Chip randomly working incorrectly.

[cdavis]

[Edit] Reposted by Don due to hosting problems.

I recently bought a zx-24, and began to test it at home using my basicx dev board. I assumed the devboard would work because the zx and bx chips are pin compatible, and it did work. I wrote random programs here and there to test various functions, but something I noticed was quite odd: Sometimes I could run a program and it would behave as expected over and over again, but occasionally, the same code I just ran reliably over and over, would do something different and unexpected. This doesn't make sense to me.

Here is the code:

[dkinzer]

I assert that the program, as written, will produce varying results and is thus working correctly. This is just as true when run on a BX-24 as when run on a ZX-24 but the results may well be different between the two. The results may even be different between two ZX-24 chips or at different times of the day. The results may be different as well depending on what other equipment nearby happens to be running.

The reason that this is so is that the GetPin() function does not return the output state of a pin. Rather, it reconfigures the pin to be an input and then reads the input state. This behavior, described in the documentation, is identical to that of the BX-24. The difference in the actual result is almost certainly due to the effect of reading a floating input - the result can be different from one millisecond to the next because the high impedance input acts like an antenna. It can easily be affected by the local electromagnetic fields.

Also, different chips may exhibit slightly different behavior even if they're sitting side-by-side in the same environment. That's because the impedance and other electrical characteristics of an input pin vary from chip to chip. Two chips that were adjacent on the wafer when manufactured probably will exhibit similar characteristics but a chip from one side of a wafer and one from the opposite side might be slightly different.

The test program can be "fixed" in one of two ways. If you really wanted to read the output state, make the pin an output and add a call to a function like the one below that returns the current output state of a pin. If you intended to read the input state, make sure the pin is an input and connect the pin in question to external devices that provides a signal for the ZX to read. The simplest external device is a pullup/pulldown resistor. Adding a 10K resistor from the pin to either +5 or ground will ensure that consistent values are read. Note that, by default, all ZX input pins are high-impedance inputs (i.e. internal pullup not enabled).

There are two things to note about the code below. Firstly, it is specific to the ZX-24. To use it with the ZX-40 or ZX-44 it would have to be modified to correspond to the pin arrangements of those devices. Secondly, if PinState() is called giving a pin that is not then an output, the result is meaningless. (Actually, the result indicates whether the input pin has its internal pullup enabled or not.)

[cdavis]

Don,

After considering what you said, and reading a bit more no the forum, it made sense why my chip was acting that way. The pin I was attempting to read the state of was not configured as an input. Since I applied that idea to my programs everything works perfectly. Thanks for the help.


Casen