$2 for 1-8 layer PCBs, $60 coupons for new customers: jlcpcb.com/?from=SDG
Xinfrared T2S+ Thermal Camera: www.banggood.com/custlink/DvG...
Pico Technology 5000 Series Scope:
www.picotech.com/oscilloscope...
Support the Channel with Patreon: / sdgelectronics
----------------------------------------------------------------------------------------------------------------------
Zigbee Stuff
mmWave Sensor: s.click.aliexpress.com/e/_Dl7...
Sonoff USB Dongle: s.click.aliexpress.com/e/_Dmi...
Zigbee Hub: s.click.aliexpress.com/e/_DBp...
----------------------------------------------------------------------------------------------------------------------
SolderKing UK Made Solder: solderking.com/
SolderKing UK Made solder available at Kaisertech:
UK JBC Distributor kaisertech.co.uk/ Discount Code: "SDG5"
----------------------------------------------------------------------------------------------------------------------
0:00 Introduction
4:45 Metal Core PCBs from JLCPCB
9:07 Reflow and Assembly
12:33 Testing
----------------------------------------------------------------------------------------------------------------------
3D Printers:
Ender 3 S1 Pro on AliExpress: bit.ly/3Tfiz22
Geeetech Mizar S on AliExpress: bit.ly/3MN0c0M
Artillery Genius at Banggood: bit.ly/32iG2dv
Geeetech Mizar S on AliExpress: bit.ly/3MN0c0M
Geeetech A20M at Banggood: bit.ly/2YooVll
Geeetech A20M at AliExpress: bit.ly/3oBqo2I
Geeetech PLA+ Filament: amzn.to/3gelSX1
Ender 3 V2 at Banggood: bit.ly/36lN8Ny
Ender 3 V2 at AliExpress: bit.ly/3j1C2CO
-------------------------------------------------------------------------------------------------
Soldering Tools and Equipment:
Aixun T420D On AliExpress: bit.ly/3cQXXhh
Aixun T3B on Aliexpress: bit.ly/3CHerj1
Aixun T3B on Aliexpress: bit.ly/3cDkWIY
Aixun T3A on Aliexpress: bit.ly/30LDcgG
AiXun T3A on Aliexpress: bit.ly/2VdRVhU
Yihua 853A Preheater: www.banggood.com/custlink/DGD...
Youyue 946C Hotplate at Banggood: bit.ly/2MuLwqM
Youyue 946C on AliExpress: bit.ly/374v3Tp
Solder Paste: bit.ly/2E1P5jR
Solder Flux: bit.ly/2MQn7we
Quick 857DW+ Hot Air Station: bit.ly/2OWDvf4
T12 Soldering Station: www.banggood.com/custlink/3KK...
AC Version: www.banggood.com/custlink/3GK...
T12 Iron Tips: www.banggood.com/custlink/3DG...
Daniu Tip Cleaner: bit.ly/31VHwWM
KSGER T12-BCM3 Tip: bit.ly/2xgYXCQ
-------------------------------------------------------------------------------------------------
Microscope Equipment:
Eakins Barlow Lenses: bit.ly/3aBY5tR
Eakins Black Microscope: bit.ly/2VEcYaW
Eakins Camera Microscope: bit.ly/2x86hnT
Camera Adaptor for Microscopes: bit.ly/2KBZQMQ
Amscope Microscope Head (US): amzn.to/2YmZWNx
Amscope Trinocular Microscope (US): amzn.to/2YdBE8k
Amscope Trinocular Microscope (UK): amzn.to/2V8w9Gh
Swift SS41-WF20 Binocular Microscope US: amzn.to/2VLWEpb
Swift SS41-WF20 Binocular Microscope UK: amzn.to/2GYGYFV
Visit my website: www.sdgelectronics.co.uk/
-------------------------------------------------------------------------------------------------
#ElectronicsCreators
$2 for 1-8 layer PCBs, $60 coupons for new customers: jlcpcb.com/?from=SDG Xinfrared T2S+ Thermal Camera: www.banggood.com/custlink/DvGtMvqLMi Pico Technology 5000 Series Scope: www.picotech.com/oscilloscope/5000/flexible-resolution-oscilloscope Support the Channel with Patreon: www.patreon.com/sdgelectronics
For LEDs without the heatsink pad, you can try the Vishay ThermaWick thermal jumpers. It looks like a SMT resistor that is thermally conductive but electrically isolated. You can place close to LED from anode or cathode to the copper base. It helps significantly on FR4 boards due to the fiberglass prepreg. With the couple mil insulator on copper core pcbs (that's ~1-2W/mk) it is less beneficial, but can still be useful in lowering junction temps.
I have been itching to find a project to use those thermal jumpers on, just because I think they’re cool for some reason! :p
Cool, expensive and kinda useless - for LEDs just use the appropriate substrate or reduce power density - very rarely you need that lowest thermal resistance.
Glad to see someone working with the Rubix. I bought several a few months ago, bonded to some sort of ceramic PCB (I think) ... never got around to mounting them on a heatsink for evaluation.
This reminds me of the Pentair Microbrite swimming pool lights I have waiting to be repaired. They use the Luxeon Lumileds, in 3 groups of 4 on a 30mm circular aluminium PCB. Each group is powered by a ZXLD1362 driver + supporting components and the whole thing attaches to the power and control part of the light through the same pin header you have used. I've had a couple of these fail - drivers and diodes burning out - but one needs a new LED too. I have no idea how I'm going to get tiny blobs of paste on the pads for the LEDs, bu I have been successful with the other components, using a Miniware hot plate to do the repairs. Since the lights are $400 each, it's worth giving it a go!
Interesting the green has higher thermal resistance too. You are right you basically need active cooling if not water cooling to run these things near their max spec for sustained time periods. Its not as bad but similar to how FETs specify current at 25C die temp.
The green is rather warm 🔥. Interesting to know we can easily get different thinness for the stencil's 👍👍
It's amazing how cheap aluminium PCBs are these days. I was expecting you to mount the connector upside down so that the body was through the PCB leaving you with a nice flat top in case you decide to mount a diffuser or something like that.
I had considered it, but the legs have a slight taper on them and I was wasn't sure if it would short on the copper core.
Nice job.
Interesting. One of the yellow leds I purchase was backwards, surprise. Guess the K is usually😊 the heat sink and the other track can be smaller. Maybe be possible to alter stencil layer and make smaller pad or several small dots.
I tried reflowing some of these rubix a few months back, all of mine sat at varying degrees of wonky too. I pushed the cyan up to 2.5A before it went fap, on an al backed board. I've got 180 on starboards on order. One of the whites goes up to 3A if i recall, but the cri was something like 65
For the copper PCBs you show at 5:50, how did you specify the cutout in the dielectric layer in design files? How do they produce it? Do they mill the cutout before laminating to the copper?
JLCPCB will confirm the appropriate heatsink pads, or you can add another mech layer and write "HS" on the pads
Do you recommend a specific reflow oven for this type of stuff? I will need to be doing a lot of high power led stuff for wearable devices and heat sinking is number one on the issues list.
Did you used high thermal conductivity solder paste (eg bismut based) Steve ? Or is it just regular SnPb or even Lead free ? thx
It's just 63/37 paste
An almost ideal solution for a changeable spectrum is to use phosphor-based colour RGB LEDs, so that you get 'overlapping"RGB spectra, avoiding problems of "peaky" spectrums in mixed white...if you know what I mean?
If you want tunable white - just use multiple white LEDs of your choosing - anything RGBW without optical feedback will just lead to color shift and is hard to implement in general.
Well...if all you want is "tunable white", maybe, but you lose spectral flexibility, and end up having more troubles, with more LEDs to be "tuned"...the RGB driver developed at National Semiconductor, for me, has a wide range of tuning possibilities, including per-LED temperature compensation, and 12-bit calibration per LED, which is then multiplied by the PWM setting (it has 8 bit current setting, and 8 bit PWM which also has linear I bit to logarithmic 12- bit PWM values LP5523 I2C control, 9 Channels, on board charge pump, ADC to do self-Test and ambient temperature or brightness, 3 engines that can run programmed sequences,/cycles, and, and...
@@lohikarhu734 you want a synchrotron, not a tunable color engine at this point, besides the hard part is not the source or drivers, its the feedback.
I tried these LEDs and their thermal resistance sucks. You should have a look at the CREE Element G LEDs. Same thing but they feature an isolated thermal pad.
Thanks I'll look at those. Pretty much all of the Lumileds LEDs are either too small, or the larger ones aren't rated for high current.
This is great! I'm working on a project using L1I0-0850060000000 (Lumiled IR) and trying to design a driver for them. I've learned a lot from your channel. Still no luck on the driver but I'll keep trying.
What kind of specifications are you looking at for the driver?
@@sdgelectronics we're doing high speed video capture. Trying to use the lm3409HV to drive 10 of them at 2A (10us pulse). So about 64w.
Resistor is a driver, don't be shy using simple solutions - for low duty cycles voltage source + FET + resistor is quite thing.
I may be not thinking straight, but can't you just reduce the voltage to the green LED?
For ease of driving, he’s using a single current-limited supply for the whole string. That means each section (one or two LEDs in parallel) shares the same current. In this configuration, there’s no way to make the green have a different current to the rest. BTW, LEDs are current driven devices, not voltage driven, and their operating voltage is only partly dependent on that current, mostly being defined by the diode forward voltage which differs depending on the emission colour.
I could limit the PWM duty cycle into the green LED which overall reduces the power dissipated by the green LED. But I still need to address the stencil issue as it should be possible to keep the temperature lower.
@@simontillson482 Thanks for your explanation, just learning, but picking up more every day.
@@sdgelectronics I see, thanks for your reply, Steve.
SO tiny
Thats what SHE said