Neptune SKY LED

[Jon's Reef]

So where are the real life units on the axis?

As to thickness guess there is some importance.
Optimizing Heat Sink Base Spreading Resistance to Enhance Thermal Performance | Advanced Thermal Solutions

See lower limit in link.

They are referring to the thickness of the metal baseplate of the heat sink, not the thickness of the thermal paste.

Thermal paste has a lower thermal conductivity than metal to metal contact. The ideal thickness depends on the surface finish of the 2 surfaces being joined. Generally a 50 micron RMS roughness on each results in an ideal thickness of ~100 microns. The thermal paste is only there to fill the air voids due to micro scratches. it is not intended as a gap filler. In this case the surfaces should be mechanically lapped. In no circumstance should it be so thick that it prevents metal on metal contact.

 

[Mywifeisgunnakillme]

Actually their main beef was it was cool, showing poor heat transfer.
That's what I took out of it.

Surface area is much more important.
Active cooling can " fix" a lot of ills.

I agree. They also commented the black boxes (with license plates) were cheaply made.

 

[Mywifeisgunnakillme]

From: https://www.google.com/url?sa=i&url=https://www.semikron.com/dl/service-support/downloads/download/semikron-application-note-thermal-paste-application-en-2018-01-19-rev-00/&psig=AOvVaw0Czb9ujv5faLpdfInkYc6n&ust=1629843285575000&source=images&cd=vfe&ved=2ahUKEwjQy_DBlcjyAhVLYs0KHX-UAX0Qjhx6BAgAEAs

Also from the cree doc reference above:
"though a TIM may have better thermal conductivity than air, its conductivity will not be nearly as good as metal’s, so the approach is not to add material between metal components but rather to fill the voids that are typically occupied by air."


What is the total thickness of the heat sink and aluminum backed plate? Also what is the difference in the size of the heat sinks? Lastly, what is the CFM of the fans used in the BB light?

Whether the thickness of the heat sink is included in the overall thickness depends on the resistance of the plate to heat sink junction, which without testing cannot be determined. Heat transfer is also significantly guided by dT across the thermal junction. So again without a temperature measurement it is simply speculation.

I don't know these answers. I just prefer that when i spend $900.00 and buy a light--that when it's taken apart and reviewed by people not affiliated by the company--those people are like "yeah, wow, well constructed" and "Really impressed" and "definitely nothing questionable here" and "really quality parts and design here"....

The fact there is even a debate as to construction and materials tells me pretty much all i need to know.... In other words, for $900, there should not be a debate on how well it is constructed.... For $900 construct it well... Obviously, there can always be debate on how well it performs... (which appears to be the same as the gen 4 XR 30).

For example--no one I know of says GHL's are not made well. They might say other things... But you're getting value as to construction and the hardware... Same with say Stratons from what i hear... People may not prefer them--but no one is saying they're cheaply made like many people are saying about Skys...

 

[oreo54]

They are referring to the thickness of the metal baseplate of the heat sink, not the thickness of the thermal paste.

Thermal paste has a lower thermal conductivity than metal to metal contact. The ideal thickness depends on the surface finish of the 2 surfaces being joined. Generally a 50 micron RMS roughness on each results in an ideal thickness of ~100 microns. The thermal paste is only there to fill the air voids due to micro scratches. it is not intended as a gap filler. In this case the surfaces should be mechanically lapped. In no circumstance should it be so thick that it prevents metal on metal contact.

You are pretty optimistic in the micro structure of finished metals.
If anodized in its face it's already err "lumpy" at the micro level.
General photo:

We actually we agree mostly but each has it's own tradeoffs.

In this discussion. I'd prefer phase change thermal tape over metal to metal BUT only as a guess.

All of these builds need real world testing. Thermal managements is an art and science.

Phase Change Thermal Interface Material | Custom Die-Cuts | Parafix

Phase change thermal interface material, for superior surface wetting & minimal thermal contact resistance. Available in custom sheets and die-cut shapes.

parafix.com

Your def. of "gap filler" is a bit loose.

Bridgelux eb strips don't require heat sinks btw..

EB Series™ | Bridgelux, Inc. LED Lighting

www.bridgelux.com

LED Heat Dissipation: An Optimization Guide | SimScale

Learn the key aspects of optimizing lighting design to improve LED heat dissipation performance and maintain a lower junction temperature.

www.simscale.com

Regardless of the means of fabrication, any manufactured part can contain imperfect surfaces, with wavy and rough profiles. When two surfaces are placed in contact, there is normally only a small area of direct physical contact at the interface. And, heat transfer occurs only at these small areas of surface contact.

 

[Mywifeisgunnakillme]

You are pretty optimistic in the micro structure of finished metals.
If anodized in its face it's already err "lumpy" at the micro level.
General photo:

We actually we agree mostly but each has it's own tradeoffs.

In this discussion. I'd prefer phase change thermal tape over metal to metal BUT only as a guess.

All of these builds need real world testing. Thermal managements is an art and science.

Phase Change Thermal Interface Material | Custom Die-Cuts | Parafix

Phase change thermal interface material, for superior surface wetting & minimal thermal contact resistance. Available in custom sheets and die-cut shapes.

parafix.com

Your def. of "gap filler" is a bit loose.

Bridgelux eb strips don't require heat sinks btw..

EB Series™ | Bridgelux, Inc. LED Lighting

www.bridgelux.com

What lights to do you run?

 

[telegraham]

Actually their main beef was it was cool, showing poor heat transfer.
That's what I took out of it.

Surface area is much more important.
Active cooling can " fix" a lot of ills.

If that was their main beef, it was a Beyond Burger take on reality. The VS, set to their settings, pulls less than 90W. It's cool because it's not dropping that much heat into the 1.5mm licence plate for the heat sink/fans to draw out.

 

[oreo54]

What lights to do you run?

I build them..

 

[telegraham]

You are pretty optimistic in the micro structure of finished metals.
If anodized in its face it's already err "lumpy" at the micro level.
General photo:

We actually we agree mostly but each has it's own tradeoffs.

In this discussion. I'd prefer phase change thermal tape over metal to metal BUT only as a guess.

All of these builds need real world testing. Thermal managements is an art and science.

Phase Change Thermal Interface Material | Custom Die-Cuts | Parafix

Phase change thermal interface material, for superior surface wetting & minimal thermal contact resistance. Available in custom sheets and die-cut shapes.

parafix.com

Your def. of "gap filler" is a bit loose.

Bridgelux eb strips don't require heat sinks btw..

EB Series™ | Bridgelux, Inc. LED Lighting

www.bridgelux.com

LED Heat Dissipation: An Optimization Guide | SimScale

Learn the key aspects of optimizing lighting design to improve LED heat dissipation performance and maintain a lower junction temperature.

www.simscale.com

The heat sink is anodized. Even the mating face.

 

[Jon's Reef]

Here are some numbers.

I set the light to 100% Coral Growth, then let it soak for 15 min. Then took the temperature at 3 locations using an electrically isolated probe thermometer. The probe was allowed to equilibrate at each point for ~2-3 min. The fans turned on very shortly after turning the light up.

Room Temp: 72 F
Removed a Skimz, temp of the board next to a white led (~5 mm from the led, not temp of the plastic diffuser, but the actual LED board): 116.5 F
Temperature of the set screw which attaches the front right mounting point to the the upper heat sink: 88.5 F
Temperature of the air in the front right exhaust: 87.0 F

Note that this is not the ideal measurement situation, but gives an approximation for the temperatures present. For reference, LED from Cree are binned with a reference junction temp of 85 C (185 F). (Reference)

It may be that the combination of large spread in the layout, good thermal adhesion to the primary heat sink (the Al backing plate) and close contact to the secondary heat sink results in the required cooling capacity. In addition, it is important to note that cooling beyond this may be detrimental as it would result in condensation.

 

[oreo54]

The heat sink is anodized. Even the mating face.

Normally I'd say that was a bad thing as to the fact the anodizing increases surface area by being lumpy BUT it may not be the case.
Interesting stuff..

https://www.mdpi.com/2073-8994/13/3/433/pdf

Obviously not a paste or no paste study.
What I always found sort of funny is people that worry their heatsinks are hot when that just shows they are working..


Therefore, as more heat was transferred to the outside through the heat sink, the temperature between LED and heat sink decreased and the surface temperature increased.

 

[Mywifeisgunnakillme]

I build them..

I really liked the days where more people built stuff... This is a hobby after all.... This hobby now appears to be losing a DIY aspect that I think is to its detriment when it comes to innovation and improved success keeping corals. I think Telegram hit the nail on the head when he said that there really has not been any real improvement in LED's in years. Costs go up. Marketing plays a bigger role. Quality is hit or miss. And trending in the wrong direction?

I think that is one reason i am irritated by Neptune. No innovation of there own here--other than maybe nice software--but even then they limit the channels you can control with the LEDs which seems like step backwards (standard default settings yes--options yes--denying ability to tune for advanced users--NO!).... Then they grab Phillips technology, put it under a license plate, add other people's LEDs, and put it in a plastic box, call it "revolutionary," hype the hell out it, and charge almost $900 under the false claim it replaces multiple units of its main competitor. NO WAY is that revolutionary. They can't even innovate their own mount for god sakes... I'd say that standardizing mounts would be revolutionary, kinda, if i believe for a moment that's why they are using eco tech's mounting pattern.

Anyway, when i switch out T5's some day because they're extinct--i hope you're still in the hobby making the latest and greatest DIY. I think that would be a fun project that is in the spirit of why i started into this hobby in the 2000's.

 

[Mywifeisgunnakillme]

I'll mention too--if someone is looking at the Sky--and thinks they might like it--just get the coral care gen 2... It's a super beefy, well built light. No one would pick up or take apart that light and say its not well constructed. It's not that heavy either and not bulky over any tank over 50 gallons IMO... And it has the same diffusion panel...

I have seen it in person twice... looks sweet and it's appears way better made, same performance, and cheaper.

 

[oreo54]

I really liked the days where more people built stuff... This is a hobby after all.... This hobby now appears to be losing a DIY aspect that I think is to its detriment when it comes to innovation and improved success keeping corals. I think Telegram hit the nail on the head when he said that there really has not been any real improvement in LED's in years. Costs go up. Marketing plays a bigger role. Quality is hit or miss. And trending in the wrong direction?



Anyway, when i switch out T5's some day because they're extinct--i hope you're still in the hobby making the latest and greatest DIY. I think that would be a fun project that is in the spirit of why i started into this hobby in the 2000's.

With a formal release of ATI's rails to North America still a complete unknown,

Want to use someone elses or have none?
Sorry just a ironic coincidence.

ATI Straton LED Light - Comments, Review, PAR, Coverage, Discuss...

I didn't see if you owned the ATI Straton? At any rate, you should pick up a Straton and a Kessil A500X to play around with. I have picked up most of the new lights as I normally do when a new one comes out. Gives you a chance to try them for yourself and see what works for your setup. A...

www.reef2reef.com

Building stuff is easy.. building stuff that looks cool is not.

As in all things.. "This too shall pass"..

 

[Mywifeisgunnakillme]

Building stuff is easy.. building stuff that looks cool is not.

I think it's doable. And that's what makes it fun. This is what i "built" for my 50 gallon lagoon as hybrid fixture (T5, reefbrite, kessil 360's (started with 160's)). I am working on something a 200 gallon just about ready for water. Only on the 200--it's X500's mounted in the ceiling with narrow reflectors and T5's/reefbrites that automatically lower for part of the day (electric window curtain roller) and recess in the ceiling when viewing in the evening (25 plus inches of clearance above the tank...).. Totally unnecessary--but totally fun to build i hope...

That all said, i have seriously looked at the sky's, radions, and stratons, etc. for the 200 gallon. The idea of one cord is attractive!! But i am not seeing "revolustionary" improvement over t5's. I am not even sure they are remotely better than T5's....

And i've had so much more fun building the 50 gallon light--that i want do it all over again for the 200 gallon. And don't me wrong--nothing i've put together here is building an LED--it's just minor carpentry, but i think i could get into LED making too.

 

[KimG]

So, the engineer in me has an itch that I cannot scratch. Does anyone know why the diffusion panel has a gasket on only 3 sides? Anyone know what is it for?

 

[Jon's Reef]

I am guessing that Reef Builders had the correct reason:
"the only reasoning we can come up for this gasket is for spacing the LEDs and the mixing chamber at an ideal distance."

 

[Mywifeisgunnakillme]

I am guessing that Reef Builders had the correct reason:
"the only reasoning we can come up for this gasket is for spacing the LEDs and the mixing chamber at an ideal distance."

Or the gasket was useful/designed to be used on say a Gen 1 phillips coral care and the sky uses that same mixing panel but the sky design does not incorporate the need for a gasket. Cheaper just to leave it in there than design their own without a gasket.

 

[Jon's Reef]

Or the gasket was useful/designed to be used on say a Gen 1 phillips coral care and the sky uses that same mixing panel but the sky design does not incorporate the need for a gasket. Cheaper just to leave it in there than design their own without a gasket.

The CC gen 1 panels were not 1 piece.

pic from reefs.com article:

 

[Mywifeisgunnakillme]

The CC gen 1 panels were not 1 piece.

pic from reefs.com article:

Did they have gasket?

Just adding to Telegram's theory here. Maybe Neptune will chime in, but probably not since it's all top secret.

 

[Jon's Reef]

I responded in the reef builders thread. The diffuser design requires a spacer between the diffuser plate and the led board. This is documented in the patent. The size looks to coincide with the gap, so the gasket is likely a spacer.