Go Down

Topic: Reference: ShiftOut() versus SPI.h (Read 3879 times) previous topic - next topic

wossname

The OEs on both chips are wired directly to GND as per the schematic.  With regard to fading the LEDs, I think that's venturing outside the remit of the article we're trying to clarify.  I think we should keep it simple for now.

Another problem I found with that original ShiftOut article is that it says use a 0.1"f capacitor between the latch pin (SS in this case) and GND.  However the schematic says a 1µf.  This is something we need to clarify also.  Personally I've tried both and neither type make any apparent useful difference.  The 1µf one actually prevents anything from being displayed on the output pins!


CrossRoads

PWM, your call.

With OE tied to ground, whatever is in the output register will show up at the output at power on.
Maybe an RC circuit and a transistor to take it high on power up, and go low as the cap charges up & turns the transisto on.

I can't fathom why there would be an capacitor on the SS pin that would just degrade the signal.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

Graynomad

Quote
original ShiftOut article is that it says use a 0.1"f capacitor between the latch pin (SS in this case) and GND

That's a known error and has been there forever. Why it isn't fixed is one of those great mysteries in life as it trips over evey newbie doing their first shift register project.

There shouldn't be a cap at all.

______
Rob
Rob Gray aka the GRAYnomad www.robgray.com

CrossRoads

How can we get some permissions set up so we can fix those? We can fix things in the playground, why not the Learning section as well?
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

wossname

#19
Feb 21, 2011, 08:46 pm Last Edit: Feb 21, 2011, 08:48 pm by wossname Reason: 1
Perhaps a wiki-style system would be appropriate?

To be blunt, might it be such a bad idea if the entire set of reference pages was torn up and redone from scratch by those people who have the proper time to spend on good quality documentation?  It seems that it was initially written by several over-excited hackers in a rush to build their next project and skimming over the details in order to get in more soldering time :)  Can't say I blame them, I'd rather be building circuits than writing about how to do it.


Graynomad

Yes I've found the documentation wanting on several occasions. I used to do the tech pubs for a large computer company but won't be putting my hand up for a rewrite, I'd rather be building new stuff as well :)

______
Rob
Rob Gray aka the GRAYnomad www.robgray.com

David Cuartielles

just FYI the documentation site is a wiki page, the exact same as the playground. It is just that the documentation site is curated. I think that the effort you guys have made in this forum topic is remarkable. If you can translate the results into a playground page, it would take no time to push it into the documentation site.

This is the flow we have chosen for the documentation to be generated:

* First people ask questions that get answered in the forum
* If the answers get really good, like the one shown here, someone decides to make a playground article out of it
* We monitor the playground and see what should become an example in the IDE (we also monitor the forum)

Somehow the playground articles are more finished than the ones in the forum.

Finally, if you find errors in the documentation site, just post them in this board of the forum, and I will redirect the task to someone in the team. I am getting pretty good in delegating :-)

Go Up