Capacitator on reset

Go To Last Post
26 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am desiging a small experiment board for hobby use. I am taking my time and searching around. Mostly, I am striving to understand the what and why. Hence an earlier question by me on leds and resistors. ( https://www.avrfreaks.net/index.p... )

Now, I've read the Atmel note on reset circuitry, and they suggest a pull-up resistor on the reset and a capacitator. After google'ing I understand the resistor (explained here: http://www.seattlerobotics.org/e... )

But I do not understand the need for the capacitator. Why would I want a cap on the reset? Isn't a cap meant to combat spikes in the system in case there is a sudden draw on power? I can't imagine that to be an issue on a reset.

Attachment(s): 

Code, justify, code - Pitr Dubovich

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You do not need the capacitor. In fact it is a positive disadvantage for debugWire AVRs.
You do not need the resistor. All AVRs have an internal weak pull-up.

OTOH, if you stick an aerial on the RESET pin and move to Crystal Palace or some other high power transmitter, you could pick up spurious noise. This may or may not reset your AVR.

Other controllers e.g. traditional 8051 actually need some form of start-up 'delay'. Hence they might have a capacitor (or an external RESET controller chip).

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The RC circuit avoid unwanted reset in case, if you want to connect a programmer cable on it.
Since this cable works as an antenna.
If not, then connect RESET directly to VCC.

Peter

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

You do not need the resistor. All AVRs have an internal weak pull-up.

Having taken part on a support case regarding this, I'd say that an external resistor is still a damned good idea.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

+1, because an internal pullup resistor is 30..200 k which makes the RESET pin (even itself) a good microwave antenna. A 100..1000 pF capacitor is also not that bad idea.

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I also prefer to use RC-circuit as Low Pass Filter. I believe it's better to spend a little much money than to have a strange surprise.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Having the C gives delays coming out of reset sometime after power is applied. This gives everything internal a chance to initialize.
Tau=R*C which is the time for the voltage on the reset line to rise to 63% of its final value. Fortuitously ViH for the reset pin is approx 0.6-0.7Vcc, so you can calculate the time delay before resetting very easily from tres=RC
I reckon it is a damned good idea to have the C there in all MCU's that go into service, whether you decide to use an external pull-up or not.
If you do need to use dWire at anytime, just temporarily remove the cap while debugging.

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

LDEVRIES wrote:
Having the C gives delays coming out of reset sometime after power is applied...

Oh, thank you for such detailed clarification. What about me I didn't pay much attention to reset circuit. Now I know something :lol:

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Attention Englishmen: Is Capacitator one of those Limey spellings like Aluminium?

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
If not, then connect RESET directly to VCC.

If you do that you will never be able to program the chip, unless you have a chip that has a JTAG port.
I would suggest a 10K pull-up as per your schematic, but make the cap 1 or 10nF range. I have had a few times that making it 100nF caused problems specially while debugging using debug wire, so I make the capacitor 10nF now a days and that seems to work.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For these it is best to read AVR documentation, like AVR Hardware Design Considerations appnote AVR042. It seems it is updated last year so good to read the updates.

It seems for DebugWire to work, the pull-up must be 10k or larger, and capacitor 1nF or smaller (those conditions are tested for several Atmel programmers).

Those values should be good enough.

I think I have succesfully used 100nF and 4k7 with my AVRISP, but I have not used DebugWire ever. Some programmers have weak drivers on reset pin or expect fast response on reset, so they cannot discharge 100nF quickly enough so AVR does not reset properly, or do not wait long enough for it to charge, so they talk to the chip too soon while it is still in reset.

The appnote also explains why a diode should be put on reset pin to VCC (no internal protection because of high voltage programming, and of course it is not available if there is a diode), and why there should be a 330R resistor in series to outside world or at least to a reset button (buttons bounce on/off several times rapidly, short circuit current from capacitor is large, and because of (stray) inductance, large current changes in short time period causes large voltage changes that may go outside safe operating values of reset pin).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

bobgardner wrote:
Attention Englishmen: Is Capacitator one of those Limey spellings like Aluminium?

English spelling is Capacitor

How do Americans spell Aluminium ?

I always understood that you spelled it Aluminium but simply have difficulties with your pronunciation.

A bit like new.

Your friend George W. was the expert in inserting extra syllables into regular words.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

I always understood that you spelled it Aluminium but simply have difficulties with your pronunciation.

Nope they really spell it Aluminum

Good to see that while Wikipedia list both spellings the article is titled the British way ;-)

http://en.wikipedia.org/wiki/Alu...

There is an entire section about:

http://en.wikipedia.org/wiki/Alu...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How about all the other elements like Caesium, Thorium, magnesium, ... ?

I just look at the "American Way" as a sort of special allowance for educationally challenged people.

Like their pint, gallon, hundredweight, ton, ...

Mind you, some of their spellings are more intuitive than the British versions.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Let's not talk about the americanisation of a lot of latin ( or is it greek) words used in medicine etc. Pedophile vs paedophile. Pedophile suggests someone interested in feet to me.Then we have words like fetal and apnea. Seems they don't like oe and ae sounds.
Call me a new traditionalist.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Capacitator make me hungry. I start thinking about a Big Potato.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Capacitators are okay. Incapacitators will incapacitate you.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
How about all the other elements like Caesium, Thorium, magnesium, ... ?

David.

Like Platinum, Molybdenum, Tantalum.

Greg

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think the reset pin on the Xmega may be different. It is used for PDI and too much pullup or filtering will incapacitate it. I don't use anything on the Xmega reset pin and neither does Atmel on the Xplain board.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Like Platinum, Molybdenum, Tantalum.

The very point made in that Wikipedia article.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

STK and ISP dont like pullups less than 10K... that's too much for the programmer to pull down and hold in reset.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

bobgardner wrote:
STK and ISP dont like pullups less than 10K... that's too much for the programmer to pull down and hold in reset.

You mean we must always use pullup more than 10K whenever we're going to use ISP? Then, capacitor is calculated as C = tau / R... OK?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yep. Sounds right to me.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

And if you don't use an external pull-up resistor, there is still an internal one of about 30K so
C=tau/30 remember that tau is in seconds, R is in M ohm & C is in uFarad.

Edit. changed to "internal"

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

Last Edited: Fri. Feb 10, 2012 - 10:54 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

LDEVRIES wrote:
And if you don't use an external pull-up resistor, there is still an external one of about 30K

internal? :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

:oops: If I was a y gener, I would have to say "my bad" :wink:

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?