BTU Shocker Triple MT-G2 with a twist -- Aiming for >100Watt ~9000Lumens -- With external 2S power pack, handle etc...

377 posts / 0 new
Last post
LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Bit more planning stuff, this time on how I will be tackling the emitter mounting and how I’ll approach sinking some heat into that massive reflector.

Front view showing the noctigons in place and the crazy pattern of necessity for my copper spacers :).

The idea is to have the reflector back sitting flush on these copper spacers and sucking some heat off the pill.
Would love to be able to machine a nice copper spacer/gasket for this but I’ll have to make do with what I can find.
So far I’ve found some 15×15mm copper squares on ebay available in various thickness so I’ve ordered a bunch of different packs to combine and hopefully cover my spacing needs.
Going by my measurements and Noctigon/Cree datasheets I should need a spacer height of 2.3mm to just clear the top of the MT-G2 baseplates with the reflector. (The emitters pcbs themselves will probably be glued down)
So I’ve ordered 2mm, 1mm, 0.3mm and 2× 1.2mm sets of 10 spacers so I should be able to cover any distance that’s 2mm, 2.2mm —> 2.5mm in 0.1mm increments and even reach to 2.7mm if get some crazy tall noctigons for some reason. Silly

Of course they’re all separate little squares so once I’m happy with the fit I’ll probably try reflowing them together. Maybe in a kind of press or something that can be heated while keeping the components under good pressure to get a nice tight single piece copper disk. This sounds like a nightmare by I’ll give it a go Smile

In the end I just hope to be able to sink some heat into the massive bulk of the BTU reflector, because in stock form it looks like there is very little thermal pathway between it and the body/pill.

For the emitter wiring I’ll need to drill 6 holes through the pill surface to get the wires close enough to each pcb solder point. I figure drilling at an angle from the places shown on my model I should just about be able to drill through the thinner part of the pill and gain access to the driver cavity without going through too much solid sidewall.
In there I’ll have to see about adding some more thermal mass that’ll bolt to the bottom of the mcpcb mounting plate.
Depends on how much space I need for drivers and such.

For the drivers I’ll be running a separate set of 7135s/(slave driver) to each emitter so while the whole circuit will still be in parallel I should gain some per emitter regulation and avoid any issues with differing vF or resistances.

So that’s it, let me know what you guys think and if anyone has attempted something similar that I can learn from.
Especially with regard to those copper spacers and reflowing stuff like that together.
Cheers

MRsDNF
MRsDNF's picture
Offline
Last seen: 28 min 25 sec ago
Joined: 12/22/2011 - 21:18
Posts: 13296
Location: A light beam away from the missus in the land of Aus.

Still here reading. Interesting.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

Tom E
Tom E's picture
Offline
Last seen: 7 hours 4 min ago
Joined: 08/19/2012 - 08:23
Posts: 12880
Location: LI NY

Wow! I was thinking of getting a copper screw to replace the standard screw that goes into the reflector, thinking it will heat sink from the pill top to the reflector. Getting direct contact from the pill top to the reflector is a little beyond my capabilities, but I gave it some thought, wondering if even thermal pads stuck in there and compressed on contact would help, but don't think they are very efficient.

Still not sure how you will get the "perfect" spacing from those copper inserts to the reflector - you want the reflector sitting down to get the best output of the emitters for sure, not to mention how perfectly level they must be. Maybe you will use some quality thermal paste in there?

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland
Tom E wrote:
Still not sure how you will get the “perfect” spacing from those copper inserts to the reflector – you want the reflector sitting down to get the best output of the emitters for sure, not to mention how perfectly level they must be. Maybe you will use some quality thermal paste in there?

yup it may all go horribly wrong! Wink

The biggest problem will be getting those copper pieces together and perfectly flat for sure. I imagine there will be a lot of cutting, sanding and cursing at whoever stenciled out the pieces in the first place.

If I get it just right the reflector will be fully seated on the spacers and the emitter domes will be as far into the reflectors as possible without themselves actually being pressed down on by the reflector rim. I want the pressure, if any to be on the solder points on the pcb. If I then find I need to move the reflector out a bit more because the focus of the emitters is wrong I should have the ability to do that by changing around or adding on some more of those copper spacers.

And yes there will definitely be a generous amount of thermal paste on all the contact surfaces. Smile

With regard to the copper screw, that’s a really good idea. I need to look at that, maybe I can find a high-copper brass bolt to replace the one that’s in there. Can’t hurt Smile
You wouldn’t happen to know the thread size and length of the stock bolt would you?

Cheers

Tom E
Tom E's picture
Offline
Last seen: 7 hours 4 min ago
Joined: 08/19/2012 - 08:23
Posts: 12880
Location: LI NY

Crap! Actually I worked on a Shocker mod'ed by someone else and they left the screw out! I found one that fit well, but think it was a US size. All I had to do was cut it down. Hhmm, can't recall the thread size... Was thinking of brass too, but like you said, high copper content would be better I assume.

Running these Shockers at 6A to each emitter does get warm pretty quick, so would think getting that reflector to do more heat sinking is good, but your amps have to be up there, because 4A is no problem with the pill and pill area.

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

No worries, I’ll take some measurements once I get the real thing apart.

Silicon Bronze seems like it would be a great option for a bolt. Not cheap though even if you can find the right size Smile

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Bit of an update on the battery pack.

I received the lipos today and decided to get the battery pack assembled to a servicable/testable state.
I ended up going a slightly different route for the lipo packs, rather than the two Turnigy 5000mah 2s1p 20c softcase packs I chose to get two Turnigy 6000mah 2s2p hardcase packs with a 25c discharge rating.


Turnigy 2s2p 6000mah hardcase

I am actually not too keen on parallel wired cells in preassembled packs because I like being able to see directly how each cell is doing, but in this case it ended up being the best option.
This was mostly down to size and capacity, with the longer but thinner 5000mah packs I could have theoretically fit 4 in there but it would have been too tight to realistically do the wiring without soldering the packs in permanently. With the 6000mah 2s2p packs I have more space for connectors and wiring at the top while also filling up the vertical space fairly efficiently and getting some more capacity in there with less hassle (compared with two 5000mah packs).
The hardcase is also quite nice and means I don’t have to worry about potentially damaging the foil packages of the bare cells when jamming the lid shut. It’s quite a tight squeeze to get things closed up. Smile

Here’s the assembled pack as it currently stands, the two 2s 6000mah lipos connected up in parallel with deans t-connectors.
Above it are the twin relays also hooked up in parallel. They feature 6v coils, with a rating of 12Amps each and are triggered on by the master toggle switch on the outside of the pack.

Still to come is the temperature monitor/alarm and a circuit for automatic low voltage cutoff. Possibly also a short circuit detection, we’ll see.
But for now it’s ready to test.

Another difference to the last time I showed the pack is the much thicker coiled power cord. The 18awg cable I was using before just wasn’t up to the task and at 15A test currents it was wasting a stupid amount of energy through high resistance. This new cord is based on a 14awg copper stranded speaker cable and is performing much much better. The material of this speaker cable wasn’t keen to coil under heating so I had to shrink wrap it first them do the coiling and that worked out really well. It’s quite a stiff coil that likes to return to this position, much more so than my first attempt with the thinner cable.

It also weighs a heck of a lot more, but well… the lipos aren’t exactly light either! Smile

Speaking of which the total weight of the pack is now 1.36kg! Which means you’ll definitely notice its presence on the belt, and also means I need to beef up that belt clip to something more substantial.

Finally a couple more renders showing how I plan to run the switch wiring down the back of the light.

The parts are starting to trickle in now so I can’t wait to start assembling the final thing.

Cheers
Linus

lionheart_2281
lionheart_2281's picture
Offline
Last seen: 3 hours 53 min ago
Joined: 10/25/2012 - 18:32
Posts: 3023
Location: Brisbane, Australia

I’m more excited for the outcome of this build more than any other build I’ve seen on here, bring it on!!!

RMM
RMM's picture
Offline
Last seen: 3 months 1 day ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

Feeding Chain:
MEGA Shocker (this build)
Super Shocker
BTU Shocker

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Did some battery pack run-in tests and here are the results.

They’ll do just fine at 15Amps, in fact that’s what I’m using to do these “gentle” run-in discharge cycles because it’s just a little above 1C load for each lipo. Basically nothing, which means there’s hardly any voltage sag.

In the chart I measured battery voltage based off the meter built into the pack and also took the voltage at the end of the coiled power cord using my DMM. This is important here because of the fairly substantial resistance losses in the cable and I need to know what voltage range is actually going to be available to the drivers.
You’ll see I still drop around half a volt across the length of the cable but it’s miles better than the 18awg wire I was using before, that was dropping over 1.5v in the same test scenario.

At half a volt I should be well within the working range of my target drive current for fairly long regulated runtimes even at ~5Amps per led.

I’m using the vF/I graph from the Mt-g2 torture test done by djozz to get an idea of where I need to be to get the results I’m after and if those graphs are reliable (which I’m sure they are) I should be well on track for >15A high and around 9000lumen emitter output.
Possibly with some source voltage headroom to go even higher for Turbo mode, 7Amps per emitter sounds entirely possible but probably not for very long.
I also need to keep driver and emitter wiring losses in mind but the cable is by far the biggest hurdle to overall performance.

In any case as long as my source voltage stays above 7v I should be able to achieve 5A per emitter quite comfortably and as this test shows, the voltage at the end of the cable is still 7.16v at 13mins into the test.

I didn’t want to discharge the batteries too far on this first test so I called it a day at 13mins, but I’ll do further discharge tests once I’ve cycled the packs a few more times and see how long I can maintain this voltage range before dropping below 7v.

RMM: There looks to be a place at the top of that list for an ULTRA Shocker, who’s on the case? :bigsmile:

Linus

MRsDNF
MRsDNF's picture
Offline
Last seen: 28 min 25 sec ago
Joined: 12/22/2011 - 21:18
Posts: 13296
Location: A light beam away from the missus in the land of Aus.

Nice work Linus. One MTG-2 at 5 amps pumps heat like you would not believe. Cooling will be the biggest headache here. Keep it up. Smile

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Oh I’m sure heat will be my biggest problem. I fully expect my turbo mode to be a short burst only mode. Smile

I’m beginning to realise just how much 100watts of heat is and how difficult it is to dissipate even on my substantial resistor bank. Had to bring in a 12cm fan to keep things in check for the test. Silly

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

What’s this!. A real life/non digital flashlight in this mod thread! Never thought I’d see the day Silly

The BTU host finally arrived along with most of the parts I needed for the handle. So I got stuck into assembling the handle and finalizing the trigger switch before putting it all together. I was really keen to see if this thing would actually fit as I had anticipated.

So far so good, and thanks to the 3d model no real nasty surprises.
Here is where it’s at right now. Smile

The handle is temporarily mounted to the battery tube using a two part barrel scope mount with picatinny rail top, thanks for this goes out to the inept Kaidomain shipping department who sent me this completely by mistake instead of a plastic reflector I had ordered. Thanks guys, it actually came in much handier than the reflector would have as the actual rail part that I plan to use is still in transit. Smile
Ultimately with the proper rail in place the handle will sit about 15mm lower than it currently is but other than that it’s very close to how I want it.
I’m still debating whether to mount that handle rail to the tailcap as in my renderings or directly to the battery tube. The latter would give me more options for a secure attachment while the tailcap option would probably make boring out the tube for 4×18650 cells easier. The tailcap is even less spacious that I expected and drilling two m4 holes to mount that rail may not be possible anyway so it could be an easy choice in the end. We’ll see.

-


For mounting the switch I cut the case down to size, added a metal plate to the top with m3 threaded hole and drilled a countersunk hole through the top handle rail to bolt it securely into place. Sounds easy and straightforward doesn’t it. Silly It really wasn’t and gave me all sorts of headaches but in the end I got it potted with epoxy and very securely mounted. Smile
Works well but I really hope the switch won’t fail anytime soon because getting it back out to replace it would be a nightmare…!


In any case the handle feels really good and offers a steady hold of the bulky light, the trigger switch is easily accessible by index finger and with this placement of the handle and the bare torch host the whole thing is also quite nicely balanced.
Really pleased with it so far and I think it looks pretty cool as well, it’s clearly cobbled together but still looks like it means business Smile


Still lots to do before I even think about the guts of the light. Next up is the tailcap and figuring out how best to mount the XLR socket inside. Should be fun Silly

Cheers
Linus

RMM
RMM's picture
Offline
Last seen: 3 months 1 day ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

Wicked. Totally wicked.

J)

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

Ramblings
Ramblings's picture
Offline
Last seen: 1 week 2 days ago
Joined: 01/09/2013 - 04:05
Posts: 757
Location: Australia

I hope those cable ties aren't your perma solution.

Going the full nine yards, you should be able to find a way to get some short metal tube (prob around half the length of the original battery holder), thread it and mount that handle with some bolts.

You are the light of the world. Like a city on a hill, you cannot be hidden.
People do not light a lamp and then hide it away, instead they put it where it gives lig

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

There’s no problem that can’t be solved by cable ties! Wink

The whole handle mounting is totally temporary, like I said even the rail mount I’m currently using isn’t what I will end up with. It’s just a scope mount from Kaidomain that happened to arrive in the mail by accident and I put it to use since it had a short picatinny rail on top that I could easily mount my handle riser to for testing. Works surprisingly well though and gives me a chance to try out different handle mount configurations before I commit and start drilling into the BTU.

Ultimately I will be bolting a short, relatively flat picatinny rail onto the body of the light which I will then mount the handle riser to directly. Still not sure if I will try to attach this rail further back on the tailcap as I planned in my 3d models or if I will drill the bolt holes directly into the battery tube as in the current handle configuration.

Or… as you say I could use the current scope mount, ditch the temporary zip ties and figure out a better mounting for it that would be sturdy but also non destructive to the BTU body. I was thinking along he lines of replacing the zip ties with thin metal rods with threaded ends that I would bend around the battery tube/body and bolt/tension to the scope mount plates with a set of lock nuts.

Lots of options at the moment Smile

unknown00101
Offline
Last seen: 5 months 3 weeks ago
Joined: 08/16/2013 - 11:53
Posts: 4342
Location: USA

Through all the BTU threads/posts/pics every time I think to myself “that is a big light… and I need one.”

But that picture of you holding onto the handle put it more into perspective.

I did a double take and it made me remember this ridiculously large shoulder-able halogen spotlight I was given some 5-7 years ago. There isn’t really resemblance, seeing a flashlight with a handle just made me remember it.

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

hehe, yeah the BTU is a BIG light. It’s kind of silly really, first time I picked it up it felt like I had just grabbed a deliberately oversized prop from a movie or something. Something like Hellboy’s personal flashlight! Wink

With a handle in this kind of “lantern” format it makes a bit more sense to me.

unknown00101
Offline
Last seen: 5 months 3 weeks ago
Joined: 08/16/2013 - 11:53
Posts: 4342
Location: USA

Definitely makes sense to me for sure. I have tiny hands; fingertips can’t touch when holding a D maglite. Silly

Your mod looks great btw, even in mock-up.

MRsDNF
MRsDNF's picture
Offline
Last seen: 28 min 25 sec ago
Joined: 12/22/2011 - 21:18
Posts: 13296
Location: A light beam away from the missus in the land of Aus.

Very nice. Your mod is looking real good.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

Ledsmoke
Ledsmoke's picture
Offline
Last seen: 3 days 2 min ago
Joined: 08/08/2011 - 16:05
Posts: 1995
Location: Denmark

That looks awesome! Really cool stuff here!

~ Ledsmoke ~

Dutch humor:

[quote=djozz]

 I do not think that the BLF-community ben

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Oh look there’s a baby flashlight on my flashlight!

Thanks again goes to kaidomain for this scope mount that I didn’t know I needed, had no idea how useful it could be for this build! hehe Smile

unknown00101
Offline
Last seen: 5 months 3 weeks ago
Joined: 08/16/2013 - 11:53
Posts: 4342
Location: USA

There’s no such thing as too much.

Now you need to mount 3 more handles to the btu and 6 or 7 rails on the bezel so you can do something truely effective. Like this, but in a flashlight:

Wink

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

hehe Smile

feilox
Offline
Last seen: 6 months 5 days ago
Joined: 07/21/2012 - 04:42
Posts: 840
Location: wisconsin, usa

Great job wut a monster!

Don wrote:
It sounds like the XM LEDs won’t really be suitable for flashlight use. Pity…

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Well, a small update in the grand scheme of things but I succeeded in flashing my first nanjg 105c driver with custom firmware today!
Took a fair bit of reading, fiddling with SOIC clips and hitting dead ends with eXtreme AVR Burner before I finally got things working with AVRDude.
I couldn’t have done it without the great wiki resources and threads offered by users of this forum so thanks a lot to all involved Smile

Being able to flash custom firmware will be super handy in getting the precise driver modes and operation I want in this flashlight. Not to mention any other light. I’m pleased I have that hurdle crossed.

Cheers

MRsDNF
MRsDNF's picture
Offline
Last seen: 28 min 25 sec ago
Joined: 12/22/2011 - 21:18
Posts: 13296
Location: A light beam away from the missus in the land of Aus.

Well done. Congratulations is in order. Beer

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

Hikelite
Offline
Last seen: 18 hours 38 min ago
Joined: 07/13/2011 - 16:18
Posts: 3746
Location: RO

Bomb build, nice!

LinusHofmann
LinusHofmann's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/28/2013 - 08:27
Posts: 974
Location: Switzerland

Thanks guys!

This might save someone some frustration when attempting to program atmel chips with a SOIC clip from ebay. Like this one

I had some communication issues at first and it was frustrating because I wasn’t sure if I had messed up the pinout or if it was a software issue.
Turns out it was because of the clip. The little copper tabs/pins on my clip tend to move and slide inwards if pushed, and it’s very easy to accidentally push them in when trying to make a good contact on a chip that’s soldered to the driver board. Once they’re no longer at the very tip of the clip grooves they won’t make good contact anymore and you’ll often fail to connect to the chip.

Solution that I came up with in frustration was to dab a touch of super glue into the pin guides inside the clip. First pull/push the pins forward as far at they will go and then apply a small drop of contact adhesive where you see the copper pins run inside the clip. Careful to avoid getting any on the contacts or at the back where they mate with the cable pins. I accidentally seeped some glue in to one wire contact so that one is now permanently attached…whoops Silly
Anyway the glue holds the pins in place solidly and I can now clip on with perfect contact every time, even on tight pcbs layouts like the nanjg 101c Smile

Looking at the clip, maybe it’s just mine but it seems these pins aren’t held in there particularly well, looks like the plastic has just been melted down around them.
So if you have this similar problem just glue the buggers down Smile

southland
southland's picture
Offline
Last seen: 1 month 1 week ago
Joined: 11/04/2012 - 07:15
Posts: 3204
Location: Georgia, USA

First in line for production model.

Pages