Arduino TFT (A000096) very dim on Due

Afternoon all,

Hope you can shed some ligth on the but following the instructions on the following link gives a very dim TFT display.

http://arduino.cc/en/Guide/TFTtoBoards

This shows a jumper from the +5V connection on the TFT to the BL connection, both being supplied from the 3V3 connection on the arduino. In this orientation it was very dim.

I'd noticed a few statements about the backlight could be dimmable by PWM, but couldn't find any reference to it in a breif flick thought he header files. I removed the jumper between BL and +5 and the screen gets notably brighter! How should this be wired?

Cheers,

Wes.

Ok, more investigation in advance of help!

Page 7 of the spec sheet specs the back light as an LED with a forward voltage of 3.0 - 3.4V and current of 30mA. http://arduino.cc/en/uploads/Main/HTF0177SN-01-SPEC.pdf

The following schematic specifies an n-channel mosfet on the BL line of the board. It appears the intention is that the BL line doens't pull much current and the current for the LED is drawn from the +5 line which is modulated by a signal from BL which drives the gate of the mosfet and so modulating the supply to the LED. http://arduino.cc/en/uploads/Main/GLCD_sch.pdf

There a P channel mosfet on the board - PMV48XP (marked KNH) http://www.nxp.com/documents/data_sheet/PMV48XP.pdf The N-Channel on the drawing seems to be a mistake or the P-Channel on the board?

Looking at the component when I've removed the BL signal is the power for the backlight is being taken by the mosfet's body diode?

Probably rather obvious I'm a novice with electronics so using this as a learning experience as much as anything, but do need to understand this better.

Online tutorials seem to have P-Channel mosfets the other way around?

So after reading this:

http://hackaday.com/2011/12/06/reverse-voltage-protection-with-a-p-fet/

The set up looks like reverse voltage protection with the added signal from the arduino going to the gate. So signal removed and 3.3V on the +5 and 0V on BL gives a voltage drop on the diode of around 0.82V giving a Gate-Source voltage of -2.48V as the gate is pulled down to ground. This turns the mosfet on, which chucks the full 3.3V at the LED backlight.

Add 3.3V to the gate from the arduino and the gate source voltage raises to 0.82V which turns the mosfet off dropping the voltage down to 2.48V on the backlight leaving it very dim.

This seems counter intuative, particually with regards to the wiring recomendations.

Is this opperation intentional?

Would it not make more sense to ditch the mosfet completely and seperate the supply to the back light so we can feed it with the 5V out form the Due board without effecting the 3.3V requiement for the Due's logic?

No one able to explain/confirm what's going on or how the back light is supposed to work?

As you pointed out, the backlight is rated for about 3.3 V, so for use with a Due I just shorted out the 220 Ohm current-limiting resistor. That way I'm feeding 3.3 V straight to the LED (transistor shouldn't drop much voltage). The brightness looks pretty reasonable now, and all the magic blue smoke is still on the inside where it belongs.