Looks like 2 linear AMC7140 parts (700mA max each). A microcontroller and switch and coax power input.
I'd guess the 2 amc parts are in parallel so 1400mA total current and the MC-E is likely wired 2p2s and the show is running from something like 2 li-ion cells.
Presumably a headlamp type 'solution' with the uC doing PWM or similar to 'dim' the amc parts.
Originally Posted by georges80
Cheers George I missed a pic so added below
it is all dies Parallel
yes the battery is a 7.2 li ion pack
the only thermal contact is with the screw in bezel touching the mcpcb from the top
so not even a metal to metal contact
Without analyzing the actual circuit in detail, I can only make general statements based on some logical assumptions.
In addition to the heat from LEDs, it looks like they're using linear regulators to go from 7.4v down to the Vf the MCE. That's a lot of additional heat being generated by the linear regulators. The regulator will only be about 50% efficient which means they'll generate just as much heat as the main power LED. I don't see how that'll even work running at the full power rating of the regulators. It makes me think this is always running at a very low current or a very low PWM duty cycle. Otherwise, I don't see how the components would survive.
They are also using a lot of solder connections as mechanical connections. Generally speaking that's a very risky practice and usually results in long term failure of the solder joints. Especially for something with a lot of thermal cycles from getting hot and cooling back down.
Well, with the MC-E in 4P then those poor linear regs are going to be VERY hot dropping 2 li-ion down to 4P Vf. At mtbmacgyver states, efficiency would be around 50% given the voltage drop that needs to occur.
Looks like a pretty nasty design - especially the right angle boards and solder connections. One should NEVER use solder as a structural support. The whole thing looks hand made - especially evident in the soldering 'quality'.
Anyhow, lots of bad work in that product, as you point out, the thermal path is dreadful.
Quick non-technical summary would be "YUK! Toss it in the bin"
Thanks for that Guys much appreciated .
only measuring I could do was the draw from the battery .
on hi 1300 Ma
med 590 Ma
low 330 Ma
the led mcpcb got too hot to hold in 20 seconds and when back in the case the housing was only body heat in 4 minutes so how hot must the led have been .
Because it's a linear regulator, the input currents you measured will be equal to the actual LED drive current. All the excess power is just turned into heat in the linear regulators.
Originally Posted by troutie-mtb
It may be worth watching the current over time as the light heats up. Most linear regulators have built in thermal protection to keep from burning up. There's a good chance the currents will drop once the regulators start to get overheated. I didn't look at the datasheet for these particular parts, but it's a pretty standard feature.
I took a quick peek at the datasheet. The regulator does have thermal protection, but it sounds like the chip shuts down when it gets too hot. The datasheet doesn't go into much detail.
Originally Posted by MtbMacgyver
It is pretty craply put together and designed so no wonder each die is only getting 325 Ma
the beam is bad too .
I will pass on Georges stirling advise "YUK! Toss it in the bin"
plus their web address is no good