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

Ah, I see. (RE: the flickers)

Sorry to make observations which do not make a lot of sense. Sometimes mentioning crazy stuff is poisonous information: everyone starts seeing the same false thing. Hopefully we can avoid that and get to the bottom of things with enough measurements. I hope to re-attack testing for that problem myself.

I do not recall what I was using to turn on the 7135(s) for that test. I would say that your suggestion is a good one (giving them 6v directly). The required Vdd is quite low according to the ADDtek datasheet: 2.7v.

Another thing of note in the datasheet, check out the OUT CURRENT vs. OUT_DROPOUT VOLTAGE graph. It shows an odd blip upwards right as dropout is hit. Maybe this explains your odd performance around 8.3v w/ the iCharger.

EDIT: the datasheet also claims 200uA of supply current consumption… so 48x would be <10mA, which should be no problem for the ATtiny’s output pins. I don’t know what speed they can switch that load at though, capacitance is also not mentioned by I assume it’s incredibly low.

great work Linus!

thanks for sharing. :beer:

Truly amazing! I applaud you for all your efforts!

Yes very interesting, that could well explain some of the weird behaviour. Also I may be missing something but it doesn’t state at what temperature that test was done, but I’d assume that’s done in ideal conditions showing a best case scenario.
What I’d love to see in the datasheet is more graphs based on temperature, since there’s no mention anywhere about the throttling back behavior we observe when temperatures get too high.

Surely if that was an active control system to protect the chip they’d mention it in the datasheet? More likely it’s simply a side effect of the circuitry not working right at high temperatures right?
Makes me wonder what else is happening inside the chip on the way up to those high temperatures… :stuck_out_tongue:

Dropout voltage graphed against out-current like it is in the datasheet but done at various temperatures would be very interesting. I suspect it can vary a fair bit as the chip heats up.
Could temperature have been a factor in your dropout observations do you think? More chips, more amps, more temperature and a higher dropout as a result?

They strongly advise keeping operating temperature below 85 degrees (max junction temp is given as 150) but I’d love to see how higher temps affect dropout voltage and max out current directly.
Only one way to find out I guess! :wink:

-

“What is AMC hiding! It’s all a conspiracy, I have seen the light…stop using 7135s you brainwashed sheeple, it’s all a big cover up. Wake up!” :open_mouth:

We really need a tinfoil hat emote… :slight_smile:

I don’t think so. As I said, it was <8 chips. I was operating them on a pair of those 4x7135 PCBs I think (in free air). I will re-test. Maybe I have stripboard that the 7135s will fit on nicely.

Cool I look forward to seeing your results on that.

I’m not sure I have the right equipment to really test this stuff accurately, but I’ll have a go at characterizing the thermal behaviour. Maybe something like soldering a few 7135s to a block of copper and heating the copper to various stable temperatures to see how that affects their regulated output and dropout behavior. I don’t want to rely on just letting the little buggers heat themselves up from their own junction temperature but see how they respond once heatsoak has set in.

From my testing in the light it seems heat is always a factor in how the chips perform. Even when out of regulation completely the output current seems to drop linearly as things get hotter. Wonder if I can replicate that in an isolated test.

Probably so. I think sense resistors are actually characterized in their datasheets RE: how much the value changes with temp.

Yeah, right! Seems strange not to include something like that in a datasheet for a device that’s almost invariably going to get really hot, just by it’s nature of being a linear controller.

The blower fan I ordered a while ago finally arrived today, so I sellotaped it to the light and re-ran a couple tests to see how things stack up with active cooling.

The plan was to direct all of the airflow from the blower over the majority of the meager heatsink fins available on the BTU. I just did this with electrical tape for now, keeping the airflow contained and running between the fins top and bottom then leaving through an exhaust port on the opposite side.

The fan is designed for 12v so it doesn’t quite run as fast as I’d hoped, but it still shifts some decent air and makes quite a racket on 8v. :stuck_out_tongue:
Ultimately I think I’ll mount the blower flush on the side of the battery tube and fashion some kind of a 90degree duct to redirect the airflow through the fins. All mounted on a removable modular picatinny rail setup of course. Maybe even make it temperature controlled if I can be bothered. :slight_smile:

The following graphs are simple reruns with the blower on full compared against the previous tests. Most obvious difference is I could easilly run the test to 10mins before getting worried about heatsking temps exceeding 70degrees.

Nothing too surprising here, although I was hoping to see some plateauing of the temperature at some point, guess I need two fans after all. :stuck_out_tongue:
It’ll probably keep up better when the light is running off batteries.

Also of note is that a much lower heatsink temperature seems to only really affect the drivers, ie by delaying the heat related behaviour where they drop output current. The leds don’t seem to produce noticeably higher output at these lower temperatures and are primarily handicapped by the drivers throttling back drive current.

Although granted I don’t trust my light meter readings between tests enough to show subtle differences in output regarding temperatures at the leds. Sometimes it can be as much as 5klux out at turn on and when I compensate the graphs to match performance at the same drive current draw they line up perfectly again.
I’d need a decent integrating sphere to really see what differences in output actually are.

I also decided to go resitance hunting and started by reinforcing the twisty interface contacts a bit. Just for peace of mind and to see if I could gain any noticeable voltage overhead there. The contact pressure had declined slowly since I made the thing and that was bugging me, it was nice and tight to begin with but with all the testing and reopening/closing and general wear it wasn’t quite as reassuringly tight as it was. The springs where still springy but the tolerances had just opened up some in general.
The positive path also needed a bit of beefing up and improving to put my mind at ease :slight_smile:

I added an small disk of copper to the top of the positive brass post to increase the height a bit, then ran an additional length of 16awg wire through a hole in the carrier disk and soldered it directly it to the thick positive wire coming from the connector. I also jammed some spare silicone tubing under the spring to ensure contact pressure was really high for a solid connection.

The negative posts were also overhauled a bit and I replaced copper braid in two springs (which had started failing) with 18awg wires. These should make good contact.
Ultimately as a result of all this I almost couldn’t tighten down the battery tube, the contact pressure and friction was considerably higher than before…and the positive post showed a nice polished arc where it mated with the brass recess on the contact board! Great

I ran tests at 8.5 and 8.7v after all this and compared against the before graphs, that showed no noticeable improvements at all…nothing! The bumps and dips on the current readings all lined up and so did everything else, so little difference in fact I won’t even bother posting the graphs.
A bit disappointing but I think it’s safe to say the contacts where likely good enough to begin with and I can at least rule this out as a major resistance bottleneck in the light.

Finally just a couple shots of the whole light after a hard day strapped into the test apparatus. 8)
Shows how it sits in the hand as well. It’s really a very comfy grip once you get used to the unnatural weight of the thing :wink:

Urgh, I’ve been playing in my 3d sandbox again…and that usually means new things need to be ordered from china very soon! :wink:

This time I wanted to cook up a functional, decent looking and completely removable blower module.

Here’s the result using various random bits of metal and plastic (see if you can spot the other half of the AC plug pack case that I also used to make the trigger housing… no waste here! :wink: ) and a nice black aluminum lens hood to act as a duct to contain and direct the airflow across the pill section.


The duct covers the entire surface of the pill and should offer some nice options for directing the airflow. Maybe it will work well enough in this configuration, with the air going around a bit before blowing out through the gap between the heatsink fins and the back lip of the hood. Could have part of the exhaust area blocked off to control the flow better… or I could seal off the hood all the way around and drill some exhaust vent slots/holes on the opposite side to the blower. Not sure yet.

Everything is picatinny rail mounted of course for easy removal. Power will be provided by a socket on the tailcap and the usual spiral cable to make the connection. There’s also plenty of room inside the improvised elbow joint to house some electronics to control the fan. I envisage a basic control pot sticking up the top of the assembly at the elbow for starters.

And of course you’ll notice the shorty configuration of the light has made a comeback.
I think I have a good way of making this configuration work now so when my spare BTU parts finally arrive via the rusty bucket express I’ll give it a go. :slight_smile:


This quick image slap-up shows the shorty look on the real thing. This one’s actually a bit too short but you get the idea, it’s much more compact and quite a lot lighter.

-

Btw if you’re a bit insane like me, have a BTU shocker gathering dust and fancies the look of what’s going on here. I’m contemplating building a few more of these lights (hopefully much improved upon this prototype) over the next couple of months.

I’m having a lot of fun building and testing this thing but there’s only so much money/time I can justify spending on it. Unfortunately having built this one I see so many ways of improving upon it should I build another. If I can build a few lights for other people I have an excellent excuse to keep tinkering with this thing! :bigsmile:

Here’s a quick breakdown of what I’d try to do for a V2.0 of this light. (Copied from a PM I sent to another member)

-Shorty Body configuration like above but with power indicator led built into trigger switch housing. Same reverse clicky with off-time driver UI, my personal favorite combo atm.

–9000mah 3s LiFePO4 9.9v battery pack (LiFePO4 chemistry for the better suited voltage overhead and stable discharge performance) Plus constant >9v availability for a VERY wicked DD mode. XLRs will probably be replaced with Speakon high power connectors to handle DD turbo currents. Also twin power relays in pack will be replaced by power mosfets for smaller resistance loses and lower power consumption.
The pack will also have a nice new display on the outside showing pack voltage and current draw. Otherwise the same as prototype.

-Drivers based on Wights Linear 7136 Driver with DD turbo.
Modes: High, Med Low etc tuned to users preferences and stable long run performance. Driver running off-time STAR firmware modified to handle the DD bypass.
Turbo: With the extra LiFePO4 voltage on tap I can easily see maxed out Mt-g2s on DD. Over 12Amps per emitter for short bursts (heat limited short runs of course but >12,000lumens @ ~300watts should be possible). That’s also not just on a fresh charge but you’ll have a very impressive turbo up your sleeve basically all the time until the battery is dead! (this is a rough idea, the drivers and packs need to be tested to see precisely how they behave…but I’m certain this power source and driver combination (provided the driver tests ok) is going to be a better way to go than my 7135s and 2s lipos).

-Active external cooling module as shown above. If I can figure out a nice way of having it temperature controlled I will. Otherwise a pot to dial in fan speed. (Note: The light in general won’t be particularly water proof but this part will certainly not hold up to getting wet!) This cooling system also won’t be able to keep up with turbo mode temps (nothing much would beyond a dunk in an icy lake!) but it will certainly help with the cooling down periods between blasts! Which to me is frankly almost as important. And of course it will enable longer stable running on the “lower” modes.

-Improved heatsinking mass behind driver cavity. Given the new driver approach I should have a bit more space to work with to improve the thermal paths for both the led shelf and the driver mosfets themselves.

So shoot me a PM if you’re at all interested and we can discuss things. Warning: parts cost alone is close to 450dollars so the term “budget” doesn’t really apply here. :wink:

Cheers
Linus

:beer:

you are :open_mouth:

I pretty closely skimmed your post (gtg for now!). I’ll come back and read it at a more leisurely pace soon.

In the meantime I wanted to say this: why are you looking at an external blower? It seems to me that integrating a blower inside the unused space of the battery tube makes the most sense. The concept of being able to convert back to a fully-handheld/no-external-batteries light seems like nonsense to me. This a seriously powerful build and will require a remote battery pack to be useful/interesting. You may as well treat it that way and install a blower internally.

There’s plenty of space in there. Here’s how I think it could go down (but I don’t own a BTU Shocker): If you modify how power goes through the head and into the battery tube & minimize driver size you’ll have enough room to drill/cut vents at the bottom of the fins. After that you route the power wires as far off to one side as possible (top side, towards the handle). Use a small blower with the intake facing the front of the light, then seal the tube around/behind it. Vent radially around the blower. You’ll only be able to fit maybe a 30-35mm blower in there, but I suspect that would make a big difference. You could also look at using an axial fan with a high static pressure in the same configuration.

That’s all my two cents and it’s not my build of course… it’s your show! I just noticed that you hadn’t mentioned considering this option and kind of wondered why.

Is an intercooler being incorporated with the turbo blower? If so this would greatly increase the density in the battery increasing efficiency and greater lumen output.

Great scott! You’re onto something there! I will of course be adding a heat exchanger sleeve to the spiral cable to keep temps and resistances at an absolute minimum. Keeping the electron charge pressures high and temps low will certainly improve total max lumens when the …erm… PWMs? hit the redline…or something :stuck_out_tongue:

Let me stop you right there! What’s this “sense” thing you’re talking about? I’m pretty sure I’m lacking it :wink:

An internal fan would probably work quite well in the shorty config as you say and keep things much neater, but there’s plenty of reasons I’ve gone this route (for now anyway, it’s a rough and ready first pass concept).

Primarily though I just liked the idea of strapping a large blower to the side of a powerful light, to me at least it looks pretty sweet hanging off the side like that. Goes with the modular/tactical nonsense design of the rest of the light. Not much actual sense involved in that decision and preferences can vary.
I had to seriously restrain myself from adding a second blower onto the other side you know! haha :slight_smile:

Apart from the sense deficiency I’m also lacking in machining equipment/skill so I tend to shy away from anything that involves too much external machining on the light itself. The thought of a lot of drilling into an anodized BTU pill or body (they’re getting rare as hens teeth you know :wink: ) without having the option of covering mistakes with a bolt head doesn’t sound particularly appealing to me.

Finally I do want to maintain some basic water proofing on the base light. It’s never going to go underwater unscathed but it should survive a quick drop in a puddle or sitting in some melting snow (for emergency cooling :stuck_out_tongue: ) without something internal going fizz. That’s also why I want the blower module detachable in case it’s raining outside or whatever. Not to mention if the external fan goes pop (don’t know how many million operating hours are on my particular sample since it came from a server) it’s a five minute job to swap it out.

Having said that the XLR plug on my current light isn’t yet waterproof but a bit of silicone on the inside grounding tab opening and an oring around the fixture will take care of that. I also have other unsealed lights and I hate the condensation issues you get on the inside of the lens when it’s cold outside. Can’t be having any of that here.

What I really like about the idea of an internal blower is a way of shifting some air directly over the driver components. But I think as long as I can maximize the heat sink path of the critical internal components to the pill then the blower will do a good job of removing a decent chunk of that heat from the surface. It’s not ideal because of the limited surface area but should do a decent enough job, especially to more quickly return the light to a safe temperature after a turbo stint. The light has so much mass that left to it’s own devices just sitting there it takes an absolute age to cool back down after it’s hit a soak at 70deg.

Thanks for the comment though, I’ll probably try an implementation of your internal fan idea (easy enough to try on the 3d prototype) and see how it could work.

Linus

Well well, I may have to apologize to the humble 7135s…

I just reconfigured the driver and ran a test with the MCU (Attiny13A) completely bypassed. I simply tapped into the MCUs positive voltage supply (4.3v zener controlled) and fed that directly in the Vcc harness controlling the 48x 7135s. This would force them into a constant-on (high only) mode of operation.

The results are very telling. WE FINALLY HAVE PROPER REGULATION!!! and at a much lower voltage overhead than before. :open_mouth:

-

This graph shows the results of running the MCU-less setup off the iCharger at 8.7v, faded graph is the older test at the same voltage for comparison. Everything connected up the same way.

I thought my current meter had locked up for a moment there, couldn’t believe what I was seeing, solid stable regulation for almost 5minutes! At the full theoretical current limit of 17.85A, with the MCU active I had to feed the thing 9.1v! before it would show me those figures and then it only maintained it briefly.

-

Of course I thoroughly verified my results with state of the art measuring equipment…

Yep, she’s flat alright :bigsmile: :wink:

SOOOooo Attiny 13A, what the heck are you doing?? But certainly you’re not switching those 7135s with enough authority or something else fishy is going on. Maybe the Attiny doesn’t like the heat? Need to study it’s data sheet a bit and see what could be causing this. I assume we can trust that the STAR firmware is doing a good job at 100% duty cycle? I can’t wait for my little Rigol scope to finally get here so I can have a better look into what’s actually going on on the PWM line.

Thoughts welcome guys, I’m a little stumped as to what’s going on once again. But the finger is firmly pointed at the Attiny13A at this stage…
I am rather pleased to finally see some regulation and more expected performance from this setup, it just too bad I had to lose all my lower modes to get there… :stuck_out_tongue:

Cheers
Linus

PS. I just checked and the annoying Warm up flicker is also completely gone running in this MCU-less configuration. Just solid stable output at much lower battery pack voltages.

Keeps getting more and more amazing

Bravo. Love your problem solving and thanks for including the second thing I understand about in this total build. Mines a fair bit longer though and not quite as straight.

Thanks guys, you think I should hit the MCU with the “measuring equipment” and see if it works better after? :stuck_out_tongue:

-

I’m totally delighted with the performance of this light at the moment, just been running down the batteries a bit outside. Granted there’s no modes so I have to aim it far away to avoid blinding myself, but the output is rock solid (not a hint of a flicker anywhere) and even with a pack voltage down by 7.5v it’s still kicking butt on the output, seeing 67klux @ 3.35m where as before it would have been way below 50k at turn on.
Before it was whimpering and output was noticeable down as soon as the pack dropped below 8v.

This is much more like what I expected to see.

Maybe the LifePO4s aren’t necessary after all…hmm, now I just need to figure out how to get my UI and modes back at this performance level. :stuck_out_tongue:

I think you could use a P-channel MOSFET driving the 7135’s with the ATtiny13A running the gate on the MOSFET. You’d need to invert the PWM for the modes, that’s not a big deal though.

That sounds very interesting, bit over my head at the moment though (just had to read up the difference between P and N mosfets). By invert the PWM you mean that 255 in the firmware would be equivalent to 0% duty cycle in the standard setup and 0 would be 100% duty cycle? Or is there something more complicated and “frequency” that I don’t understand yet. Hoping to learn a lot when I finally have a decent oscilloscope and can test/verify simple stuff like this.

First I’m wondering if simply increasing the supply voltage to the MCU from 4.3v to 5.5v will make any difference, that I can test easily enough when the parts for your linear driver arrive in a couple of days.

My thinking is that for whatever reason the MCU isn’t able to sufficiently saturate the gates of all the 7135 internal mini fets (please correct me if I’m using nonsense terms here :stuck_out_tongue: ), so they’re not actually fully on even when they shouldn’t be regulating (in the linear range I believe it’s called?) which leads to the large increase in voltage dropout because of higher internal resistance, which in turn leads to much higher temps and flickering and all the rest of the weirdness.

Of course it may just be that my particular MCU board isn’t working 100%, that’s the first thing I’ll try to eliminate. Flash up a fresh MCU and zener board and see if it makes any difference.

Your understanding of inverting the PWM is correct, 255/255 would be fully off and 0/255 would be fully on.

I won’t speculate much about the rest. As far as each individual 7135 is concerned it’s almost certainly a supply voltage issue (too low). As far as the MCU is concerned… I don’t know if it’s low supply voltage (and therefore low output voltage to the 7135’s) or limited ability to provide current (and therefore sagging / low output voltage to the 7135’s).