10K . . . = arbitrary pull down value . . ??
The project that raised this question for me was a custom Proto-board for my picAxes (up to 14 pins)
So I plan on using this for a variety of applications, but I also wanted to get your view on the value and possible detractions of the practice for future circuit design.
My question is specific to a project that
DOES have a DL circuit.
Sorry for the late reply, I thought I was subscribed to this thread.
Your adding a resistor that is not necessary ?
"not necessary" perhaps. . .
For a large percentage of projects this would not be necessary (if 32k works for the other inputs)
(I suppose application makes all the difference, but if you like 32k for your most vulnerable input, wouldn't you also use this value[or similar] for all inputs?)
I guess part of my question is: "
how arbitrary is 10k valued pull downs for "typical" noise"
If it's common to have problems above this value then it would seem shrewd to
automatically use this for all inputs (including / especially the Ser-In)
(so to not use the additional 10k would be the exception, instead of the rule)
No, this may stop it working all togther.
What do you mean Technical? The additional 10k shown could possibly "stop it" . . . as an input pin?
It would work with a 'red' 100k....but...
99% of users add the download circuit, the 10k and 22k to every circuit. Then you can program the PICAXE in place.
I've seen users raise the pull down resistors for power efficiency at the cost of noise protection.
My thinking is:
If I was going to pick one pin not to "skimp" on pull down stability, I would pick this pin since it has potential to wreck the entire function of the chip
If you don't want to add the download circuit, but do want to add the red 10k, then why? Its only one more resistor to do it properly!
That's my take, of course I would use a 10k if there was no in circuit programming circuitry, since my current project is a proto-board and I think it would be practical to add it permanently, even with the DL circuit.
It's easy to damage pins removing from a breadboard to re-program, why not use one of the programming adaptor boards that just plugs into the breadboard, it's only about £2 and makes life so easy, I have a bit of stripboard on headers with the 2 resistors and programming socket on mine (cost about 30p) and use jumpers to connect to whatever picaxe is on the board at the time.
I work on the idea that projects are never right first time and even if it is you'll think of somethine else to add later anyway.
I've got a couple of these, I like them for the breadboard, but I'm aiming to clean it up with my own proto-board design.
I think the easiest solution is to simply make a connection on the 3.5mm connector from 'b' to the inner pin 'c'. This way the pull-down on pin 2 will be 22K instead of 32K. (of course without making connection between the outer 'c' and the inner 'c')
Or even better, make a connection between PICAXE pin2 and the inner pin 'c' of the 3.5mm connector. Then pin2 is grounded when there is no plug connected.
This would eliminate all possible noise, but also the ability to program . . .
Since my current project is a proto-board I may have to make this temp connection when not programming.
My latest project was in a noisy environment that kept resetting the 08M2 until I grounded pin2 using this connection to the plug.
Interesting, obviously in this environment the 32k pull down provided by your programming circuit was not sufficient. (assuming there was one)
a question for you then:
Were you using other inputs on the 08M2? If so what value pull down did you use? ( < 10K?)
I'm sure it will depend on the application, but almost certainly couldn't it be said that '
the Ser-In port should generally be the lowest value pull down'?
So . . . if all of your
other inputs have 10K resistors . . . why wouldn't you put one on the seemingly
most vulnerable?
Right?
Thanks for your input!