20X2 C.6 input only debacle

[RBruceM]

I wasted a lot of time on schematic and PCB artwork, and troubleshooting on the 20X2 C.6 input only issue. An output on that pin simply doesn't work... and I know from the Forum posts now that this is a Microchip design "feature".

I can't seem to find it anywhere in the manuals or X2 datasheet. Rev-Ed does a great job of documentation, but this one slipped by me. I think it ought to be noted more prominently, though perhaps I just missed it???

Best Wishes,
Bruce M
Snowflake AZ

 

[Lorian]

Manual 1 page 27.

Note pin C.6 is input only on the 20X2 part. This is due to the internal silicon
design of the chip and cannot be altered.

 

[RBruceM]

Thanks, I'm going to put that up over my desk as a reminder. I missed it!

 

[westaust55]

The fact that 20X pin C.6 is input only is also mentioned in many places in manual 2 including:

Page 100 - Input command - C.6 = fixed input

Page 112 – Let DirsA/DirsB/Dirsc/DirsD commands
”Note that some pins are fixed as inputs/outputs and so using this command will have no affect on these pins, which are . . . 20X2 A.0 (output) C.6 (input)”

Page 122 Output command
All pins are configured as inputs on first power-up (unless the pin is a fixed
output). Fixed pins are not affected by this command. These pins are:. . . 20X2 A.0 = fixed output C.6 = fixed input”

Page 164 Reverse command –
“All pins are configured as inputs on first power-up (unless the pin is a fixed
output). Fixed pins are not affected by this command. These pins are: . . . 20X2 A.0 = fixed output C.6 = fixed input"

 

[hippy]

It's possibly a bit late now but this is one of the reasons it's recommended to breadboard the circuit and test preliminary software before committing to a PCB design and its production.

It's common to have theory not met in practice, something unexpected or overlooked arising when you come to actually do it. That can be pins which cannot do the job expected, too much current draw on particular ports or even type of input ( TTL or ST ).

Even with due diligence and a lot of research first it's still possible to run into unexpected behaviour. Even if the circuit and software is perfect it might not work with particular external hardware and require change. It's common in commercial settings therefore to breadboard the proof of concept, build a prototype PCB, test that and fix issues, then commit to a final PCB design.

Lack of getting it right first time is just a fact of life, something which has to be accepted. I doubt there's anyone who has created anything who hasn't found some issue with their first attempt and then plans the Mark 2 version to overcome those issues, be it go kart, cake or anything else.

 

[Goeytex]

Thanks, I'm going to put that up over my desk as a reminder. I missed it!

Goood Idea.

When working with a particular device whether it is a micro or anything else
I create and keep my own application notes for that device that covers all the "gotchas " that are discovered either in the manuals or by testing.

Rev-Ed covers every model of their chip in a single manual which sometimes makes it a challenge to find what you need, so it can be easy to miss something if you are not very careful.

On one project I was on, the CTO, against the advise of several engineers and techs, ordered 1,500 boards based upon a design that was not properly peer reviewed, prototyped, and tested/debugged . All ended up in the dumpster because of something similar to what you experienced.

Goey