Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Novice needs help with power

Status
Not open for further replies.

NickK1066

New Member
I have a Light Engine that I wish to power. I have a PSU that should provide enough power - however just connecting directly blew the LE up after about 24 hours of operation at the lowest light level.
Understandably - this being the replacement I'm anxious not to repeat this again and want to check (a) if the power seems ok and/or if I should make a power regulator (with some help from the folks here).

I can solder but I have little understanding of power.

The Light Engine is a bit confusing however I read the power requirements as 12V <5A. There's a PSU onboard (which blew last time) that provides a dimmer system for the LED array (using PWM I assume).
Specifications here:
**broken link removed**

The datasheet that comes with this in the box indicates 12V for blue, whereas white etc require 9V. Is it worth attempting to run this a 9V just incase the datasheet is wrong?

My PSU is a 12V 5A supply. This is switched mode.
Specifications here:
RS | Power Supplies | Power Supplies, Inverters, DC/DC Converters & Generators | External Power Supplies | Switch Mode - RS

Should I just be able to connect the LE to the PSU (thinking last time was something defective) or would I be wise to regulate the to a lower voltage and place some resistors in the way along with a fuse to ensure the previous problem doesn't occur again.

The previous blown LE was a mess - the incoming resistor, a large percentage of the power regulators had burst or burnt. The PCB is slightly warped too! :eek:

Help! :D
 
Sounds like it may have overheated. Was the cooling fan running when it blew? Is it open so cooling air can get to it. It can't be mounted in a closed box.

According to the data sheet 12V should be okay for the blue unit. You might contact the manufacturer to be sure. Did you measure the output of the supply to make sure it's 12V?

Adding a fuse would be a good idea.
 
Sounds like it may have overheated. Was the cooling fan running when it blew? Is it open so cooling air can get to it. It can't be mounted in a closed box.

According to the data sheet 12V should be okay for the blue unit. You might contact the manufacturer to be sure. Did you measure the output of the supply to make sure it's 12V?

Adding a fuse would be a good idea.

I've just measured the voltage from the PSU: 12.3 so it's within the ±2.5%.

Before the LE was basically just sitting mounted open to air, the fan on the heatsink was working fine too but I did notice the resistor R10 was very hot but didn't think anything of it (especially as it's operating temp is +45degC).

I've attached a photo of the blown board for your 'enjoyment'.

It makes me think that something is attempting to draw too much power or the 12V is too high.

I'm considering running it with 9V instead. Just in case the datasheet is wrong. Is this possible/wise?
 

Attachments

  • IMG_0924.JPG
    IMG_0924.JPG
    88.3 KB · Views: 169
Last edited:
I'm considering running it with 9V instead. Just in case the datasheet is wrong. Is this possible/wise?
I would think that would be okay but, not knowing the circuit, it would be best if you could confirm that with the manufacturer first. Since they say to run on 12V it could act up at a lower voltage. Does the data sheet list a minimum operating voltage?
 
Apparently 12 volts is the minimum for blue and UV, etc. 9v for red.

Goodness, what a bunch of stuff. No fewer than six SOT-223's and a DPAK... five BSP373 N-channel MOSFETS with a .16Ω Rds on (assuming Vgs actually = 10V) all running in parallel... with the damaged 33 ohm resistor connected to... what, the source leads?

Hmm, what a very strange way of doing it. Seems any significant current would cause such an I*R drop on that 33Ω resistor that the gate-source voltage would suffer as a result. And the one visible trace (from R42) showing how Q2-6 may be driven is snaking its way suspiciously back to that Atmel microcontroller, instead of to a proper gate driver.

Nick, just to satisfy my curiosity, could you flip the board over and see if R10 is really connected to the third leg of those transistors? And could you read the number on R45?
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top