Hi all. I purchased a 74HCT245 chip on the advice of +Robert Atkins to

Hi all. I purchased a 74HCT245 chip on the advice of +Robert Atkins to convert my Due’s 3.3v output to the voltage required for some WS2811 12v pixels.

I have no idea how to wire it up though. The datasheet isn’t very helpful and I can’t find anything on the electric googles.

Can anyone help? Thanks.

Hi Mark, just look for “74hct245 datasheet”. You will find a number of places from where to download a pdf file

DIR and Vcc on +5V, Gnd and OE to Gnd, 3.3V input to A7, 5V output at B7.

It is an active low OE… so OE to gnd !!!

Yes, I corrected it.

http://www.pjrc.com/teensy/td_libs_OctoWS2811.html

Scroll down to “Signal Quality”

Very good Paul !
A picture is worth a thousand words !!!

Ahh, that’s brilliant! Thanks! The thing that was confusing me was the lack of a 3.3v reference that other chips have. I’ll try this.

Internally, the “T” chips have different ratios of mosfet transistor channel widths, which causes the switching threshold to be different.

Normal CMOS chips are designed with the N and P channel transistors to have approximately equal transconductance, or change in the “on” resistance of the transistor with the change in voltage at the gate. As the voltage change from low to high, the N-channel transistor turns off at about the same rate as the P-channel transistor turns on, which results in a logic switching threshold roughly half way between the power supply and ground. This results in the widest noise margin, and the best performance.

The HCT and other “T” chips have a wider (more conductive with voltage increase from GND) N-channel transistor, and a less conductive (with voltage between Vdd) P-channel one. When designing the chip, it’s easy to just make the width of the transistor different. Doubling the width is just like placing 2 transistors in parallel, except in board-level design you aren’t guaranteed the transistors will closely match in performance. Inside the chip, it’s all the same silicon, so you can play with the transconductance of the transistors very easily by changing the amount of area allocated.

That’s how the “T” versions have a lower switching threshold, and slightly slower propagation delay than the normal version. No reference voltage is needed. It’s just a simple change to the design of the transistors inside the chip.

Blimey! Electrons an’ that… Thanks for the information!