Making a list on what I need for my quad Seoul, stuck on heatsink

Robban

Enlightened
Joined
Jul 6, 2004
Messages
849
Location
Sweden
For the last couple of weeks I've been thinking about and looking at the different options there are for building one of them sweet quad Seouls. I've (finally...) settled on the powersource, host (Mag2C), reflectors etc but I'm not sure what to do when it comes to the heatsink.

I've been looking at the PES2 but as I understand it the Seouls are slightly too large to fit into the indentations since it was made for Luxeons. Problem is I don't own nor have access to any type of tools to make those bigger (a Dremel doesn't really count here). So are there any alternatives out there that I could use instead?

I'm sorry if this has been asked before but I haven't found any real info on this so far.
 
Does it have to be a quad? will you settle for a Tri??

I just did a Tri-Seoul and it was really easy.....
 
cqbdude said:
Does it have to be a quad? will you settle for a Tri??

I just did a Tri-Seoul and it was really easy.....
I've considered going with a tri-star sink but I figured what's worth doing is worth over doing 😉
 
I have done both quad and tri Cree and have not noticed a considerable difference.
Maybe Seoul will be different...Im finding out that its the reflector that is making a difference. And of course how much you drive the LED's .
Good Luck, Im sure someone will suggest a heatsink. Im thinking you might have to modify an existing heatsink..
 
Ok. Maybe I should start out with a tri first since it'll be easier to build. I can always build another one 🙂
 
Robban said:
Ok. Maybe I should start out with a tri first since it'll be easier to build. I can always build another one 🙂

That is true....I have a few now....and they all Rock......
 
I'm building three quad-Seoul 2C Mags, all with Shark converters and external pots. Two are being built with IMS20s and one with McR20s.

I'm using PES2s, and originally planned on mounting the Seouls at 45 degrees from the existing quad-Lux pattern. I made a scale template on my computer, printed it out, and cut it with an Xacto knife. Two of the Seouls would have slightly overlapped the existing pockets of the tri-Lux pattern, but I believe the heatsinking would still have been adequate.

I elected to send the sinks out to a machinist to have the surfaces refaced to flat. That is where I'm at currently. I'll use the template to locate the emitters when epoxying. The IMS have been shaved both at the legs and back of the reflector. I did this by mounting my Dremel bit in my drill press, and carefully adjusted the table height for cutting depth. I put a Post-it on the face of the reflector to protect it, and slid it around the table under the bit. These builds will be easy; epoxy the legs to the heatsink and individually center on the emitters.

The McR20s have had 0.030" shaved off the rear, but now have a centering issue. I designed a reflector holder, basically a thin alum disk with four holes that locates the reflectors in the bore of the head. As long as the emitters are accurately located, the reflector holder will center all the reflectors and epoxy to the inside of the head. This is another part being machined now. For this light I'm trying to get Seoul-sized pockets for alignment machined into the resurfaced sink face. Hopefully the pockets will be 2-stepped, one to locate the surround, and one slightly deeper to isolate the slug. Tight fitting single-step holes that locate the slug only concern me with the Seoul. Isolation may be an issue since the slug can contact the sides of the holes as well as the bottom. I'd rather just mount it to a flat surface using a template.

There have been some builds where the McR20s are super-glued to the emitter, but I think there may be a durability issue if the light were to receive a lateral shock as when dropped on its side. The loosest reflector may come free and shear the emitter dome. The reflectors are only held by pressure from the lens and the contact point at the emitter, and not all reflectors will be held tightly in compression.

I've taken a no-compromise approach to my current builds, but I'm confident that simply locating the four Seouls on the heatsink in a position of maximum slug contact would have been fine. Using modded IMS20s (0.030" off legs and back) with legs epoxied to the heatsink, and the thick (2.84 mm) UCL lens, the build would be relatively easy and dimensionally correct (bezel seals with correct gap).

By refacing the heatsinks, I may have a fit issue with the bezel. It may bottom out on the head before the lens and O-ring seat. In this case I'll find a thicker bezel O-ring. I won't unscrew the PES2 to raise it though; I want max heat transfer to the finned head which means compound and a tight fit. If I do need to do this I'll make a shim that gets sandwiched between the sink and head bulkhead. The PES2 is beefy but 4 Seouls at 1 amp each will probably generate some heat.

Another issue I discovered is that the reflectors now sit lower in the head than with Luxes, and interfere with the bore of head. Slight filing of the reflectors at the contact point would solve this, but I am having the heads bored instead.

I have both quad and tri-LuxIII 2C Mags, and though the brightness of the extra emitter is not blaring, it is easily noticeable with my meter or eyes. Those are running 3 X 123, or 2 X 18650, Fatmans, and external pots.

I'm hope to post a step-by-step when I finish the lights, but the above are some tips that you may find helpful with your decision.
 
LED Zeppelin said:
I'm building three quad-Seoul 2C Mags, all with Shark converters and external pots. Two are being built with IMS20s and one with McR20s.
<snip>
I'm hope to post a step-by-step when I finish the lights, but the above are some tips that you may find helpful with your decision.
Very helpful indeed!
A couple of questions though. When you say slight overlap, how slight is it? I'm planning on running these puppies as hard as the Shark and a pair of 18650s can push them so keeping them cool is sorta important 🙂

Will 4 IMS20s fit in a C-mag? I thought 17s where the max?

Sounds like your project is very similar to mine; 2x18650, Shark, 2C Mag (planning on getting a water dropped one from RcatR), IMSXXs, Seoul, PES2 🙂
 
Robban,

When I line up my template, there is one slug in particular that will overlap the existing PES2 pockets the most, by perhaps 20% of the surface area. If you fill the pocket with Arctic Alumina, there should still be adequate heatsinking ability IMO.

Four IMS20 will fit in a 2C Mag no problem. They fit nice and snug. With Luxeons the reflectors sit just high enough to clear the slight chamfered edge just inside the head, next to where the lens meets. With the modded IMS, the reflectors sit 0.030" lower, just enough to touch the chamfer. A couple strokes with a file on the side of the IMS where they meet the head would be enough to give adequate clearance. The 3 legs will not only meet the heatsink face, but 2 can be aligned to touch the short wall surrounding the heatsink. Epoxying the legs will secure them quite well. If it were a single reflector light, I'd insist on a McR reflector as the IMS can have some artifacts. But with four, I think any artifacts will be washed out by the other emitters.

In short, the IMS build, mainly due to having the legs on the reflectors for centering and securing, is the way to go. Once I'm finished, I'll do some beamshots and lux readings with the 2 different reflector setups and see if there is any appreciable difference. But I'll guess that the additional cost and difficulty centering the McR20s is not worth the trouble.

I said I trimmed the legs and back of the IMS 0.030", but that's a generalization. Actually I found the overall height of the Seouls to be slightly less than a Luxeon, and wasn't sure if the recommended 0.030" adjustment was from the back of the reflector in contact with the emitter, or from the base of the slug. I ended up splitting the difference, and trimmed 0.035" from the legs. I use the legs as the locating surface, and trimmed the back of the reflector enough to clear the emitter surround.

FYI I measured my IMS to have an overall dia. of 0.802". Amazingly enough, I recalled enough geometry to figure that the center to center distance of opposing emitters should be 1.134" (reflectors just touching each other). My template is basically a circle that fits inside the PES2, with 4 holes in it where the emitters locate. I sized the holes larger than the actual emitters so that I can visually center them, and also any excess epoxy doesn't squeeze out onto the template.

As I mentioned, without modding the heatsink face, the dimensions are right so that a 2.84 mm thick UCL will seat the bezel and lens properly without much gap. If you were to shave the face of the heatsink to clean up the pockets (any shop should be able to do this easily), the reflectors would sit just a bit lower. I still think the bezel will seat properly, but at worst a thicker O-ring will do.

If you're planning on 2 X 18650, you'll need a deep tailcap (Modamag has some gold ones still available), or an extension ring (Mirageman). 2X 18500s will fit without any length issues, just a sleeve.

I'm partial to external pots, but a cool build would be to make a can that houses the converter board and (+) contact on the underside of the PES2. Do away with the stock switch, and use a tailcap switch like Modamag's new release. It would be too bright so a multi-level converter or some other means (2-stage tail?) of dropping the output for close-up use would be almost a necessity. You'd have to make sure the tube end was sanded and butt securely against the heatsink for the ground path, and a grounding screw for the converter/emitter somewhere on the heatsink would be necessary. If I ever finish these current builds, I might try this with either a single or multi-emitter mod.
 
LED Zeppelin said:
Robban,

When I line up my template, there is one slug in particular that will overlap the existing PES2 pockets the most, by perhaps 20% of the surface area. If you fill the pocket with Arctic Alumina, there should still be adequate heatsinking ability IMO.
<snip>
Again, lots of useful info, thanks a lot 🙂
Good to hear about the reflectors as that will help tighten up the hotspot a bit.

As for trimming the reflectors I think I'll skip that at first (I know, I'm working against what I said just above) just to see what the spot will be like. Who knows I might like it that way. I swapped the emitter in my EDC U60 and I ended up sanding down the reflector just weee bit instead of going for the tightest focus. I'm quite happy with it.

So you're saying I'm gonna have to dig up the long lost geometry and math skills huh? To think I could do that stuff in my sleep just a few years ago *shakes head*
Regarding the overlap. I was thinking about filling it up with compound like you suggested. It's certainly not the way it was meant to be used but I'm sure it'll work better than an air pocket 🙂

I was thinking about just de-anodizing the tailcap and finding a smaller spring to drop in there to make room for the cells. Same idea as this.

I haven't decided what I want to do regarding external pots or other solutions yet. I've thought about having unlimited levels with a pot but on the other hand it would be convenient to have a few set levels to choose from with a switch. The "problem" is that I would like to keep the light as close to stock as possible on the outside so adding a clunky switch would ruin that. I've tried coming up with some way of integrating said switch/pot into the head so that turning the head would change the output but haven't come up with any good ideas so far. I'm thinking it will be easier to find a solution once I have the parts in hand so I can see what I have to work with.
 

Latest posts

Back
Top