TLC5926-Q1: Thermal issue at 100% duty

Hi Youhei,

Our expert is OoO and he will reply to you once he is back to office, thanks for your patience.

BR, Jared

Hi Jared,

Is there any update on this topic?

Regards,

Youhei

Hi Youhei,

First, when OE=H, is there any output current from OUT pin?

If no, that's still reasonable if there is power consumption because the supply current is still there even though the OUT is off.

BR, Gracie

Hi Gracie, 

Can you check customer calculation?

VDD=3.3V, IDD≒22mA(∴REXT=270Ω、OUT0-OUT15=off), RθJC(top)=23.4℃/W

So, Δt(top) = 3.3V * 22mA * 23.4℃/W ≒ 1.7℃

That is far from their measurement result. IC surface rise 7℃ on their test.

Youhei Miyaoka said:

IC surface temperature rises by about 7°C.

Also, their PCB pattern is here. Seeing from this, I can't understand why temperature rise so much.

Regards,

Youhei

Hi Youhei,

Do you mean there is no output current when OE=H, but the temp rise is far from theory? Do you see all chips function this way?

BR, Gracie

Hi Gracie,

Sorry for long no update.

When OEn=H, the LED is off and no current is flowing to the LED.
The same temperature rise was observed in both of 2x LED drivers.

OEn=H is input to TLC5926 and signals are input to SDI, CLK, LE.
(When input OEn=L, OUT0 to 15 LEDs will light up)

Regards,

Youhei

Hi Youhei,

TLC5926 is a quite old device and I currently don't have EVM for it. Is it possible that customer uses TLC6c5716-Q1 instead of TLC5926 if it's still in design phase?

BR, Gracie

Hi Gracie,

It is difficult because they need 70mA per channel.

Regards,

Youhei

Hi Youhei,

We even don't have EVM for this one. 

So do you know where this thermal rise comes from? Do you know what's the VCC and ICC when channel is off?

BR, Gracie