74HC595 query

[cowana]

I'm trying to construct a circuit that controls 7 segement LEDs using a 74HC595.

I constructed the circuit, but none of the LEDs turned on.

I then built the minimum circuit using a new 74HC595, leaving the outputs floating.

However, the outputs stay at 0v - whatever the inputs do.

As shown in the below image, Vcc = 5v, pin 10 (reset) = 5v.

Output enable is usually 0v, spiking high whenever the data changes.

The code being run is known to work, so it can only be a hardware error. All the signals shown in the attached image were measured directly on the chip's legs - so what could be wrong? I can only assume the 74HC595 is faulty - but I have tried replacing it with a new one twice, with no improvement. There is a tiny chance it was a bad batch of chips I ordered, but it seems very unlikely.

Andrew

 

[3v0]

There is little magic about this chip. If the code is right and you have it wired right it should work.

Step through the program and find out what is going wrong. Trust nothing.

 

[crutschow]

Post the circuit diagram.

Are all the unused inputs tied to ground or V+? Any floating inputs can cause problems with CMOS circuits.

 

[cowana]

The program (not written by me) works on the writers hardware. From my logic analyser, I can see the data that should be written into the 595 - eg in the third attachment of my post, it should:
Turn off the outputs
Shift in the data: 0 0 0 1 1 1 1 1
Pulse the latch low, transfering the above data to the latches
Turn the outputs back on

I can't see any reason why it isn't doing that - the correct voltages are all there at the correct time!

 

[cowana]

The project is Addressable serial controlled PWM RGB LED Driver

And the circuit diagram (for both the controller and display) are: Addressable serial controlled PWM RGB LED Driver

crutschow: I'm using all of the inputs on the '595 - there aren't any floating ones to tie down!

 

[cowana]

Another oddity.

When the system is turned on, OE is high, so the outputs are in high impedance state. This shows up as high in my logic analyser - it has a very weak pullup.

When OE first goes low, the outputs all change to 0v. They then stay like this, even when OE goes high again. This is not what the datasheet says to expect - it should be that when OE goes high, all outputs go high impedance. Mine don't seem to...

Andrew

 

[cowana]

Another oddity!

When the supply voltage is turned right down (from 5v to 3.9v, the third digit comes on (flickering and displaying nonsense). As the supply is lowered, at 3.8v, the sencond digit comes on (also flickering and displaying nonsense).

I think this must be due to the 74HC595 shutting down (or at least operating unreliably) at a low voltage - although I would have expected it to go on working to 2v, rather than the 4v is starts outputting data at...

Andrew

 

[RMMM]

Are you using bypass/decoupling caps?

again, a schematic would help

 

[cowana]

I've got a 100nF cap across the regulator, and another 100nF cap across the 595.

The schematic is linked to in post 5.

Andrew