LED Par fall off

[goldhillreefer]

1. When PAR starts to degrade from an LED fixture, is the problem the LED, PCB, Lens, or driver?

2. For you Radion users, How long have you had your lights before par starts to drop off?

 

[arcangelreef]

1. When PAR starts to degrade from an LED fixture, is the problem the LED, PCB, Lens, or driver?

2. For you Radion users, How long have you had your lights before par starts to drop off?

There are G5, G4 and G3 radions in use all over the place with quite literally 10's of thousands of hours on them. I think the unit will just outright fail in other ways before you ever have to worry about any kind of significant "par drop off" from the leds.

If you are new to the hobby, or maybe leds. I will tell you straight up. This is not something you need to concern yourself with.

 

[braaap]

There are G5, G4 and G3 radions in use all over the place with quite literally 10's of thousands of hours on them. I think the unit will just outright fail in other ways before you ever have to worry about any kind of significant "par drop off" from the leds.

If you are new to the hobby, or maybe leds. I will tell you straight up. This is not something you need to concern yourself with.

Well you are wrong. And it is ok to be wrong. Lots of studies done on it. PAR drop off is a real thing. And it does matter in the long term.

https://www.hennings-miniriff.ch/vortrag-sindelfingen-2013/messungen/verluste-von-led-modulen/
http://hennings-miniriff.jimdo.com/vortrag-sindelfingen-2013/module/vorstellung-module/
http://hennings-miniriff.jimdo.com/vortrag-sindelfingen-2013/module/vorstellung-strahler/

Degradation can be a mix of everything. Drivers, diodes, resistance in wires reducing power to diodes etc.

All that said yes there are TONS of older black boxes and radions etc still working just fine. If you want the best growth it would be justifiable to replace the light/diodes around 30,000 hours.

 

[goldhillreefer]

There are G5, G4 and G3 radions in use all over the place with quite literally 10's of thousands of hours on them. I think the unit will just outright fail in other ways before you ever have to worry about any kind of significant "par drop off" from the leds.

If you are new to the hobby, or maybe leds. I will tell you straight up. This is not something you need to concern yourself with.

I'm not new to either the hobby or LEDs and this has not been my experience. I once had my lights mounted 16" off the surface and had optimal sps par with my lights running at 60%. Now I can barely get 200 par with the lights maxed out and moved to 6" off the surface.

 

[goldhillreefer]

Well you are wrong. And it is ok to be wrong. Lots of studies done on it. PAR drop off is a real thing. And it does matter in the long term.

https://www.hennings-miniriff.ch/vortrag-sindelfingen-2013/messungen/verluste-von-led-modulen/
http://hennings-miniriff.jimdo.com/vortrag-sindelfingen-2013/module/vorstellung-module/
http://hennings-miniriff.jimdo.com/vortrag-sindelfingen-2013/module/vorstellung-strahler/

Degradation can be a mix of everything. Drivers, diodes, resistance in wires reducing power to diodes etc.

All that said yes there are TONS of older black boxes and radions etc still working just fine. If you want the best growth it would be justifiable to replace the light/diodes around 30,000 hours.

Your comments are more in line with my experience. So, if a fixture is only going to produce optimal par levels for 3 years, does it really make sense to buy thousand dollar lights?

 

[BeanAnimal]

Your comments are more in line with my experience. So, if a fixture is only going to produce optimal par levels for 3 years, does it really make sense to buy thousand dollar lights?

The lifespan and output of an LED is mostly dependent on thermal management and to an extent for some colors, phosphor lifespan. The harder they are driven, the faster they will degrade. While other components can fail, they usually cause a system failure, not degraded output.

Most fixtures are not run at full intensity for the full photo period. Better fixtures would generally have Better thermal management.

It is hard to make generalizations for any fixture, let alone do so without data about the actual usage and environment.

That said, if you are replacing LED fixtures in 3 years, something is very wrong.

 

[oreo54]

I'm not new to either the hobby or LEDs and this has not been my experience. I once had my lights mounted 16" off the surface and had optimal sps par with my lights running at 60%. Now I can barely get 200 par with the lights maxed out and moved to 6" off the surface.

You really should add more context to that..
I mean for all anyone would know is the light was a home made led with stars glued to plastic instead of a proper heat sink..and this is 5 years later
Current quality leds are more than capable of 50000 hours based on a cut out point of 80% of original output.
Now of course that is JUST the led proper..

I can toast one in a few minutes without proper cooling ..or have them last a decade..

 

[braaap]

Your comments are more in line with my experience. So, if a fixture is only going to produce optimal par levels for 3 years, does it really make sense to buy thousand dollar lights?

My math isn’t great but 30,000/12hrs a day = 2500/365days = 6.8 years to get to that point.

The lifespan and output of an LED is mostly dependent on thermal management and to an extent for some colors, phosphor lifespan. The harder they are driven, the faster they will degrade. While other components can fail, they usually cause a system failure, not degraded output.

Most fixtures are not run at full intensity for the full photo period. Better fixtures would generally have Better thermal management.

It is hard to make generalizations for any fixture, let alone do so without data about the actual usage and environment.

That said, if you are replacing LED fixtures in 3 years, something is very wrong.

You really should add more context to that..
I mean for all anyone would know is the light was a home made led with stars glued to plastic instead of a proper heat sink..and this is 5 years later
Current quality leds are more than capable of 50000 hours based on a cut out point of 80% of original output.
Now of course that is JUST the led proper..

I can toast one in a few minutes without proper cooling ..or have them last a decade..

All of this. Degradation of LEDs is affected by 100s of variables. Only thing you can really do is monitor par. Take a yearly measurement and see what your falloff is. ¯\_(ツ)_/¯

Most people quit the hobby before it matters. Or they upgrade to the newest/coolest tech.

 

[arcangelreef]

Well you are wrong. And it is ok to be wrong. Lots of studies done on it. PAR drop off is a real thing. And it does matter in the long term.

https://www.hennings-miniriff.ch/vortrag-sindelfingen-2013/messungen/verluste-von-led-modulen/
http://hennings-miniriff.jimdo.com/vortrag-sindelfingen-2013/module/vorstellung-module/
http://hennings-miniriff.jimdo.com/vortrag-sindelfingen-2013/module/vorstellung-strahler/

Degradation can be a mix of everything. Drivers, diodes, resistance in wires reducing power to diodes etc.

All that said yes there are TONS of older black boxes and radions etc still working just fine. If you want the best growth it would be justifiable to replace the light/diodes around 30,000 hours.

I never said it doesn't happen. I was simply stating its really not worth worrying about.
There are a million and one situations to consider when looking at lifespan and output of an Led light. There is ZERO point in debating this with such a broad stroke.
So ill just say it again. Its not worth worrying about.

"If you want the best growth it would be justifiable to replace the light/diodes around 30,000 hours."

I rest my case... 30000 hours? thats 7 years on a 12hr light cycle.

 

[braaap]

I never said it doesn't happen. I was simply stating its really not worth worrying about.
There are a million and one situations to consider when looking at lifespan and output of an Led light. There is ZERO point in debating this with such a broad stroke.
So ill just say it again. Its not worth worrying about.

"If you want the best growth it would be justifiable to replace the light/diodes around 30,000 hours."

I rest my case... 30000 hours? thats 7 years on a 12hr light cycle.

Your case sucks. Because you said not to worry about it. But you absolutely should. My fixture is 8+ years old. I just upgraded/replaced the diodes due to PAR falloff. So it is a justified concern.

 

[BeanAnimal]

All of this. Degradation of LEDs is affected by 100s of variables

I don't know about thousands, but in general thermal management vs output is the major variable for emitter life. If we are talking about the fixture and the rest of the components then there is much more to consider when talking about failure.

I would not consider (for any reasonable unit) anything but thermal management and how hard the emitters are being driven compared to their L80 ratings if considering PAR degradation estimates. I would (if money allowed) use enough fixtures to now have to run them at 100% output.

. Only thing you can really do is monitor par. Take a yearly measurement and see what your falloff is. ¯\_(ツ)_/¯

I don't think that is bad advice at all.

 

[braaap]

I don't know about thousands, but in general thermal management vs output is the major variable for emitter life. If we are talking about the fixture and the rest of the components then there is much more to consider when talking about failure.

I would not consider (for any reasonable unit) anything but thermal management and how hard the emitters are being driven compared to their L80 ratings if considering PAR degradation estimates. I would (if money allowed) use enough fixtures to now have to run them at 100% output.


I don't think that is bad advice at all.

I said 100s lol

But it really is. How often do they clean it? How hard do they drive it? How hot is the ambient environment? Does it get salt spray?

You literally can go on forever. In the perfect lab environment you are correct. It’s thermal management and drive level for the most part. But in our settings the environment matters to a point.

 

[gbroadbridge]

1. When PAR starts to degrade from an LED fixture, is the problem the LED, PCB, Lens, or driver?

2. For you Radion users, How long have you had your lights before par starts to drop off?

As someone else stated, for most folks it's not worth worrying about.

There is certainly a decrease in light output over time, however that is dependent on too many things to list here.

These things are considered during the design phase of reputable light fixtures.

For the most part light fixtures in an environment such as unsubmerged aquarium lighting would be designed with a life of about 10 years based on a 30% duty cycle - with end of life being < 80% initial output.

That's certainly the basic premise that I have used professionally as a design engineer when designing commercial LED fixtures.

It's really unlikely that any fixture would still be in use after that period of time, which is why I say it's not worth worrying about.

 

[VintageReefer]

I used a photon v1, black box style for 5+ years and noticed no degradation. I upgraded to a v2 for more control and a more sleek fixture

I still have the 5 year old photon 48 v1 and I wouldn’t hesitate to use it again.

Context it was used 5+ years at roughly 10-12 hours a day with channels in the 30-40% power range and easily lit my mixed reef with sps.

A prime 16hd being driven with channels at 100% or even 130% like many people do? Yea. I wouldn’t think that will make it 5+ years without degradation or issues

 

[gbroadbridge]

A prime 16hd being driven with channels at 100% or even 130% like many people do? Yea. I wouldn’t think that will make it 5+ years without degradation or issues

Running at 100% 24 hours a day, I agree it wouldn't be designed to last 5 years.

100% @ 12 hours a day, if designed within spec, I'd expect 5 years would be about end of useful life.

But I don't expect folks would run them like that except in coral farming facilities, where they would already have been written off anyway.

 

[monfilsi]

LEDs and generally all semiconductors lose efficiency over time. So over time the LED needs more current to generate the same amount of light. The lose in efficiency in this case is accelerated primarily due to the combination of damp and heat at the diode and/or the current drivers.

Better thermal management will increase the lifetime of the led but the damp environment plays a significant impact.

We accelerate the aging of lasers and leds in my industry to estimate their lifetime instead of running them for years.

Damp heat is a killer of semiconductors over time.