I have an S70S and the fins doesn’t really get warm. But the switch area definitely warm. That thread in head is the heat bottleneck.
After how much time?
I can go turbo on my L6 for about a minute and the switch area gets really hot, but the upper fins are just warm. This is a one piece head.
The S70s might need to be on high for 5 or 10 minutes to see if the heat travels smoothly past the threads or if there is a bottle neck.
I think we’ve all dealt with threaded pills before, the heat seems to transfer well with them.
We can ask Sofirn about making a one piece head, but that is a huge change from their current design.
Repost of pics:
.
2 minutes. Made a video right now. It is uploading. Switch area is always 2-3 degree celsius higher than the first fin abowe thread.
Remember that if the heat were not dissipating, it would be the same all over.
Heat dissipation is like a river running downhill, the source is the highest and the outlet (fins) by definition are the lowest point.
If fins are dissipating heat well they will be significantly cooler than areas closer to the source of the heat regardless of how efficient the heat path is.
Hot fins are a sign that the cooling is not keeping up, not cool fins. The switch area being warm and the fins not tells us nothing about whether the heat path is efficient.
But when two fins next to each other has 3 deg difference ant the thread we questioning is between them there is something wrong. It can work but try be better if it is possible. It will be a 6-7000 lumens monster. This is critical. If it will be an XHP35 HI thrower than no big deal. 2000 lumens don’t make problems in a big light.
Here is the video:
To bad the light is not using Narsil or Andruil! The UI is perfect and a lot of time and knowladge went into it and the drivers. I think there is nothing better on the market (for a comparable price).
The look of the light is ok. But from a thermal point of view there can be done much more.
What i do not like is the head diameter. It is so close to already existing lights. Like a few people before me, i first thought the plan is a 90mm reflector 90mm (that would be nice!) but it is the head diameter.
[quote=ZozzV6]
But when two fins next to each other has 3 deg difference ant the thread we questioning is between them there is something wrong. quote]
This does not take into account that the cooler fins are not only further away from the heat source, but also a much larger diameter.
For 2 good reasons they should be cooler, and 3 degrees does not seem significant enough to show that the heat transfer is an issue at all, never mind pinpointing it to the threads.
Remember if you separated those two sets of fins and isolated them, and added exactly the same amount of heat to the centre of each, the larger diameter fins would certainly be cooler, especially at the outer surface never mind that the smaller set is dissipating heat before passing the leftovers on to the larger set.
I don’t see a 3°C difference between two fins next to each other. I see less than 1°C difference which is perfectly normal. Look at sections :34 to :39 when you do a slow sweep with the thermometer.
If we look at the picture above I do see a 3°C difference between points 1 or 2 and 5 or 6. That’s normal.
Near the end of the video I think the thermometer goes from point 4 to point 2 which is 3 degrees, but still perfectly normal.
I don’t think 2 minutes is long enough for the heat to spread. Maybe 5 or 10 minutes on high would be a better test?
No 1 & 2 need more fins, like said before.
If it were a BLF light it might use NarsilM or Anduril, but this is Sofirns light. They wanted to use their own UI.
At first I was hoping the head diameter would be 75mm like the S70 or L6. That is a good size. This light would be like an updated and high powered version of those.
I was not too happy they went 15mm bigger in head width at first. On the other hand it should give a tighter hot spot and more throw, so I guess it’s not so bad.
10 minutes at very high powereven the worst heat sinked light geat up evenly. We need to focus the first 2-3 minutes or with 6000 lumens the first minute to get that heat away fast.
Those saying that the heat path on this light is horrible are 100% correct.
The heat will have no easy way of using the fins for cooling and it will end up being not real improvement from existing lights thermally.
Doesn’t make sense to me when such minor changes would not cost anything.
You want the body of the light to follow the reflector outline as close as possible, getting rid of that ridge on the reflector would be good as well.
The upper fins are virtually useless unless the heat can get to them.
Think of the heat like a fluid, you want the freest flowing path possible or it will be limited in how quickly it can move from point A (the LED) to point B (the fins).
Well, if TA says so :+1:
And why do they want to use a clicky? It will have an impact on the max output (even the best switches increase the resistance). It will increase the price (for machining and parts) and it will be harder to waterproof the light. Keep all that money and spend it somewhere on the light where it has bigger/better impact.
I don’t think the fins come into play in only 2 minutes of being on. In this short amount of time we need the heat to go from the led to the MCPCB and from there into the aluminum mass. This is all conductive heat transfer.
The fins (basically just more surface area) don’t start working until the temperature gradient between them and the ambient air start to increase. The higher the gradient, the more effective they become. This is where convective heat transfer comes into play.
Since we do have quite a bit of mass in Sofirn’s design it might take 3 to 4 minutes before the fins start to get hot enough to effectively start transfering heat.
There will be no way a light this size will be able to sustain 5,000+ lumen/100+ watts continously. It will have to step down in power. We might get 3,000 to 4,000 lumen continously, but we won’t know for sure until we try.
I would like to hear from Sofirns designer/engineers on why they made the decisions they did. If they have degrees in thermodynamics and understand how much heat will be generated by this led then they probably made the right choices. We are basically amateur scientists after all.
On the other hand, it’s possible their designer/engineers don’t have degrees in Thermodynamics. In which case, maybe we can show them a thing or two about improving their design.
Sorry for being rude, but what’s the point of this light then?
Might as well buy the s70 if it does not perform better.
I think you are exaggerating. The heat path is not horrible. It’s not ideal, but not horrible. In fact, it’s pretty typical.
Existing lights seem to do pretty well thermally. There always has to be certain compromises between form and function. Just look back when Sofirn showed us the plunger shaped head. This made more sense thermally because it had even more surface area than the stepped design. Unfortunately form won out with people saying the plunger design looked ugly and the stepped design looking nicer.
Even the high end companies such as Acebeam and Nitecore, etc… know they have to find a good compromise if they want the product to sell.
There is always a penalty. I know they are already concerned about the weight of the head making the light unbalanced. There may be other reasons such as the raw material costs going up, shipping costs going up, machining time going up, etc… They also may not want to spend a lot of extra time tweaking things to get a little better thermal performance. Maybe we can convince them to change a few things to get a better product.
The reflector ridge is necessary for the machining process. Perhaps we can convince them to leave some extra material where the weight distribution penalty is not great.
Yes, but they look nice. 
These are the basics of conduction. In theory you want the freest flowing path possible, but in the real world it’s not necessary. It’s not exactly like water where a bottleneck can really hurt the flow. With the medium of aluminium, bottlenecks are only minor slowdowns.
We already have the best conduction from the led by using a large DTP copper mcpcb which transfers the heat quickly into the shelf. The shelf and everything around it has great thermal mass. A lot of lights don’t even have such good thermal designs.
Also, it really isn’t Sofirns goal to make the best thermodynamically designed light. They want sales and to make profit. They have to keep costs down as much as they can so maybe we can convince them to make improvements in a few certain areas.
Why do you say it will not perform better?
It is better in many areas. Much higher output, much higher throw distance, much better UI. All for a great price. No modding necessary. :+1:
Costs? Remove less material saving machining time. Costs lowered. That’s all. Every time we want something like stainless bezel people tell they will pay more if it is happening. I will pay more a light with the best possibble heat sinking and stainless bezel……
If this make as it is I will pay 65$ max but with better design and stainless bezel I will pay 75$ or more.
BLF lights mostly about performance and if that is great then have a reasonably good price beside what we get.
This was decided long ago.
I have tested my L6 both with and bypassing the rear switch. I get the exact same output in lumens. My light draws 17A and I’ve had the FET driver in it a year. I’ve had the light almost 2 years and I use it daily at work. Whatever switch Simon used in the L6 is a really good one. I don’t know if Sofirn will use the same exact switch or not, but at least we know there are good switches out there that can hold up to severe abuse.