Cheap LED Shop lights

[slow-poke]

I’ve had these for 3+ years and had one stop working. Turned out it was a loose connection.

When they start to fail, one section of LED's starts flashing erratically, a few weeks later the entire assembly goes dark. I have these in the garage, and I just looked (lost another one), down to 6 of 16 they are less than 3 years old.

 

[Dan Dubeau]

When they start to fail, one section of LED's starts flashing erratically, a few weeks later the entire assembly goes dark. I have these in the garage, and I just looked (lost another one), down to 6 of 16 they are less than 3 years old.

How many hours would you figure you got form them before failure?

 

[JustaDB]

I've had multiple failures of LED bulbs in the house, these are the standard screw in type.

Us, as well. I've taken to writing the install date on 'em w/ a Sharpie & have saved the receipts. 'Course, none have failed since I started doing that...

 

[DPittman]

Us, as well. I've taken to writing the install date on 'em w/ a Sharpie & have saved the receipts. 'Course, none have failed since I started doing that...

Yup me too, I'm P'O at the lies the manufactures make on the life span of the lights. MAYBE in a vacuum of space where the lights are never turned off and on might they approach the life span they claim but who realistically ever has those ideal conditions? The compact fluorescent bulbs were even worse for life span for me.
Rant over, now back to my regular Easter morning activities .

 

[slow-poke]

How many hours would you figure you got form them before failure?

Not many, because the garage is mostly just used to park the cars and not a lot else, so I would guess an hour a day tops on average, so perhaps 1000 hours.

 

[Dan Dubeau]

Not many, because the garage is mostly just used to park the cars and not a lot else, so I would guess an hour a day tops on average, so perhaps 1000 hours.

Thanks. That's not a very good sign. I initially purchased these ones base on the reviews, the bad ones, as they tell a more accurate story I think. Aside from a few early failures, they seemed to have better longevity than other similar lights.

Mine are sometimes 3-4 hours a night through the week, and usually all day on the weekends with a few off here and there. No failures or partial or complete yet, but I'll report when they do. Hopefully a long time in the future. I think is was October 2023 when I put them up.

 

[Stuart Samuel]

I can speak a little bit to LED lifespan, from a professional perspective. When an LED, or product with LED’s in it, claims a certain lifespan, even if it’s from a major, up front manufacturer, they’re typically stating the duration after which the lamps were producing either 70% or more commonly 50% of their original output (typically meaning 30% or 50% of the chips are dead).

Heat is a major factor. Run the LEDs at a lower intensity, they produce less heat and last longer. (My neighbour who runs a glassblowing shop has issues with this, unsurprisingly, because of the high ambient temperature)

At work we will often buy higher intensity LEDs and run them at 75% (give or take) of their maximum output, to ensure greater lifespan. (With applications like church lighting, ie high ceilings ensuring higher ambient temperatures, and difficulty in accessing for replacement, we try to overbuild)

I am far from an expert on this. Not to throw him under a bus (of sorts) but if anyone wants to pick my coworker Dann’s brains about this, he’ll probably be at the Ontario meet-up on Saturday. He runs our assembly department, and just enjoys electronics, generally. (Worked at Active Surplus downtown for years)

 

[JustaDB]

Heat is a major factor. Run the LEDs at a lower intensity, they produce less heat and last longer.

Ah. So, during the cold streak this winter when it was 26F in my shop, I actually saved LED life. Good to know...

 

[Stuart Samuel]

I can speak a little bit to LED lifespan, from a professional perspective. When an LED, or product with LED’s in it, claims a certain lifespan, even if it’s from a major, up front manufacturer, they’re typically stating the duration after which the lamps were producing either 70% or more commonly 50% of their original output (typically meaning 30% or 50% of the chips are dead).

Heat is a major factor. Run the LEDs at a lower intensity, they produce less heat and last longer. (My neighbour who runs a glassblowing shop has issues with this, unsurprisingly, because of the high ambient temperature)

At work we will often buy higher intensity LEDs and run them at 75% (give or take) of their maximum output, to ensure greater lifespan. (With applications like church lighting, ie high ceilings ensuring higher ambient temperatures, and difficulty in accessing for replacement, we try to overbuild)

I am far from an expert on this. Not to throw him under a bus (of sorts) but if anyone wants to pick my coworker Dann’s brains about this, he’ll probably be at the Ontario meet-up on Saturday. He runs our assembly department, and just enjoys electronics, generally. (Worked at Active Surplus downtown for years)

Further thoughts:
Manufactures aren’t running their chips for 50,000 hours to determine lifespan. They’re doing the math and running them hotter, with an understanding of how much that should accelerate the failure. (Sort of like salt spray testing is used to determine durability of coatings, to accelerate corrosion)

In more demanding applications, it’s not uncommon for a client to specify a certain minimum lumen output after X number of hours/years. We built around 1,000 fixtures for the West Block of Parliament, and that was one of the conditions. (I don’t remember the specific duration, just that we had obligations, measured at desk height, and they wanted a lot of output, all pointing up and bouncing off the ceiling)

 

[slow-poke]

I have not diagnosed the actual failures, but I did take one apart and if I had to guess.......

LED's can fail open or shorted, and luminaries like these typically have many LED's in series, so as the LED's progressively short, their individual voltage drop goes from the nominal say 2.5V to near 0V. Higher quality luminaries use constant current power supplies so a shorted LED just causes less light but the same current. The power supply in these luminaries is really really cheap (forget about constant current) most likely the only thing that limits the current is the series summation of the 50 or so LED's, so every time one fails the current goes up a bit and then all the LED's run a bit hotter at some point the 20 cent power supply can't take it anymore and fails. Great design if you want lowest possible cost and guaranteed failures in the not too distant future to invoke future sales for replacements.

 

[jcdammeyer]

IIRC there were 186 lamps with 6 strips of 32 LEDs (192 LEDs per lamp, or about 35,000 LEDs). They were called necklace lights. Run on 110VAC.


And look like this:


They were installed in the summer of 2009 just before the 2010 Winter Olympic Games. They only go on at dusk and off in the morning. I figure even just 8 hours per day * 365.25 days * almost 15 years is 43,000 hours.

Be interesting to see how much longer they last before they have to replace them. Or maybe I'll get a PO and won't retire quite yet.

 

[slow-poke]

Looks like a high side switcher (5 pin device)? Will be interesting to see how long the electrolytics hang in there, doing well at > 40,000 hours. It's nice to see something you have designed just keep on working!

I did a couple of luminary designs for bill boards.

 

[140mower]

IIRC there were 186 lamps with 6 strips of 32 LEDs (192 LEDs per lamp, or about 35,000 LEDs). They were called necklace lights. Run on 110VAC.
View attachment 46215

And look like this:
View attachment 46216View attachment 46217

They were installed in the summer of 2009 just before the 2010 Winter Olympic Games. They only go on at dusk and off in the morning. I figure even just 8 hours per day * 365.25 days * almost 15 years is 43,000 hours.

Be interesting to see how much longer they last before they have to replace them. Or maybe I'll get a PO and won't retire quite yet.

I was just looking at those yesterday on the way to a hockey game, and remembered that you had a hand in them.... They weren't on at the time, but we weren't going very fast if you know what I mean......

 

[jcdammeyer]

Looks like a high side switcher (5 pin device)? Will be interesting to see how long the electrolytics hang in there, doing well at > 40,000 hours. It's nice to see something you have designed just keep on working!

I did a couple of luminary designs for bill boards.

The electrolytics were the most expensive longest hours devices we could find and are, in addition to the LEDs themselves the weak spot. I was looking at file dates and initial R&D started in 2008. The first LEDs we tried were from China and lasted a month before they were at 50% intensity and then it just got worse. Then we changed to Nichia LEDs from Japan.

Mechanically because there are six strips in each lamp with 16 of the 20 degree beam angle LED pointing out in a slightly different angle it's quite possible that some lamps have dead strips. But if you aren't at that view angle you won't see that.

Oh and 8 pin device MXHV9910

 

[jcdammeyer]

The early lights used a different driver and only 16 LEDs. We decided it wasn't bright enough.


Although still bright.



But just not bright enough. These photos are from October 2007.



Then tried a different driver and directly off 120VAC. I used my mill to make the 3 prototype brackets out of plastic.



Special jig for aligning the LEDs as the boards went through the wave soldering machine.



It was a fun project.

 

[Tom Kitta]

I remember the early stuff from China was horrible - they either quickly went dead (like a month of little use) or started to flicker. This was more than 10 years ago.

I have shop lights in my shop for around 5 years now and I have not noticed any degradation. Through we are talking here about maybe 100 hours per month, so far they have seen 6000h or so.

We went far from the early days - when some super over worked Japanese guy invented blue LED & got like $200 bonus for his invention (!) That Japanese company was super nasty to the guy that made them millions.

 

[jcdammeyer]

We went far from the early days - when some super over worked Japanese guy invented blue LED & got like $200 bonus for his invention (!) That Japanese company was super nasty to the guy that made them millions.

As my Dad was working for the Alberta Research Council at the time it's unlikely he could have patented this idea but he did get an article on it published. This was before LEDs and the 'concept' of an opto-isolator chip like a 4N26. I wonder if he'd patented the concept how wealthy we would have been. Assuming the idea hadn't already been patented.

 

[slow-poke]

Job, well done, those off the line switchers can be pretty handy for applications like this, we used them in a smart meter application. Conducted EMI can require a bit of tinkering;-)

 

[jcdammeyer]

Job, well done, those off the line switchers can be pretty handy for applications like this, we used them in a smart meter application. Conducted EMI can require a bit of tinkering;-)

Yes the AC/DC interface is mostly filters.
The LED lights for the rings at YVR ran of 24V toroidal transformer supplies. No electrical emissions which was important as they were right by the airport.

The ones on the barge ran off 48V batteries, Genset and fuel cells. Different power supply in the lamps due to up to 68V charging voltage. Float plane pilots did not like the noise in their radios.

 

[MrWhoopee]

I guess I got lucky, 3 years and no failures.

Update: I got to looking closely at my leds yesterday. Two of them have one of the individual leds out. At this rate, they will last longer than I will.