Toggle switches to Analog board?

[Al Skierkiewicz]

It should look like this...

[sircedric4]

Originally Posted by Al Skierkiewicz

It should look like this...

If I am looking at this correctly, does this mean I can't use a SPST switch to do what I want to do? I need a SPDT?

[Al Skierkiewicz]

If you are going to the analog board, it does not have pull up resistors so you have to use the SPDT to switch between +5 and zero volts. If you use a digital input, the pullup would allow you to simply open or close a connection to the common to give two states.

[Alan Anderson]

Originally Posted by Al Skierkiewicz

If you are going to the analog board, it does not have pull up resistors so you have to use the SPDT to switch between +5 and zero volts. If you use a digital input, the pullup would allow you to simply open or close a connection to the common to give two states.

Sorry, Al, but I have to disagree. If you put a 10k pullup resistor between the analog signal and +5 power pins, a simple SPST switch between signal and ground will give the desired results. The schematic Darren posted is exactly how I would do it.

[JamesBrown]

Originally Posted by Al Skierkiewicz

If you are going to the analog board, it does not have pull up resistors so you have to use the SPDT to switch between +5 and zero volts. If you use a digital input, the pullup would allow you to simply open or close a connection to the common to give two states.

Attached is a picture for SPST (Al, please correct me if I am wrong but this should give him two states (Closed =0v Open=~5v) using a Single Pole Single throw switch on the Analog inputs which seems to be what he wants to do.

[Al Skierkiewicz]

Yes, You guys are right. I have to stop answering these questions while I am working on interfaces here. Alan, can you add something about programming so that switch bounce doesn't confuse the readings?

[Alan Anderson]

Switch bounce would affect switches connected to a digital input at least as much as it would analog inputs. Most of the time, the programming is going to be such that it simply doesn't care whether the switch gets detected several times in quick succession.

For reading a switch on an analog input, the WPILib "analog trigger" function is what you want. That rejects glitch-type noise as part of its implementation.

[JamesBrown]

Edit: Looks like I need to type faster if I am going to post in the same threads as Al and Alan. But what I posted may still be usefull so I will leave it up.

Originally Posted by Al Skierkiewicz

Yes, You guys are right. I have to stop answering these questions while I am working on interfaces here. Alan, can you add something about programming so that switch bounce doesn't confuse the readings?

I am not Alan but I should be able to help here any way. To debounce a switch in software what you want to do is ensure that the press that was detected was a true press, not just noise or bounce. The easiest way to do this is once a press is detected start a counter, if the counter reaaches a certain value (10ms is probably acceptable for FRC Applications) then acknowledge that the switch is pressed. If the switch is released before counter times out the reset the counter and ignore the bounce.

Psuedo Code
Create Counter- Initialize to Zero

Assuming Switch sampling occusr every ~1ms

If (Software_representation of switch = not-pressed)

If switch is not pressed
Reset Counter
Set Software representation of switch to not-pressed
else (switch is pressed)
increment counter
end if

if counter > 10
Set software representation of swtch to pressed
end if

Then you can reverse the switched/not-switched cases for then the butto is pressed if you need to detect the release as well.

[sircedric4]

Cool discussion, I am happy I can use my SPST switches. I already had a bunch purchased and laying around. I also appreciate the pseudocode. Thanks again.

[Al Skierkiewicz]

Darren,
Sorry I led you down the rosy path earlier.