EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405

Arfa

On 02/01/2015 14:56, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa

There is a shop in town that is all LED strip lighting. So far no
failures noticed, but I will estimate the total number of LEDs and
monitor over time. I would guess , sitting here, something like 20,000
5mm LEDs in total (not the higher powered types). I feel like running a
red/orange felt-tip along all
the cover-strips of the LED runs, as its that horrible stark blue-white.
Doesn't take much pen ink to give a much warmer tone (to CFL bulbs anyway)

This article has soldering as a dying art along with trug making etc

http://www.theguardian.com/culture/2...heritage-craft

perhaps they meant soldering as in tin-smithing.
But on the other hand how much hands-on soldering rework of modern day
mass electronics production is there?

N_Cook wrote:

This article has soldering as a dying art along with trug making etc

http://www.theguardian.com/culture/2...heritage-craft

perhaps they meant soldering as in tin-smithing.

http://i.ebayimg.com/00/s/NDUzWDY0MA==/z/CMgAAOxyLm9TDdVl/$_12.JPG

But on the other hand how much hands-on soldering rework of modern day
mass electronics production is there?

Probably just rework on any dead items at the end of the line (if has
sufficient value) or attaching the big lumpy stuff like connectors and
transformers.

"N_Cook" wrote in message
...
On 02/01/2015 14:56, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa

There is a shop in town that is all LED strip lighting. So far no failures
noticed, but I will estimate the total number of LEDs and monitor over
time. I would guess , sitting here, something like 20,000 5mm LEDs in
total (not the higher powered types). I feel like running a red/orange
felt-tip along all
the cover-strips of the LED runs, as its that horrible stark blue-white.
Doesn't take much pen ink to give a much warmer tone (to CFL bulbs anyway)

This article has soldering as a dying art along with trug making etc
http://www.theguardian.com/culture/2...heritage-craft

Just more mindlessly silly journalism.

There might be only one person still making denim
on that soggy little island, but plenty in china.

perhaps they meant soldering as in tin-smithing.

Corse they did and they're wrong about that too.

But on the other hand how much hands-on soldering rework of modern day
mass electronics production is there?

Quite a bit actually. Forget what they were making but
someone posted a link to a youtube of a factory in china
making something like a video camera or something.

Lots of hand soldering going on.

On 02/01/2015 15:19, N_Cook wrote:

There is a shop in town that is all LED strip lighting. So far no
failures noticed, but I will estimate the total number of LEDs and
monitor over time.

Our office has been up-and-running for less than 6 months. I don't know
when it was fitted out, but the LED lights in the conference rooms are
on clever circuits so they come on only when they are used - and our
floor was empty.

One of them has died. Not the LEDs, but the driver - it's a square
luminaire with LEDs on each side, and one side flashed.

Andy

"Vir Campestris" wrote in message
o.uk...
On 02/01/2015 15:19, N_Cook wrote:

There is a shop in town that is all LED strip lighting. So far no
failures noticed, but I will estimate the total number of LEDs and
monitor over time.

Our office has been up-and-running for less than 6 months. I don't know
when it was fitted out, but the LED lights in the conference rooms are on
clever circuits so they come on only when they are used - and our floor
was empty.

One of them has died. Not the LEDs, but the driver - it's a square
luminaire with LEDs on each side, and one side flashed.

Andy


A while back, they changed the car park lights to LED at my local Co-op
store. One of them started flashing in fairly short order, and has continued
to do so ever since, despite my repeatedly bringing it to their attention. I
suppose in the days of the metal halide lights that they had previously, the
maintenance guy would just have shinned up a ladder and replaced the lamp in
the fitting. To sort out this LED one would require the entire unit to be
unbolted from the wall, and replaced. As to the failure on the one in your
place being the driver rather than the LEDs themselves, I think that goes
along with what the guy is suggesting in the EE Times article.

Arfa

Well back in the days when i could see, I used to build stuff and used
solder, but increasingly even then automation for mass produced products
meant that surface mount and wire wrap techniques abounded. I'd imagine with
the banning of leaded solder its more likely that machines can make better
joints than humans even if they could use soldering stations easily.
Back then the only folk still soldering in industry were the development
types. If a pcb failed in test, then it was taken out junked and replaced.
Gone are the days when simple hands on component substitution was cost
effective it seems.
As for led lamps, I imagine they are still in the early stages of
evaluation. Nobody has had them in service long enough to refine colour temp
and reliability methods. It will probably only get better though as the
efficiency is so much better than alternatives, they are going to make it
work.
Brian

--
From the Sofa of Brian Gaff Reply address is active
"N_Cook" wrote in message
...
On 02/01/2015 14:56, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa

There is a shop in town that is all LED strip lighting. So far no failures
noticed, but I will estimate the total number of LEDs and monitor over
time. I would guess , sitting here, something like 20,000 5mm LEDs in
total (not the higher powered types). I feel like running a red/orange
felt-tip along all
the cover-strips of the LED runs, as its that horrible stark blue-white.
Doesn't take much pen ink to give a much warmer tone (to CFL bulbs anyway)

This article has soldering as a dying art along with trug making etc

http://www.theguardian.com/culture/2...heritage-craft

perhaps they meant soldering as in tin-smithing.
But on the other hand how much hands-on soldering rework of modern day
mass electronics production is there?

On the subject - what *is* the best way to drive LEDs?

Seems to me that whilst convenient and in line with my earlier comments
on standardisation, putting little 230V PSUs in every lamp that get hot
and blow up is not the best way forward.

Does a 12V supply offer any advantages in terms of minimising on board
electronics? 12V SELV is at least standard.

If an LED has a Vf (forward voltage drop) of x volts, is it considered
good form to put 12/x LEDs in series across the supply with no other
limiting circuitry?

Or is there a really simple 2 pin current regulator on a chip available?

Old style 0.2" 20mA LEDs weren't that bothered, but I'm not au fait with
high power Crees and the like.

On 03/01/15 10:13, Tim Watts wrote:
On the subject - what *is* the best way to drive LEDs?

Seems to me that whilst convenient and in line with my earlier comments
on standardisation, putting little 230V PSUs in every lamp that get hot
and blow up is not the best way forward.

Does a 12V supply offer any advantages in terms of minimising on board
electronics? 12V SELV is at least standard.

If an LED has a Vf (forward voltage drop) of x volts, is it considered
good form to put 12/x LEDs in series across the supply with no other
limiting circuitry?

Or is there a really simple 2 pin current regulator on a chip available?

Old style 0.2" 20mA LEDs weren't that bothered, but I'm not au fait with
high power Crees and the like.


Anyone?

"Tim Watts" wrote in message
...
On 03/01/15 10:13, Tim Watts wrote:
On the subject - what *is* the best way to drive LEDs?

Seems to me that whilst convenient and in line with my earlier comments
on standardisation, putting little 230V PSUs in every lamp that get hot
and blow up is not the best way forward.

Does a 12V supply offer any advantages in terms of minimising on board
electronics? 12V SELV is at least standard.

If an LED has a Vf (forward voltage drop) of x volts, is it considered
good form to put 12/x LEDs in series across the supply with no other
limiting circuitry?

Or is there a really simple 2 pin current regulator on a chip available?

Old style 0.2" 20mA LEDs weren't that bothered, but I'm not au fait with
high power Crees and the like.

Anyone?

www.google.com/search?q=cree+teardown

Brian Gaff wrote

Well back in the days when i could see, I used to build stuff and used
solder, but increasingly even then automation for mass produced products
meant that surface mount and wire wrap techniques abounded. I'd imagine
with the banning of leaded solder its more likely that machines can make
better joints than humans even if they could use soldering stations
easily.

They still do in china as the recently posted youtube
video of one of the production lines in china shows.

Back then the only folk still soldering in industry were the development
types.

Not in china.

If a pcb failed in test, then it was taken out junked and replaced.

Not in china.

Gone are the days when simple hands on component substitution was cost
effective it seems.

Not in china.

As for led lamps, I imagine they are still in the early stages of
evaluation.

Particularly with the higher powered ones.

More production than evaluation tho.

Nobody has had them in service long enough to refine colour temp and
reliability methods.

I doubt that is true of Cree.

It will probably only get better though

Absolutely certainly.

as the efficiency is so much better than alternatives, they are going to
make it work.

They already have with quite a bit of LED stuff.

"N_Cook" wrote in message
...
On 02/01/2015 14:56, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa

There is a shop in town that is all LED strip lighting. So far no
failures noticed, but I will estimate the total number of LEDs and
monitor over time. I would guess , sitting here, something like 20,000
5mm LEDs in total (not the higher powered types). I feel like running a
red/orange felt-tip along all
the cover-strips of the LED runs, as its that horrible stark blue-white.
Doesn't take much pen ink to give a much warmer tone (to CFL bulbs
anyway)

This article has soldering as a dying art along with trug making etc

http://www.theguardian.com/culture/2...heritage-craft

perhaps they meant soldering as in tin-smithing.
But on the other hand how much hands-on soldering rework of modern day
mass electronics production is there?

On 02/01/2015 14:56, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa


The electronics don't have 100 components as stated and as LED lighting
becomes more widespread the support circuits will be integrated into
fewer components.

While the support electronics may be less reliable than the LED itself
other factors such as running the devices at elevated temperatures is
more likely to shorten the life. Elevated temperatures will occur if
the bulb is in an inappropriate fitting (inappropriate for a LED bulb)

If you applied the authors logic to LED backlit TVs 1 in 4 would have
already failed.


--
mailto: news {at} admac {dot] myzen {dot} co {dot} uk

His basic premise makes sense -- more components = lower reliability -- but
the fact is that one can easily find electronic devices 50 and 60 years old
that have never been serviced that continue to work. Members of this group
probably own them.

A failure rate of 1 in 10,000 (per year?) strikes me as unduly pessimistic,
even for devices used in vacuum tube equipment. Solid-state -- which almost
always operates at lower voltages and temperature -- should be even more
reliable.

I own devices that contain far more components than an LED bulb -- yet they do
not drop like flies. Flat-panel TVs are a good example. CU says the reported
breakdown rate is extremely low (3% for a few brands is on the high end), and
advises against purchasing service contracts.

Though the point he raises is valid, and not only deserves, but requires,
study, you can't assume these products fail prematurely simply because they
contain "too many parts". Why they failed is more important. I wouldn't be
surprised if it was due to SMD soldering failure.

William Sommerwerck wrote:

His basic premise makes sense -- more components = lower reliability -- but
the fact is that one can easily find electronic devices 50 and 60 years old
that have never been serviced that continue to work. Members of this group
probably own them.

A failure rate of 1 in 10,000 (per year?) strikes me as unduly pessimistic,
even for devices used in vacuum tube equipment. Solid-state -- which almost
always operates at lower voltages and temperature -- should be even more
reliable.

I own devices that contain far more components than an LED bulb -- yet they do
not drop like flies. Flat-panel TVs are a good example. CU says the reported
breakdown rate is extremely low (3% for a few brands is on the high end), and
advises against purchasing service contracts.

Though the point he raises is valid, and not only deserves, but requires,
study, you can't assume these products fail prematurely simply because they
contain "too many parts". Why they failed is more important. I wouldn't be
surprised if it was due to SMD soldering failure.


I see a lot of LED traffic lights with groups of dead LEDs. The one
closest to me is on the fourth red LED lamp in the last couple years and
already has some groups that are quite dim.


--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.

"Michael A. Terrell" wrote in message
...

William Sommerwerck wrote:

His basic premise makes sense -- more components = lower reliability --
but
the fact is that one can easily find electronic devices 50 and 60 years
old
that have never been serviced that continue to work. Members of this
group
probably own them.

A failure rate of 1 in 10,000 (per year?) strikes me as unduly
pessimistic,
even for devices used in vacuum tube equipment. Solid-state -- which
almost
always operates at lower voltages and temperature -- should be even more
reliable.

I own devices that contain far more components than an LED bulb -- yet
they do
not drop like flies. Flat-panel TVs are a good example. CU says the
reported
breakdown rate is extremely low (3% for a few brands is on the high end),
and
advises against purchasing service contracts.

Though the point he raises is valid, and not only deserves, but requires,
study, you can't assume these products fail prematurely simply because
they
contain "too many parts". Why they failed is more important. I wouldn't
be
surprised if it was due to SMD soldering failure.

I see a lot of LED traffic lights with groups of dead LEDs.

Can't say I have ever seen even one.

The one closest to me is on the fourth red LED lamp in the last
couple years and already has some groups that are quite dim.

Presumably just a very bad design.

On Fri, 02 Jan 2015 19:51:18 -0500, "Michael A. Terrell"
wrote:

I see a lot of LED traffic lights with groups of dead LEDs.

I haven't. Locally, we have some LED traffic and street lights. I
have yet to see one malfunction. However, it might be simply because
the traffic department is good about quickly replacing any failures.
Most LED lights include a remote monitoring feature.

I'm not sure what might be causing the failures that you've observed.
Any sign of overheating? Power glitches? Bullet holes?

Los Angeles Saves Millions With LED Street Light Deployment
http://www.forbes.com/sites/justingerdes/2013/01/25/los-angeles-saves-millions-with-led-street-light-deployment/
After 36 months of initial operation, for instance, high-intensity
discharge (HID) fixtures in Los Angeles recorded an average failure
rate of 10%; the average failure rate for LED fixtures, according
to the latest figures, is 0.2% (189 of 98,000 installed).

The one
closest to me is on the fourth red LED lamp in the last couple years and
already has some groups that are quite dim.

Got an IR temperature gun? Get as close as you can and get a
temperature reading. My guess(tm) is that it's running hot, even with
some blown lights.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

Jeff Liebermann wrote:

On Fri, 02 Jan 2015 19:51:18 -0500, "Michael A. Terrell"
wrote:

I see a lot of LED traffic lights with groups of dead LEDs.

I haven't. Locally, we have some LED traffic and street lights. I
have yet to see one malfunction. However, it might be simply because
the traffic department is good about quickly replacing any failures.
Most LED lights include a remote monitoring feature.

I'm not sure what might be causing the failures that you've observed.
Any sign of overheating? Power glitches? Bullet holes?


This is Central Florida, so it is hot, and the electricity is crap,
since it went from Florida Power, to Progress Energy, then to Duke
Energy. They are specified and maintained by FDOT, or one of their
contractors. I've seen a few spots where they went back to the
incandescent lamp and colored lens.


Los Angeles Saves Millions With LED Street Light Deployment
http://www.forbes.com/sites/justingerdes/2013/01/25/los-angeles-saves-millions-with-led-street-light-deployment/
After 36 months of initial operation, for instance, high-intensity
discharge (HID) fixtures in Los Angeles recorded an average failure
rate of 10%; the average failure rate for LED fixtures, according
to the latest figures, is 0.2% (189 of 98,000 installed).

The one
closest to me is on the fourth red LED lamp in the last couple years and
already has some groups that are quite dim.

Got an IR temperature gun? Get as close as you can and get a
temperature reading. My guess(tm) is that it's running hot, even with
some blown lights.


No, all I have is the small fob type of contactless IR thermometer.


--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.

On 02/01/2015 19:22, William Sommerwerck wrote:
His basic premise makes sense -- more components = lower reliability --
but the fact is that one can easily find electronic devices 50 and 60
years old that have never been serviced that continue to work. Members
of this group probably own them.

A failure rate of 1 in 10,000 (per year?) strikes me as unduly
pessimistic, even for devices used in vacuum tube equipment. Solid-state
-- which almost always operates at lower voltages and temperature --
should be even more reliable.

I own devices that contain far more components than an LED bulb -- yet
they do not drop like flies. Flat-panel TVs are a good example. CU says
the reported breakdown rate is extremely low (3% for a few brands is on
the high end), and advises against purchasing service contracts.

Though the point he raises is valid, and not only deserves, but
requires, study, you can't assume these products fail prematurely simply
because they contain "too many parts". Why they failed is more
important. I wouldn't be surprised if it was due to SMD soldering failure.


The Haswell-E die is composed of 2.6 billion transistors. You have to
achieve phenomenal component reliability for any of them to work as they
leave the factory, let alone years later! Adding up everything in a
typical modern PC including the display leads to even huger numbers of
components. Yet we see many of them struggle on for many years until
they are replaced, all too often, due to inadequate computing power (or
not being able to justify the complete re-install of an updated OS on an
old box) rather than component failure.

--
Rod

On 03/01/15 13:03, polygonum wrote:
On 02/01/2015 19:22, William Sommerwerck wrote:
His basic premise makes sense -- more components = lower reliability --
but the fact is that one can easily find electronic devices 50 and 60
years old that have never been serviced that continue to work. Members
of this group probably own them.

A failure rate of 1 in 10,000 (per year?) strikes me as unduly
pessimistic, even for devices used in vacuum tube equipment. Solid-state
-- which almost always operates at lower voltages and temperature --
should be even more reliable.

I own devices that contain far more components than an LED bulb -- yet
they do not drop like flies. Flat-panel TVs are a good example. CU says
the reported breakdown rate is extremely low (3% for a few brands is on
the high end), and advises against purchasing service contracts.

Though the point he raises is valid, and not only deserves, but
requires, study, you can't assume these products fail prematurely simply
because they contain "too many parts". Why they failed is more
important. I wouldn't be surprised if it was due to SMD soldering
failure.


The Haswell-E die is composed of 2.6 billion transistors. You have to
achieve phenomenal component reliability for any of them to work as they
leave the factory, let alone years later! Adding up everything in a
typical modern PC including the display leads to even huger numbers of
components. Yet we see many of them struggle on for many years until
they are replaced, all too often, due to inadequate computing power (or
not being able to justify the complete re-install of an updated OS on an
old box) rather than component failure.

The deeper analysis asks the question 'what fails, and why?'

In general a chip once made, wont degrade catastrophically. Its
thermally stable, and any manufacturing faults show up on test or early on.

Yes, RAM and other chops do age, but there is a wide tolerance before
they go so far out of spec they are useless.

By far the greatest killer is heat: heat accelerates ageing., death
occurs in microseconds at 180C, decades at 30C



--
Everything you read in newspapers is absolutely true, except for the
rare story of which you happen to have first-hand knowledge. €“ Erwin Knoll

"polygonum" wrote in message ...

The Haswell-E die is composed of 2.6 billion transistors. You have
to achieve phenomenal component reliability for any of them to
work as they leave the factory, let alone years later!

One might argue that all the transistors are created simultaneously in a
single processing sequence, and that the chip is, technically, a single
component.

On 03/01/15 15:03, William Sommerwerck wrote:
"polygonum" wrote in message ...

The Haswell-E die is composed of 2.6 billion transistors. You have
to achieve phenomenal component reliability for any of them to
work as they leave the factory, let alone years later!

One might argue that all the transistors are created simultaneously in a
single processing sequence, and that the chip is, technically, a single
component.
In general its also a tested component.

so it works to spec or it doesn't.

*Failure* of a system that worked to start with implies change over
time: The relevant point is what changes happen to circuits over time
and what law if any, they follow.



--
Everything you read in newspapers is absolutely true, except for the
rare story of which you happen to have first-hand knowledge. €“ Erwin Knoll

On 03/01/2015 15:03, William Sommerwerck wrote:
"polygonum" wrote in message ...

The Haswell-E die is composed of 2.6 billion transistors. You have
to achieve phenomenal component reliability for any of them to
work as they leave the factory, let alone years later!

One might argue that all the transistors are created simultaneously in a
single processing sequence, and that the chip is, technically, a single
component.

One might argue that is the case for the 100-component circuit referred
to in the EE Times article. I simply do not know what goes on in the
making of "an LED plus its associated circuitry". Is it made as one,two
or 101 components?

--
Rod

"polygonum" wrote in message ...
On 03/01/2015 15:03, William Sommerwerck wrote:

One might argue that all the transistors are created simultaneously in a
single processing sequence, and that the chip is, technically, a single
component.

One might argue that is the case for the 100-component circuit referred to
in the EE Times article.

One //might not// argue that. The LED lamp is made of discrete components that
are manufactured separately, and individually soldered in place.

"Huge" wrote in message ...
On 2015-01-02, William Sommerwerck wrote:

His basic premise makes sense -- more components = lower reliability
-- but the fact is that one can easily find electronic devices 50 and 60
years old that have never been serviced that continue to work. Members
of this group probably own them.

This is a category error. Yes, we all have 'n' year-old electronic devices,
because we have thrown away the ones that have failed.

That's logically correct. But I have 40 to 50 year old Sony and KLH products
that work fine. Whereas Sony stuff from the last 15 years is gradually falling
apart.

On 01/03/2015 7:00 AM, William Sommerwerck wrote:
"Huge" wrote in message ...
On 2015-01-02, William Sommerwerck wrote:

His basic premise makes sense -- more components = lower reliability
-- but the fact is that one can easily find electronic devices 50 and 60
years old that have never been serviced that continue to work. Members
of this group probably own them.

This is a category error. Yes, we all have 'n' year-old electronic
devices,
because we have thrown away the ones that have failed.

That's logically correct. But I have 40 to 50 year old Sony and KLH
products that work fine. Whereas Sony stuff from the last 15 years is
gradually falling apart.

And I have a lovely Phillips reel-to-reel tape player (1960s) that has
had all the internal drive belts and idler tires turn to goo...a real
pain to get running again! If I ever get around to it. The electronics
all appear just fine.

Not to mention one of my first jobs in the 60s was scrapping rotted foam
sound insulation from the inside cases of IBM punch card printers, etc.
- around 1967 as I recall.

So, what can fail often has little to do with electronics, there is all
that support stuff that goes bad after its BBD (Best Before Date).

John :-#)#

--
(Please post followups or tech inquiries to the newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

Huge wrote
William Sommerwerck wrote

His basic premise makes sense -- more components = lower
reliability -- but the fact is that one can easily find electronic
devices 50 and 60 years old that have never been serviced that
continue to work. Members of this group probably own them.

This is a category error.

No.

Yes, we all have 'n' year old electronic devices, because
we have thrown away the ones that have failed.

And yet cars are in fact MUCH more reliable now even tho
they have a lot more components than they used to have.

Same with other stuff like TVs etc too.

Rod Speed udtrykte præcist:
Huge wrote
William Sommerwerck wrote

His basic premise makes sense -- more components = lower reliability --
but the fact is that one can easily find electronic devices 50 and 60
years old that have never been serviced that continue to work. Members of
this group probably own them.

This is a category error.

No.

Yes, we all have 'n' year old electronic devices, because we have thrown
away the ones that have failed.

And yet cars are in fact MUCH more reliable now even tho they have a lot more
components than they used to have.

My Yaris ran on 3 cylinders. First I changed spark plugs, because that
what usually worked on old cars. But even with a gap of 1.5mm, the
spark was fine.
So I hooked up the OBD-2 reader: Engine misfire cylinder 1.
I exchanged two "spark plug caps" which is really the ignition coil and
some electronics, one unit per cylinder.
Now "Engine misfire cylinder 3"
So a "new" used unit from a junk yard gotthe car running again.
The car only had run 460000km, not sure when, if ever, the spark plugs
had been changed.

So new cars are MUCH more reliable.

Leif

--
Husk kørelys bagpå, hvis din bilfabrikant har taget den idiotiske
beslutning at undlade det.

Leif Neland wrote
Rod Speed wrote
Huge wrote
William Sommerwerck wrote

His basic premise makes sense -- more components = lower
reliability -- but the fact is that one can easily find electronic
devices 50 and 60 years old that have never been serviced that
continue to work. Members of this group probably own them.

This is a category error.

No.

Yes, we all have 'n' year old electronic devices, because
we have thrown away the ones that have failed.

And yet cars are in fact MUCH more reliable now even tho
they have a lot more components than they used to have.

My Yaris ran on 3 cylinders. First I changed spark
plugs, because that what usually worked on old cars.
But even with a gap of 1.5mm, the spark was fine.

Because of all those other components that old cars didn't have.

I haven't even bothered to change mine after 10 years.

So I hooked up the OBD-2 reader: Engine misfire cylinder 1.
I exchanged two "spark plug caps" which is really the
ignition coil and some electronics, one unit per cylinder.

Very unusual way to do things.

Now "Engine misfire cylinder 3"
So a "new" used unit from a junk yard gotthe car
running again. The car only had run 460000km, not
sure when, if ever, the spark plugs had been changed.

So new cars are MUCH more reliable.

Yep. Because they have vastly more components.

Rod Speed sendte dette med sin computer:
Leif Neland wrote

So I hooked up the OBD-2 reader: Engine misfire cylinder 1.
I exchanged two "spark plug caps" which is really the ignition coil and
some electronics, one unit per cylinder.

Very unusual way to do things.


Here is the product page:

http://www.thansen.dk/product.asp?pn=-831477966

Price is USD 160, I got an used one for 24USD.

Alas, now it needs a new back door, because wife didn't see I parked
the Landcruiser outside the garage :-(

Leif

--
Husk kørelys bagpå, hvis din bilfabrikant har taget den idiotiske
beslutning at undlade det.

On 05/01/2015 20:57, Leif Neland wrote:
The car only had run 460000km, not sure when, if ever, the spark plugs
had been changed.

That's pretty good - nearly half a million on one set of plugs.

Andy

On 1/2/2015 6:56 AM, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa
I've never had a CFL failure that I could trace to the CFL.
It's always the electronics driving it. A power glitch can take
out a LED just as easily as a CFL.

Lifetime guarantee. If you can find where to send it and pay
more in shipping it back than the cost of replacement.
And it's not the lifetime of the light. It's the lifetime
of the supplier.

Home lamps are designed for cheap,cheap,cheap,
not for reliability. But it will get better over time.

I always giggle at flashlights that say you never replace the LED,
over 20,000 hour life, etc.
The elephant in the room is that in most cheap designs, they
put the LED directly across three batteries and hope the battery
resistance increases faster than the time it takes to melt the light.
And the better designs have a dozen high-stress parts in
some regulator.

I sawed a free harbor freight flashlight in half and soldered it
back together to make an extension. Put an 18650 in an extended light.
Really bright. Replacing three 1.5V with one 4.2V oughta work, right?
Not for long...but far longer than I expected. Put one ohm in series
and it looks like life might be ok. ;-) Still pretty bright.
There's a reason they ship with heavy duty batteries.

Home Depot has some really great deals if you watch closely.
I bought 44 40W LED's when they were 4 for $5.05 and replaced
most of my CFL. Turns out that two 40's in a Y-adapter was
way cheaper than a 60W and had better light dispersion.
Then they had 60W at 3 for $6.97 to replace 12 of the 40's.
Then they had 75W at ONE CENT. I bought all 8 of them to replace
some of the 60's.

There's no possible financial reason for doing that.
I tell myself that the LED's should last longer in applications
where you switch them on and off a lot. But I put LED's in fixtures
I haven't turned on in years. Don't judge me...I'm saving the planet...

In the attic, I have about a hundred used incandescents, right
next to the hundred CFL's, right next to all the spare LED's
I bought but never used.
Right next to the box of flashlights I couldn't help buying.
And I still mine Home Depot for LED deals. I need to go
to rehab for LED abuse.

Often we do stuff because we can, not because we should.

If I remember correctly, if you combine the MTBF of all the components,
the probability of system failure within the system MTBF is 37%.

Are we having fun yet?

In sci.electronics.repair mike wrote:
On 1/2/2015 6:56 AM, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa
I've never had a CFL failure that I could trace to the CFL.
It's always the electronics driving it. A power glitch can take
out a LED just as easily as a CFL.

I've had a couple CFLs burn out at the bulb from old age. Most had
electronic failures, or jsut broke.

The longest lasting CFL may be one in my bathroom with 10 years of
on-off use. I'd have to remove it to checked the purchase date I write on
them though.

In sci.electronics.repair Cydrome Leader wrote:
In sci.electronics.repair mike wrote:
On 1/2/2015 6:56 AM, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa
I've never had a CFL failure that I could trace to the CFL.
It's always the electronics driving it. A power glitch can take
out a LED just as easily as a CFL.

I've had a couple CFLs burn out at the bulb from old age. Most had
electronic failures, or jsut broke.

The longest lasting CFL may be one in my bathroom with 10 years of
on-off use. I'd have to remove it to checked the purchase date I write on
them though.

I've had a few CFL's fail due to electronics failure, but the newer
bulbs last longer. Unfortunately as they age they get dimmer. Looking
at an aged bulb I noticed that the phosphor layer has turned a brownish
color.

I did have one failure that when I took it apart, 1 of the wires to to
the flourescent was disconnected & the others were questionable.
Soldered them all & re-assembled the bulb and it's still in use in the
basement bathroom.

In sci.electronics.repair Jerry Peters wrote:
In sci.electronics.repair Cydrome Leader wrote:
In sci.electronics.repair mike wrote:
On 1/2/2015 6:56 AM, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa
I've never had a CFL failure that I could trace to the CFL.
It's always the electronics driving it. A power glitch can take
out a LED just as easily as a CFL.

I've had a couple CFLs burn out at the bulb from old age. Most had
electronic failures, or jsut broke.

The longest lasting CFL may be one in my bathroom with 10 years of
on-off use. I'd have to remove it to checked the purchase date I write on
them though.

I've had a few CFL's fail due to electronics failure, but the newer
bulbs last longer. Unfortunately as they age they get dimmer. Looking
at an aged bulb I noticed that the phosphor layer has turned a brownish
color.

I'd have to agree that the newer ones seem a bit better. They've figured
out how make them as simple as possible now is my guess.

I did have one failure that when I took it apart, 1 of the wires to to
the flourescent was disconnected & the others were questionable.
Soldered them all & re-assembled the bulb and it's still in use in the
basement bathroom.

they're still useless for outdoor cold use. The ones outside by the stairs
are like nightlights when the temps drop to freezing. It's amazing they
even start.

I'm not a fan of the warm up period they take, even indoors.

"Cydrome Leader" wrote in message
...
In sci.electronics.repair Jerry Peters wrote:
In sci.electronics.repair Cydrome Leader wrote:
In sci.electronics.repair mike wrote:
On 1/2/2015 6:56 AM, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa
I've never had a CFL failure that I could trace to the CFL.
It's always the electronics driving it. A power glitch can take
out a LED just as easily as a CFL.

I've had a couple CFLs burn out at the bulb from old age. Most had
electronic failures, or jsut broke.

The longest lasting CFL may be one in my bathroom with 10 years of
on-off use. I'd have to remove it to checked the purchase date I write
on
them though.

I've had a few CFL's fail due to electronics failure, but the newer
bulbs last longer. Unfortunately as they age they get dimmer. Looking
at an aged bulb I noticed that the phosphor layer has turned a brownish
color.

I'd have to agree that the newer ones seem a bit better. They've figured
out how make them as simple as possible now is my guess.

I did have one failure that when I took it apart, 1 of the wires to to
the flourescent was disconnected & the others were questionable.
Soldered them all & re-assembled the bulb and it's still in use in the
basement bathroom.

they're still useless for outdoor cold use. The ones outside by the stairs
are like nightlights when the temps drop to freezing. It's amazing they
even start.

I'm not a fan of the warm up period they take, even indoors.

I like it with the bedhead light. You don’t get the same
blinding effect when you turn it on in the dark.

Cydrome Leader wrote:
In sci.electronics.repair Jerry wrote:
In sci.electronics.repair Cydrome wrote:
In sci.electronics.repair wrote:
On 1/2/2015 6:56 AM, Arfa Daily wrote:
EE Times article that came to me by email today

http://www.electronics-eetimes.com/e...s_id=222923405


Arfa
I've never had a CFL failure that I could trace to the CFL.
It's always the electronics driving it. A power glitch can take
out a LED just as easily as a CFL.

I've had a couple CFLs burn out at the bulb from old age. Most had
electronic failures, or jsut broke.

The longest lasting CFL may be one in my bathroom with 10 years of
on-off use. I'd have to remove it to checked the purchase date I write on
them though.

I've had a few CFL's fail due to electronics failure, but the newer
bulbs last longer. Unfortunately as they age they get dimmer. Looking
at an aged bulb I noticed that the phosphor layer has turned a brownish
color.

I'd have to agree that the newer ones seem a bit better. They've figured
out how make them as simple as possible now is my guess.

I did have one failure that when I took it apart, 1 of the wires to to
the flourescent was disconnected& the others were questionable.
Soldered them all& re-assembled the bulb and it's still in use in the
basement bathroom.

they're still useless for outdoor cold use. The ones outside by the stairs
are like nightlights when the temps drop to freezing. It's amazing they
even start.

I'm not a fan of the warm up period they take, even indoors.

Interesting, my outside CFLs are over 35yrs old and start down to -8C
without fail. My newer CFLs, die like flies in comparison, particularly
in glass globe fittings. I reckon some CFLs have as short a life as
filament bulbs, even when running cool..

"Cydrome Leader" wrote in message ...

I'm not a fan of the warm up period they take, even indoors.

Recent models come to full brightness quickly.

On 02/01/2015 21:09, Jerry Peters wrote:


I've had a few CFL's fail due to electronics failure,


+1
http://www.admac.myzen.co.uk/bulb/


--
mailto: news {at} admac {dot] myzen {dot} co {dot} uk

mike wrote:

I've never had a CFL failure that I could trace to the CFL.


** Really? That puts you in a minority of one.

The most common CFL failure is the tube heaters going open - most are rated for only 1 to 2 thousand switch on cycles. Ones that light up quickly and tolerate low temps better seem to be the poorest at this.

All other failures I have seen are the result of the electronics section running too hot causing capacitor failure, particularly electro and EMI suppression caps.

A great many ceiling fixtures are not well suited to CFL or LED lamps cos they trap heat.


It's always the electronics driving it.

** Not true in general.

A power glitch can take
out a LED just as easily as a CFL.

** Power glitch = what ??

A line voltage spike lasting less than a mS or a hit by lighting ?

The typically 4 to 22uF electro in the DC supply absorbs brief spikes nicely and most indirect effects of lightning too.

..... Phil

On Friday, January 2, 2015 at 11:29:24 AM UTC-8, mike wrote:

In the attic, I have about a hundred used incandescents, right
next to the hundred CFL's, right next to all the spare LED's
I bought but never used.

Use up the CFL's. Those have a small amount of radioactive fill
gas as their 'starter' and it has a half-life. It lasts maybe one
decade.