blown up 7812 and 7805

[m3r]

hi
i use this pcb for control RGB led.but my problem is blown up 7812 when sudden increase bright from min to max.
i use industrial switching power (24V 15A) and current of led is about 5a.
what is problem?

sorry for my bad ٍenglish

 

[ftsolutions]

What are you using for heatsinks on the 7812 and 7805?
You are feeding the 7805 from the 7812, so all current passes through both devices.
With 24V input and ~5A of output, you are dropping 60W in that 7812, and a considerable amount in the 7805.
That is alot of power to dissipate and requires substantial heatsink/cooling. Your thermal pad interface needs to be low thermal impedance also.

 

[KlausST]

Hi,

a lot of mistakes.

No GND plane.
No capacitors at XTAL
No fast ceramics capacitor at EACH IC power supply pin (solely C3 is placed well)
wrong signal flow from input to 7812 then capacitor: It should be: input-capacitor-7812 (to prevent voltage spikes to reach 7812 input)
No capacitor at node 7812_output to 7805_input (maybe this is the most critical problem)
(i doubt 7812 is needed at all)

I assume there are even more problems.

Klaus

- - - Updated - - -

Hi,

With 24V input and ~5A of output, you are dropping 60W in that 7812, and a considerable amount in the 7805.

5A is LED current. But LED current doesn't flow through 7812 and 7805... only uC and wireless_module current.

Klaus

- - - Updated - - -

My recommendation:


* first add the missing componentis in your schematic.
when schematic is finished...
* wire input + and - directely to a bulk capacitor. Short thick traces.
use those two capacitor connections as star point. This means use seperate traces for power path (thick, short) and seperate traces for control path.

even if there is the bulk capacitor, expect spikes.
therefore a series R in the + line to the control path is a good idea.


Not urgent, but personally I´d replace the 7812 with a couple of zeners.
I asumme your control path including wireless_module will mot consume more than 100mA.
If so, then use 1W zeners with less than 5V.

Then place a smaller bulk capacitor directely at the input of the first voltage regulator.

...then the next steps...

Klaus

 

[pjmelect]

Although I agree with most of the comments by KlausST I don't think that these are the cause of the problem of the 7812 blowing up. The 7812 is a tough device and can withstand a lot of abuse, but what it can't stand is input voltages higher than its maximum rating. The maximum input voltage of a 7812 is 35V and you may be getting voltage spikes when you switch the high current on and off. As suggested adding input and output capacitors to the 7812 may help. Look at the voltage input with a scope, consider the ground wiring to the controller, maybe a 30V zener wired to the input of the 7812.