understand your point of view.
Which of these statements do you disagree with.

1) Most LED lights run way to hot and fail early because of it.

2) The LED lights above would last way longer if they didn't run them so hot.

3) Making them run at under 50C is trivial (be lowering output wattage) and would add next to nothing to product cost. Obviously cost to consumer would be a little higher as they might have to buy more lights or bigger light get same output, but the product lasting for so much longer would make up for that by many times.

4) Most companies are never going to voluntary make changes that cost next to nothing but make longer lasting products because it's going to reduce profit. (unless there's a major shift in public buying habits and people stop trying to save a dollar now even if it cost $5 later)

5) The best way to fix this is for the government to pass some sort of law that prevents companies making consumable products that perform badly when there's no technical reason for it.  Or by forced warranty length in industries where there are these problems, like lighting.
I know the exact law and how it works would need to have a lot of thought. And implementing it is not that easy but i think it's totally doable.



I suspect your issue is with point 5 and I suspect its due to you being very antigovernment regulation.
Which I assume is because the government where you are is maybe not working for the people.
And if so, i would put forward that the issue is then with the government not working, rather than an issue with a law to stop or help reduce planned obsolescent.   Apologies if one of my assumptions in this paragraph is wrong, it's just a guess.

[Psi]

What do you call planned obsolescence?
My Ikea bulbs have been working for almost 10 years. Only recently one started flickering.
Going for $1 bulbs is not planned obsolescence, you know the parts inside have John's Goodenough QA.
Bought a chinese bulb as a temporal fix, lasted 4 days 

Yep, there is a wide range of quality levels and some are ok.

A law to stop planned obsolescence does not need to take the form of regulating how things are made, it can simply be laws that block importing and selling of crap quality products that fail early or forcing warranty lengths.
However that does put a lot of pressure on sellers though, If for example a law said that all LED lights sold must have a 10 year warrantee for example. Then it's almost impossible for a retailer to know if what they are buying is going to last that long when they buy it.

[tautech]

I think we as consumers also should share some of the blame when instead we should be insisting on product MTBF specifications.

[Psi]

I think we as consumers also should share some of the blame when instead we should be insisting on product MTBF specifications.

Yep absolutely. The core of the problem is most people are pretty stupid, or usually just uneducated, and they're pushing money into companies that produce trash.
There's also the people living pay check to pay check who will always buy the cheapest version they can get their hands on.
If cheap rubbish didn't exist they would perhaps have to deal with less light in their house until they could afford a 2nd light but overall they would be much better off and have more money (not having to replace it every year). And E-waste would be reduced.

[tooki]

This is why we need legislation against planned obsolescence, either intentional or unintentional.

How could unintentional planned obsolescence even exist? Planned obsolescence by definition means planning the obsolescence; you can’t unintentionally plan something, since planning requires forethought and intent.

[John B]

I think we as consumers also should share some of the blame when instead we should be insisting on product MTBF specifications.

Is that even fair? The average consumer can really only read the packaging which says the lights will last 20,000 - 50,000hrs or whatever BS figure, and due to the length of time involved it's exceptionally hard to make a claim of false advertising stick.

I agree with the theme of the thread. I'm tired of hearing that companies need an unlimited benefit-of-the-doubt pass until the end of time considering how long this discussion has been going on across just about every aspect of consumer products.

At the very least, every major company has had an engineering meeting where the balance of cost, repairability and longevity were discussed and clearly the latter 2 have been given the lower priority more often.

[Psi]

This is why we need legislation against planned obsolescence, either intentional or unintentional.

How could unintentional planned obsolescence even exist? Planned obsolescence by definition means planning the obsolescence; you can’t unintentionally plan something, since planning requires forethought and intent.

Maybe a better name for the concept is Unnecessary Obsolescence.

I was just trying to include situations where the company was making a product with an unnecessary short life-span but there was no planning involved, or at least not planning to make it fail sooner. 
To avoid the company side-stepping any rule by claiming they never planned to make their product become obsolescence sooner and are therefor not guilty of planned obsolescence.

[Marco]

Lamp life should have a similar grading system to energy efficiency. Part of the grade based on say a couple months high temperature high humidity stress testing, part based on best guess component longevity in standard luminaires by the manufacturer (with clear fraud being prosecuted down the line).

Using film capacitors, hermetic sealing and low power density you could probably make LED lamps which outlive me at this point and it wouldn't have to be very expensive.

[Kleinstein]

There is a kind of defacto > 20khr claimed lifetime for LED lamps. So they design for that time frame and as a lifetime estimate is not easy they sometimes end up failing earlier, e.g .with a higher than standard temperature. A design for a given MTBF is normal and in some cases (parts with predictable aging) leads to a fixed time to failure. The point is just if that time frame is given and reliable.

Some grading system for the lifetime would be ideal, but it is essentially impossible to test that reliable up-front and with possible minor product changes over time. It worked with the old incandecent lamps with usually 1000 h and some 2000 h and few 100/300 h lamps as that is still a timeframe that can be tested.  20 kHz are some 2.5 years and this is longer than many LED lamps are on the market.
If parts fail early, you know which brand not to buy in the future.

A point that could help would be a thermal rating for the sockets / lamp housings, so that one know what power one could use without running them extra hot and thus a reduced lifetime. That part would at least be a part that could be measured relatively easy and is thus well defined. With higher temperature use the lifetime is naturally small - still some 10000 or just 5000 hours would not be that bad.

So far I can not complain much about failing LED lamps: most of them last long and only a few (e.g. 1 out of 10) early failures.

[gnuarm]

So the only possibilities are inferior products that fail as soon as the warranty is up, or products that last forever? 

I think I need to stop reading EEVblog.  This place is insane!

I wish i could understand your point of view.
Which of these statements do you disagree with.

1) Most LED lights run way to hot and fail early because of it.

You are stating that as if it were fact, which it is not.  I have LED bulbs that burn for years without problem, because they are not enclosed and have clear airflow.  When the same bulb brands have restricted airflow, the bulbs burn out within a year typically. 

2) The LED lights above would last way longer if they didn't run them so hot.

Again, your supposition, not a fact.  It is not valid logic to make an assumption, then use that to prove the assumption is valid.

3) Making them run at under 50C is trivial (be lowering output wattage) and would add next to nothing to product cost. Obviously cost to consumer would be a little higher as they might have to buy more lights or bigger light get same output, but the product lasting for so much longer would make up for that by many times.

Or, you can use them with proper airflow so they don't get hot.  LED bulbs don't use much power, so they will not run hot in an open lamp or other fixture with adequate airflow.

4) Most companies are never going to voluntary make changes that cost next to nothing but make longer lasting products because it's going to reduce profit. (unless there's a major shift in public buying habits and people stop trying to save a dollar now even if it cost $5 later)

If the problem were the design, this might be valid.  I don't care.  I buy name brands with a warranty.  The store will also give me a refund if the bulbs burn out prematurely, but I don't recall that happening since I removed the air flow restrictions.

5) The best way to fix this is for the government to pass some sort of law that prevents companies making consumable products that perform badly when there's no technical reason for it.  Or by forced warranty length in industries where there are these problems, like lighting.

LOL!!!  There's absolutely no response needed to this.

I know the exact law and how it works would need to have a lot of thought. And implementing it is not that easy but i think it's totally doable.

And pointless.

I suspect your issue is with point 5 and I suspect its due to you being very antigovernment regulation.
Which I assume is because the government where you are is maybe not working for the people.
And if so, i would put forward that the issue is then with the government not working, rather than an issue with a law to stop or help reduce planned obsolescent.   Apologies if one of my assumptions in this paragraph is wrong, it's just a guess.

LOL  No, I am fully in support of government regulation where it is useful.  This is not one of those situations.  Regulation is not going to make your bulbs last longer, because very few people are going to return a light bulb to the store and the store is going to eat the cost and throw the bulb in the trash.

I'm done discussing this with you.  Bye.

[gnuarm]

I think we as consumers also should share some of the blame when instead we should be insisting on product MTBF specifications.

Take a crap in the box, but put a warranty on the outside!  Yeah, good idea. 

These are lightbulbs, not components in a design.  Almost no one returns a burnt out light bulb and the few that do get returned are tossed in the trash by the store.  No one is held accountable, no matter what it says on the outside of the box!

[Marco]

Some grading system for the lifetime would be ideal, but it is essentially impossible to test that reliable up-front and with possible minor product changes over time.

A documented estimate based on projected MTBF of components in standardized luminaires (which for GU10 should be recessed ceiling lights with a glass cover) would still be useful. If their documentation is for polymer/film caps and they threw in wet electrolytics, that's just plain fraud. If they just swapped parts and their best effort estimate stays the same or better, then let them keep using the same type number.

I don't need a guarantee, just a best effort estimate. It will be easy enough to see if they are taking the mickey, BOMs are small and failure modes predictable.

If you start doing it for complex appliances, then it would be far harder to judge honesty.

[TimFox]

Last I heard, a fundamental thermal problem with LED bulb-shaped objects that fail prematurely is that the heat-producing and thermally-sensitive electronics are mounted near the base while the actual LEDs are spaced apart in the bulbous globe.

[PlainName]

There's also the people living pay check to pay check who will always buy the cheapest version they can get their hands on.
If cheap rubbish didn't exist they would perhaps have to deal with less light in their house until they could afford a 2nd light but overall they would be much better off and have more money (not having to replace it every year).

That assumes they could afford the upfront cost of the 'better' light in the first place. People living in a precarious financial state don't have the funds to invest in the future. Buy a good light and something else doesn't get bought. Or buy a cheap light and the other thing. So it's going to cost them more in the future because they'll buy many cheap lights instead of one good one, but when you don't have the funds to invest that's what happens. Same applies to food (and everything else) - they pay more for their food because they can only afford single-portion sizes and/or poor quality, whereas someone with money will buy in bulk and save overall.

[nctnico]

Last I heard, a fundamental thermal problem with LED bulb-shaped objects that fail prematurely is that the heat-producing and thermally-sensitive electronics are mounted near the base while the actual LEDs are spaced apart in the bulbous globe.

This isn't a problem for LED lamps specifically. Some halogen lamps also suffer from this problem leading to extremely short lives for the lamps.

What worries me more is that some LED lamps (or better put: light fixtures) can not be seperated from their fitting at all. So if the light fixture fails, it needs to be replaced entirely instead of just changing the lamp part.