To verify that your programming ("burning" the flash) works, try runninmg "blinky" does it work?

I'm quite sure I'm getting my programs loaded. With my setup I can load code into flash by flipping a switch and running the associated "loader" program. Then i just flip the switch back to run.
When I make changes that change the test LED's sequence i can verify that they change.

1) 4k7 is way too high for a "flying lead" IIC bus, try 1k if your uC can sink 5ma (see darasheet).

Is it possible that this will work without any pullups? because I've tried all possible ways and nothing has changed.
this is the only device I'm testing with at this point so basically the bus goes from my '51 to the appropriate pins on the max6956.

When I was reading through the "IIC Bible" it says the bus could be up to 10 meters in length, surely with my jumpers being just 16 inches long the signal is getting there.

2) recode your timings to make the speed, say, 1/10 of what it is right now and see what happens.

I'm not sure I can get them any lower. I've got them set up as routines I can "lcall" and they just use a "djnz". The 1 I call "bitdly" does a djnz from 10. I've tried with as low as 1 and high as 255, makes no difference.
Is it possible for this '51 to even run fast enough to do this communication? When looking through some of the specs I noticed some time references to 350 ns.


I put a transistor and an LED on both the SDA and SCL so that I could see if anything was actually happening on the pins. a) that will kill the bus and

Connecting the SDA or SCL to the base of a transistor will "kill the bus"? I've got the collector connected to the V+ of my breadboard (5 volts) and the emitter connected to the anode of my led and then the cathode grounded (I hope I've stated anode and cathode correctly). That's how I monitor the pins. I've also probed the pin with my logic probe and see it pulsing when it should be.

b) your eyeballs can not follow a 100kHz blink.
When I increased my delay times I could see the blinks, just to verify that they were actually changing, then dropped them back down.



To verify that your programming ("burnin" the memory) works, try runninmg "blinky" does it work?

I didn't try "blinky" because before I moved to this point I had originally hooked a 5x7 LED to P1 and P3.
I then designed my font using this setup.
I even had it display the proper character based on whichever key was pressed.
My '51 is working if that's what you wanted to verify.